Natriuretic Peptides

Natriuretic peptides are hormones that make the kidneys excrete sodium and water. The name comes from natrium, sodium, and the Greek ouresis, urination. Two of them, atrial natriuretic peptide (ANP) and B-type natriuretic peptide (BNP), are made by the heart muscle itself, and their discovery showed that the heart is also an endocrine organ[1]. A third, C-type natriuretic peptide (CNP), works mainly as a local signal in blood vessels and in growing bone[2].

The heart releases these peptides when its walls are stretched, and they act against the hormones that retain salt and water. That makes them a message from the heart about how hard it is working, and a message can be measured. This page follows that idea from the first experiments with heart extracts to the blood test now used in emergency departments, and then to the drugs that try to strengthen the signal.

The heart as a gland

The first clues came from the microscope. In 1956 Kisch noticed that cells of the atria, but not of the ventricles, had well-developed Golgi networks like those of secretory cells. In 1964 Jamieson and Palade found dense granules in the same atrial cells[2]. At about the same time, physiologists saw that stretching the atria of dogs with a balloon increased urine output[2]. The link was made by Adolfo de Bold and colleagues in 1981. They injected extracts of rat atria into other rats, and within moments sodium and chloride excretion rose more than thirtyfold and urine volume tenfold. Extracts of ventricles had no such effect[1].

Several groups soon purified the active peptide, and it was given several names, among them atrial natriuretic factor, atriopeptin, cardionatrin and cardiodilatin; atrial natriuretic peptide, ANP, is the name that stuck[2]. A second peptide was isolated in 1988 from pig brain and was called brain natriuretic peptide. It later turned out to be far more abundant in the heart ventricles of patients with heart failure, and it is now called B-type natriuretic peptide, keeping the abbreviation BNP[2]. CNP was also first found in pig brain, through its ability to relax smooth muscle[2].

Three peptides, one ring

ANP, BNP and CNP are products of three separate genes, but they share a common design: a ring of 17 amino acids closed by a disulfide bond between two cysteines[2]. The ring is essential for activity[1]. Eleven of its 17 positions are identical in all three human peptides, while the tails on either side of the ring differ. CNP has no tail at all after the ring, and this is why it does not cause natriuresis[1].

Each peptide is first made as a longer precursor. The ANP gene, NPPA, on chromosome 1, gives a 151-residue preproANP. Removing the signal sequence leaves the 126-residue proANP, which is stored in the atrial granules. On release it is cut by corin, a protease on the surface of heart muscle cells, into the active 28-residue ANP[2]. People with loss-of-function mutations in the corin gene have a condition consistent with too little ANP activity, such as high blood pressure[1]. The mature ANP of humans differs from that of rats at only one position[2].

BNP is handled differently, and the difference matters for what follows. Its gene, NPPB, lies next to NPPA on chromosome 1[1] and gives a 134-residue preproBNP and then a 108-residue proBNP[2]. Cleavage of proBNP releases the active 32-residue BNP and an inactive N-terminal fragment, NT-proBNP, in equal amounts[3]. Some BNP is stored in atrial granules together with ANP, but in the ventricles it is not stored at all: it is made as needed when the heart is under strain, for example from volume overload[2]. How much of the BNP in heart failure comes from the ventricles and how much from the atria is still debated. De Bold's group argues that the atria remain the main source even then[1].

CNP comes from a gene on another chromosome, NPPC on chromosome 2. In tissues it acts mostly as a 53-residue form, and in blood a 22-residue form dominates. The 22-residue CNP is identical in humans, chimpanzees, dogs, mice and rats[2].

Receptors and actions

Natriuretic peptides act through three receptors on the cell surface. Two of them, NPR-A and NPR-B, are enzymes: when a peptide binds outside the cell, a guanylyl cyclase domain inside makes cyclic GMP (cGMP), which acts on protein kinases, ion channels and phosphodiesterases[1]. NPR-A prefers ANP and BNP, and NPR-B prefers CNP[2]. The third receptor, NPR-C, has only 37 amino acids inside the cell and no enzyme activity. It takes the peptides into the cell and destroys them, and it accounts for about 94% of the ANP binding sites on cells lining blood vessels[2].

Through NPR-A, ANP and BNP increase the filtration of blood in the kidney, reduce the reabsorption of sodium, lower the release of renin and aldosterone, relax blood vessels and let fluid pass from the blood into the tissues[1]. Together these actions lower blood volume and pressure, the opposite of the renin–angiotensin–aldosterone system[4].

