Gene Expression of Natriuretic Peptide Receptors in Myocardial Cells
Abstract Atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP) are cardiac hormones that serve to unload the heart through their effects on the kidney and vasculature. Whether the heart itself represents a site of action for these peptides is currently the subject of debate. Although functional studies indicate that ANP has some effects on isolated myocytes, several studies have been unable to detect binding of the hormone to these cells. The present study demonstrates that the genes for all three natriuretic peptide receptor (NPR) subtypes, NPR-A, NPR-B, and NPR-C, are expressed in the rat heart. For microlocalization of the receptor mRNAs in myocytes and nonmyocytic cells, a combination of cell isolation and reverse transcription–polymerase chain reaction (RT-PCR) was used. Cardiac myocytes were isolated by enzymatic dissociation of rat ventricular tissue, purified by density gradient centrifugation, and collected as single cells under microscopic control. Analysis by RT-PCR revealed the presence of transcripts for NPR-A as well as NPR-B and NPR-C. However, cGMP generation in purified myocytes was stimulated only by ANP and BNP, which specifically bind to NPR-A, whereas C-type natriuretic peptide (CNP, an NPR-B agonist) was ineffective. Therefore, rat ventricular myocytes appear to produce predominantly NPR-A. The expression of NPR-B may be low or even absent. The mRNAs for all three NPRs were also found in cultures of fibroblasts from the rat heart. In contrast to the myocytes, large increases in cGMP were observed in response not only to ANP but also to CNP. Cardiac fibroblasts may therefore synthesize functionally relevant amounts of both NPR-A and NPR-B. The expression of NPR-C in these cells is suggested by binding studies demonstrating that >70% of the maximum binding of [125I]ANP can be displaced by the NPR-C agonist C-ANP 4-23. These observations may help to resolve whether the heart muscle cells themselves are targets for natriuretic peptides. The finding that in addition to myocytes, cardiac fibroblasts are capable of expressing the NPR genes raises the interesting possibility that natriuretic peptides are involved in the structural remodeling of the heart.
- natriuretic peptide receptors
- cardiac fibroblasts C-receptor DNA
- heart muscle cells
- polymerase chain reaction
The natriuretic peptides are a family of vasoactive substances that are produced in a number of organs, most notably the heart.1 2 3 Three members of the family are presently known. They are designated ANP, BNP, and CNP.1 4 5 An increase in intravascular volume causes the release of ANP from atrial myocytes. Once in the circulation, ANP induces natriuresis, diuresis, and vasorelaxation.1 2 3 BNP is primarily a cardiac hormone secreted from the ventricles. Similar to ANP, it has natriuretic, diuretic, and hypotensive properties.4 CNP is synthesized in vascular endothelial cells and appears to play a role in the local regulation of vascular tone.6 The three natriuretic peptides are recognized by a family of receptors. Two of these receptors have guanylyl cyclase activity and are called NPR-A and NPR-B.7 8 9 NPR-A interacts with both ANP and BNP, whereas NPR-B is selectively stimulated by CNP.10 11 A third NPR, which does not possess guanylyl cyclase activity but serves to remove ANP from the plasma, has been designated a clearance receptor, or NPR-C.9 12 13 14
Apart from acting as hormones, ANP and BNP may have local functions at the sites of their synthesis. This is suggested by the observation that in addition to the heart, these peptides are produced by many other tissues but in amounts far too low for inducing endocrine effects.15 The idea of a paracrine or autocrine action of ANP or BNP in those organs is supported by the presence of NPRs near or even on natriuretic peptide–producing cells.16
There is some controversy about whether the heart itself represents a site of action for natriuretic peptides. Early experiments in isolated heart preparations failed to demonstrate direct effects of ANP on cardiac performance.17 18 This would be in accordance with autoradiographic data suggesting that binding of natriuretic peptides in the heart is confined to the endocardium and absent in cardiac myocytes.19 20 21 22 Evidence supporting the view that NPRs also exist on heart muscle cells comes from experiments in isolated myocytes, the physiological functions of which were characteristically altered by the administration of ANP.23 24
In the present study, molecular techniques have been used to both localize and characterize NPRs in the heart. Northern blot analysis was used for specific detection and quantification of receptor mRNAs in the whole heart. For their microlocalization in myocytes and nonmyocytic cells, a combination of RT and PCR was used. For analysis in cardiac muscle cells, myocytes were isolated by enzymatic dissociation of ventricular tissue, purified by density gradient centrifugation, and collected as single cells under microscopic control. The results of these experiments indicate that the genes for NPR-A, NPR-B, and NPR-C are expressed in both myocytes and nonmyocytic cells of the rat heart. This would be consistent with the view that the cardiac natriuretic peptides act not only as hormones but also have local paracrine and autocrine functions within the heart itself.
