Perspectives on C‐Type Natriuretic Peptide in Cardiometabolic Disease

Abstract

Although CNP (C‐type natriuretic peptide) was first identified in brain tissue, its systemic expression and release is mainly located to the endothelial lining of the vasculature. As such, CNP should not be viewed as a classic blood‐borne hormone but rather as a local factor that supports the integrity of vascular function. Albeit CNP is structurally related to both ANP and BNP (A‐ and B‐type natriuretic peptide), the main signaling pathway is through a specific membrane‐bound receptor without crosstalk from endocrine ANP and BNP. CNP expression and CNP‐mediated effects can thus be uniquely targeted without interference from the related cardiac NPs. In this perspective, we present the CNP system as a largely overlooked target for potential intervention in cardiometabolic disease, hypertension, and diabetes before onset of complications and overt heart failure. To fully unleash this potential, with this perspective, we hope to fuel renewed interest with a focus on use of human in vitro and ex vivo systems as well as experimental medical approaches and human cohort data.

Nonstandard Abbreviations and Acronyms

CNP

C‐type natriuretic peptide

In an evolutionary context, CNP (C‐type natriuretic peptide) is structurally related to the cardiac hormones ANP and BNP (A‐ and B‐type natriuretic peptide). CNP biology and its role in human disease, however, differs markedly from that of its cardiac counterparts. CNP is mainly produced outside the heart and even has a specific receptor. ¹ In terms of therapy, the CNP system therefore represents a unique vascular target without involving the hemodynamic effects known for ANP and BNP, effects that may even be considered harmful in some patients. ² , ³ In this perspective, we briefly discuss key elements of CNP‐specific biology that have gone somewhat under the radar in comparison with the cardiac NPs. We also point to disease scenarios where CNP‐targeted therapy may be of future relevance. Finally, shortcomings in today’s CNP translational science is highlighted with the hope that the needs will be addressed and solved.

CNP ENTERS THE STAGE OF BIOLOGICALLY ACTIVE PEPTIDES

Biologically active peptides often form families of peptides with similar structures. After identifying human ANP, researchers conducted thorough searches in both rat and human atrial tissues using an ANP‐specific antibody and an ANP cDNA probe without finding new peptides. Rather, CNP was discovered somewhat by chance in the process of searching for new peptides using porcine brain tissue. ⁴ Biochemical screening of tissue extracts (40 kilos of brain tissue) was crudely monitored for bioactivity in terms of muscle relaxation of chick rectum strips, a bioassay that already had proved instrumental for the previous identification of BNP. ⁵ From this screening, further chromatographic separation of the bioactive material led to a substance with a molecular mass of approximately 3000 daltons. One fraction contained relaxant activity yet devoid of BNP, which was then finally identified as a 22 amino acid residue peptide structure that is now known as CNP or CNP‐22 (Figure). A radioimmunoassay for the structure was developed and used to further identify a larger CNP form in porcine brain: CNP‐53. ⁶ In addition to the identification, the first use of the CNP immunoassay suggested that CNP expression markedly differs from that of ANP and BNP with only minor amounts produced in cardiac tissue. ⁷ , ⁸ Several studies have since then reported on CNP expression in a variety of tissues, and these reports are also based on measurement of another peptide fragment from the biosynthetic precursor, proCNP. ⁹ , ¹⁰ , ¹¹ Importantly, measurement of N‐terminal proCNP fragments for instance comes with a high degree of analytical specificity compared with bioactive CNP peptides, as these fragments are not structurally related to similar fragments from the other NPs.

Figure 1. From discovery to molecular forms.

A, A simplified version of the original figure from the discovery of CNP. Adapted from Sudoh et al. ⁴ with permission. Copyright ©1990, Elsevier. Three principal modalities were used for the original identification: ion exchange high pressure liquid chromatography elution monitored by absorbance, a bioassay using muscle relaxation, and an assay for following BNP immunoreactivity (for exclusion purposes). B, The principal peptide forms stemming from proCNP. The proCNP_51‐81 fragment in brackets is still a theoretical structure that has not been identified in circulation. C, The complete pre‐proCNP structure in 4 mammalian species. Note the striking conservation throughout the molecule: CNP‐22 is located at the C‐terminal end. BNP indicates B‐type natriuretic peptide; and CNP, C‐type natriuretic peptide.

