96 (13) 7403-7408
Natriuretic peptides (NPs), mainly produced in heart [atrial (ANP) and B-type (BNP)], brain (CNP), and kidney (urodilatin), decrease blood pressure and increase salt excretion. These functions are mediated by natriuretic peptide receptors A and B (NPRA and NPRB) having cytoplasmic guanylyl cyclase domains that are stimulated when the receptors bind ligand. A more abundantly expressed receptor (NPRC or C-type) has a short cytoplasmic domain without guanylyl cyclase activity. NPRC is thought to act as a clearance receptor, although it may have additional functions. To test how NPRC affects the cardiovascular and renal systems, we inactivated its gene (Npr3) in mice by homologous recombination. The half life of [125 I]ANP in the circulation of homozygotes lacking NPRC is two-thirds longer than in the wild type, although plasma levels of ANP and BNP in heterozygotes and homozygotes are close to the wild type. Heterozygotes and homozygotes have a progressively reduced ability to concentrate urine, exhibit mild diuresis, and tend to be blood volume depleted. Blood pressure in the homozygotes is 8 mmHg (1 mmHg = 133 Pa) below normal. These results are consistent with the sole cardiovascular/renal function of NPRC being to clear natriuretic peptides, thereby modulating local effects of the natriuretic peptide system. Unexpectedly, Npr3 −/− homozygotes have skeletal deformities associated with a considerable increase in bone turnover. The phenotype is consistent with the bone function of NPRC being to clear locally synthesized CNP and modulate its effects. We conclude that NPRC modulates the availability of the natriuretic peptides at their target organs, thereby allowing the activity of the natriuretic peptide system to be tailored to specific local needs.
Get alerts for new articles, or get an alert when an article is cited.
The natriuretic peptides (NPs) play important roles in cardiovascular homeostasis. Three isoforms, atrial natriuretic peptide (ANP), B-type natriuretic peptide (BNP), and C-type natriuretic peptide (CNP), constitute the natriuretic peptide family (reviewed in ref. 1). ANP and BNP are mainly produced in the cardiac atria and ventricles, respectively; both are present in the circulation; and they directly influence blood pressure and body fluid homeostasis (reviewed in ref. 2). CNP is most strongly expressed in the brain but also is produced in vascular endothelial cells and in other tissues; its normal level in the circulation is very low; and it may have a paracrine/autocrine role. The biological functions of the natriuretic peptides are mediated by two receptors, natriuretic peptide receptor A (NPRA) [also known as guanylyl cyclase (GC) A] (3) and NPRB (GC-B) (4), which have cytoplasmic GC domains that are stimulated when the receptors bind ligand. NPRA responds to ANP and, to a 10-fold lesser degree, to BNP; NPRB responds primarily to CNP. NPRA is strongly expressed in the vasculature, kidneys, and adrenal glands, and its stimulation mediates vasorelaxant and natriuretic functions and decreases aldosterone synthesis. NPRB is strongly expressed in the brain, including the pituitary gland, and may have a role in neuroendocrine regulation. A third natriuretic peptide receptor (NPRC) has only a short cytoplasmic domain with no GC activity; it is generally thought to act as a clearance receptor and remove natriuretic peptides from the circulation (5), although several reports have suggested roles in addition to this clearance function (reviewed in ref. 6). NPRC interacts with all three natriuretic peptides in the order ANP > CNP > BNP (7). NPRC is the most widely and abundantly expressed natriuretic peptide receptor with a tissue distribution that includes many but not all tissues that express a guanylyl cyclase receptor; for example, kidney glomeruli (8) strongly express both NPRA and NPRC whereas Leydig cells in the testis strongly express NPRA but not NPRC (9). To gain a better understanding of the relationship between the receptors and their ligands and to test how NPRC affects the cardiovascular and renal systems, we have inactivated its gene (Npr3) in mice by homologous recombination. We find strong evidence that NPRC, in addition to being involved in the systemic clearance of circulating natriuretic peptides, plays an important role in controlling local effects of the natriuretic peptide system. Its complete absence produces unexpected skeletal abnormalities.
Portions of Npr3 (the mouse gene coding for NPRC) were cloned from mouse strain 129 genomic DNA fragments by using a probe based on the mouse Npr3 exon 1 sequence (D. G. Lowe, personal communication). The 5′ region of homology in the targeting construct (Fig. 1a) was a 5.9-kilobase (kb) Xba I-Spe I fragment from upstream of exon 1. The 3′ homology region was a 0.7-kb Xho I-Hin dIII fragment that includes parts of exon 1 and intron 1. Four electroporations of strain 129 embryonic stem cells were carried out as described (10). Candidate targeted clones were identified by PCR amplification using primer C (5′-ACGCGTCACCTTAATATGCG-3′) and primer D (5′-TCGGCTCCTTCCTCTATCTA-3′) (Fig. 1a). Targeting was confirmed by the presence of a 6.3-kb hybridizing band in addition to an 8-kb endogenous band in Southern blots of DNA digested with Hin dIII and hybridized to probe A (a 0.7-kb Bgl II-Spe I fragment) (Fig. 1b). Chimeric mice carrying the nonfunctional allele were generated and mated to C57BL/6 females to yield F1 heterozygotes, which were intercrossed to obtain Npr3 +/+, +/−, and −/− F2 animals for use in the present studies. Animals were handled under University of North Carolina-approved protocols.
