Renal receptors for atria1 and C-type natriuretic peptides in the rat J. BROWN Physiological
AND Z. ZUO Laboratory,
Cambridge
CB2 3EG, United Kingdom
or intravenously (2, 14, 16, 35, 39). Brown, J., and 2. Zuo. Renal receptors for atria1 and broventricularly C-type natriuretic peptides in the rat. Am. J. Physiol. 263 Consequently, these two groups of peptides may have (Renal Fluid ElectroZyte Physiol. 32): F89-F96, 1992.-Recep- complimentary physiological and pathological effects. tors for a-atria1 natriuretic peptide (cu-ANP) and C-type natriCNPs have so far been isolated from porcine brain as uretic peptide [CNP-( l-22)] were quantified in kidneys from a 22-amino acid form [CNP-(l-22)] (37) and as a adult Wistar rats by in vitro autoradiography. 1251-labeled 53-amino acid form [CNP-(l--53)], which includes the cu-ANP(100 PM) bound reversibly to glomeruli, outer medullary at its COOH-terminal (26). vasa recta, and inner medulla with an apparent dissociation sequence of CNP-(l-22) Molecular cloning techniques suggest that rat and porconstant (&) of 3-6 nM. The presence of 10 PM descine CNPs are identical (18). The intraand extraceren18,Ser1g,Gly20,Leu21,Gly22]ANP-(4-23) (C-ANP), a spe[Gl of CNPs are only now being defined cific ligand of the ANPR-C subtype of a-ANP receptor, inhib- bra1 distributions ited -50% of the glomerular binding of 1251-cz-ANP,and this (20, 38), and it is not clear to what extent CNPs might moiety of glomerular binding was also inhibited by CNP- compliment actions of the other natriuretic peptides. (l-22) with an apparent inhibitory constant (Ki) of 10.47 t However, CNP-( 1 - 22) may not share all of its receptors 7.59nM. C-ANP and CNP-( l-22) showedlittle affinity for the with the ANPs and BNPs (19). Molecular cloning has medullary binding sites of cr-ANP. 1251[TyrO] CNP- (I - 22) identified three receptors for natriuretic peptides. Two (110 PM) bound solely to glomeruli and was competitively dis- of these, ANPR-A and ANPR-B, are proteins of -120placedby increasingconcentrations of [TyrO]CNP-( 1- 22) with guanylate cyclase (2, 10, 11, an apparent Kd of 1.42 t 0.48 nM. Binding of increasing con- 130 kDa with constitutive dimer of centrations (25 pM to 1 nM) of 1251-[Tyro]CNP-(1-22) in the 31). The third, ANPR-C, is a disulfide-bridged guanylate presenceor absenceof 1 PM [TyrO]CNP-(1-22) also demon- 66 kDa units. This receptor has no intrinsic strated a high affinity (Kd of 0.41 t 0.07 nM) for the glomerular cyclase, and it probably binds and clears its ligands by (2,22,29). It, or closely similar proteins, binding of f251-[Tyro]CNP-(1-22). Bound 1251-[Tyro]CNP- internalization (l-22) could be displacedby excesscu-ANP and excessCNP- may also be facultatively linked to second messenger (1- 22), both with high affinities. The glomerular binding of systems (1, 15). ANPR-A has strict structural require1251-[Tyro]CNP-(1-22) was also prevented by 10 PM C-ANP. ments of its ligands so that it has high affinity for Guanosine3’,5’-cyclic monophosphateproducedby isolatedglo- cu-ANP but little or no affinity for synthetic analogues meruli was measuredby radioimmunoassay.cr-ANP increased of cu-ANP such as [TyrO]ANP-(5-25) and desproduction powerfully, but CNP-( 1-22) causedonly a slight n1S,Ser1g,Gly20,Leu 21 , Gl~~~lANP-(4--23) (C-ANP) stimulation with much lower affinity. Theseresults suggestthat W (2, 6, 7, 25, 29). ANPR-C is much less conservative and Wistar rat kidneys expressdetectablequantities of the ANPR-C affinities for all of these ANPs (2, 6, 7, and ANPR-A but not the ANPR-B subtypesof natriuretic pep- has significant 22, 25). In addition, both ANPR-A and ANPR-C bind tide receptor. guanylate cyclase-coupledreceptors; clearancereceptors; auto- avidly to a variety of BNPs such as porcine BNP(l-26) and rat BNP-(l-45) (3-5,23,34). ANPR-B has radiography C-typenatriureticpeptides (CNPs) share many properties of the atria1 and brain natriuretic peptides (ANPs and BNPs, respectively). The CNPs, ANPs, and BNPs arise respectively from posttranslational modifications of the products of three different genes (2, 18, 24, 26, 30), but their structures all share a characteristically similar 17-amino acid disulfide-bridged ring (2, 17, 18, 26, 35-37). As a result, the peptides also partly share the same receptors (3-5, 8, 10-13, 19, 23, 31, 34), and all are to some extent natriuretic, diuretic, and vasorelaxant (2,35, 37). ANPs were first isolated from the heart, from which they enter the circulation, principally as the 2%amino acid a!-ANP, to exert a variety of actions on the cardiovascular system and on fluid homeostasis (2). ANPs are also synthesized in the central nervous system, where they probably modulate the nervous control of the circulation and body fluids (14). In contrast, BNPs were first isolated from the brain, but it is now known that, like ANPs, they are also synthesized by the heart and released into the circulation (17, 30). ANPs and BNPs share receptors both intra- and extracerebrally (3-5, 13, 23, 34), and they have similar actions when injected either intracereTHERECENTLYDISCOVERED
been identified by molecular cloning, and because it has had no known high-affinity ligands [for example, having virtually no affinity for a-ANP (10,19,31)], nothing has been discovered of its in vivo expression. Recently, however, ANPR-B has been expressed in transfected cells and shown to have high affinity for CNP-( l-22) (19). CNP-( 1 - 22) has also been found to have significant affinity for ANPR-C but none for ANPR-A in such cells (19). Thus ANPR-B is a guanylate cyclase-coupled receptor that seems to be relatively specific for CNP(l-22), and its in vivo distribution may have important functional consequences. Here, we use quantitative autoradiography to examine the endogenous receptors of CNP-( l-22) and CV-ANP in rat kidney. The results demonstrate that CNP-( l-22) binds avidly to glomerular ANPR-C expressed in vivo. However, no other high-affinity renal receptors for CNP-( l-22) were found. This suggests that rat kidney expresses ANPR-A and ANPR-C in vivo, but does not express detectable quantities of ANPR-B. METHODS
Autoradiographic studies. Three groups, each of six adult male Wistar rats (260-330 g; Charles Rivers Breeding) were 0363-6127/92 $2.00Copyright 0 1992the American Physiological Society F89
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
F90
RENAL
ANP
AND
maintained with free access to water and food (Labsure PRD, William Lillico). The rats were killed by rapid exsanguination. Kidneys were quickly removed and snap-frozen in isopentane, cooled by dry ice. Ten-micrometer-thick cryostat sections were prepared for autoradiography as previously described in detail, except for the use of the additional peptidase inhibitor phosphoramidone (6, 7). Briefly, after preincubation at ZOOC for 10 min in 30 mM phosphatebuffer (pH 7.18) containing 120 mM sodium chloride, 40 pg/ml bacitracin, 2 hg/ml leupeptin, 100 pg/ml phenylmethylsulfonyl fluoride and 5 pg/ml phosphoramidone (Sigma Chemical), sections were incubated with either 3-[1251]iodo-28-tyrosyl rat atria1 natriuretic peptide 125I-aANP-( 1- 28), sp act 2,000 Ci/mmol; Amersham International] or 3-[ 1251]iodo-0-tyrosyl rat C-type natriuretic peptide 1251[TyrO] CNP- (I- 22)) sp act 1,398 Ci/mmol; Peninsula Laboratories) in fresh preincubation buffer plus 0.5% bovine serum albumin (fraction V, proteasefree; Sigma) at 20°C for 15 min. The competitive inhibition of 1251-~-ANPbinding by unlabeled peptides was examined in consecutive sections from the first group of six rats by coincubating with various concentrations (1 pM to 10 PM) of unlabeledrat a-ANP, rat CNP-( I - 22), or rat C-ANP [Peninsula Laboratories (Europe)]. Sections were also incubated with 1251-a-ANPplus 10PM CNP-(1-22) and I MM cu-ANP, or with 1251-a-ANPplus 10 PM CNP-(1-22) and 10 PM C-ANP. The competitive inhibition of 1251[TyrO] CNP(l-22) binding was similarly examined by coincubating sections from the secondgroup of six rats with 1 pM to IO PM of unlabeled rat [TyrO]CNP-(l-22), rat CNP-(l-22), rat ar-ANP, or C-ANP (Peninsula). Sections were also incubated with the radioligand plus 1 PM CNP-( l-22) and 10 PM a-ANP, or radioligand plus 1 PM CNP-( l-22) and 10 PM C-ANP. Finally, the saturation of the binding of increasing concentrations (25 pM to 1 nM) of 1251-[Tyro]CNP-(l-22) to consecutive sectionsin the absenceand presenceof 1 PM unlabeled [TyrO]CNP-( l-22) was examined in the third group of six rats. The specificity of the inhibition of the binding of both radioligandsby natriuretic peptides was assessed in the presence of the unrelated peptides angiotensin II, vasopressin (Sigma), or gastrin (Peninsula) (all 10 PM). After incubation, sectionswere rinsed, washedand dried for exposureto Hyperfilm 3H (Amersham International) for 4-21 days, and autoradiographswere developedin Kodak D-19 (Kodak), as described (6, 7). After exposure,sectionswere fixed in formaldehyde and stained with hematoxylin and eosin. Preliminary experiments showedthat the specifically reversible binding of 1251[TyrO] CNP- (I - 22) did not increasesignificantly with incubations beyond 15 min, aspreviously shownfor 1251-~-ANP(7). The stability of 1251-[Tyro]CNP-(1-22) under the conditions of incubation was also checked in preliminary experiments by gamma counting of fractions collected from high-pressureliquid chromatography (HPLC) of the incubation fluid using a Cl, column (Novapak Radialpak, Millipore) and a O-60% linear gradient of acetonitrile in water containing 0.04% trifluoroacetic acid (Fisons Scientific Equipment) run over 40 min, again as we have describedfor 1251-a-ANP(5). Regionalbinding of the radioligandswasmeasuredin femtomolesper squaremillimeter in individual glomeruli of the outer third of the renal cortex and in the medulla by comparing the regional optical densities of the renal autoradiographs with those producedby 1251standards aspreviously described(6, 7). Specific gravity and Lowry protein content values determined in blocks of renal cortex and medulla from six further male Wistar rats (250-300 g) were usedto convert values of bound ligand from femtomolesper squaremillimeter to femtomolesper milligram protein, as described(6, 7). The number of binding sitesof different affinities for a given radioligand, their apparent dissociation constants (Kd) and their maximum binding capacities (B,,,) on particular structures were derived separately in each individual using the
CNP
RECEPTORS
LIGAND iterative model-fitting computer program (28). The inhibitory constant (Ki) and B,,, for the binding of a nonhomologousunlabeledpeptide which inhibited radioligand binding were also determined separately for each individual by LIGAND. Guanosine 3’,5’-cyclic monophosphate (cGMP) assay. cGMP production wasmeasuredin glomeruli isolatedfrom finely diced cortex by sieving (27). Glomeruli werepooledfrom both kidneys of each of sevenmale Wistar rats (250-300 g). Aliquots of 200 glomeruli were preincubated for 2 min at 20°C in Hanks’ buffered salt solution [containing (in mM) 137 NaCl, 5.4 KCl, 0.4 0.34 Na2HP04, 1.26 CaC12, 4.17 NaHC03, 0.44 J&ml, KH2P04, and 0.49 MgCl,] plus 5.56 mM glucose,10 mM N-2hydroxyethylpiperazine-N’-2-ethanesulfonic acid, 0.2% bovine serumalbumin (fraction V, proteasefree), and 1 mM isobutylmethylxanthine (Sigma), centrifuged at 300 g for 3 min, and then resuspendedin the samesolution with a range of concentrations of either rat cu-ANPor CNP-(1-22) (both O-l PM) or with 1 PM C-ANP. Nine glomerular aliquots were incubated separately at each concentration of both peptidesfor 10 min at 20°C. Incubations were terminated by addition of ice-cold trichloroacetic acid to a final concentration of 6%. The aliquots were then centrifuged at 3,000g for 10 min, and the supernatants were extracted four times, each time with thrice their volume of water-saturated ether. cGMP in the supernatant was then measured for each aliquot by radioimmunoassayafter acetylation (Amersham). Statistics. Results are means * SE. The fits of single and multiple site modelsof ligand binding were comparedboth by the extra sum of squarestest and the runs test using LIGAND (28). Other comparisonswere by paired or unpaired Student’s t tests or by Fisher’s variance ratio tests as appropriate (33). RESULTS
Competitive inhibition of 1251-labeledcu-ANP binding.