Mice lacking single genes show what each part does[2]:

Human genetics agrees. People with two non-working copies of the NPR-B gene have a rare form of dwarfism, acromesomelic dysplasia type Maroteaux, and carriers of one copy are on average shorter[2]. Common variants near the NPPA and NPPB genes that raise plasma ANP and BNP are linked to a 15% lower risk of high blood pressure. Variants that lower ANP are linked to hypertension, while variants affecting BNP are not, which suggests that ANP is the main peptide controlling blood pressure[1].

The peptides do not last long. ANP has a half-life in human blood of about two minutes, BNP of about twenty[2]. Both are removed by NPR-C and by neprilysin (also called neutral endopeptidase), a protease that breaks down several other peptides as well[2][1]. Neprilysin will reappear below, because blocking it became the most successful way of turning these peptides into medicine.

The blood test

Because the heart releases BNP in proportion to the stretch of its walls, the amount in blood is a measure of how strained the heart is. In practice the test measures either BNP or the inactive NT-proBNP, which is released with it but lasts longer in blood, about 70 minutes against about 20[3]. Of the three validated markers, BNP, NT-proBNP and MR-proANP (a fragment of the ANP precursor), each has its own normal range and cut-offs, so values cannot be converted from one to another[4].

The test answers a common and difficult question: is a breathless patient breathless because of the heart? Clinical examination is unreliable here. Signs such as crackles in the lungs, swollen neck veins and leg oedema detect heart failure in only 50–60% of cases[4]. In the Breathing Not Properly study of 1,586 patients with acute breathlessness, published in 2002, a BNP cut-off of 100 pg/mL detected 90% of the patients who had heart failure[3]. Later studies with NT-proBNP, the largest of them ICON with 1,256 patients, showed that a value below 300 pg/mL ruled out acute heart failure with a sensitivity of 99%[3]. The European Heart Failure Association recommends the following cut-offs[4]:

In people with obesity, cut-offs about 50% lower are recommended[4].

The test is best at excluding heart failure. A low value makes heart failure very unlikely, while a high value only says that the heart is under strain, not why[4]. Roughly one in five patients with acute breathlessness falls in the grey zone, and about half of them turn out to have acute heart failure[4]. The guidance therefore stresses that the result must never be used on its own, and that an elevated value calls for imaging of the heart, usually echocardiography[4].

Many things move the value. It is raised by atrial fibrillation, pulmonary embolism, pulmonary hypertension, sepsis and kidney failure[4][3]. Kidney function and body mass are the two strongest confounders[4]. The link with the kidneys is often misunderstood. Only about a quarter of the peptides are cleared by kidney filtration; the rest are removed by other organs with a high blood flow, so raised values in kidney disease partly reflect real strain on the heart[4]. Values are lower in obesity, for reasons that are not fully understood[4][1], and they are higher in women than in men[3]. They can also be falsely low when the problem lies upstream of the left ventricle, as in mitral stenosis, or when breathlessness starts within an hour, because the heart stores little BNP and has not yet had time to make more[4]. In patients with shock, the test cannot tell a failing heart from sepsis, although a high value still predicts a worse outcome[4]. Intensive-care patients have no validated cut-offs at all[5].

The laboratory side has its own pitfalls. BNP assays from different manufacturers use different antibodies, and results for the same sample can differ by up to 43% between methods; most NT-proBNP assays use the same antibodies and differ by about 9%[3]. The assays also detect the uncleaved prohormone, which predominates in chronic heart failure, so measured values depend in part on the method[1]. A change has to be large, around 50% or more, before it can be taken as a real change in the patient[4].

Whether adjusting treatment to bring the value down improves outcomes is less clear. The largest trial, GUIDE-IT, with 894 patients, found no benefit of NT-proBNP-guided treatment over usual care, although a pooled analysis of all trials suggests some benefit in heart failure with reduced ejection fraction[4]. The European guidance calls this use "controversial"[4]. Several of its authors have received funding or fees from companies that make the tests[4].

