Electron micrograph of atrial graules in myocyte of rat heart atrium. Granules limited by a single membrane and with a dense core are seen near a fenestrated Golgi cisterna (G). Margin of G shows interconnecting tubules and budding Golgi vesicles. Arrows point to thickening of the membrane on certain “buds,” suggesting the Golgi origin of atrial granules, rb, residual body. Magnification, × 57,000.

The discovery of atrial natriuretic factor. Intravenous administration of crude atrial extract (heavy lines) but not ventricular extract (light lines) into rats led to major increases in urine flow rate (V) and the excretion of sodium (U_NaV) and chloride (U_ClV). Atrial extract also led to a more modest but significant increase in urinary excretion of potassium (U_KV).

Isolation of natriuretic and vasorelaxant peptides from rat atria. Atria from 200 rats were homogenized in 1 N acetic acid containing protease inhibitors and the supernatant was lyophilized and reconstituted in the same medium. A. Gel filtration on Sephadex G‐50. Aliquots of each fraction were lyophilized, reconstituted in phosphate buffer, and assayed for natriuretic activity in isolated perfused rat kidney and for vasorelaxant activity in Ang II‐contracted rabbit aortic rings. Two broad regions of activity were identified, corresponding to molecular weights of >10 kd and 2–5 kd (bracket). The latter fractions were pooled, lyophilized and then further purified by C₁₈ reversed‐phase HPLC, resulting in a single broad region of activity (Step 2, not shown). B. Rechromatography of low mol. wt. fraction by reversed‐phase HPLC. Using a shallow gradient of acetonitrile (Solvent B), active fractions were resolved into three principal components with coincident natriuretic and vasorelaxant activity. These were subsequently purified to homogeneity from a total of 1400 rat hearts by additional chromatographic steps (not shown).

Amino acid sequences of ANF peptides isolated from rat and human atria. The peptides isolated are compared to the amino acid sequence of the C‐terminal portion of the rat ANF precursor (proANF). These peptides were identified in studies by several investigators: rat Ala⁹²‐Tyr¹²⁶ 405,558; Leu⁹⁴‐Tyr¹²⁶ 405,589,769; Ala⁹⁵‐Tyr¹²⁶ 405,589,769; Gly⁹⁶‐Arg¹²⁵ 558; Ser⁹⁹‐Tyr¹²⁶ 236,405; Arg¹⁰¹‐Tyr¹²⁶ 769; Arg¹⁰²‐Tyr¹²⁶ 57,405,557,558; Arg¹⁰²‐Arg¹²⁵ 57; Ser¹⁰³‐Ser¹²³ 177; Ser¹⁰³‐Arg¹²⁵ 177; Ser¹⁰³‐Tyr¹²⁶ 282; and human Ser⁹⁹‐Tyr¹²⁶ 408. The latter sequence is identical to that of the rat peptides except for substitution of Met for Ile at position 110.

Deduced amino acid sequences of ANF precursors (preproANF) from several mammalian species. Amino acid residues are numbered according to the predicted 625,968 and subsequently proven 235,406,409 cleavage point of the signal peptide (i.e., in humans, proANF comprises residues 1 to 126, and the signal peptide comprises residues −1 to −25). Entire human sequence is shown, and only residues that differ in other species are indicated. Shorter signal peptides in some species (23 residues in the dog, 24 residues in the cow, rat, and mouse) are aligned to maximize amino acid homology, with dashes indicating residues omitted. Amino acid sequences were initially deduced from cDNA nucleotide sequences of humans 588,625,992, dog 624, rabbit 624, and rat 409,423,519,772,968,992, and were inferred from genomic DNA sequences of mouse 770 and cow 904. The caret (⁁) between positions Lys¹⁶‐Asn¹⁷ and Arg¹²⁵‐Tyr¹²⁶ indicate the points at which intervening nucleotide sequences (introns) occur in the genes of each species examined to date 48,299,518,597,770,771,904.