Materials and Methods
Male Sprague-Dawley rats (Ivanovas, Kissleg, Germany) weighing 180 to 200 g were used throughout for studies in rat tissue. The animals were anesthetized with ketamine and killed by decapitation. The lungs, kidneys, adrenal gland, liver, heart, and brain were removed for RNA extraction. The hearts were divided into atria and ventricles.
Bovine hearts were obtained from a local slaughterhouse. Right and left atria, pieces of right and left ventricles, valves, papillary muscles, Purkinje fibers, and coronary vessels were dissected and immediately frozen in liquid nitrogen.
Preparation of Heart Muscle Cells
Rat heart muscle cells were prepared by following a procedure reported by Piper et al.25 In brief, animals were anesthetized with ketamine, and the hearts were quickly removed and mounted on a Langendorff perfusion system. Perfusion was started with buffer 1 alone (mmol/L: KCl 2.5, KH2 PO4 4.7, MgSO4 1.2, NaCl 110, NaHCO3 25, and glucose 11), followed by buffer 1 containing 42 μmol/L CaCl2 and 30 mg/mL collagenase A (Boehringer). After 30 to 40 minutes of perfusion, the ventricles were separated from the atria, minced, and incubated in buffer 1 containing 1.5% bovine serum albumin. The tissue was dissociated by gentle trituration using a siliconized pipette, and CaCl2 was slowly added to the resulting cell suspension to gradually increase the calcium content up to a final concentration of 1 mmol/L.
Isolated cells were centrifuged at 60g for 2 minutes. The resulting supernatant was used for the preparation of nonmyocytic cells (see “Cell Culture”). The pellet consisting mainly of heart muscle cells was resuspended in a small volume of buffer 1 containing 1 mmol/L CaCl2 and loaded onto a Percoll gradient (40% [vol/vol] Percoll [Pharmacia] in buffer 1 containing 1 mmol/L CaCl2; centrifugation, 18 000g for 40 minutes) to separate cardiac myocytes from other cells. After centrifugation at 650g for 4 minutes, the fraction of intact myocytes, visible as a distinct band ≈1 cm above the bottom of the tube, was carefully removed.
The myocytes were pelleted and washed several times with medium 199. For isolation, they were suspended in the same medium at a concentration of ≈10 to 20 cells per milliliter and plated on culture dishes. Single myocytes were then collected under visual control by using an inverted microscope and a glass pipette (inner diameter of the tip, ≈50 to 100 μm) (Fig 4⇓). Only myocytes free of attached cellular debris were selected. Care was taken to keep the volume aspirated with each cell as low as possible to avoid contamination with material from outside the field of visual control.
The supernatant from the 60g centrifugation was centrifuged at 350g for 10 minutes to pellet the nonmyocytic cells. After resuspension, cells were grown in DMEM containing 10% fetal bovine serum. Characterization was carried out after four or five passages by immunofluorescence staining using anti-desmin (muscle cells), anti-vimentin (fibroblasts), and anti-human factor VIII (endothelial cells) antibodies.
RNA Extraction and Northern Blot Analysis
RNA was extracted from cells or tissue using guanidine thiocyanate–acid phenol.26 Poly(A)+-containing RNA was enriched via two cycles of oligo(dT)-cellulose column chromatography (type 7, Pharmacia).
RNA was denatured at 55°C for 30 minutes in glyoxal/dimethyl sulfoxide and loaded in glycerol/sodium phosphate buffer to a 1.2% agarose gel. After electrophoresis, RNA was transferred onto nylon membranes (Nytran N, Schleicher & Schüll).
Digoxigenin-labeled cRNA probes were used for hybridization (Boehringer). Labeling was performed according to the manufacturer’s protocol by in vitro transcription of the respective cDNAs [cloned in pGEM4, pGEM7 f(+), or pSP65] using digoxigenin-UTP and SP6 or T7 RNA-polymerase. The membranes were prehybridized for 2 hours at 68°C in hybridization solution consisting of 50% formamide, 5% SSC, 2% blocking agent (Boehringer), 0.1% N-laurylsarcosine, and 0.02% SDS. Hybridization was carried out overnight at 68°C by using 100 μg digoxigenin-labeled RNA probe per milliliter hybridization solution (see above). Membranes were washed with 2× SSC/0.1% SDS for 10 minutes at room temperature and with 0.1× SSC/0.1% SDS for 30 minutes at 68°C.