As drugs

A hormone that lowers blood pressure and removes excess fluid looked like an obvious treatment for heart failure. The direct approach was to infuse the peptide. Carperitide, synthetic ANP, was approved in Japan in 1995 for acute decompensated heart failure[2][1]. Nesiritide, recombinant BNP, was approved in the United States in 2001 for the same purpose[2]. After approval it was used in more patients than had taken part in any trial, and two meta-analyses in 2005 suggested that it worsened kidney function and increased the risk of death[2]. The large ASCEND-HF trial later found that nesiritide caused hypotension, and it is no longer recommended for routine use[1]. Trials of anaritide, a shortened ANP, in acute kidney failure failed for a similar reason: no benefit, and frequent severe drops in blood pressure[2]. As infused drugs, the peptides also have to be given continuously into a vein, because they are broken down within minutes[1].

The approach that worked was to slow the breakdown of the body's own peptides. Sacubitril blocks neprilysin and is combined with valsartan, which blocks the angiotensin receptor[1]. In the PARADIGM-HF trial in patients with heart failure and reduced ejection fraction, the combination reduced deaths from cardiovascular causes by 20% and hospital admissions for heart failure by 21% compared with enalapril[1]. Its main side effect is low blood pressure, which occurred in 14% of patients against 9.2% on enalapril[1]. Because neprilysin also breaks down other peptides, such as bradykinin and amyloid-β, there were concerns about swelling of the tissues and about Alzheimer disease; the latter has not been seen in humans[1]. In heart failure with preserved ejection fraction, the PARAGON-HF trial found no significant benefit over valsartan alone[1].

The drug changed the blood test too. Neprilysin breaks down BNP but not NT-proBNP[1], so patients taking sacubitril can show a rise in BNP even as they improve, while NT-proBNP falls[3]. NT-proBNP is therefore the preferred marker in these patients[4]. The rise in ANP is larger and more consistent than that in BNP, and some researchers argue that ANP, not BNP, is what mediates the drug's benefit[1].

Other ideas are still at the stage of animals and small trials. These include ANP and BNP fused to albumin or antibody fragments to make them last longer, and designer peptides that activate both NPR-A and NPR-B[1]. CNP points in a different direction. Since mice lacking it are dwarfs and extra CNP partly restored bone growth in a mouse model of achondroplasia, the most common form of human dwarfism, CNP analogues were proposed as a treatment for short stature[2]. That proposal has since reached patients. Vosoritide, a CNP analogue that acts on NPR-B and is injected under the skin, was approved in the United States in 2021 for children with achondroplasia whose growth plates are still open. In the main trial of 121 children, those treated grew on average 1.57 cm more in a year than those given placebo. The approval was accelerated, and a further study must show whether treatment increases final adult height[6].

Some 45 years after de Bold's experiments in rats, the natriuretic peptides are used more as a message than as a medicine. Infusing ANP or BNP proved hard, because the body clears them in minutes and they lower blood pressure too much. The treatments that worked took a detour: one slows the breakdown of the body's own peptides, and the other is a modified CNP that acts on bone rather than on the heart. Measuring the peptides, meanwhile, gave doctors a direct reading of strain on the heart.

References

1. Goetze JP, Bruneau BG, Ramos HR, Ogawa T, de Bold MK, de Bold AJ (2020). Cardiac natriuretic peptides. Nature Reviews Cardiology 17:698–717. DOI: 10.1038/s41569-020-0381-0
2. Potter LR, Yoder AR, Flora DR, Antos LK, Dickey DM (2009). Natriuretic peptides: their structures, receptors, physiologic functions and therapeutic applications. Handbook of Experimental Pharmacology 191:341–366. DOI: 10.1007/978-3-540-68964-5_15
3. Lee DJW, Aw TC (2023). Natriuretic peptides in clinical practice: a current review. Journal of Immunological Sciences 7(1):28–34.
4. Mueller C, McDonald K, de Boer RA, Maisel A, Cleland JGF, Kozhuharov N, Coats AJS, Metra M, Mebazaa A, Ruschitzka F, Lainscak M, Filippatos G, Seferovic PM, Meijers WC, Bayes-Genis A, Mueller T, Richards M, Januzzi JL Jr (2019). Heart Failure Association of the European Society of Cardiology practical guidance on the use of natriuretic peptide concentrations. European Journal of Heart Failure 21:715–731. DOI: 10.1002/ejhf.1494
5. Olarewaju A, Adebowale A, Odutola P, Arnold A (2026). Use of natriuretic peptides in critically ill patients: a narrative review. Journal of Clinical Medicine 15:5244. DOI: 10.3390/jcm15135244
6. US Food and Drug Administration (19 November 2021). FDA approves first drug to improve growth in children with most common form of dwarfism. fda.gov