Schematic representation of human ANF gene, messenger RNA (mRNA), and precursor (pro‐ANF). In gene structure, horizontal lines represent 5′ and 3′ flanking regions and intervening sequences (introns) that separate the three coding sequences (exons I, II, and III). Features typical of eukaryotic genes include a “TATA” box (T) upstream from the transcription initiation site in the 5′ flanking region, and a polyadenylation signal (A) near the 3′ end; in addition, a sequence found in the second intron (GR) is highly homologous to the consensus sequence for the glucocortocoid response element. Arabic numerals refer to the amino acid sequence of the final translated product (pro‐ANF), indicating the locations of the corresponding coding sequences in the gene and mRNA. The narrow and wide horizontal bars indicate, respectively, the untranslated (ut) and translated portions of the mRNA. The signal peptide (cross‐hatched region derived from exon I, residues −1 to −25), typical of secreted proteins, is presumably cleaved cotranslationally. In addition, in certain other species the primary transcript codes for an additional C‐terminal dipeptide derived from exon III (arg¹²⁷‐Arg¹²⁸, not shown) which is thought to be cleaved prior to formation of mature granules storing pro‐ANF (see text and Fig. 5). Posttranslational processing of stored pro‐ANF leads to formation of the mature ANF peptide(s) at the C‐terminus (shaded region); hydrolytic cleavage occurs predominantly at Arg⁹⁸‐Ser⁹⁹ in atrial cardiocytes but may occur at other peptide bonds in extracardiac tissues (see text and Fig. 7 for details).

Amino acid sequences of human proANF and ANF. Standard one‐letter abbreviations for amino acids are used. In the proANF structure, bold arrow indicates principal hydrolytic cleavage site during posttranslational processing in cardiac atria; dashed arrows indicate possible or proven alternate cleavage sites in extracardiac tissues (see text). In the structure of mature ANF 99–126 (ANF_1–28), bold arrow indicates major degradative cleavage site demonstrated in vivo and in vitro, and fine arrows indicate other possible metabolic sites demonstrated in vitro (see later section on PHARMACOKINETICS AND METABOLISM for details).

Amino acid sequences of mature ANF (ANF‐28) and of brain natriuretic peptide (BNP‐32) and C‐type natriuretic peptide (CNP‐22) from porcine brain. Alternate nomenclature (in brackets) indicates the corresponding residues of the respective precursors, from which each peptide is derived from the C‐terminus. Boldface indicates amino acid homology between two or more of these peptide families, and underlined amino acids (within the ring structure) are fully homologous in all three families.

Effect of acute volume expansion on (a) right atrial pressure and (b) plasma immunoreactive ANF in male Wistar rats. Pentobarbital anesthetized animals were instrumented with a right atrial catheter via the jugular vein and with a femoral venous catheter for saline administration. Volume expansion was produced by infusion of either 2 or 8 ml saline over a 1 min period. Blood samples were collected by aortic puncture at 1, 5, or 10 min following termination of the volume load, (a) Representative right atrial pressure tracings. (b) Plasma immunoreactive ANF (expressed as picogram equivalents of ANF 103–126 [atriopeptin III] used as standard) in control animals (CO, n = 10) and in animals receiving either 2 ml (solid bars) or 8 ml (open bars) saline (n = 10 for each time point in each group). * P < 0.05 vs. control animals.

Effect of stepwise increases in dietary sodium on plasma atrial natriuretic peptide (ANP), aldosterone, and renin (PRA), and on urinary sodium excretion rate and cumulative sodium balance in normal human subjects. Six normal volunteers (age range 19–21 yr) were placed, following two days observation on their normal sodium intake, on a low‐sodium intake of 10 mmol/day. After 4 days equilibration, sodium intake was progressively increased by supplemental increments of 50 mmol/day, reaching a total daily intake of 350 mmol by day 13 of the study. Hormone measurements and sodium excretion rates were monitored daily. Open symbols, left‐hand ordinate; closed symbols, right‐hand ordinate.