Comparison of autoradiographs with their corresponding stained sections revealed specifically reversible binding of wANP to glomeruli, intrarenal arteries, outer medullary vasa recta bundles, and inner medulla (Fig. 1). The unrelated peptides angiotensin II, vasopressin, and gastrin displaced none of this binding. Analysis of the competitive inhibition of the binding of 1251-~-ANP by unlabeled cu-ANP on glomeruli, vasa recta bundles, and inner medulla was consistent with reversible binding sites for cu-ANP of uniform affinity on each structure. The Kd and B max of these sites were, respectively, 3.15 t 0.82 nM and 583 t 122 fm o l/ mg protein on glomeruli, 3.61 t 0.39 nM and 341 t 46 fmol/mg protein on vasa recta bundles, and 5.59 t 1.55 nM and 913 t 280 fmol/mg protein on inner medulla. These results are similar to those of previous investigations of the distribution and binding constants of cu-ANP receptors in rat kidney (6, 7, 24). C-ANP (10 PM), which is known to almost fully saturate ANPR-C without significantly binding to renal guanylate cyclasecoupled receptors (6, 7,25, 29), inhibited 51.3 t 10.7% of the specifically reversible binding of 1251-a-ANP to glomeruli. However, it did not significantly alter the specifically reversible radioligand binding to vasa recta bundles (inhibiting 7.0 t 7.0%) or to inner medulla (inhibiting 10.6 t 11.7%), just as in previous studies (6, 7, 25). Increasing concentrations of CNP-( 1 - 22) progressively inhibited the binding of 12~I-cr-ANP to glomeruli, and this inhibition was consistent with sites of a single affinity for CNP-( 1-22) characterized by a Ki of 10.47 t 7.59 nM and B,,, of 342 t 118 fmol/mg protein. This B max was significantly less than that of 583 t 122 l
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
RENALANPANDCNPRECEPTORS
F91
Fig. 1. Autoradiographs of binding of 100 pM 1251-labeled a-atria1 natriuretic peptide (‘251-a-ANP) (A-D) or 110 pM 12sI-labeled C-type natriuretic peptide ( 1251-[Tyro]CNP-(1-22)] (E-H) in rat kidney. Binding is shown in absence of competing unlabeled ligands (A and E) and in presence of 1 (B) or 10 PM (F) unlabeled wANP, 10 PM CNP-(l-22) (C and G), and 10 FM C-ANP (D and H). g, glomeruli; vr, vasa recta bundles; im, inner medulla.
fmol/mg protein for the glomerular < 0.01, paired t test). CNP-(1-22) less affinity for the binding sites recta bundles and inner medulla 13.3% of the reversible binding recta bundles (P < 0.01) and 59.8
binding of cr-ANP (P competed with much of 1251-a-ANP on vasa (Fig. 2). Thus 59.7 f. of 1251-a-ANP to vasa f 7.3% of that to inner
medulla (P < 0.001) remained bound even in the presence of 10 FM CNP-(1-22). The inhibition of radioligand binding produced on any structure by 10 PM CNP(l-22) plus 1 PM a-ANP was not significantly different from that produced by 1 PM cu-ANP alone. Similarly, the inhibition of radioligand binding on any structure
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
F92
RENAL
Vasa
Glomeruli
recta
ANP AND CNP RECEPTORS
Inner
bundles
medulla
Fig. 2. Competitive inhibition of specifitally reversible binding of 1251-a-ANP by increasing concentrations of CNP_ (1 - 22) in glomeruli (A), vasa recta bundles (B), and inner medulla (C) of kidneys of 6 rats.
8 6c3 z 50 za 45 3 m G 2cLLI 3 ’ = o0
lb-”
lb-9
lb-’
(j
1 o-5
16-11
CONCENTRATION
16-9
lb-7
OF
l&5
CNPo-,,,
(M)
produced by 10 PM CNP-( l-22) plus 10 PM C-ANP was not significantly different from that by 10 PM CNP(l-22) alone. Competitive inhibition binding. Autoradiographs
of
1251-[Tyro]CNP-(l -22)
revealed that 110 pM lz51[TyrO] CNP-( I- 22) bound to glomeruli and did not bind significantly to any other renal structure (Fig. 1). This binding was not altered by the unrelated peptides angiotensin II, vasopressin, and gastrin. However, it was progressively inhibited by increasing concentrations of unlabeled [ TyrO] CNP- (1 - 22)) consistent with glomerular binding sites of uniform affinity for this ligand. The Kd and B,ax values for these glomerular sites were 1.42 t 0.48 nM and 67 t 17 fmol/mg protein, respectively. a-ANP (10 PM), 10 PM CNP-(l-22), and 10 PM C-ANP all completely prevented this specifically reversible binding of 1251-[Tyro]CNP-(l-22) (Fig. 1). The Ki and B,,, values for the inhibition of the specifically reversible globy a-ANP merular binding of 1251-[Tyro]CNP-(l-22) were 1.24 t 0.18 nM and 78 t 22 fmol/mg protein, respectively, and those for the corresponding inhibition by CNP-(l-22) were 0.83 t 0.13 nM and 73 t 24 fmol/mg protein, respectively (Fig. 3). The Ki for CNP-( l-22) was significantly smaller than that for wANP (P < 0.05). There were no significant differences among the B,,, values for the binding of [TyrO]CNP-(l-22), wANP, or CNP-( l-22) as assessed by their displacement of lz51- [TyrO] CNP- (1 - 22). However, these were all significantly smaller than the B,,, of 342 t 118 fmol/mg protein for the glomerular binding of CNP-( l-22) assessed from its ability to inhibit the binding of lz51-a-ANP (all P < 0.01). Finally, there was no significant difference between the inhibition of the binding of lz51- [TyrO] CNP(l-22) produced by 10 PM CNP-( l-22) plus 1 PM cu-ANP or by 10 PM CNP-(l-22) plus 10 PM C-ANP. Saturation of ‘“51_[Tyr”]CNP-(1 -22) Ibinding. Increasproduced a ing concentrations of 1251-[Tyro]CNP-(l-22) progressive increase in the glomerular binding of the radioligand. The specifically reversible component of this
binding was consistent with saturable ligand binding sites of uniform affinity. The Kd and B,,, values for these sites were 0.41 t 0.07 nM and 29 t 4 fmol/mg protein, respectively. No specifically reversible radioligand binding was detected to structures other than glomeruli at any of the concentrations (up to 1 nM) of 1251-[Tyro]CNP(l-22) that were used. Recovery of 1251-[Tyro]CNP-(l -22). The HPLC of incubation fluids recovered after exposure to the renal sections showed that 62.1 t 8.5% of the total counts added were recovered in a chromatographic peak with the retention time of 33 min, which was identical to that of authentic lz51- [ Tyr”] CNP- (1 - 22). Two further HPLC peaks, each containing -15% of the total counts, were at retention times of 20 and 22.5 min, respectively. Production of cGMP. Basal production of cGMP, in the absence of added natriuretic peptides, was 0.22 t 0.04 pm01 . mg protein1 min -l. The effects of cr-ANP and CNP-(l-22) are shown in Fig. 4. wANP markedly increased cGMP production, an effect that tended to saturate above 10e7 M cu-ANP. For example, lo-” M cu-ANP increased cGMP production to 1.51 t 0.40 pmol. mg protein-‘. min-l, which was significantly different from the basal rate (P < 0.001). Only lop7 and lo-” M CNP(1 - 22) significantly increased cGMP production, and there was no indication that this effect was saturated by 10m6 M CNP-(l-22). For example, 10m7 M CNP-(l-22) increased cGMP production to 0.34 t 0.04 pmol mg protein-l min-‘, which was significantly higher than basal (P < 0.05). cGMP production in the presence of lo+ M C-ANP was 0.22 t 0.04 pmolmg protein-l min-l, which was not different to basal. l
l
l
DISCUSSION
The findings described here confirm previous radioligand binding and receptor cross-linking studies which show that the clearance-related receptor (ANPR-C) comprises 50-80% of the glomerular receptors for cu-ANP and is not detectable on vasa recta bundles or in the inner
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
RENAL
F93
ANP AND CNP RECEPTORS
-oCANP --+CN~,-,,, -- tyr”-CNP,,e,,,
-- 0 -O-
a
+
CW,-,a
-ANP
0
,o-~'
,o-'o
,o-9
1o-8
1o-7
,o-6
LIGAND CONCENTRATION (M) Fig. 4. Effects of wANP and CNP-( l-22) on basal (+) cGMP production by isolated glomeruli. Each point is mean of 9 determinations, each using glomeruli pooled from ‘7 rats.