A CRITICAL APPRAISAL ON CNP MEASUREMENT IN PLASMA

Measuring peptides in tissue and plasma is most often based on methods involving antibody recognition of the peptide in question. As CNP is structurally related to ANP and BNP, caution must always be applied as to which method is used for quantitating CNP in circulation. Even a minor degree of cross‐reactivity to ANP and BNP will affect measurement and thus the results in terms of interpretation. As an example, a commercial CNP method was evaluated for cross‐reactivity to the related BNP and shown to cross‐react. ¹² At BNP concentrations of 100 pmol/L, the cross‐reactivity of that CNP assay was ∼5%. Thus, studies on circulating CNP concentrations in, for instance, patients with heart failure and markedly increased BNP concentrations should be scrutinized with caution, as the reported data are likely to also report BNP concentrations. To complicate molecular quantitation further, CNP circulates in exceptionally low concentrations, which fundamentally challenges the sensitivity of immunological methods. Ideally, the method should be able to detect concentrations around 0 to 1 pmol/L, which often will be at the border of sensitivity for most methods. Finally, the molecular heterogeneity of the CNP system needs to be appreciated. Known forms today are CNP‐22, CNP‐53, and fragments from the N‐terminal part of proCNP (Figure). Although the C‐terminal peptides (the 22 and 53 forms) both are bioactive, that is, bind to the same receptor, ¹³ it is still unknown whether the intact proform, proCNP, possesses bioactivity. From an analytical perspective, there is currently no quality control system or a regulatory body that has reported guidelines or other helpful tools for assay validation: Thus, researchers engaging in CNP/proCNP measurement are strongly recommended to reach out to experts in the field with analytical experience in accurate measurement of both CNP and proCNP.

FOCUSING ON ENDOTHELIAL EXPRESSION

Measuring CNP in tissue extracts includes other cells in the tissue than the parenchymal cells of the organ in question. A vascular component will always be present in such extract analyses. CNP gene expression was identified in the vasculature early on, namely the endothelial cells lining the vasculature. ¹⁴ In that report on vascular CNP expression, the authors detected CNP immunoreactivity in both cultured endothelial cells (storage) and in the cell medium (release). Qualification of the CNP contents, likely to be CNP‐22 and CNP‐53, were further examined using high‐performance gel permeation chromatography. The endothelial expression was markedly stimulated by TGF‐β (transforming growth factor‐β) and taken together, a concept of a vascular NP system was introduced. From there, functional studies on endothelial CNP synthesis and release were initiated. Most studies have been performed using animal models, where different approaches to create vascular disease have been used. One example is an early report on rat carotid arteries subjected to intimal lesions induced by air drying. ¹⁵ Air drying causes intimal lesions that undergo thickening in the days after the lesion occurs. Administration of intravenous CNP markedly reduces the intimal thickening, which suggests that CNP may be a useful peptide in situations where intimal thickening is unwanted, as for example after coronary angioplasty. In parallel, rabbit models of vascular smooth muscle proliferation have also been used. ¹⁶ In this report, the native CNP receptor, NPR‐B (NP receptor B), was found both before and after injury. Interestingly, the NPR‐C receptor was upregulated after introducing the lesion, suggesting that not all effects of CNP are mediated via NPR‐B and the intracellular cGMP messenger system. Another study based on injured rabbit carotid arteries reported that adenovirus‐based CNP administration to the lesion affected local inflammation and shear‐stress‐related thrombosis and prevented neointima formation. ¹⁷ Finally, a recent report on mice devoid of vascular CNP expression has shown that atherosclerosis is augmented with diminished vascular regeneration after an ischemic event, ¹⁸ which is in line with a recent finding in mice where enhanced plague stability and alleviated macrophage inflammatory responses were observed with CNP treatment. ¹⁹ Moreover, a study of vascular smooth muscle cell (VSMC) phenotype at the single‐cell level showed that phenotypic modulation of contractile VSMCs to chondrocyte‐like plaque cells during atherogenesis is associated with a switch from ANP/NPR‐A to CNP/NPR‐B signaling. ²⁰ In addition, silencing of the CNP/NPR‐B axis in VSMCs results in an increase of chondrocyte‐like plaque cells. These results suggest that the CNP/NPR‐B/cGMP pathway plays an important role as an atherosclerosis‐protective regulator of modulated VSMCs, limiting their transition to chondrocyte‐like cells. Taken together, the CNP system seems to be involved in vascular biology beyond the systemic effects from cardiac NPs. ²¹ For a more elaborate presentation of vascular CNP, we refer to a recent comprehensive review on the topic. ²² One important note for now is that most CNP‐related biology in the vasculature is based on animal models. Thus, the transition from rodents to humans still constitutes a gap in CNP‐related research that needs attention. In simple terms, an induced lesion in an animal artery followed for days cannot readily be translated to human disease that develops over decades.