Reversed‐phase HPLC of an extract of pooled normal human plasma. Plasma was collected on ice in EDTA and was extracted on C₁₈‐silica (Sep‐Pak) cartridges 155. Extract was injected (arrow) onto a 0.39 × 30 cm μBondapak C₁₈ column and eluted with a linear gradient of 10%–60% acetonitrile (CH₃CN) in trifluoroacetic acid over 50 min. Immunoreactive ANF was assayed in each fraction as described 155,223. The major form of immunoreactivity has a retention time (26 min) identical to that of synthetic human ANF 99–126. The earlier eluting peak (retention time 21 min) is found consistently under a variety of sampling conditions but is not formed when synthetic ANF is added to plasma in vitro, suggesting that it is produced endogenously. The structure of this presumed metabolite is uncertain. Note that no significant immunoreactivity is detected as the retention time of intact proANF (40 min) in normal plasma.

Complete amino acid sequence of mature forms of biological and clearance receptors of atrial natriuretic factor. Primary amino acid sequence of rat brain B_A‐ANF receptor 149 and of bovine aortic smooth muscle C‐ANF receptor 250 are aligned to show sequence homology. Conventional single‐letter abbreviation is used to designate the amino acids. Numbers on the right indicate the amino acid position in the sequence. Stars on top of each row indicate identical residues. There is a 33% homology between the extracellular domains of B_A‐ANF and C‐ANF receptors. Heavy underline indicates predicted single transmembrane domains for each receptor 149,250. C‐ANF receptors have a very short cytoplasmic tail of 37 amino acids 250, whereas, B_A‐ANF receptors contain a large cytoplasmic domain and within this domain a guanylate cyclase‐like sequence [underlined; 149]. In addition, the 256 amino acid sequence in the cytoplasmic domain of the B_A‐ANF receptor adjacent to the membrane is 31% homologous with the tyrosine kinase domain of the platelet derived growth factor receptor 149.

Schematic representation of cellular functions of two main classes of ANF receptors. B‐ANF receptors (R) are biological receptors proper of ANF and mediate known effects of the hormone, via the generation of cGMP. Main cellular effects of cGMP on stimulation (+) of membrane Ca²⁺‐ATPase, inhibition (−) of amiloride‐sensitive Na⁺ uptake and stimulation of furosemide‐sensitive Na⁺‐2Cl⁻–K⁺ cotransport are depicted by solid arrows. Other possible mechanisms of cGMP‐mediated decrease in cytosolic calcium ion activity () are depicted by a thick broken arrow. C‐ANF receptors (R) have as a main function the clearance of ANF from the circulation via receptor‐mediated endocytosis, delivery of ANF to lysosomes where they undergo hydrolysis to amino acids (aa), and recycling of the receptor to the cell membrane. Putative effects on other second messengers, either mediated directly by B‐ANF or C‐ANF receptors, or indirectly by cGMP, are depicted by broken arrows with? marks. Conventional abbreviations are used for nucleotides, second messengers, ions, and some enzymes. PK and GC, protein kinase‐like domain and guanylate cyclase‐like domain of the B‐ANF receptor; AC, membrane adenylate cyclase; PDE, phosphodiesterase; EV, endocytic vesicle; E, endosome; CURL, compartment for uncoupling receptor and ligand; PL, primary lysosome; SL, secondary lysosome. See text for description and references.

C‐ANF receptor‐mediated endocytosis, lysosomal hydrolysis, and receptor recycling in cultured bovine aortic smooth muscle (BSVM) cells. Confluent monolayers of BSVM cells were equilibrated with saturating concentrations of ¹²⁵I‐ANF_1–28 at 4°C. Then cells were warmed to 37°C in the continuous presence of saturating concentrations of the radioligand. The density of C‐ANF receptors in BSVM cells is approximately 250,000 receptor sites/cell. Radioactivity bound to the cell membrane (panel A) is a measure of the density of membrane C‐ANF receptors at each time of incubation at 37°C (abscissa). Radioactivity in the cell interior (panel B) and [¹²⁵I]monoiodotyrosine (TCA soluble radioactivity) in the medium measure internalization and lysosomal hydrolysis of ¹²⁵I‐ANF_1–28, respectively. Experiments were performed under control conditions (open circles), or in the presence of 10 mM of the lysosomotropic agent, NH₄Cl (closed circles), or in the presence of 0.1 mM of the protein synthesis inhibitor, cycloheximide (closed squares). The results show that after an initial burst of internalization, membrane receptors rapidly return to the cell surface even in presence of NH₄Cl or cycloheximide. This demonstrates that C‐ANF receptors are rapidly internalized and recycled to the cell surface. NH₄Cl almost completely blocked the hydrolysis of ¹²⁵I‐ANF_1–28 (panel C) and, consequently augmented the accumulation of radioactivity in the intracellular compartment (panel B). This demonstrates that internalized ¹²⁵I‐ANF_1–28 is delivered to lysosomes where it undergoes hydrolysis. Form these and other experiments (reported in reference 613) it can be calculated that the entire population of surface C‐ANF receptors is internalized and recycled every hour.