in the absence of a specific, high-affinity ligand for ANPR-B, it has not hitherto been possible to document the expression of this receptor by different tissues in situ. CNP-(l-22), [Tyr*]CNP-(l-22), and 1251-[Tyro]. -2, --- 0 CNP-( l-22) all behave as high-affinity ligands that are d 4. 0I 1 strongly selective for ANPR-B and ANPR-C over 0 ANPR-A when the human forms of these receptors are CONCENTRATION OF UNLABELLED LIGAND (M) introduced by vectors into cells which do not normally Fig. 3. Competitive inhibition of specifically reversible binding of 1251express such receptors (19). Consistent with this in the [TyrO]CNP-( l-22) to renal glomeruli in 6 rats caused by increasing present experiments in rats, CNP-( 1 - 22) displaced 1251concentrations of unlabeled ligands [TyrO]CNP-(l-22), ar-ANP, and CNP-(1-22). a-ANP most effectively from renal sites known to express ANPR-C, namely glomeruli. CNP-(l-22), 10 PM, bound to all the glomerular cu-ANP receptors that bound 10 PM C-ANP, because there was no significant difference medulla of the rat (2, 6, 7, 25). The same radioligand binding and receptor cross-linking studies, coupled to between the inhibition of the specifically reversible glostudies of cGMP production (2), have suggested that the merular binding of 1251-~-ANP produced by 10 PM CNP(l-22) and that produced by 10 PM CNP-(l-22) plus 10 residue of wANP receptors on these renal structures, which do not bind C-ANP and therefore differ from PM C-ANP. Moreover, the B,,, for the binding of CNP(1 - 22) to glomeruli was -60% of that for the glomerular ANPR-C, are guanylate cyclase-coupled receptor proteins of -120-130 kDa. Molecular cloning studies have idenbinding of wANP. This was similar to the proportion of tified two such guanylate cyclase-coupled receptors in glomerular clearance receptors for cu-ANP (51.3 t 10.7%) both the human and rat genome. The first of these, assessed from the inhibition of the binding of f251-~-ANP ANPR-A, has high affinity for a-ANP when clones from caused by 10 PM C-ANP in the same rats. CNP-(l-22) binding either species are expressed (10, 11,19,31). The second of also displaced 1251-~-ANP from its high-affinity these receptors, ANPR-B, is virtually without affinity for sites in the medulla, which does not contain ANPR-C, a-ANP, regardless of its specific origin (10, 11,19,31). On but it did this with a greatly lower affinity than in glothe basis of this wide difference of affinities for a-ANP, it meruli, which do express ANPR-C. Because 10 PM CNP plus 1 PM unlabeled a-ANP displaced no more radioliis likely that high-affinity guanylate cyclase-coupled binding sites for wANP in rat kidney at least partly gand binding than 1 PM unlabeled a-ANP alone, these results suggest that CNP-( l-22) has a significantly represent ANPR-A. However, low-affinity, specifically higher affinity for ANPR-C than for the remaining reversible binding sites for wANP also occur in rat kidwANP receptors of the rat kidney, which conform not ney (7), and these could represent ANPR-B. Porcine only to the rat and human forms of ANPR-A in their high BNP-( l-26) was originally reported as a specific ligand affinity for wANP and poor affinity for C-ANP (10, 11, of ANPR-B over ANPR-A (lo), but subsequent studies have not found such a wide difference in its binding and 19, 31), but also at least to the human form of ANPR-A in its poor affinity for CNP-( 1 - 22) (19). activation of ANPR-A and ANPR-B (19, 31). Therefore, ,;-11
$10
$9
&-a
$7
10-6
lo+
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
F94
RENAL
ANP
AND
The high affinity specifically reversible binding of 1251[ TyrO] CNP- (1 - 22) further strengthened this conclusion. No specifically reversible binding of 1251-[TyrO] CNP(l-22) occurred in the renal medulla, a region without ANPR-C but rich in guanylate cyclase-coupled natriuretic peptide receptor sites, even when sections were incubated with up to 1 nM of the radioligand. Instead, the high-affinity binding of 12?I- [ TyrO] CNP- (1 - 22) was confined to glomeruli, and here it was completely inhibited by excess C-ANP and had high affinity for cu-ANP (Ki = 1.24 t 0.18 nM), confirming that it was to ANPR-C rather than ANPR-B. However, not all ANPR-C seemed to bind CNP-( 1-22) with high affinity. The specifically reversible binding of 1251-[Tyro]CNP-(l-22) to ANPR-C was of high affinity (& = 0.41 t 0.07 nM), and the affinities of CNP-(1-22) and [TyrO]CNP-(1-22) for these same binding sites were also high (affinity constants of 0.83 t 0.13 and 1.42 t 0.48 nM, respectively). Nevertheless, the estimates of B,,, for these glomerular sites were uniformly low, however they were measured. Thus the B,,, values derived from the inhibition of the binding of 110 pM 1251-[Tyro]CNP-(l-22) by CNP(l-22) and [TyrO]CNP-(1-22) were 73 t 24 and67 t 17 fmol/mg protein, respectively, and the B,,, value measured from the saturable binding of up to 1 nM 1251[TyrO]CNP-(1-22) was 29 t 4 fmol/mg protein. In contrast, the binding of excess C-ANP suggested that ANPR-C comprises more than 50% of the glomerular B 17naX for cu-ANP of 583 t 122 fmol/mg protein, confirming previous estimates of the total density of glomerular ANPR-C (6, 7). It is possible, therefore, that only a fraction of ANPR-C, as defined by binding C-ANP, have a sufficiently high affinity for analogues of CNP-( I- 22) to bind subnanomolar concentrations of 1251-[TyrO] CNP(l-22) appreciably. On the other hand, much higher concentrations (up to 10 PM) of CNP-(1-22) competed measurably at -60% of the glomerular binding sites of 1251-ar-ANP. Although, as in the medulla, some of this competition may have been with low affinity at glomerular ANPR-A, most of it was at ANPR-C, because there was no significant difference between the inhibition of the glomerular binding of 1251-a-ANP achieved by IO PM CNP-(l-22) or by 10 PM CNP-(1-22) plus 10 PM C-ANP. This implies that up to 10 PM CNP-(1-22) could also inhibit binding of 1251-~-ANP to ANPR-C that did not bind subnanomolar concentrations of 1251[TyrO]CNP-( 1 - 22) significantly. It is not certain whether, compared with other ANPR-C, ANPR-C that apparently did not bind subnanomolar 1251-[TyrO] CNP(l-22) significantly also had a lower affinity for CNP(l-22), but such heterogeneous affinities of ANPR-C for CNP-( 1 - 22) might explain the relatively low affinity and variability of the gross Ki value of 10.47 t 7.59 nM derived from the single site analysis of the inhibition of f251-~-ANP binding by CNP-(1-22). This Ki value was an order of magnitude higher than the Ki value of 0.83 t 0.13 nM for the inhibition of 1251-[Tyro]CNP-(l-22) binding by CNP-( 1 - 22). Nevertheless, the data actually did not establish binding of CNP-( 1 - 22) to glomerular cu-ANP receptors with significantly more than one affinity for CNP-(l-22), although the concentration inter-
CNP
RECEPTORS
vals of CNP-(1-22) used to inhibit 1251-a-ANP binding were not spaced to address this point in detail. Moreover, some degradation of 1251-[TyrO] CNP- (1 - 22) did occur during the present incubations despite the use of conditions of incubation and peptidase inhibitors that have previously been shown to prevent significant degradation of 1251-a-ANP by renal sections (7). Although -62% of the 1251-[Tyro]CNP-( l-22) remained intact by the end of the incubations, the B,,, for its binding would be underestimated in proportion to the reduction of its effective concentration. Consequently, quantitative comparison of results involving 1251-a-ANP and 1251-[TyrO] CNP(l-22) should be cautious, and further experiments will be necessary to settle the issue of subtypes of ANPR-C. This notwithstanding, the main conclusion of the present study is that no high-affinity binding sites for either CNP-(1-22) or 1251-[TyrO] CNP- (1 - 22) were detected in the rat kidney apart from those related to ANPR-C by their affinities for C-ANP and a-ANP, and no sites compatible with ANPR-B were found. This conclusion was strengthened by functional data. We found that CNP-( l-22) did not significantly stimulate cGMP production by isolated glomeruli unless concentrations of 10e7 M or more were used. This confirms a preliminary report (12), and it is not consistent with the presence of detectable quantities of an ANPR-B for which CNP-( l-22) is a high-affinity ligand. ANPR-C may also stimulate cGMP production in some circumstances (15). However, there was a substantial disparity between the high affinities of CNP-( l-22) and its analogues for glomerular binding sites related to ANPR-C and the high concentrations of CNP-( l-22) required to stimulate glomerular guanylate cyclase. Instead, slight stimulation of glomerular guanylate cyclase activity at the highest concentrations of CNP-( 1 - 22) is consistent with the data that CNP-( 1 - 22) is a low-affinity ligand of ANPR-A. Thus both binding and functional data are consonant with the lack of detectable quantities of ANPR-B in adult male Wistar rat kidney. Nevertheless, mRNA for ANPR-B may occur in adult rat kidney (3 1). It is known that mRNA levels do not necessarily reflect the degree of expression of the encoded protein (21). However, interpretation of the published Northern analysis showing renal mRNA for ANPR-B (31) should be cautious, because the small quantity of mRNA in the renal electrophoretic lane was adjacent to a very much larger quantity of mRNA from lung, and contamination between lanes was not excluded. In the absence of a renal guanylate cyclase-coupled receptor for CNP-( l-22), CNP-( l-22) could still have effects by binding ANPR-C coupled to production of other second messengers. For example, C-ANP and other specific ligands of ANPR-C can inhibit CAMP production (1). However, C-ANP itself seems to have no acute effects on the function of the isolated rat kidney (22) and does not modify the effects of cu-ANP on isolated rat renal tissues (2,22,29). Instead, the available evidence actually suggests that the short-term role of ANPR-C in the rat kidney is to clear bound ligands (2, 22). Consistent with this, the acute excretory effects of CNP-( l-22) are -loo-fold less potent than those of cu-ANP in the rat
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
RENAL
(37). Over longer periods, however, synthetic ligands specific for ANPR-C can also modulate the growth of cultured rat aortic smooth muscle (9). CNP-( 1 - 22) seems not to circulate (38), but it may nevertheless occur in the rat kidney since local CNP-like immunoreactivity has been found there (20). Our results suggest that CNP(I -22) and its close analogues principally bind to ANPR-C in the kidney. Therefore these findings taken together raise the important possibility that intrarenal CNP-( l-22) may interact at local ANPR-C to regulate tissue growth. This work was supported by the National Kidney Research Fund and the British Heart Foundation. Address for reprint requests: J. Brown, Physiological Laboratory, Downing St., Cambridge CB2 3EG, UK. Received 19 November 1991; accepted in final form 18 December 1991. REFERENCES 1. Anand-Srivastava, M. B., M. R. Sairam, and M. Cantin. Ring-deleted analogs of atria1 natriuretic factor inhibit adenylate cyclase/cAMP system. J. BioZ. Chem. 265: 8566-8572, 1990. 2. Brenner, B. M., B. J. Ballermann, M. E. Gunning, and M. L. Zeidel. Diverse biological actions of atria1 natriuretic peptide. Physiol. Reu. 70: 665-699, 1990. 3. Brown, J., and B. Cheung. Binding of rat brain natriuretic peptide (BNP) to atria1 natriuretic peptide (ANP) receptors in rat kidney (Abstract). J. Physiol. Lond. 446: 91P, 1992. 4. Brown, J., and A. Czarnecki. Autoradiographic localization of atria1 and brain natriuretic peptide receptors in rat brain. Am. J. Physiol. 258 (Regulatory Integrative Comp. Physiol. 27): R57-R63, 1990. 5. Brown, J., and A. Czarnecki. Binding of atria1 and brain natriuretic peptides in brains of hypertensive rats. Brain Res. 512: 132-137, 6.