To detect the digoxigenin-labeled probes, after it was washed, the membrane was incubated for 30 minutes at room temperature with a 2% solution of blocking agent (Boehringer) in 100 mmol/L Tris-HCl (pH 7.4) and 150 mmol/L NaCl and then for 30 minutes at room temperature in the same solution containing 1:10 000 dilution of a polyclonal anti-digoxigenin sheep antibody Fab fragment conjugated to alkaline phosphatase (Boehringer). The membrane was washed twice with 100 mmol/L Tris-HCl (pH 7.4) and 150 mmol/L NaCl and then incubated for 10 minutes in 1× PBS containing 0.1% Tween 20 and 0.2% I-block reagent. After the membrane was washed with 100 mmol/L Tris-HCl (pH 7.4) and 150 mmol/L NaCl, alkaline phosphatase activity was determined by addition of the substrate CSPD (50 μL [in 5.0 mL of 0.1 mol/L diethanolamine] and 1 mmol/L MgCl2, Serva) and exposure of the membrane to an x-ray film.
The film was developed after 5 to 120 minutes of exposure, depending on signal intensity. The relative quantities of mRNAs were determined spectrophotometrically.
Probes for Hybridization
Specific cDNA probes for rat NPR-A and NPR-B were prepared by cDNA synthesis and PCR amplification by using RNA from rat heart tissue. For details see next section.
The PCR products obtained were size-fractionated by electrophoresis through a 2% agarose gel. The bands corresponding to the length predicted from the respective receptor DNA sequence (NPR-A, 780 bp; NPR-B, 716 bp) were excised, and the DNA was extracted by using glassmilk (Geneclean, BIO 101 Inc). After blunt-ending with T4 DNA polymerase and the addition of EcoRI linkers, the receptor DNAs were ligated into the EcoRI site of pGEM7 PZ(+) for sequencing and generation of cRNA probes (see above).
The complete cDNA sequence for the rat NPR-C is currently not known. To obtain a specific probe for this receptor, a cDNA library from rat brain (Clonetech) was screened by using an 880-bp DNA fragment carrying the sequence from 1167 to 2047 of bovine NPR-C cDNA.12 The fragment was amplified by RT-PCR using mRNA extracted from bovine kidney. The primers used for PCR were as follows: forward, 5′-ACC AAA GAC TTG GAT CTG GAG GAC; reverse, 5′-CCG TAA CTC CCG ATG TTT TCC. Of several original hybridizing plaques, one was completely characterized. After subcloning into pGEM7 fZ(+), DNA sequencing was performed by the dideoxy chain-termination method.27 The clone contained a 1367-bp fragment, which was found to encode the amino-terminal part of NPR-C. The nucleotide sequence is shown in Fig 1⇓. A translation initiation codon at position 38 defines the beginning of an open reading frame. The nucleotide sequence shows >90% identity with the human and bovine NPR-C cDNA.12
A 500-bp fragment of rat pre-pro ANP cDNA (kindly provided by C. Seidman, Harvard Medical School, Boston, Mass) cloned into p SP65 was used for ANP mRNA detection.28 A 560-bp fragment of human β-actin (kindly provided by M. Moos, Department of Neurobiology, University of Heidelberg [Germany]) cloned into pGEM4 was used for hybridization of β-actin.29
Analysis of NPR Expression by RT-PCR
For RT of extracted mRNA, 2 μg of total RNA was denatured at 65°C for 5 minutes. After they were cooled, the following components were added: 5 μL of 5× RT buffer, 8 μL of 2.5 mmol/L deoxynucleotide mixture, 2 μL oligo-dT primer, 0.5 μL of 0.1 mol/L DTT, 1 μL RNase inhibitor, and 1 μL Moloney murine leukemia virus reverse transcriptase (BRL). After 10 minutes of incubation at room temperature and 60 minutes at 42°C, reverse transcriptase was inactivated by heating the mixture to 95°C for 5 minutes.
The PCR was performed as follows: to 20 μL of the RT reaction were added 4 μL of 20 μmol/L forward primer, 4 μL of 20 μmol/L reverse primer, 10 μL of 10× amplification buffer, 6 μL H2O, and 1 μL Taq polymerase (Amersham). One hundred microliters of mineral oil was overlaid to prevent evaporation. The thermal cycler program consisted of an initial incubation at 94°C for 3 minutes, followed by 40 cycles (93°C, 60 seconds; 52°C, 120 seconds; and 72°C, 180 seconds) and a final extension at 72°C for 10 minutes.
The receptor primers for rat NPR-A and NPR-B were designed from published sequences7 8 : for NPR-A: forward, 5′-GAC TTG CAG CCC AGC AGC CTG; reverse, 5′-CAG GTG GCT CTG CAG ATC CAT (predicted length of the PCR product, 780 bp); for NPR-B: forward, 5′-TCA AAC ACA TGA GAG ATG TTC; reverse, 5′-TAT TGG CAT ACT GTT CCA TGC (predicted length of the PCR product, 716 bp).