Effects of atrial natriuretic factor on renal function and renal vascular resistance in the isolated perfused rat kidney. A. Isolated rat kidneys were perfused with normal (2 mM; groups A and B) or low (0.2 mM; groups C and D) calcium in presence of atrial extract (AE; groups A and C) or ventricular extract (VE; groups 2 and 4). Results are expressed as relative changes in renal function parameters relative to control periods in which the isolated kidneys were perfused in the absence of AE or VE (mean ± SE; *, P < 0.01). The results demonstrate that AE has direct hemodynamic and excretory actions in the kidney. In presence of normal calcium concentrations (groups A and B), AE but not VE slightly but significantly increases renal vascular resistance (RR) and leads to major increases in GFR and filtration fraction (FF). This indicates that ANF has a preferential efferent arteriolar vasoconstrictive effect (see text). AE but not VE markedly increases urine flow rate (V), absolute and fractional excretion of sodium (U_NaV and FE_Na, respectively), and absolute and fractional excretion of potassium (U_KV and FE_K, respectively). The AE‐induced changes in renal hemodynamic and excretory functions were almost completely abolished when kidneys were perfused with a low calcium concentration (group C). These experiments unveiled the overall renal hemodynamic actions of ANF and led to the postulate that its natriuretic action is due, at least in part, to an increase in GFR and a decrease in inner medullary hypertonicity (see text). B. Effects of ANF on renal vascular resistance (RVR) is isolated rat kidney perfused under control conditions (C) or after the addition of vasoconstrictive substances to the perfusate. Results are mean ± SE for each treatment (*, P < 0.01). ANF increased RVR slightly but significantly in control kidneys but markedly decreased RVR in kidneys preconstricted with hormonal (angiotensin II, norepinephrine, vasopressin) or nonhormonal (ouabain, tetracaine) substances. The results of these experiments indicate that ANF has a weak agonist (vasoconstrictive) effect on its own and is a powerful antagonist of vasoconstriction (see text). Based on data from references 57,121, and 510.

Steady‐state effects of ANF_4–27 (auriculin A) on renal functions in anesthetized dogs. After control periods (C), ANF was infused at a dose of 0.1 μg·min⁻¹.kg⁻¹ for 60 min. Renal function parameters were determined in experimental periods (E) and 30–40 min after the suspension of ANF infusion (R). Results are mean ± SE (*, P < 0.05 vs C). MBP, mean arterial blood pressure; RPF, renal plasma flow; RVR, renal vascular resistance; FF, filtration fraction; GFR, glomerular filtration rate; V, urine flow rate; U_OSM, urine osmolal concentration; , free water clearance; FL_Na, fractional excretion of sodium; U_KV, urinary excretion of potassium. See text for description. Based on data from Maack et al (ref 512).

Effect of ANF_4–27 (auriculin A) on single nephron and whole kidney GFR of the rat. ANF_4–27 was infused i.v. at a rate of 0.3 μg·min_‐1·kg_‐1 and single nephron GFR was measured in anesthetized rats at end convoluted proximal tubule (open circles) and early distal (filled circles) sites. Whole kidney GFR was determined in anesthetized (open circles) and awake (filled squares) rats. These experiments demonstrate that superficial single‐nephron GFR increases in the same proportion as whole kidney GFR, suggesting that ANF does not lead to a redistribution of renal blood flow and GFR in the rat kidney.