7.
8.
9.
10.
1990.
Brown, J., S. P. Salas, and J. M. Polak. Renal atria1 natriuretic peptide receptor subtypes in spontaneously hypertensive rats. Am. J. Physiol. 259 (Renal Fluid Electrolyte Physiol. 28): F605-F612, 1990. Brown, J., S. P. Salas, A. Singleton, J. M. Polak, and C. T. Dollery. Autoradiographic localization of atria1 natriuretic peptide receptor subtypes in rat kidney. Am. J. Physiol. 259 (Renal Fluid Electrolyte Physiol. 28): F26-F39, 1990. Brown, J., and 2. Zuo. Receptor binding of C-type natriuretic peptide CNP-(1-22) in rat brain (Abstract). J. Physiol. Lond. 446: 44OP, 1992. Cahill, P. A., and A. Hassid. Clearance receptor-binding atria1 natriuretic peptides inhibit mitogenesis and proliferation of rat aortic smooth muscle cells. Biochem. Biophys. Res. Commun. 179: 1606-1613, 1991. Chang, M., D. G. Lowe, M. Lewis, R. Hellmiss, E. Chen, and D. V. Goeddel. Differential activation by atria1 and brain natriuretic peptides of two different receptor guanylate cyclases. Nature
Lond.
341: 68-72,
1989.
Chinkers, M., D. L. Garbers, M. S. Chang, D. G. Lowe, H. Chin, D. V. Goeddel, and S. Schulz. A membrane form of guanylate cyclase is an atria1 natriuretic peptide receptor. Nature Lond. 338: 78-83, 1989. M., M. Takehisa, Y. Minamitake, Y. Kitajima, Y. 12. Furuya, Hayashi, N. Ohnuma, T. Ishihara, N. Minamino, K. Kangawa, and H. Matsuo. Novel natriuretic peptide, CNP, potently stimulates cyclic GMP production in rat cultured vascular smooth muscle cells. Biochem. Biophys. Res. Commun. 170: 201-208, 1990. T., K. Higuchi, M. Ohashi, R. Takayanagi, H. 13. Hashiguchi, Matsuo, and H. Nawata. Effect of porcine brain natriuretic peptide (pBNP) on human adrenocortical steroidogenesis. CZin. Endocrinol. 31: 623-630, 1989. 14. Imura, H., and K. Nakao. Central nervous system actions of atria1 natriuretic peptide. In: Atria1 Natriuretic Peptides, edited by W. K. Samson and R. Quirion. Boca Raton, FL: CRC Press, 1990, p. 221-230. 11.
F95
ANP AND CNP RECEPTORS
15. Ishido, M., T. Fujita, M. Shimonaka,
T. Saheki, S. Ohuchi, T. Kume, I. Ishigaki, and S. Hirose. Inhibition of atria1 natriuretic peptide-induced cyclic GMP accumulation in bovine endothelial cells with anti-atria1 natriuretic peptide receptor antiserum. J. BioZ. Chem. 264: 641-645, 1989. 16. Itoh, H., K. Nakao, T. Yamada, G. Shirakami, K. Kangawa, N. Minamino, H. Matsuo, and H. Imura. Anti-dipsogenic action of a novel peptide, brain natriuretic peptide, in rats. Eur. J. Pharmacol. 150: 193496, 1988. 17 Kambayashi, Y., K. Nakao, H. Itoh, K. Hosoda, Y. Saito, T. Yamada, M. Mukoyama, H. Arai, G. Shirakami, S. Suga, Y. Ogawa, M. Jougasaki, N. Minamino, K. Kangawa, H. Matsuo, K. Inouye, and H. Imura. Isolation and sequence determination of rat cardiac natriuretic peptide. Biochem. Biophys. Res. Commun. 163: 233-240, 1989. 18. Kojima, M., N. Minamino, K. Kangawa, and H. Matsuo. Cloning and sequence analysis of a cDNA encoding a precursor for rat C-type natriuretic peptide (CNP). FEBS Lett. 276: 209-213, 1990. K. lg. Koller, K. J., D. G. Lowe, G. L. Bennett, N. Minamino, Kangawa, H. Matsuo, and D. V. Goeddel. Selective activation of the B natriuretic peptide receptor by C-type natriuretic peptide (CNP). Science Wash. DC 252: 120-123, 1991. Y ., K. Nakao, S. Suga, Y. Ogawa, M. 20. Komatsu, Mukoyama, H. Arai, G. Shirakami, K. Hosoda, 0. Nakagawa, N. Hama, I. Kishimoto, and H. Imura. C-type natriuretic peptide (CNP) in rats and humans. Endocrinology 129: 1104-1106, 1991. 21. Latchman, D. Gene Regulation. A Eukaryotic Perspective. London: Unwin Hyman, 1990, p. 269. 22. Maack, T., M. Suzuki, F. A. Almeida, D. Nussenzveig, R. M. Scarborough, G. A. McEnroe, and J. A. Lewicki. Physiological role of silent receptors of atria1 natriuretic factor. Science Wash. DC 238: 675-678, 1987. 23. Maeda, T., M. Niwa, K. Shigematsu, M. Kurihara, Y. Kataoka, K. Nakao, H. Imura, H. Matsuo, H. Tsuchiyama, and M. Ozaki. Specific [ 1251] brain natriuretic peptidebinding sites in rat and pig kidneys. Eur. J. Pharmacol. 176: 341-350, 1990. 24. Maekawa, K., T. Sudoh, M. Furusawa, N. Minamino, K. Kangawa, H. Ohkudo, S. Nakanishi, and H. Matsuo. Cloning and sequence analysis of cDNA encoding a precursor for porcine brain natriuretic peptide. Biochem. Biophys. Res. Commun. l
157: 410-416,
1988.
25. Martin, E. R., J. A. Lewicki, R. M. Scarborough, and B. J. Ballermann. Expression and regulation of ANP receptor subtypes in rat renal glomeruli and papillae. Am. J. Physiol. 257 (Renal Fluid Electrolyte Physiol. 26): F649-F657, 1989. 26 . Minamino, N., K. Kangawa, and H. Matsuo. N-terminally extended form of C-type natriuretic peptide (CNP-53) identified in porcine brain. Biochem. Biophys. Res. Commun. 170: 973-979, 1990. 27 . Misra, R. 0. Isolation of glomeruli from mammalian kidneys by graded sieving. Am. J. Clin. Pathol. 58: 135-139, 1972, 28 Munson, P. J., and D. Rodbard. LIGAND; a versatile com. puterized approach for characterization of ligand binding system. Anal. Biochem. 107: 220-239, 1980. D., R. M. Scarborough, J. A. Lewicki, and T. 2g* Nussenzveig, Maack. Clearance receptors of atria1 natriuretic factor (C-ANP receptors) in isolated glomeruli and mesangial cells in culture (Abstract). Kidney Int. 33: 279P, 1988. 30. Saito, Y., K. Nakao, H. Itoh, T. Yamada, M. Mukoyama, H. Arai, K. Hosoda, G. Shirakami, S. Suga, N. Minamino, K. Kangawa, H. Matsuo, and H. Imura. Brain natriuretic peptide is a novel cardiac hormone. Biochem. Biophys. Res. Commun. 158: 360-368, 1989. 31. Schulz, S., S. Singh, R. A. Bellet, G. Singh, D. J. Tubb, H. Chin, and D. L. Garbers. The primary structure of a plasma membrane guanylate cyclase demonstrates diversity within this new receptor family. CeZL58: 1155-1162, 1989. J. J., A. Arfsten, J. A. Miller, P. Lindquist, R. 32. Seilhamer, M. Scarborough, J. A. Lewicki, and J. G. Porter. Human and canine gene homologs of porcine brain natriuretic peptide. Biochem. Biophys. Res. Commun. 165: 650-658, 1989.
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
F96
RENAL
ANP
AND
33. Snedecor, G. W., and W. G. Cochran. Statistical Methods (6th ed.). Ames: Iowa State Univ. Press, 1974. 34. Song, D. L., K. P. Kohse, and F. Murad. Brain natriuretic factor. Augmentation of cellular cyclic GMP, activation of particulate guanylate cyclase and receptor binding. FEBS Lett. 232: 125-129, 1988. 35. Sudoh, T., K. Kangawa, N. Minamino, and H. Matsuo. A new natriuretic peptide in porcine brain. Nature Land. 332: 78-81, 1988. 36. Sudoh, T., N. Minamino, K. Kangawa, and H. Matsuo. Brain natriuretic peptide 32: N-terminal six amino acid extended form of brain natriuretic peptide identified in porcine brain. Biochem. Biophys. Res. Commun. 155: 726-732, 1988.
CNP
RECEPTORS
37. Sudoh, T., N. Minamino, K. Kangawa, and H. Matsuo. C-type natriuretic peptide (CNP): a new member of natriuretic peptide family identified in porcine brain. Biochem. Biophys. Res. Commun. 168: 863-870, 1990. M. Aburaya, K. Kangawa, S. Mat38. Ueda, S., N. Minamino, sukura, and H. Matsuo. Distribution, and characterization of immunoreactive porcine C-type natriuretic peptide. Biochem. Biophys. Res. Commun. 175: 759-767, 1991. 39. Yamada, T., K. Nakao, H. Itoh, G. Shirakami, K. Kangawa, N. Minamino, H. Matsuo, and H. Imura. Intracerebroventricular injection of brain natriuretic peptide inhibits vasopressin secretion in conscious rats. Neurosci. Lett. 95: 223-228, 1988.