CNP IN CARDIOMETABOLIC DISEASE

Cardiometabolic disease as a concept sums up the long‐term effect of diabetes and obesity. In this context, diabetes refers to type 2 diabetes, which is often but not always associated with obesity and sedentary lifestyle. Although modern handling of type 2 diabetes has the pharmacological tools needed to restore near‐normal glucose levels, there are still major challenges with the long‐term complications. Notably, some of these complications involve both the macro‐ and microvasculature. In this context, it is surprising that metabolic disease and its effects on vascular CNP expression have not been eagerly pursued. Obesity and type 2 diabetes are conditions characterized by insulin resistance, which may partly be a result of reduced capillarization in skeletal muscle. ²³ , ²⁴ , ²⁵ Given the role of CNP as an essential regulator of angiogenesis and vascular remodeling, enhancing CNP signaling may hold the potential to improve insulin sensitivity in cardiometabolic disease, which is in line with reduced insulin levels observed in obese rats treated with CNP (unpublished results). To what extent CNP modulates insulin sensitivity in humans remains to be further explored.

There are few CNP biomarker studies where the analytical concerns raised earlier still need to be part of the interpretation. For obesity as such, it seems that CNP/proCNP concentrations in plasma decrease, a finding that parallels that of ANP and BNP. ²⁶ For diabetes without cardiac complications, there is a complete lack of published studies. In fact, diabetes seems to be a so‐called confounder in clinical cohort studies involving CNP/proCNP measurement, and later statistical analyses have adjusted for this factor. It thus seems reasonable to conclude that our present knowledge of the CNP system in relation to diabetes is a field in need of attention, a possible golden opportunity. Some simple questions need to be addressed first. To study the net effect of diabetes before manifest complications, patients initially receiving the diagnosis might offer such a possibility. Also, it would be interesting to evaluate type 2 diabetes in patients with “normal” body weight with standardized CNP/proCNP measurements. It should be doable: Data from the UKPDS (United Kingdom Prospective Diabetes Study) cohort suggest that 1 in 3 patients in the United Kingdom with a new diagnosis of type 2 diabetes has a body mass index <25 kg/m². ²⁷ In parallel, animal models of diabetes need to be evaluated for basic expression of the CNP system, as does the possible effect on circulating CNP and proCNP in human patients with type 2 diabetes related to the regulation of blood sugar. A focus on vascular function and CNP expression in patients with diabetes could increase our understanding of the activity of the system and, hence, whether treatment targeted towards this system might be something to pursue in future long‐term diabetes therapy beyond blood glucose regulation.

CNP IN ESSENTIAL HYPERTENSION

The CNP system has received scientific interest in connection with hypertension. CNP from vascular endothelial cells and NPR‐B from VSMCs have been assumed to be involved in the regulation of vascular tone. The functional significance of this system in the regulation of blood pressure in vivo using vascular endothelial cell‐specific CNP knockout mice and VSMC–specific NPR‐B knockout mice has been studied. Endothelial cell‐specific CNP knockout mice exhibit significantly increased blood pressures, enhanced endothelin‐1 gene expression, and increased plasma endothelin‐1 concentrations, thereby inducing a greater reduction in blood pressure in response to an endothelin receptor antagonist compared with their control littermates. By contrast, VSMC–specific NPR‐B knockout mice exhibited blood pressures similar to those in control mice. These results indicate that endothelium‐derived CNP contributes to the chronic regulation of vascular tone and systemic blood pressure by maintaining endothelial function independently of vascular smooth muscle GC‐B (guanylate cyclase‐B). ²⁸ In accordance, endothelial‐derived CNP has been shown to act on pericytes to regulate microcirculatory flow and blood pressure in mice. ²⁹ It is currently not known to which extent these mechanisms, reported in preclinical models, are also operating in humans.