Autoradiographic localization of specific binding of ANF in kidney and adrenal. Dark field photomicrograph of autoradiograms of guinea pig kidney and adrenal sections incubated with 150 pM of ¹²⁵I‐ANF_1–28 (panel a) and its serial pair (panel b) incubated with 150 pM of the radioligand plus 1 μM of unlabeled ANF_1–28 (nonspecific binding). Radioligand binding appears as white silver grains. In the kidney section, the high density of specific binding is present in glomeruli (GA), renal arteries (RA), and outer medulla (OM). In the adrenal cortex, the zona glomerulosa (ZG) is clearly outlined by a dense array of silver grains in the adrenal cortex.

Effect of atrial natriuretic factor (ANF) on the secretion of sodium and chloride (upper panel) and on transepithelial voltage (lower panel) in isolated perfused inner medullary collecting ducts from the rat. ANF (6 nM), significantly (*, P<0.01) increased the bath‐to‐lumen flux of sodium and chloride, and led to a more negative lumen potential, effects that were fully abolished by the addition of 1 μM of furosemide (FURO) to the bathing solution. The ANF‐induced effects were fully restored once furosemide was removed from the bathing solution. This experiment demonstrated that ANF stimulates tubular secretion of sodium and chloride by activating a furosemide‐sensitive Na⁺–2Cl⁻–K⁺ cotransporter localized in the basolateral membranes of inner medullary collecting duct from the rat. This effect may explain in part the high urinary sodium concentrations observed after the administration of ANF.

Schematic representation of postulated sites of action of ANF in the mammalian nephron. See text for description and references. G, glomerulus; AA, afferent arteriole; EA, efferent arteriole; PCT, proximal convoluted tubule; DHL, thin descending limb of Henle's loop; AHL, thin ascending limb of Henle's loop; TAL, thick ascending limb of Henle's loop; MD, macula densa; DCT, distal convoluted tubule; CCT, cortical collecting tubule; IMCT, inner medullary collecting tubule. Numbers in circles represent postulated effects of ANF (+, stimulatory; −, inhibitory): 1, increase in GFR; 2, direct or indirect inhibition of sodium reabsorption, particularly in deep nephron PCT; 3, inhibition of AT II stimulated sodium reabsorption; 4, increase in sodium load to Henle's loop; 5, decrease in hypertonicity of the medullary interstitium; 6, decrease in passive water efflux; 7, decrease in passive sodium efflux; 8, increase in load to MD; 9, inhibition of renin secretion; 10, inhibition of sodium reabsorption due to ANF‐induced decrease in plasma aldosterone levels; 11, inhibition of thiazide‐sensitive Na⁺‐Cl⁻ cotransport; 12, increase in sodium load to IMCT; 13, inhibition of amiloride‐sensitive sodium reabsorption; 14, stimulation of forosemide‐sensitive Na⁺‐2Cl⁻‐K⁺ cotransport in basolateral membranes of IMCT. U_NaV, urinary sodium excretion.

Decrease in renal perfusion by clamping the renal artery abolishes renal hemodynamic and natriuretic effects of ANF. Upper panel. “Early” clamp experiments in anesthetized dogs. Clamping of the left renal artery prior to the infusion of ANF_4–27 (auriculin) such as to preclude its hemodynamic effects and the increase in GFR completely abolished the natriuretic effect (U_NaV) of the intravenously infused peptide but not of furosemide (solid lines). The right unclamped kidney responded normally to the intravenous infusion of ANF_4–27, increasing GFR and U_NaV (dotted lines). Lower panel. “Late” clamp experiments in anesthetized rats. Clamping of the left renal artery during the infusion of ANF_4–27 (auriculin), such as to return renal hemodynamic parameters and GFR to values not significantly different from those prior to the infusion of the peptide, abolished the natriuretic (U_NaV) effect of ANF. The right unclamped kidney served as a control and shows a persistent increase in GFR and U_NaV.

Effect of constant infusion of synthetic ANF on renin secretion rate (RSR), plasma renin activity (PRA), plasma aldosterone, and plasma cortisol in intact, instrumented, anesthetized dogs. Synthetic ANF 102–125 [auriculin A; 57] was administered as a 1 /μg/kg bolus followed by infusion of 100 ng/kg per min for 60 min (horizontal bar). RSR was determined by simulaneous measurement of arteriovenous differences in PRA across the kidney and of para‐aminohippurate clearance rate to estimate renal plasma flow. The data demonstrate that ANF induces a prompt decrease in RSR followed by more gradual reductions in peripheral PRA and plasma aldosterone (*P < 0.05), whereas plasma cortisol was unchanged.