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
AND Z. ZUO Laboratory,
Cambridge
CB2 3EG, United Kingdom
or intravenously (2, 14, 16, 35, 39). Brown, J., and 2. Zuo. Renal receptors for atria1 and broventricularly C-type natriuretic peptides in the rat. Am. J. Physiol. 263 Consequently, these two groups of peptides may have (Renal Fluid ElectroZyte Physiol. 32): F89-F96, 1992.-Recep- complimentary physiological and pathological effects. tors for a-atria1 natriuretic peptide (cu-ANP) and C-type natriCNPs have so far been isolated from porcine brain as uretic peptide [CNP-( l-22)] were quantified in kidneys from a 22-amino acid form [CNP-(l-22)] (37) and as a adult Wistar rats by in vitro autoradiography. 1251-labeled 53-amino acid form [CNP-(l--53)], which includes the cu-ANP(100 PM) bound reversibly to glomeruli, outer medullary at its COOH-terminal (26). vasa recta, and inner medulla with an apparent dissociation sequence of CNP-(l-22) Molecular cloning techniques suggest that rat and porconstant (&) of 3-6 nM. The presence of 10 PM descine CNPs are identical (18). The intraand extraceren18,Ser1g,Gly20,Leu21,Gly22]ANP-(4-23) (C-ANP), a spe[Gl of CNPs are only now being defined cific ligand of the ANPR-C subtype of a-ANP receptor, inhib- bra1 distributions ited -50% of the glomerular binding of 1251-cz-ANP,and this (20, 38), and it is not clear to what extent CNPs might moiety of glomerular binding was also inhibited by CNP- compliment actions of the other natriuretic peptides. (l-22) with an apparent inhibitory constant (Ki) of 10.47 t However, CNP-( 1 - 22) may not share all of its receptors 7.59nM. C-ANP and CNP-( l-22) showedlittle affinity for the with the ANPs and BNPs (19). Molecular cloning has medullary binding sites of cr-ANP. 1251[TyrO] CNP- (I - 22) identified three receptors for natriuretic peptides. Two (110 PM) bound solely to glomeruli and was competitively dis- of these, ANPR-A and ANPR-B, are proteins of -120placedby increasingconcentrations of [TyrO]CNP-( 1- 22) with guanylate cyclase (2, 10, 11, an apparent Kd of 1.42 t 0.48 nM. Binding of increasing con- 130 kDa with constitutive dimer of centrations (25 pM to 1 nM) of 1251-[Tyro]CNP-(1-22) in the 31). The third, ANPR-C, is a disulfide-bridged guanylate presenceor absenceof 1 PM [TyrO]CNP-(1-22) also demon- 66 kDa units. This receptor has no intrinsic strated a high affinity (Kd of 0.41 t 0.07 nM) for the glomerular cyclase, and it probably binds and clears its ligands by (2,22,29). It, or closely similar proteins, binding of f251-[Tyro]CNP-(1-22). Bound 1251-[Tyro]CNP- internalization (l-22) could be displacedby excesscu-ANP and excessCNP- may also be facultatively linked to second messenger (1- 22), both with high affinities. The glomerular binding of systems (1, 15). ANPR-A has strict structural require1251-[Tyro]CNP-(1-22) was also prevented by 10 PM C-ANP. ments of its ligands so that it has high affinity for Guanosine3’,5’-cyclic monophosphateproducedby isolatedglo- cu-ANP but little or no affinity for synthetic analogues meruli was measuredby radioimmunoassay.cr-ANP increased of cu-ANP such as [TyrO]ANP-(5-25) and desproduction powerfully, but CNP-( 1-22) causedonly a slight n1S,Ser1g,Gly20,Leu 21 , Gl~~~lANP-(4--23) (C-ANP) stimulation with much lower affinity. Theseresults suggestthat W (2, 6, 7, 25, 29). ANPR-C is much less conservative and Wistar rat kidneys expressdetectablequantities of the ANPR-C affinities for all of these ANPs (2, 6, 7, and ANPR-A but not the ANPR-B subtypesof natriuretic pep- has significant 22, 25). In addition, both ANPR-A and ANPR-C bind tide receptor. guanylate cyclase-coupledreceptors; clearancereceptors; auto- avidly to a variety of BNPs such as porcine BNP(l-26) and rat BNP-(l-45) (3-5,23,34). ANPR-B has radiography C-typenatriureticpeptides (CNPs) share many properties of the atria1 and brain natriuretic peptides (ANPs and BNPs, respectively). The CNPs, ANPs, and BNPs arise respectively from posttranslational modifications of the products of three different genes (2, 18, 24, 26, 30), but their structures all share a characteristically similar 17-amino acid disulfide-bridged ring (2, 17, 18, 26, 35-37). As a result, the peptides also partly share the same receptors (3-5, 8, 10-13, 19, 23, 31, 34), and all are to some extent natriuretic, diuretic, and vasorelaxant (2,35, 37). ANPs were first isolated from the heart, from which they enter the circulation, principally as the 2%amino acid a!-ANP, to exert a variety of actions on the cardiovascular system and on fluid homeostasis (2). ANPs are also synthesized in the central nervous system, where they probably modulate the nervous control of the circulation and body fluids (14). In contrast, BNPs were first isolated from the brain, but it is now known that, like ANPs, they are also synthesized by the heart and released into the circulation (17, 30). ANPs and BNPs share receptors both intra- and extracerebrally (3-5, 13, 23, 34), and they have similar actions when injected either intracereTHERECENTLYDISCOVERED
been identified by molecular cloning, and because it has had no known high-affinity ligands [for example, having virtually no affinity for a-ANP (10,19,31)], nothing has been discovered of its in vivo expression. Recently, however, ANPR-B has been expressed in transfected cells and shown to have high affinity for CNP-( l-22) (19). CNP-( 1 - 22) has also been found to have significant affinity for ANPR-C but none for ANPR-A in such cells (19). Thus ANPR-B is a guanylate cyclase-coupled receptor that seems to be relatively specific for CNP(l-22), and its in vivo distribution may have important functional consequences. Here, we use quantitative autoradiography to examine the endogenous receptors of CNP-( l-22) and CV-ANP in rat kidney. The results demonstrate that CNP-( l-22) binds avidly to glomerular ANPR-C expressed in vivo. However, no other high-affinity renal receptors for CNP-( l-22) were found. This suggests that rat kidney expresses ANPR-A and ANPR-C in vivo, but does not express detectable quantities of ANPR-B. METHODS
Autoradiographic studies. Three groups, each of six adult male Wistar rats (260-330 g; Charles Rivers Breeding) were 0363-6127/92 $2.00Copyright 0 1992the American Physiological Society F89
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
F90
RENAL
ANP
AND
maintained with free access to water and food (Labsure PRD, William Lillico). The rats were killed by rapid exsanguination. Kidneys were quickly removed and snap-frozen in isopentane, cooled by dry ice. Ten-micrometer-thick cryostat sections were prepared for autoradiography as previously described in detail, except for the use of the additional peptidase inhibitor phosphoramidone (6, 7). Briefly, after preincubation at ZOOC for 10 min in 30 mM phosphatebuffer (pH 7.18) containing 120 mM sodium chloride, 40 pg/ml bacitracin, 2 hg/ml leupeptin, 100 pg/ml phenylmethylsulfonyl fluoride and 5 pg/ml phosphoramidone (Sigma Chemical), sections were incubated with either 3-[1251]iodo-28-tyrosyl rat atria1 natriuretic peptide 125I-aANP-( 1- 28), sp act 2,000 Ci/mmol; Amersham International] or 3-[ 1251]iodo-0-tyrosyl rat C-type natriuretic peptide 1251[TyrO] CNP- (I- 22)) sp act 1,398 Ci/mmol; Peninsula Laboratories) in fresh preincubation buffer plus 0.5% bovine serum albumin (fraction V, proteasefree; Sigma) at 20°C for 15 min. The competitive inhibition of 1251-~-ANPbinding by unlabeled peptides was examined in consecutive sections from the first group of six rats by coincubating with various concentrations (1 pM to 10 PM) of unlabeledrat a-ANP, rat CNP-( I - 22), or rat C-ANP [Peninsula Laboratories (Europe)]. Sections were also incubated with 1251-a-ANPplus 10PM CNP-(1-22) and I MM cu-ANP, or with 1251-a-ANPplus 10 PM CNP-(1-22) and 10 PM C-ANP. The competitive inhibition of 1251[TyrO] CNP(l-22) binding was similarly examined by coincubating sections from the secondgroup of six rats with 1 pM to IO PM of unlabeled rat [TyrO]CNP-(l-22), rat CNP-(l-22), rat ar-ANP, or C-ANP (Peninsula). Sections were also incubated with the radioligand plus 1 PM CNP-( l-22) and 10 PM a-ANP, or radioligand plus 1 PM CNP-( l-22) and 10 PM C-ANP. Finally, the saturation of the binding of increasing concentrations (25 pM to 1 nM) of 1251-[Tyro]CNP-(l-22) to consecutive sectionsin the absenceand presenceof 1 PM unlabeled [TyrO]CNP-( l-22) was examined in the third group of six rats. The specificity of the inhibition of the binding of both radioligandsby natriuretic peptides was assessed in the presence of the unrelated peptides angiotensin II, vasopressin (Sigma), or gastrin (Peninsula) (all 10 PM). After incubation, sectionswere rinsed, washedand dried for exposureto Hyperfilm 3H (Amersham International) for 4-21 days, and autoradiographswere developedin Kodak D-19 (Kodak), as described (6, 7). After exposure,sectionswere fixed in formaldehyde and stained with hematoxylin and eosin. Preliminary experiments showedthat the specifically reversible binding of 1251[TyrO] CNP- (I - 22) did not increasesignificantly with incubations beyond 15 min, aspreviously shownfor 1251-~-ANP(7). The stability of 1251-[Tyro]CNP-(1-22) under the conditions of incubation was also checked in preliminary experiments by gamma counting of fractions collected from high-pressureliquid chromatography (HPLC) of the incubation fluid using a Cl, column (Novapak Radialpak, Millipore) and a O-60% linear gradient of acetonitrile in water containing 0.04% trifluoroacetic acid (Fisons Scientific Equipment) run over 40 min, again as we have describedfor 1251-a-ANP(5). Regionalbinding of the radioligandswasmeasuredin femtomolesper squaremillimeter in individual glomeruli of the outer third of the renal cortex and in the medulla by comparing the regional optical densities of the renal autoradiographs with those producedby 1251standards aspreviously described(6, 7). Specific gravity and Lowry protein content values determined in blocks of renal cortex and medulla from six further male Wistar rats (250-300 g) were usedto convert values of bound ligand from femtomolesper squaremillimeter to femtomolesper milligram protein, as described(6, 7). The number of binding sitesof different affinities for a given radioligand, their apparent dissociation constants (Kd) and their maximum binding capacities (B,,,) on particular structures were derived separately in each individual using the
CNP
RECEPTORS
LIGAND iterative model-fitting computer program (28). The inhibitory constant (Ki) and B,,, for the binding of a nonhomologousunlabeledpeptide which inhibited radioligand binding were also determined separately for each individual by LIGAND. Guanosine 3’,5’-cyclic monophosphate (cGMP) assay. cGMP production wasmeasuredin glomeruli isolatedfrom finely diced cortex by sieving (27). Glomeruli werepooledfrom both kidneys of each of sevenmale Wistar rats (250-300 g). Aliquots of 200 glomeruli were preincubated for 2 min at 20°C in Hanks’ buffered salt solution [containing (in mM) 137 NaCl, 5.4 KCl, 0.4 0.34 Na2HP04, 1.26 CaC12, 4.17 NaHC03, 0.44 J&ml, KH2P04, and 0.49 MgCl,] plus 5.56 mM glucose,10 mM N-2hydroxyethylpiperazine-N’-2-ethanesulfonic acid, 0.2% bovine serumalbumin (fraction V, proteasefree), and 1 mM isobutylmethylxanthine (Sigma), centrifuged at 300 g for 3 min, and then resuspendedin the samesolution with a range of concentrations of either rat cu-ANPor CNP-(1-22) (both O-l PM) or with 1 PM C-ANP. Nine glomerular aliquots were incubated separately at each concentration of both peptidesfor 10 min at 20°C. Incubations were terminated by addition of ice-cold trichloroacetic acid to a final concentration of 6%. The aliquots were then centrifuged at 3,000g for 10 min, and the supernatants were extracted four times, each time with thrice their volume of water-saturated ether. cGMP in the supernatant was then measured for each aliquot by radioimmunoassayafter acetylation (Amersham). Statistics. Results are means * SE. The fits of single and multiple site modelsof ligand binding were comparedboth by the extra sum of squarestest and the runs test using LIGAND (28). Other comparisonswere by paired or unpaired Student’s t tests or by Fisher’s variance ratio tests as appropriate (33). RESULTS
Competitive inhibition of 1251-labeledcu-ANP binding.