The primers for rat NPR-C were designed from the nucleotide sequence of the rat NPR-C cDNA clone described above. Primers are as follows: forward, 5′-CGA CCG GGA GAG AGA GGC; reverse, 5′-CAG AAC TTT TCA CCT CCA TGG (predicted length of the PCR product, 903 bp).
Negative controls in the PCR included omission of the reverse transcriptase reaction and amplification in the absence of template, Taq polymerase, or one primer.
Amplified DNA was size-fractionated by electrophoresis on agarose gels containing ethidium bromide. Restriction enzyme digestion of the PCR products was used to confirm their identity with the respective receptor cDNAs. The predicted lengths of the restriction fragments resulting from Apa I digestion of the NPR-B product were 650 and 66 bp. The expected lengths of the restriction fragments resulting from Xho I digestion of the NPR-C product were 528 and 375 bp. The predicted sequence of the NPR-A product did not show a restriction site suited for digestion. Restriction fragment analysis was therefore not carried out for the NPR-A RT-PCR product.
Receptor Binding Assay
Confluent fibroblasts from 48-well plates were washed with PBS solution containing 1% bovine serum albumin and incubated at 4°C for 4 hours with [125I]ANP 99-126 (100 000 cpm per well) and various concentrations of unlabeled ANP 99-126 or the NPR-C–specific analogue C-ANP 4-23 (Peninsula) in 0.1 mL DMEM containing 1% bovine serum albumin. At the end of the incubation, unbound radioactivity was aspirated, and the cells were washed three times with 0.2 mL cold binding buffer. Washed cells were dissolved in 0.2 mL of 0.25 mol/L NaOH containing 0.5% SDS for 30 minutes. Cell lysates were transferred to polystyrene tubes and counted in a Beckmann gamma counter. Binding data were evaluated with computer-assisted nonlinear regression analysis (graphpad inplot, Graph PAD Software).
cGMP and AMP Experiments
Ventricular fibroblasts were grown to confluence in 24-well plates. The cells were washed twice with DMEM and preincubated for 15 minutes at 37°C in 450 μL of DMEM containing 10 mmol/L HEPES and 10−4 mol/L isobutylmethylxanthine. Natriuretic peptides were added in 50 μL to obtain final concentrations ranging from 10−11 to 10−6 mol/L. After incubation for 5 minutes, the medium was aspirated and 500 μL ice-cold 65% (vol/vol) ethanol was added to the cells. The supernatant was drawn off into test tubes. The remaining precipitates were washed with 65% ethanol, and the washings were added to the appropriate tubes. The extracts were centrifuged at 2000g for 15 minutes, transferred to fresh tubes, and dried in a vacuum centrifuge.
Enzymatically dissociated ventricular cells were loaded onto a Percoll gradient to separate cardiac myocytes from other cells as specified above. Purified myocytes (contamination with nonmyocytic cells, <5%) were suspended in medium 199 and diluted to a concentration of 5×105 cells per milliliter. Aliquots (450 μL) were made, and 50 μL isobutylmethylxanthine at a final concentration of 10−4 mol/L was added. Stimulation with ANP or CNP was started 15 minutes later. After incubation for 5 minutes, the cells were pelleted by centrifugation (500g, 5 minutes), and the supernatant was discarded. cGMP was extracted by the addition of 65% ethanol, as described above.
The intracellular levels of cGMP or cAMP were measured by radioimmunoassay after acetylation (Amersham).
Fig 2⇓ shows the tissue distribution of NPR-A, NPR-B, and NPR-C, as determined by Northern blot analysis using poly(A) RNA from various rat tissues. The NPR-A and the NPR-B cRNA probes detect an ≈3900-base mRNA in the heart atria and ventricles, lungs, adrenal gland, and kidney. No hybridization signal is apparent in the liver. The brain shows a distinct band for NPR-B but not for NPR-A. When related to the respective levels of β-actin mRNA, which in each case was hybridized on the same filters, the NPR-A mRNA is found to be most abundant in the adrenal gland; the NPR-B mRNA, in the heart atria.
At least four discrete mRNA species hybridized to the rat NPR-C cRNA probe (Fig 2⇑). Such size heterogeneity has been observed in previous studies in which a bovine cDNA probe was used for the detection of NPR-C expression in rat and bovine tissue.12 Similar to these experiments, our experimental results indicate a major band at ≈8 kb, accompanied by three additional bands at ≈3, 4, and 5 kb. Distinct hybridization signals for NPR-C are seen in rat heart atria, ventricles, lung, adrenal gland, and kidney (Fig 2⇑).