Hormonal and renal responses to graded infusion of synthetic ANF in a sodium‐depleted human subject. The subject was studied in the seated position after 5 days on a diet containing 10 mmol sodium per day. Sequential administration of synthetic ANF 102–126 [auriculin B; 57] at the four doses indicated (40 min each) was preceded by a 40 min control period and followed by a 40 min recovery period. Blood and urine samples were obtained at 20 min intervals throughout; urine parameters shown are only from the second collection (i.e., 20–40 min of each period). P_cGMP > plasma cGMP; PRA, plasma renin activity; PA, plasma aldosterone; V, urine flow rate; U_aldoV, U_cGMPV, U_NaV, urinary excretion rates of aldosterone, cGMP, and sodium. M. S. Pecker and S. A. Atlas, unpublished data.

Effect of ANF infusion on plasma renin activity (PRA) in animals with normal or reduced renal perfusion pressure. A. Effect of ANF in intact conscious or anesthetized dogs or in dogs with acute unilateral renal artery constriction. The infusion protocol was as described for Figure 22. Unilateral renal artery constriction (Clip) was produced with a snare around the left renal artery that reduced renal perfusion pressure to 80–90 mm Hg and prevented the ANF‐induced increase in GFR 815,816; ANF infusion was begun 1 h after constriction. C, last control period; E₁, E₂, 40‐ and 60‐min of ANF infusion; R, last recovery period. ANF suppressed PRA in both conscious and anesthetized dogs (*P < 0.05). Renal artery constriction stimulated PRA slightly but prevented the ANF‐induced suppression of PRA. Based on data from Maack et al. 512 and Sosa et al. 816

Inhibition of (A) ACTH‐induced or (B) angiotensin II–induced aldosterone production by synthetic ANF 99–126 in isolated rat adrenal glomerulosa cells. Data points are the mean of duplicate incubations from representative experiments.

Inhibition of angiotensin II‐stimulated aldosterone (A) and pregnenolone (B) production by synthetic ANF 102–125 [auriculin A; 57] in isolated bovine adrenal glomerulosa cells. ANF was added at 4 nM final concentration. Inhibition of pregnenolone production (B) was performed in the presence of 10⁻⁵ M trilostane, an agent that blocks further metabolism of pregnenolone.

Inhibition of norepinephrine‐induced (A) and angiotensin II‐induced (B) contraction of isolated rabbit thoracic aortic rings by increasing concentration of ANF. Each curve depicts the change in tension caused by increasing concentrations of vasoconstrictor in control rings (open circles) or in rings exposed to increasing concentrations of highly purified ANF (∇, □, Δ). The data demonstrate that ANF causes a dose‐dependent rightward shift of the norepinephrine dose‐response curve, but that, in addition, ANF markedly depresses the maximal contraction that can be elicited by angiotensin II.

Effects of ANF infusion on hematocrit and plasma volume in bilaterally nephrectomized rats. Mean ± SE values of mean blood pressure (MBP), hematocrit, and plasma volume during control periods (open bars) and during the infusion (hatched bars) of either saline (8 μl/min), sodium nitroprusside (1 to 2 μg·min⁻¹/kg body weight) or ANF_1‐28 (1 μg·min⁻¹/kg body weight) in bilaterally nephrectomized rats. * P, <0.01 vs control periods. ANF, but not sodium nitropusside (at doses that led to an equivalent fall in MBP), significantly decreased plasma volume leading to a corresponding increase in hematocrit. Since these experiments were performed in bilaterally nephrectomized rats, the ANF‐induced decrease in plasma volume is due to a shift of fluid from the vascular to the interstitial compartment.

Hemodynamic effects of ANF in conscious dogs. After control period (C), animals received an i.v. infusion of ANF_4‐27 (auriculin A) as a 3 μg/kg body weight bolus followed by 0.3 μg·min⁻¹/kg body weight for 30 min (E). This was followed by a 1 h recovery period (R). Values (mean ± SE) are shown for systolic and diastolic blood pressure (BP), heart rate (HR), cardiac output (CO), stroke volume (SV), central venous pressure (CVP), and calculated total peripheral resistance (TPR); units for the latter are dyne‐s‐cm⁻⁵. *, P < 0.05 vs. C.