Comparison of autoradiographs with their corresponding stained sections revealed specifically reversible binding of wANP to glomeruli, intrarenal arteries, outer medullary vasa recta bundles, and inner medulla (Fig. 1). The unrelated peptides angiotensin II, vasopressin, and gastrin displaced none of this binding. Analysis of the competitive inhibition of the binding of 1251-~-ANP by unlabeled cu-ANP on glomeruli, vasa recta bundles, and inner medulla was consistent with reversible binding sites for cu-ANP of uniform affinity on each structure. The Kd and B max of these sites were, respectively, 3.15 t 0.82 nM and 583 t 122 fm o l/ mg protein on glomeruli, 3.61 t 0.39 nM and 341 t 46 fmol/mg protein on vasa recta bundles, and 5.59 t 1.55 nM and 913 t 280 fmol/mg protein on inner medulla. These results are similar to those of previous investigations of the distribution and binding constants of cu-ANP receptors in rat kidney (6, 7, 24). C-ANP (10 PM), which is known to almost fully saturate ANPR-C without significantly binding to renal guanylate cyclasecoupled receptors (6, 7,25, 29), inhibited 51.3 t 10.7% of the specifically reversible binding of 1251-a-ANP to glomeruli. However, it did not significantly alter the specifically reversible radioligand binding to vasa recta bundles (inhibiting 7.0 t 7.0%) or to inner medulla (inhibiting 10.6 t 11.7%), just as in previous studies (6, 7, 25). Increasing concentrations of CNP-( 1 - 22) progressively inhibited the binding of 12~I-cr-ANP to glomeruli, and this inhibition was consistent with sites of a single affinity for CNP-( 1-22) characterized by a Ki of 10.47 t 7.59 nM and B,,, of 342 t 118 fmol/mg protein. This B max was significantly less than that of 583 t 122 l
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
RENALANPANDCNPRECEPTORS
F91
Fig. 1. Autoradiographs of binding of 100 pM 1251-labeled a-atria1 natriuretic peptide (‘251-a-ANP) (A-D) or 110 pM 12sI-labeled C-type natriuretic peptide ( 1251-[Tyro]CNP-(1-22)] (E-H) in rat kidney. Binding is shown in absence of competing unlabeled ligands (A and E) and in presence of 1 (B) or 10 PM (F) unlabeled wANP, 10 PM CNP-(l-22) (C and G), and 10 FM C-ANP (D and H). g, glomeruli; vr, vasa recta bundles; im, inner medulla.
fmol/mg protein for the glomerular < 0.01, paired t test). CNP-(1-22) less affinity for the binding sites recta bundles and inner medulla 13.3% of the reversible binding recta bundles (P < 0.01) and 59.8
binding of cr-ANP (P competed with much of 1251-a-ANP on vasa (Fig. 2). Thus 59.7 f. of 1251-a-ANP to vasa f 7.3% of that to inner
medulla (P < 0.001) remained bound even in the presence of 10 FM CNP-(1-22). The inhibition of radioligand binding produced on any structure by 10 PM CNP(l-22) plus 1 PM a-ANP was not significantly different from that produced by 1 PM cu-ANP alone. Similarly, the inhibition of radioligand binding on any structure
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
F92
RENAL
Vasa
Glomeruli
recta
ANP AND CNP RECEPTORS
Inner
bundles
medulla
Fig. 2. Competitive inhibition of specifitally reversible binding of 1251-a-ANP by increasing concentrations of CNP_ (1 - 22) in glomeruli (A), vasa recta bundles (B), and inner medulla (C) of kidneys of 6 rats.
8 6c3 z 50 za 45 3 m G 2cLLI 3 ’ = o0
lb-”
lb-9
lb-’
(j
1 o-5
16-11
CONCENTRATION
16-9
lb-7
OF
l&5
CNPo-,,,
(M)
produced by 10 PM CNP-( l-22) plus 10 PM C-ANP was not significantly different from that by 10 PM CNP(l-22) alone. Competitive inhibition binding. Autoradiographs
of
1251-[Tyro]CNP-(l -22)
revealed that 110 pM lz51[TyrO] CNP-( I- 22) bound to glomeruli and did not bind significantly to any other renal structure (Fig. 1). This binding was not altered by the unrelated peptides angiotensin II, vasopressin, and gastrin. However, it was progressively inhibited by increasing concentrations of unlabeled [ TyrO] CNP- (1 - 22)) consistent with glomerular binding sites of uniform affinity for this ligand. The Kd and B,ax values for these glomerular sites were 1.42 t 0.48 nM and 67 t 17 fmol/mg protein, respectively. a-ANP (10 PM), 10 PM CNP-(l-22), and 10 PM C-ANP all completely prevented this specifically reversible binding of 1251-[Tyro]CNP-(l-22) (Fig. 1). The Ki and B,,, values for the inhibition of the specifically reversible globy a-ANP merular binding of 1251-[Tyro]CNP-(l-22) were 1.24 t 0.18 nM and 78 t 22 fmol/mg protein, respectively, and those for the corresponding inhibition by CNP-(l-22) were 0.83 t 0.13 nM and 73 t 24 fmol/mg protein, respectively (Fig. 3). The Ki for CNP-( l-22) was significantly smaller than that for wANP (P < 0.05). There were no significant differences among the B,,, values for the binding of [TyrO]CNP-(l-22), wANP, or CNP-( l-22) as assessed by their displacement of lz51- [TyrO] CNP- (1 - 22). However, these were all significantly smaller than the B,,, of 342 t 118 fmol/mg protein for the glomerular binding of CNP-( l-22) assessed from its ability to inhibit the binding of lz51-a-ANP (all P < 0.01). Finally, there was no significant difference between the inhibition of the binding of lz51- [TyrO] CNP(l-22) produced by 10 PM CNP-( l-22) plus 1 PM cu-ANP or by 10 PM CNP-(l-22) plus 10 PM C-ANP. Saturation of ‘“51_[Tyr”]CNP-(1 -22) Ibinding. Increasproduced a ing concentrations of 1251-[Tyro]CNP-(l-22) progressive increase in the glomerular binding of the radioligand. The specifically reversible component of this
binding was consistent with saturable ligand binding sites of uniform affinity. The Kd and B,,, values for these sites were 0.41 t 0.07 nM and 29 t 4 fmol/mg protein, respectively. No specifically reversible radioligand binding was detected to structures other than glomeruli at any of the concentrations (up to 1 nM) of 1251-[Tyro]CNP(l-22) that were used. Recovery of 1251-[Tyro]CNP-(l -22). The HPLC of incubation fluids recovered after exposure to the renal sections showed that 62.1 t 8.5% of the total counts added were recovered in a chromatographic peak with the retention time of 33 min, which was identical to that of authentic lz51- [ Tyr”] CNP- (1 - 22). Two further HPLC peaks, each containing -15% of the total counts, were at retention times of 20 and 22.5 min, respectively. Production of cGMP. Basal production of cGMP, in the absence of added natriuretic peptides, was 0.22 t 0.04 pm01 . mg protein1 min -l. The effects of cr-ANP and CNP-(l-22) are shown in Fig. 4. wANP markedly increased cGMP production, an effect that tended to saturate above 10e7 M cu-ANP. For example, lo-” M cu-ANP increased cGMP production to 1.51 t 0.40 pmol. mg protein-‘. min-l, which was significantly different from the basal rate (P < 0.001). Only lop7 and lo-” M CNP(1 - 22) significantly increased cGMP production, and there was no indication that this effect was saturated by 10m6 M CNP-(l-22). For example, 10m7 M CNP-(l-22) increased cGMP production to 0.34 t 0.04 pmol mg protein-l min-‘, which was significantly higher than basal (P < 0.05). cGMP production in the presence of lo+ M C-ANP was 0.22 t 0.04 pmolmg protein-l min-l, which was not different to basal. l
l
l
DISCUSSION
The findings described here confirm previous radioligand binding and receptor cross-linking studies which show that the clearance-related receptor (ANPR-C) comprises 50-80% of the glomerular receptors for cu-ANP and is not detectable on vasa recta bundles or in the inner
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
RENAL
F93
ANP AND CNP RECEPTORS
-oCANP --+CN~,-,,, -- tyr”-CNP,,e,,,
-- 0 -O-
a
+
CW,-,a
-ANP
0
,o-~'
,o-'o
,o-9
1o-8
1o-7
,o-6
LIGAND CONCENTRATION (M) Fig. 4. Effects of wANP and CNP-( l-22) on basal (+) cGMP production by isolated glomeruli. Each point is mean of 9 determinations, each using glomeruli pooled from ‘7 rats.