The distribution of NPR-C mRNA in the heart was determined in greater detail using poly(A) RNA extracted from various parts of the bovine heart and a cRNA probe, homologous to the bovine NPR-C DNA sequence.12 Fig 3⇓ demonstrates that NPR-C gene expression takes place throughout the heart. It includes right and left atria, right and left ventricle, and papillary muscle and Purkinje fibers, a well-known site of ANP production.30 The weak positive hybridization signal in the coronary artery may arise from NPR-C expression by vascular endothelial and smooth muscle cells. The NPR-A and NPR-B mRNA levels were not measured in this experiment, because the bovine cDNA sequences for these receptors were unavailable.
To determine the site of NPR expression in the heart at a cellular level, rat heart ventricles were dissociated enzymatically, and cardiac myocytes were separated from other cell types by density gradient centrifugation. A virtually pure myocyte fraction was obtained, from which single myocytes were collected under microscopic control (Fig 4⇓). A total of ≈500 myocytes were isolated for DNA extraction and subsequent RT-PCR. To minimize possible contamination with genomic DNA-derived products, the PCR primers were chosen such that the amplified cDNAs should have encompassed several putative exons of the so-far-uncharacterized human genes.31 32 Additional controls included DNase treatment of the RNA extracts before RT-PCR. The possibility that artifactual products have been amplified was further controlled by incomplete RT-PCR reactions, including omission of either reverse transcriptase, Taq DNA polymerase, or one of the primers (not shown). Restriction enzyme analysis was used to confirm that the PCR products identified corresponded to the cDNAs of the respective receptors.
Amplification of myocyte ssDNA was observed with primers for NPR-A, NPR-B, and NPR-C (Fig 5⇓). A single DNA species was detected in each case, which corresponded to the predicted size of the respective receptor PCR product (NPR-A, 780 bp; NPR-B, 729 bp; NPR-C, 903 bp). The expected restriction fragments, as predicted from the nucleotide sequences encoding the receptors, were obtained by digestion of the NPR-B and NPR-C PCR products with Apa I and Xho I, respectively (Fig 5⇓).
To examine whether NPRs are also expressed by the connective tissue of the heart, nonmyocytic cells were enriched by differential centrifugation of the original cell suspension, obtained by enzymatic dissociation of the ventricles, and were grown in culture. After 5 to 10 passages, the cells uniformly presented the morphological appearance of fibroblasts. On immunohistochemical examination using antibodies specific for desmin (positive in muscle cells), vimentin (positive in fibroblasts), or factor VIII (positive in endothelial cells), the cultures proved to be virtually free of smooth muscle and endothelial cells (not shown).
Size fractionation and restriction enzyme analysis of the RT-PCR products generated from these cultures suggest that cardiac fibroblasts, similar to the myocytes, express all three types of NPRs (Fig 5⇑). Amplification of mRNA from interspersed nonfibroblastic cells is very unlikely to account for these RT-PCR data, because the presence of the NPR transcripts in these cultures could be confirmed by the relatively insensitive Northern blot technique (Fig 6⇓).
The presence of NPRs in the cultured fibroblastic cells was also detected in binding studies with [125I]ANP 99-126 (Fig 6⇑). Competitive displacement was obtained with unlabeled ANP 99-126, the natural ligand for NPR-A, and the truncated and internally deleted ANP analogue C-ANP 4-23, which specifically binds to NPR-C.9 ANP 99-126 displaced 100% of [125I]ANP binding to the fibroblasts with high affinity (Kd, 6×10−10 mol/L). C-ANP 4-23 inhibited 74% of maximum binding with a slightly lower affinity than ANP (Kd, 10−9 mol/L). Those binding studies could not be performed in myocytes because not enough cells could be obtained by the collection procedure.
Since the presence of mRNA transcripts alone does not prove that the encoded proteins are expressed at functionally relevant levels, the cGMP response to the natural ligands for NPR-A and NPR-B, namely ANP 99-126 and CNP 1-22, respectively, was measured in myocytes and nonmyocytic cells.
The effects of ANP and CNP on cGMP accumulation in cultured ventricular fibroblasts are shown in Fig 7⇓. Both peptides markedly increased cGMP production in a dose-dependent manner, CNP being only slightly less potent than ANP. Cardiac fibroblasts therefore appear to synthesize both NPR-A and NPR-B, as already suggested by the RT-PCR and Northern blot data.
A different pattern of response was found in myocytes. The addition of ANP 99-126 or BNP 1-32, another putative NPR-A agonist, to purified ventricular myocytes resulted in a marked accumulation of cGMP, similar to that obtained with the fibroblast cultures. In contrast, CNP 1-22 failed to exert any effect except at the highest concentration (10−7 mol/L), when a slight increase in cGMP was observed (Fig 7⇑, bottom). This inability of CNP to alter intracellular cGMP levels in myocytes was confirmed in three separate experiments. The discrepancy between our functional studies and the RT-PCR data indicates that ventricular myocytes either completely lack the ability to produce NPR-B despite the presence of the respective mRNA or that there is only a small subpopulation of myocytes capable of synthesizing this receptor.