Long‐term essential hypertension is characterized by stiffening of the arterial network, which further aggravates the hypertensive phenotype. Modern antihypertensive treatment primarily aims at lowering the systemic blood pressure, and although this also lowers the risk of complications, the unmet need that still exists is likely to reflect residual disease mechanisms not targeted by current pharmacotherapies. Here, CNP may hold a potential through its essential role in preserving vascular structure and integrity and ultimately the compliance of the vascular bed and control of systemic blood pressure. Interestingly, a genetic polymorphism in the CNP gene has been associated with early‐onset hypertension. ³⁰ In his study, the authors performed functional expression studies including the polymorphism located in the 3′UTR region of the gene. Unfortunately, they were not able to establish whether the mutation leads to, for instance, a more unstable mRNA and concordant reduced peptide expression. In biomarker studies on CNP and proCNP, hypertension often comes out as associated with increased concentrations in circulation, as for instance in an observational study on patients presenting with ST‐elevation myocardial infarction. ³¹ Although such elevations cannot directly be interpreted as only reflecting endothelial CNP expression, it does suggest that the CNP system may be involved in, or reacts to, the hypertensive phenotype. Biomarker studies in patients, however, will always include some degree of cardiac involvement, which partly could be the source of circulating CNP/proCNP measured in plasma. ³² Biomarker studies also reflect metabolism and elimination of peptides, which again may affect the plasma concentrations. In this regard, there is also an unmet need for basic studies addressing the elimination phase of the CNP/proCNP peptides beyond just establishing the general half‐life in circulation. Which organ degrades which peptide, and to what extent will organ function affect this extraction? For now, hypertension and the CNP system seem worthy of further pursuit. Biomarker studies specifically need to qualify the specific contribution of the vasculature to the circulating pool of CNP and proCNP, and this should include hypertension in various degrees of elevated blood pressure, for example, a more dose–response related approach. Also, the dynamics of the system should be assessed in patients starting antihypertensive treatment. Finally, it would be interesting to obtain arterial tissue from patients both with and without hypertension and study the response to various stimuli and lesions ex vivo.

LINKING THE CNP SYSTEM TO DIABETES

CNP biology per se has not received specific attention where diabetes is concerned. However, it has been suggested that the peptide system might be involved. From clinical studies using the specificity of proCNP measurement over CNP measurement, plasma concentrations are associated with the presence of diabetes ³¹ , ³³ and hypertension. Cohort studies have focused on cardiovascular disease and on risk of adverse events, traditional biomarker thinking. As such, clinical studies often need to adjust for the presence of diabetes and hypertension, but one could argue that the associations themselves might be worthy of pursuit. Long‐term diabetes affects many organs, which over time makes mechanistic speculation as to increased proCNP in circulation complicated in middle‐aged and older patients. Rather, the effect of diabetes on the CNP system should be evaluated in the initial stages of the disease, preferably long before organ complications become manifest. As already mentioned, the issue of overweight and obesity also needs to be isolated in diabetes, for instance in type 2 diabetes in patients with normal body mass index values. Finally, type 1 and type 2 diabetes share a few but critical phenotypic features, such as elevated glucose levels. From an exploratory point of view, it seems reasonable to examine the CNP system in type 1 diabetes, as confounders are less pronounced or even absent. To the best of our knowledge, however, no clinical studies have been published specifically on CNP/proCNP in patients with type 1 diabetes. One study measured CNP and proCNP in umbilical cord plasma of newborn infants of mothers with type 1 diabetes. ³⁴ In that study, however, CNP and proCNP were not measured in the mothers, and the findings cannot be extrapolated to patients with diabetes.

CONCLUSIONS

CNP remains the least understood of the NPs. Given its close relation to the vasculature in terms of production and release, we propose that this unique peptide system should be further explored in common vasculopathies, such as in cardiometabolic disease and in hypertension. Notably, we do not expect that the CNP system harbors clinically relevant biomarkers but rather may specify possibilities for future intervention. To make the CNP system a drug target in patients, however, we need basic information on the system in major diseases leading to vasculopathy. The biochemical tools for examining the CNP system have now been refined, and the time seems right for focusing on pathologies leading to cardiovascular disease rather than on the utility of this unique peptide system when cardiovascular disease has already precipitated.

Sources of Funding

None.

Disclosures

Jens P. Goetze has previously served as consultant for Novo Nordisk for biochemical method development. The remaining authors have no disclosures to report.