in the absence of a specific, high-affinity ligand for ANPR-B, it has not hitherto been possible to document the expression of this receptor by different tissues in situ. CNP-(l-22), [Tyr*]CNP-(l-22), and 1251-[Tyro]. -2, --- 0 CNP-( l-22) all behave as high-affinity ligands that are d 4. 0I 1 strongly selective for ANPR-B and ANPR-C over 0 ANPR-A when the human forms of these receptors are CONCENTRATION OF UNLABELLED LIGAND (M) introduced by vectors into cells which do not normally Fig. 3. Competitive inhibition of specifically reversible binding of 1251express such receptors (19). Consistent with this in the [TyrO]CNP-( l-22) to renal glomeruli in 6 rats caused by increasing present experiments in rats, CNP-( 1 - 22) displaced 1251concentrations of unlabeled ligands [TyrO]CNP-(l-22), ar-ANP, and CNP-(1-22). a-ANP most effectively from renal sites known to express ANPR-C, namely glomeruli. CNP-(l-22), 10 PM, bound to all the glomerular cu-ANP receptors that bound 10 PM C-ANP, because there was no significant difference medulla of the rat (2, 6, 7, 25). The same radioligand binding and receptor cross-linking studies, coupled to between the inhibition of the specifically reversible glostudies of cGMP production (2), have suggested that the merular binding of 1251-~-ANP produced by 10 PM CNP(l-22) and that produced by 10 PM CNP-(l-22) plus 10 residue of wANP receptors on these renal structures, which do not bind C-ANP and therefore differ from PM C-ANP. Moreover, the B,,, for the binding of CNP(1 - 22) to glomeruli was -60% of that for the glomerular ANPR-C, are guanylate cyclase-coupled receptor proteins of -120-130 kDa. Molecular cloning studies have idenbinding of wANP. This was similar to the proportion of tified two such guanylate cyclase-coupled receptors in glomerular clearance receptors for cu-ANP (51.3 t 10.7%) both the human and rat genome. The first of these, assessed from the inhibition of the binding of f251-~-ANP ANPR-A, has high affinity for a-ANP when clones from caused by 10 PM C-ANP in the same rats. CNP-(l-22) binding either species are expressed (10, 11,19,31). The second of also displaced 1251-~-ANP from its high-affinity these receptors, ANPR-B, is virtually without affinity for sites in the medulla, which does not contain ANPR-C, a-ANP, regardless of its specific origin (10, 11,19,31). On but it did this with a greatly lower affinity than in glothe basis of this wide difference of affinities for a-ANP, it meruli, which do express ANPR-C. Because 10 PM CNP plus 1 PM unlabeled a-ANP displaced no more radioliis likely that high-affinity guanylate cyclase-coupled binding sites for wANP in rat kidney at least partly gand binding than 1 PM unlabeled a-ANP alone, these results suggest that CNP-( l-22) has a significantly represent ANPR-A. However, low-affinity, specifically higher affinity for ANPR-C than for the remaining reversible binding sites for wANP also occur in rat kidwANP receptors of the rat kidney, which conform not ney (7), and these could represent ANPR-B. Porcine only to the rat and human forms of ANPR-A in their high BNP-( l-26) was originally reported as a specific ligand affinity for wANP and poor affinity for C-ANP (10, 11, of ANPR-B over ANPR-A (lo), but subsequent studies have not found such a wide difference in its binding and 19, 31), but also at least to the human form of ANPR-A in its poor affinity for CNP-( 1 - 22) (19). activation of ANPR-A and ANPR-B (19, 31). Therefore, ,;-11
$10
$9
&-a
$7
10-6
lo+
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
F94
RENAL
ANP
AND
The high affinity specifically reversible binding of 1251[ TyrO] CNP- (1 - 22) further strengthened this conclusion. No specifically reversible binding of 1251-[TyrO] CNP(l-22) occurred in the renal medulla, a region without ANPR-C but rich in guanylate cyclase-coupled natriuretic peptide receptor sites, even when sections were incubated with up to 1 nM of the radioligand. Instead, the high-affinity binding of 12?I- [ TyrO] CNP- (1 - 22) was confined to glomeruli, and here it was completely inhibited by excess C-ANP and had high affinity for cu-ANP (Ki = 1.24 t 0.18 nM), confirming that it was to ANPR-C rather than ANPR-B. However, not all ANPR-C seemed to bind CNP-( 1-22) with high affinity. The specifically reversible binding of 1251-[Tyro]CNP-(l-22) to ANPR-C was of high affinity (& = 0.41 t 0.07 nM), and the affinities of CNP-(1-22) and [TyrO]CNP-(1-22) for these same binding sites were also high (affinity constants of 0.83 t 0.13 and 1.42 t 0.48 nM, respectively). Nevertheless, the estimates of B,,, for these glomerular sites were uniformly low, however they were measured. Thus the B,,, values derived from the inhibition of the binding of 110 pM 1251-[Tyro]CNP-(l-22) by CNP(l-22) and [TyrO]CNP-(1-22) were 73 t 24 and67 t 17 fmol/mg protein, respectively, and the B,,, value measured from the saturable binding of up to 1 nM 1251[TyrO]CNP-(1-22) was 29 t 4 fmol/mg protein. In contrast, the binding of excess C-ANP suggested that ANPR-C comprises more than 50% of the glomerular B 17naX for cu-ANP of 583 t 122 fmol/mg protein, confirming previous estimates of the total density of glomerular ANPR-C (6, 7). It is possible, therefore, that only a fraction of ANPR-C, as defined by binding C-ANP, have a sufficiently high affinity for analogues of CNP-( I- 22) to bind subnanomolar concentrations of 1251-[TyrO] CNP(l-22) appreciably. On the other hand, much higher concentrations (up to 10 PM) of CNP-(1-22) competed measurably at -60% of the glomerular binding sites of 1251-ar-ANP. Although, as in the medulla, some of this competition may have been with low affinity at glomerular ANPR-A, most of it was at ANPR-C, because there was no significant difference between the inhibition of the glomerular binding of 1251-a-ANP achieved by IO PM CNP-(l-22) or by 10 PM CNP-(1-22) plus 10 PM C-ANP. This implies that up to 10 PM CNP-(1-22) could also inhibit binding of 1251-~-ANP to ANPR-C that did not bind subnanomolar concentrations of 1251[TyrO]CNP-( 1 - 22) significantly. It is not certain whether, compared with other ANPR-C, ANPR-C that apparently did not bind subnanomolar 1251-[TyrO] CNP(l-22) significantly also had a lower affinity for CNP(l-22), but such heterogeneous affinities of ANPR-C for CNP-( 1 - 22) might explain the relatively low affinity and variability of the gross Ki value of 10.47 t 7.59 nM derived from the single site analysis of the inhibition of f251-~-ANP binding by CNP-(1-22). This Ki value was an order of magnitude higher than the Ki value of 0.83 t 0.13 nM for the inhibition of 1251-[Tyro]CNP-(l-22) binding by CNP-( 1 - 22). Nevertheless, the data actually did not establish binding of CNP-( 1 - 22) to glomerular cu-ANP receptors with significantly more than one affinity for CNP-(l-22), although the concentration inter-
CNP
RECEPTORS
vals of CNP-(1-22) used to inhibit 1251-a-ANP binding were not spaced to address this point in detail. Moreover, some degradation of 1251-[TyrO] CNP- (1 - 22) did occur during the present incubations despite the use of conditions of incubation and peptidase inhibitors that have previously been shown to prevent significant degradation of 1251-a-ANP by renal sections (7). Although -62% of the 1251-[Tyro]CNP-( l-22) remained intact by the end of the incubations, the B,,, for its binding would be underestimated in proportion to the reduction of its effective concentration. Consequently, quantitative comparison of results involving 1251-a-ANP and 1251-[TyrO] CNP(l-22) should be cautious, and further experiments will be necessary to settle the issue of subtypes of ANPR-C. This notwithstanding, the main conclusion of the present study is that no high-affinity binding sites for either CNP-(1-22) or 1251-[TyrO] CNP- (1 - 22) were detected in the rat kidney apart from those related to ANPR-C by their affinities for C-ANP and a-ANP, and no sites compatible with ANPR-B were found. This conclusion was strengthened by functional data. We found that CNP-( l-22) did not significantly stimulate cGMP production by isolated glomeruli unless concentrations of 10e7 M or more were used. This confirms a preliminary report (12), and it is not consistent with the presence of detectable quantities of an ANPR-B for which CNP-( l-22) is a high-affinity ligand. ANPR-C may also stimulate cGMP production in some circumstances (15). However, there was a substantial disparity between the high affinities of CNP-( l-22) and its analogues for glomerular binding sites related to ANPR-C and the high concentrations of CNP-( l-22) required to stimulate glomerular guanylate cyclase. Instead, slight stimulation of glomerular guanylate cyclase activity at the highest concentrations of CNP-( 1 - 22) is consistent with the data that CNP-( 1 - 22) is a low-affinity ligand of ANPR-A. Thus both binding and functional data are consonant with the lack of detectable quantities of ANPR-B in adult male Wistar rat kidney. Nevertheless, mRNA for ANPR-B may occur in adult rat kidney (3 1). It is known that mRNA levels do not necessarily reflect the degree of expression of the encoded protein (21). However, interpretation of the published Northern analysis showing renal mRNA for ANPR-B (31) should be cautious, because the small quantity of mRNA in the renal electrophoretic lane was adjacent to a very much larger quantity of mRNA from lung, and contamination between lanes was not excluded. In the absence of a renal guanylate cyclase-coupled receptor for CNP-( l-22), CNP-( l-22) could still have effects by binding ANPR-C coupled to production of other second messengers. For example, C-ANP and other specific ligands of ANPR-C can inhibit CAMP production (1). However, C-ANP itself seems to have no acute effects on the function of the isolated rat kidney (22) and does not modify the effects of cu-ANP on isolated rat renal tissues (2,22,29). Instead, the available evidence actually suggests that the short-term role of ANPR-C in the rat kidney is to clear bound ligands (2, 22). Consistent with this, the acute excretory effects of CNP-( l-22) are -loo-fold less potent than those of cu-ANP in the rat
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
RENAL
(37). Over longer periods, however, synthetic ligands specific for ANPR-C can also modulate the growth of cultured rat aortic smooth muscle (9). CNP-( 1 - 22) seems not to circulate (38), but it may nevertheless occur in the rat kidney since local CNP-like immunoreactivity has been found there (20). Our results suggest that CNP(I -22) and its close analogues principally bind to ANPR-C in the kidney. Therefore these findings taken together raise the important possibility that intrarenal CNP-( l-22) may interact at local ANPR-C to regulate tissue growth. This work was supported by the National Kidney Research Fund and the British Heart Foundation. Address for reprint requests: J. Brown, Physiological Laboratory, Downing St., Cambridge CB2 3EG, UK. Received 19 November 1991; accepted in final form 18 December 1991. REFERENCES 1. Anand-Srivastava, M. B., M. R. Sairam, and M. Cantin. Ring-deleted analogs of atria1 natriuretic factor inhibit adenylate cyclase/cAMP system. J. BioZ. Chem. 265: 8566-8572, 1990. 2. Brenner, B. M., B. J. Ballermann, M. E. Gunning, and M. L. Zeidel. Diverse biological actions of atria1 natriuretic peptide. Physiol. Reu. 70: 665-699, 1990. 3. Brown, J., and B. Cheung. Binding of rat brain natriuretic peptide (BNP) to atria1 natriuretic peptide (ANP) receptors in rat kidney (Abstract). J. Physiol. Lond. 446: 91P, 1992. 4. Brown, J., and A. Czarnecki. Autoradiographic localization of atria1 and brain natriuretic peptide receptors in rat brain. Am. J. Physiol. 258 (Regulatory Integrative Comp. Physiol. 27): R57-R63, 1990. 5. Brown, J., and A. Czarnecki. Binding of atria1 and brain natriuretic peptides in brains of hypertensive rats. Brain Res. 512: 132-137, 6.