Current evidence supports an inhibitory action of ANP on adenylate cyclase in various tissues, including the heart.33 The inhibitory effect is also seen during blockade of the cGMP-inducible phosphodiesterase and is therefore not due to degradation of cAMP by this enzyme. The receptor involved is suspected to be NPR-C, which appears to be coupled in an inhibitory manner to adenylate cyclase.34 Therefore, we measured intracellular cAMP levels in isolated myocytes stimulated with various concentrations of ANP 99-126 in order to determine whether these cells produce the NPR-C protein. The peptide induced a moderate decrease in cAMP levels, which was significantly different from the control level at a concentration of 10−7 mol/L ANP (Fig 8⇓).
Our Northern (mRNA) blotting experiments indicate that the genes for all three members of the NPR family, NPR-A, NPR-B, and NPR-C, are expressed in the rat heart. This confirms in part previous studies of the expression of NPRs in the heart.
Screening human tissues for the expression of the human NPR-C by Northern blot analysis, Porter et al13 found strong hybridization signals in mRNA extracts from placenta, kidney, and fetal heart. Their report does not indicate whether adult hearts were also included in this study. In the original publication on the isolation of cDNA clones encoding NPR-A and NPR-B, Schulz et al8 were unable to detect the respective messages in the rat heart. The most likely explanation may be the low abundance of NPR mRNAs in cardiac tissue, which makes their determination in total RNA difficult. It has to be stressed that under the experimental conditions of the present study, the presence of NPR-A and NPR-B mRNA in single rat hearts can be reliably demonstrated only when poly(A)+ RNA is used. The distribution of NPR expression in the monkey was recently determined by in situ hybridization.35 In the heart, positive hybridization was found for NPR-A and NPR-C but, in contrast to the present study, not for NPR-B. It is currently unclear whether this discrepancy is due to species-specific differences between rat and primate or whether it reflects methodological problems such as lack of sensitivity.
The possibility of nonspecific hybridization of our NPR-B probe to NPR-A mRNA appears very unlikely in view of only 70% similarity between the corresponding nucleotide sequences and the high stringency of our hybridization conditions. The specificities of our NPR-A and NPR-B probes were tested in addition by using the SP-6 transcribed receptor mRNAs as controls. No cross hybridizations were observed between NPR-A and NPR-B in these experiments.
Our observation that all three NPRs are expressed in the rat heart is in complete agreement with a recent report that the transcripts for NPR-A, NPR-B, and NPR-C exist in RNA extracted from rat heart atria and ventricles.36 The powerful technique of RT-PCR was used in the present study for mRNA detection to overcome the sensitivity problems previously encountered by other groups.
Despite their identification and detailed characterization in a variety of organs, including the heart, there is still some controversy about whether receptors for natriuretic peptides exist on heart muscle cells themselves.
Several in vitro autoradiography studies indicate that binding of ANP and BNP in the hearts of rats and humans is mainly confined to the endothelial cells of the endocardium and the epicardium but is absent in the heart muscle.19 20 21 22 This is, in part, supported by an in situ hybridization study of the monkey heart, where the mRNAs for NPR-A and NPR-C were found to be expressed in the endocardium.35 The only site where, according to these experiments, muscle cells also express an NPR, namely NPR-C, is the right atria.
The complete lack of ANP receptor expression in myocytes is suggested by another report based on receptor autoradiography experiments in dissociated ventricular and atrial cells in culture.22 The authors of this study were unable to detect binding of ANP to myocytes, even to those showing ANP immunoreactivity. Significant binding was seen, however, on the fibroblastic cells. These observations do not necessarily contradict studies reporting that ANP binds to myocytes or sarcolemmal membranes and stimulates cGMP production.33 37 38 39 Because such studies have been performed in mixed cardiac cell preparations, the possibility still remains that the ANP binding sites identified were the result of contaminations with nonmyocytic cells.
Indirect evidence in favor of the expression of NPR on heart muscle cells comes from functional studies in single cells identified unmistakably as myocytes on the basis of their morphological appearance. Thus, by means of optical systems that allow the monitoring of contractions in a single cell, ANP has been demonstrated to reduce the contraction amplitude of electrically stimulated ventricular myocytes.23 Isolated myocytes have also been shown to decrease their cytosolic free calcium concentration in response to this hormone.24 In voltage-clamp studies, inhibition of voltage-gated calcium channels and stimulation of plasma membrane Ca-ATPase activity have been identified as the possible mechanisms responsible for these effects.