Comparison of hemodynamic responses to ANF infusion in anesthetized two‐kidney, one‐clip (circles, solid line), and desoxy‐corticosterone (DOC)‐salt treated (triangles, dashed line) hypertensive rats. Synthetic rat ANF 102–126 [auriculin B; 57] was administered by graded constant infusion at three doses (30 min per dose), and the data (mean ± SE) are expressed as the steady‐state change in each parameter from control values. MAP, mean arterial pressure; TPR, total peripheral resistance; HR, heart rate; SV, stroke volume. *P < 0.05 for difference between the two experimental groups. In DOC‐salt hypertensive rats, ANF induced striking dose‐dependent falls in SV and increases in TPR, whereas MAP fell significantly only at the highest dose. In contrast, in two‐kidney, one‐clip rats, ANF significantly lowered MAP at lower doses in association with a significant fall in TPR, and SV tended to fall only at the highest dose.

Effects of central blood volume expansion on plasma ANF, renin, aldosterone, and cortisol, and on urine flow rate (V), sodium excretion (U_NaV) and cGMP excretion (U_cGMPV). Central volume expansion was produced in seated normal human subjects by head‐out water immersion (at 34°C) between 1 and 4 h (horizontal bar); subjects were seated outside the immersion tank during the control (0–1 h) and recovery (4–5 h) periods.

Schematic representation of the regulation of ANF secretion, its major target organ actions, and its interrelationship with reflex inhibition of central sympathetic outflow during expansion of central blood volume. Dashed lines indicate probable negative feedback signals. Although unproven, it is likely that ANF‐induced inhibition of aldosterone and renin release, natriuresis, and extravascular fluid shifts occur at near physiological concentrations of the hormone (see text). The latter two effects, by diminishing venous return to the heart, would provide negative feedback signals for ANF secretion. Inhibition of vasopressin release and decreased vascular resistance are probably not major effects under normal physiological conditions. Antagonism of vasoconstrictor action is probably significant in certain pathological states, and the resultant decrease in vascular resistance would tend to counter the rise in blood pressure due to increased cardiac output and might also have negative feedback effects on atrial stretch. The broken lines at the asterisks indicate that the renal effects of ANF are attenuated in the edematous disorders characterized by diminished renal perfusion.

Plasma levels of immunoreactive ANF 99–126 (irANF) in normal human subjects and in patients with essential hypertension (EH), primary aldosteronism (1° Aldo), cirrhosis, congestive heart failure (CHF), and chronic renal failure (CRF). Blood samples were obtained in the seated position, and ANF was assayed as described 155.

Natriuretic responses to synthetic ANF infusion in normal human subjects and in patients with decompensated congestive heart failure (severe CHF). Synthetic rat ANF 102–126 [auriculin B; 57] was infused intravenously to seated subjects for a 60 min period, preceded by control and followed by recovery periods (60 min each); subjects stood briefly to void at 20 min periods throughout. Absolute sodium excretion rate (U_NaV) during the third experimental period (i.e., 40–60 min of ANF infusion) is plotted against the concurrent plasma ANF level. Each point in the relationship shown for normal subjects was derived from separate groups of subjects (n = 6 each) receiving, during the experimental period, either vehicle alone, or ANF at 4, 12, or 40 pmol/kg per min. Six patients with severe CHF received only the highest dose (40 pmol/kg per min); data points representing the last control period and last experimental period are connected for each patient. In this study, one patient with mild heart failure (and normal control plasma ANF) had a normal natriuretic response (not shown).

Responses of plasma renin activity (PRA) and plasma aldosterone (PA) to synthetic ANF infusion in normal subjects (n = 6) and heart failure patients (CHF, n = 7). Seated subjects received a constant infusion of synthetic ANF 102–126 (40 pmol/kg per min) during the 60 min experimental period (horizontal bracket), as described in the legend to Figure 34. ANF infusion reduced PA in both groups but significantly reduced PRA only in normal subjects (*P < 0.05).