7.
8.
9.
10.
1990.
Brown, J., S. P. Salas, and J. M. Polak. Renal atria1 natriuretic peptide receptor subtypes in spontaneously hypertensive rats. Am. J. Physiol. 259 (Renal Fluid Electrolyte Physiol. 28): F605-F612, 1990. Brown, J., S. P. Salas, A. Singleton, J. M. Polak, and C. T. Dollery. Autoradiographic localization of atria1 natriuretic peptide receptor subtypes in rat kidney. Am. J. Physiol. 259 (Renal Fluid Electrolyte Physiol. 28): F26-F39, 1990. Brown, J., and 2. Zuo. Receptor binding of C-type natriuretic peptide CNP-(1-22) in rat brain (Abstract). J. Physiol. Lond. 446: 44OP, 1992. Cahill, P. A., and A. Hassid. Clearance receptor-binding atria1 natriuretic peptides inhibit mitogenesis and proliferation of rat aortic smooth muscle cells. Biochem. Biophys. Res. Commun. 179: 1606-1613, 1991. Chang, M., D. G. Lowe, M. Lewis, R. Hellmiss, E. Chen, and D. V. Goeddel. Differential activation by atria1 and brain natriuretic peptides of two different receptor guanylate cyclases. Nature
Lond.
341: 68-72,
1989.
Chinkers, M., D. L. Garbers, M. S. Chang, D. G. Lowe, H. Chin, D. V. Goeddel, and S. Schulz. A membrane form of guanylate cyclase is an atria1 natriuretic peptide receptor. Nature Lond. 338: 78-83, 1989. M., M. Takehisa, Y. Minamitake, Y. Kitajima, Y. 12. Furuya, Hayashi, N. Ohnuma, T. Ishihara, N. Minamino, K. Kangawa, and H. Matsuo. Novel natriuretic peptide, CNP, potently stimulates cyclic GMP production in rat cultured vascular smooth muscle cells. Biochem. Biophys. Res. Commun. 170: 201-208, 1990. T., K. Higuchi, M. Ohashi, R. Takayanagi, H. 13. Hashiguchi, Matsuo, and H. Nawata. Effect of porcine brain natriuretic peptide (pBNP) on human adrenocortical steroidogenesis. CZin. Endocrinol. 31: 623-630, 1989. 14. Imura, H., and K. Nakao. Central nervous system actions of atria1 natriuretic peptide. In: Atria1 Natriuretic Peptides, edited by W. K. Samson and R. Quirion. Boca Raton, FL: CRC Press, 1990, p. 221-230. 11.
F95
ANP AND CNP RECEPTORS
15. Ishido, M., T. Fujita, M. Shimonaka,
T. Saheki, S. Ohuchi, T. Kume, I. Ishigaki, and S. Hirose. Inhibition of atria1 natriuretic peptide-induced cyclic GMP accumulation in bovine endothelial cells with anti-atria1 natriuretic peptide receptor antiserum. J. BioZ. Chem. 264: 641-645, 1989. 16. Itoh, H., K. Nakao, T. Yamada, G. Shirakami, K. Kangawa, N. Minamino, H. Matsuo, and H. Imura. Anti-dipsogenic action of a novel peptide, brain natriuretic peptide, in rats. Eur. J. Pharmacol. 150: 193496, 1988. 17 Kambayashi, Y., K. Nakao, H. Itoh, K. Hosoda, Y. Saito, T. Yamada, M. Mukoyama, H. Arai, G. Shirakami, S. Suga, Y. Ogawa, M. Jougasaki, N. Minamino, K. Kangawa, H. Matsuo, K. Inouye, and H. Imura. Isolation and sequence determination of rat cardiac natriuretic peptide. Biochem. Biophys. Res. Commun. 163: 233-240, 1989. 18. Kojima, M., N. Minamino, K. Kangawa, and H. Matsuo. Cloning and sequence analysis of a cDNA encoding a precursor for rat C-type natriuretic peptide (CNP). FEBS Lett. 276: 209-213, 1990. K. lg. Koller, K. J., D. G. Lowe, G. L. Bennett, N. Minamino, Kangawa, H. Matsuo, and D. V. Goeddel. Selective activation of the B natriuretic peptide receptor by C-type natriuretic peptide (CNP). Science Wash. DC 252: 120-123, 1991. Y ., K. Nakao, S. Suga, Y. Ogawa, M. 20. Komatsu, Mukoyama, H. Arai, G. Shirakami, K. Hosoda, 0. Nakagawa, N. Hama, I. Kishimoto, and H. Imura. C-type natriuretic peptide (CNP) in rats and humans. Endocrinology 129: 1104-1106, 1991. 21. Latchman, D. Gene Regulation. A Eukaryotic Perspective. London: Unwin Hyman, 1990, p. 269. 22. Maack, T., M. Suzuki, F. A. Almeida, D. Nussenzveig, R. M. Scarborough, G. A. McEnroe, and J. A. Lewicki. Physiological role of silent receptors of atria1 natriuretic factor. Science Wash. DC 238: 675-678, 1987. 23. Maeda, T., M. Niwa, K. Shigematsu, M. Kurihara, Y. Kataoka, K. Nakao, H. Imura, H. Matsuo, H. Tsuchiyama, and M. Ozaki. Specific [ 1251] brain natriuretic peptidebinding sites in rat and pig kidneys. Eur. J. Pharmacol. 176: 341-350, 1990. 24. Maekawa, K., T. Sudoh, M. Furusawa, N. Minamino, K. Kangawa, H. Ohkudo, S. Nakanishi, and H. Matsuo. Cloning and sequence analysis of cDNA encoding a precursor for porcine brain natriuretic peptide. Biochem. Biophys. Res. Commun. l
157: 410-416,
1988.
25. Martin, E. R., J. A. Lewicki, R. M. Scarborough, and B. J. Ballermann. Expression and regulation of ANP receptor subtypes in rat renal glomeruli and papillae. Am. J. Physiol. 257 (Renal Fluid Electrolyte Physiol. 26): F649-F657, 1989. 26 . Minamino, N., K. Kangawa, and H. Matsuo. N-terminally extended form of C-type natriuretic peptide (CNP-53) identified in porcine brain. Biochem. Biophys. Res. Commun. 170: 973-979, 1990. 27 . Misra, R. 0. Isolation of glomeruli from mammalian kidneys by graded sieving. Am. J. Clin. Pathol. 58: 135-139, 1972, 28 Munson, P. J., and D. Rodbard. LIGAND; a versatile com. puterized approach for characterization of ligand binding system. Anal. Biochem. 107: 220-239, 1980. D., R. M. Scarborough, J. A. Lewicki, and T. 2g* Nussenzveig, Maack. Clearance receptors of atria1 natriuretic factor (C-ANP receptors) in isolated glomeruli and mesangial cells in culture (Abstract). Kidney Int. 33: 279P, 1988. 30. Saito, Y., K. Nakao, H. Itoh, T. Yamada, M. Mukoyama, H. Arai, K. Hosoda, G. Shirakami, S. Suga, N. Minamino, K. Kangawa, H. Matsuo, and H. Imura. Brain natriuretic peptide is a novel cardiac hormone. Biochem. Biophys. Res. Commun. 158: 360-368, 1989. 31. Schulz, S., S. Singh, R. A. Bellet, G. Singh, D. J. Tubb, H. Chin, and D. L. Garbers. The primary structure of a plasma membrane guanylate cyclase demonstrates diversity within this new receptor family. CeZL58: 1155-1162, 1989. J. J., A. Arfsten, J. A. Miller, P. Lindquist, R. 32. Seilhamer, M. Scarborough, J. A. Lewicki, and J. G. Porter. Human and canine gene homologs of porcine brain natriuretic peptide. Biochem. Biophys. Res. Commun. 165: 650-658, 1989.
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
F96
RENAL
ANP
AND
33. Snedecor, G. W., and W. G. Cochran. Statistical Methods (6th ed.). Ames: Iowa State Univ. Press, 1974. 34. Song, D. L., K. P. Kohse, and F. Murad. Brain natriuretic factor. Augmentation of cellular cyclic GMP, activation of particulate guanylate cyclase and receptor binding. FEBS Lett. 232: 125-129, 1988. 35. Sudoh, T., K. Kangawa, N. Minamino, and H. Matsuo. A new natriuretic peptide in porcine brain. Nature Land. 332: 78-81, 1988. 36. Sudoh, T., N. Minamino, K. Kangawa, and H. Matsuo. Brain natriuretic peptide 32: N-terminal six amino acid extended form of brain natriuretic peptide identified in porcine brain. Biochem. Biophys. Res. Commun. 155: 726-732, 1988.
CNP
RECEPTORS
37. Sudoh, T., N. Minamino, K. Kangawa, and H. Matsuo. C-type natriuretic peptide (CNP): a new member of natriuretic peptide family identified in porcine brain. Biochem. Biophys. Res. Commun. 168: 863-870, 1990. M. Aburaya, K. Kangawa, S. Mat38. Ueda, S., N. Minamino, sukura, and H. Matsuo. Distribution, and characterization of immunoreactive porcine C-type natriuretic peptide. Biochem. Biophys. Res. Commun. 175: 759-767, 1991. 39. Yamada, T., K. Nakao, H. Itoh, G. Shirakami, K. Kangawa, N. Minamino, H. Matsuo, and H. Imura. Intracerebroventricular injection of brain natriuretic peptide inhibits vasopressin secretion in conscious rats. Neurosci. Lett. 95: 223-228, 1988.
Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 10, 2019.