To assess the localization of NPR transcripts on a cellular level, we have used RT-PCR, the most sensitive method currently known for specific detection of mRNAs in small tissue samples or isolated cells. With respect to the extremely high sensitivity of this technique, it was necessary to develop a procedure that permits the isolation of a distinct cardiac cell type without the slightest contamination of other cells. The technique described here for the isolation of cardiac myocytes makes use of the extraordinary size of these cells, which enables their identification and collection under the microscope. The different nonmyocytic cell types could not be separated from each other in the same way because of their uniform appearance in suspension. Therefore, they were grown as mixed cell cultures and analyzed for their content of receptor mRNA after several passages, when, according to the morphological appearance, the monolayers were almost exclusively constituted of fibroblastic cells.
The RT-PCR experiments indicate that the mRNAs for NPR-A, NPR-B, and NPR-C are present in cardiac myocytes. However, only NPR-A (and possibly NPR-C) may be synthesized in functionally relevant amounts, since cGMP accumulated only in response to ANP and BNP but not CNP. It remains to be determined whether this lack in cGMP response to CNP is a general feature of ventricular myocytes or whether there exists a certain number of NPR-B–producing myocytes that are small enough to escape detection in functional studies yet sufficiently high for identification by RT-PCR.
The RT-PCR profile obtained for the nonmyocytic cells indicates that these cells as a group also express all three types of NPRs. However, because of the heterogeneity of the nonmyocytic fraction, it is not possible to define the cell type(s) producing the receptors. When the cultures of nonmyocytic cells, which consisted of >95% fibroblasts, were screened by Northern blot analysis, strong hybridization signals for all transcripts of NPRs were found. It appears very unlikely that this was the result of contaminations with endothelial or smooth muscle cells, because the trace amounts of RNA from these cells are not expected to reach the level of detection by Northern blotting. Further evidence for the presence of NPRs on fibroblastic cells comes from our binding studies, in which binding of [125I]ANP was potently inhibited by ANP 99-126 and the NPR-C agonist C-ANP 4-23. The latter displaced >79% of maximum [125I]ANP binding, indicating that more than one half of the ANP binding sites are of the NPR-C subtype. The observation that both ANP and CNP increased intracellular cGMP levels provides additional evidence for the generation of both NPR-A and NPR-B in ventricular fibroblasts. In this respect, it is of interest to mention that high-affinity binding sites for ANP and stimulation of cGMP production by this hormone have previously been described for fibroblast cultures from rat lung.40
The presence of NPR-A and NPR-B mRNAs in heart muscle cells and fibroblasts suggests that ANP, BNP, and CNP, the natural ligand for NPR-B, play important roles in the regulation of cardiac function. With regard to the experiments in isolated myocytes, it is intriguing to speculate that the hormones ANP and BNP serve the acute adaptation to an expanded intravascular fluid volume, not only by their peripheral effects, eg, induction of fluid extravasation and diuresis, but also by their direct actions on the heart. Produced and released in response to volume-induced increases in wall stress, they may act to improve cardiac compliance through their cardiorelaxant properties and, thereby, prevent too large elevations of diastolic pressure. Long-term regulation of heart muscle elasticity through effects on the proliferation and synthetic activity of cardiac fibroblasts is another possible function of cardiac peptides, which needs to be substantiated by future experiments.
The functional significance of NPR-C in the heart is unclear. In view of its proposed clearance function, this receptor may be assumed to protect heart muscle cells by buffering excessive fluctuations of natriuretic peptide levels. On the other hand, downregulation of NPR-C is expected to make more natriuretic peptide available to the guanylyl cyclase–linked receptors, thereby enhancing its local action in the heart. Such a situation may occur in the hypertrophied heart, as suggested by a recent study in which a dramatic decrease in mRNA levels for NPR-C was observed during the development of cardiac hypertrophy.41 As far as the suspected negative coupling of NPR-C to adenylate cyclase is concerned, a reduction in the number of these binding sites may have implications on the cardiac response to substances that act through the mediation of this signal system.
In conclusion, the present study demonstrates that RT-PCR combined with single-cell isolation is a powerful tool for localizing receptor gene expression in distinct cells isolated from complex tissues such as the heart. The method is particularly useful for the identification of structurally closely related receptor subtypes, which are difficult to discriminate by conventional ligand-binding and autoradiographic techniques.
The observation that cardiac myocytes contain transcripts for NPRs may help to resolve the long-standing controversy about whether the heart muscle cells themselves are targets for natriuretic peptides. The finding that cardiac fibroblasts also express these receptor genes points to the interesting possibility that cardiac peptides have a role in the structural remodeling of the myocardium in response to an increased load. Studies in isolated cell systems as well as in vivo experiments using specific receptor antagonists are to be performed to test this hypothesis.