Targeted inactivation of the Npr3 gene. (a) The targeting strategy. (Top line) The region of the Npr3 gene that includes exon 1 (black bar). (Middle line) The targeting construct. (Bottom line) The targeted locus in which the gene is disrupted and from which 215 amino acids of the ligand-binding domain have been deleted. neo, neomycin resistance gene; tk, Herpes simplex thymidine kinase gene. Restriction sites are H, Hin dIII; N, Not I; S, Spe I; Xb, Xba I; Xh, Xho I. The positions of two probes, A and B, and of two primers, C and D, are indicated. (b) Southern blots of tail DNA from wild-type (+/+), heterozygous (+/−), and homozygous mutant (−/−) mice digested with Hin dIII and hybridized to probe A. The sizes of the hybridizing bands in kb are shown. (c) An autoradiogram showing the presence or absence of the 70-kDa band corresponding to NPRC after SDS gel electrophoresis under reducing conditions of 4-azidobenzoyl [125 I]ANP photoaffinity-labeled lung plasma membranes from +/+ and −/− mice. The 135-kDa radio-labeled band corresponding to NPRA is indicated.
Plasma membranes (200 pg) were incubated with 4-azidobenzoyl [125 I]ANP as described (11). After photolysis, samples were washed twice and were subjected to SDS/PAGE. The receptor bands were localized by autoradiography.
[125 I]ANP(rat; 1–28) was rapidly injected (0.2 μCi/mouse) into the jugular vein; blood was collected from the carotid artery as published (12) 0.5, 1.5, 2.5, 4, 8, and 16 min after administration of the labeled peptide. Plasma (25 μl) separated from the blood samples (60 μl) was precipitated with 125 μl of ice-cold 10% trichloracetic acid. 125 I-radioactivity in the pellets was determined by γ-scintillation. The residual counts at 16 min represent trichloroacetic acid-soluble radioactivity trapped in the pellet and were subtracted from the other values before calculations. The data were normalized to 10 6 cpm injected per animal. The P value of +/+ versus −/− was calculated by analysis of covariance.
The ANP assay was with a published protocol (13) using a rabbit anti-ANP antiserum (Phoenix Laboratories, Belmont, CA). The BNP assay was developed for this study. Synthetic mouse BNP conjugated to bovine thyroglobulin was used to immunize sheep. The resulting anti-mBNP antiserum cross-reacts 0.01% with rat BNP and 0% with rat ANP[1–28], CNP-22, arginine vasopressin, and angiotensin II.
Blood pressures, measured in conscious young adult male mice aged from 3 to 4 months by a noninvasive computerized tail-cuff method (14), were the means of at least six measurement sessions on each of 5 days. Animals were fed regular chow.
Blood samples were drawn from the retroorbital sinus from aged-matched anesthetized (Avertin, 2.5%, 0.3 ml/25 g body wt) male mice of the three genotypes and were analyzed with an Ektachem DT60II Analyzer (Johnson & Johnson, Rochester, NY). Hematological determinations were with a Cobas Micro Hematology Analyzer (Roche Diagnostics).
For urine studies, mice were maintained on a 12-hour light/dark cycle in metabolic cages with free access to water and food. On day 3, 24-hour water intake, urine excretion, and urinary electrolytes were measured. Urinary cGMP was measured as described (15). All values were normalized to a body weight of 30 g. To test ability to dilute urine, each mouse was gavaged with a volume of water equal to 4% of its body weight. Body weight and ensuing urine osmolalities were measured at 60 and 120 min. To test ability to concentrate urine, animals were placed in cages without food or water for 12 hours, and urine osmolalities were measured.
Radiographs of mice were taken with soft x-rays. For histology, mice aged 10 days and 2 months were killed, were fixed in 4% paraformaldehyde in PBS, and were decalcified in 10% EDTA, and samples were embedded in paraffin. Five-micrometer sections were stained for tartrate-resistant acid phosphatase as published (16) and with hematoxylin. Some sections of bone were rehydrated and stained with 0.1% Sirus Red (Sigma-Aldrich, Milwaukee, WI) in saturated aqueous picric acid (pH 2.0) for 30 min. After a brief wash with 0.01 M HCl, sections were dehydrated, were mounted in synthetic resin, and were photographed by using polarized light (17).
A 522-bp fragment from exon 1 of Npr3 was amplified by PCR with the primers 5′-GCGTAGCGTGGAGGGCAAT-3′ and 5′-CTGCCTTGGATGTAGCGCACTAT-3′ and was inserted into pBluescript II KS(+/−) plasmid (Strategene) for transcription of either a sense or antisense 35 S-labeled riboprobe. Frozen vertebra bones from 10-day-old mice were cut into 10-μm sections. Hybridization was overnight at 50°C, and washes were at 65°C. Slides were dipped in emulsion, were exposed for 2 weeks, and were counterstained with hematoxylin and eosin.
Bone marrow stromal cell cultures (18) were established with bone marrow from long bones of 2-month-old −/− and wild-type mice. After 8 days in culture, cells were evaluated for alkaline phosphatase (ALP) activity by using a commercial kit (Sigma-Aldrich). ALP-positive colonies were counted under low power magnification on 20 randomly chosen fields, and the numbers of cells in 100 consecutive colonies were counted. Osteoclast formation in vitro was assayed as described (19) after staining cultures for tartrate-resistant acid phosphatase. Positively staining cells with three or more nuclei (osteoclasts) were counted under high power magnification on 20 randomly chosen visual fields. The number of nuclei in 100 consecutive cells was counted.
To assess bone resorption, urinary excretion of pyridonoline and deoxypyridonoline was determined by hydrolyzing urine samples with HCl followed by HPLC analysis as described (20). Values were normalized by the urinary creatinine concentration.
Except when indicated, analysis of variance and pairwise comparisons were by the Bonferroni method. In some comparisons, the data were analyzed by P-stat, a correlation/permutation test in which the null distribution is approximated by Monte Carlo sampling. Information about P-stat and a