Selected Abbreviations and Acronyms
|ANP||=||atrial natriuretic peptide|
|BNP||=||brain natriuretic peptide|
|CNP||=||C-type natriuretic peptide|
|NPR||=||natriuretic peptide receptor|
|PCR||=||polymerase chain reaction|
|SSC||=||standard saline citrate|
This study was supported by a research grant from the Deutsche Forschungsgemeinschaft (La 501/4-1). The authors thank H.M. Piper, Department of Physiology, Heinrich-Heine-University, Düsseldorf, for his advice in the preparation of heart muscle cells and H.J. Grone, Department of Pathology, Philipps-University, Marburg, for providing the desmin and vimentin antibodies. The authors also greatly appreciate the expert technical assistance of S. Fischer and thank H. Könnemann for preparing the manuscript.
- Received February 9, 1995.
- Accepted July 7, 1995.
- © 1995 American Heart Association, Inc.
Brenner BM, Ballermann, BJ, Gunning ME, Zeidel ML. Diverse biological actions of atrial natriuretic peptide. Physiol Rev. 1990;70:665-699.
Suga S, Nakao K, Itoh H, Kamatsu Y, Ogawa Y, Hama N, Imura H. Endothelial production of C-type natriuretic peptide and its marked augmentation by transforming growth factor-beta. J Clin Invest. 1992;90:1145-1149.
Koller KJ, Lowe DG, Bennett GL, Minamino N, Kangawa K, Matsuo H, Goeddel DV. Selective activation of the B natriuretic peptide receptor by C-type natriuretic peptide (CNP). Science. 1991;252:120-123.
Fuller F, Porter JG, Arfsten AE, Miller J, Schilling JW, Scarborough RM, Lewicki J, Schenk DB. Atrial natriuretic peptide clearance receptor. J Biol Chem. 1988;263:9395-9401.
Lowe DG, Camerato TR, Goeddel DV. cDNA sequence of human atrial natriuretic peptide clearance receptor. Nucleic Acids Res. 1990;18:3412.
Gardner DG, Deschepper CF, Ganong WF, Hane S, Fiddes J, Baxter JD, Lewicki J. Extra-atrial expression of the gene for atrial natriuretic factor. Proc Natl Acad Sci U S A. 1986;83:6697-6701.
Mantyh CR, Kruger L, Brecha NC, Mantyk PW. Localization of specific binding sites for atrial natriuretic factor in peripheral tissues of the guinea pig, rat and human. Hypertension. 1986;8:712-721.
Neyses L, Vetter H. Action of atrial natriuretic peptide and angiotensin II on the myocardium: studies in isolated rat ventricular cardiomyocytes. Biochem Biophys Res Commun. 1989;163;1435-1443.
Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain terminating inhibitors. Proc Natl Acad Sci U S A. 1977;74:5463-5467.
Seidman C, Duby AD, Choi E, Graham RM, Haber E, Homcy C, Smith JA, Seidman JG. The structure of rat preproatrial natriuretic factor as defined by a complementary DNA clone. Science. 1984;225:324-326.
Yamaguchi M, Rutledge LJ, Garbers DL. The primary structure of the rat guanylyl cyclase A/atrial natriuretic peptide receptor gene. J Biol Chem. 1990;265:20414-20420.
Saheki T, Mizuno T, Iwata T, Saito Y, Nagasawa T, Mizuno KU, Ito F, Hagiwara H, Hirose S. structure of the bovine atrial natriuretic peptide receptor (type C) gene. J Biol Chem. 1991;266:11122-11125.
Anand-Srivastava MB, Sairam MR, Cantin M. Ring-deleted analogs of atrial natriuretic factor inhibit adenylate cyclase/cAMP system. J Biol Chem. 1990;265:8566-8572.
Wilcox JN, Augustine A, Goeddel DV, Lowe DG. Differential regional expression of three natriuretic peptide receptor genes within primate tissues. Mol Cell Biol. 1991;11:3454-3462.
Nunez DJR, Dickson MC, Brown MJ. Natriuretic peptide receptor mRNAs in the rat and human heart. J Clin Invest. 1992;90:1966-1971.
Leitman DC, Agnost VL, Tuan JJ, Andresen JW, Murad F. Atrial natriuretic factor and sodium nitroprusside increase cyclic GMP in cultured rat lung fibroblasts by activating different forms of guanylate cyclase. Biochem J. 1987;244:69-74.
Brown LA, Nunez DJR, Wilkins MR. Different regulation of natriuretic peptide receptor messenger RNAs during the development of cardiac hypertrophy in the rat. J Clin Invest. 1993;92:2702-2712.