Corin in Natriuretic Peptide Processing and Hypertension

Abstract

Corin is a serine protease originally isolated from the heart. Functional studies show that corin is the long-sought enzyme responsible for activating cardiac natriuretic peptides. In mice, lack of corin prevents the natriuretic peptide processing, causing salt-sensitive hypertension. In humans, corin variants and mutations that reduce corin activity have been identified in patients with hypertension and heart failure. Decreased plasma levels of corin antigen and activity have been reported in patients with heart failure and coronary artery disease. Low levels of urinary corin also have been found in patients with chronic kidney disease. Most recent studies show that corin also acts in the uterus to promote spiral artery remodeling and prevent pregnancy-induced hypertension. Here we review the role of corin in natriuretic peptide processing and cardiovascular diseases such as hypertension, heart disease, pre-eclampsia, and chronic kidney disease.

Introduction

Natriuretic peptides are important hormones conserved in all vertebrates [1]. In mammals, the natriuretic peptide family has three members, i.e. atrial natriuretic peptide (ANP), brain or B-type natriuretic peptide (BNP) and C-type natriuretic peptide (CNP) [2, 3]. ANP and BNP are produced in cardiomyocytes and function to regulate salt-water balance and blood pressure. Upon binding to their receptor, natriuretic peptide receptor-A (NPR-A), ANP and BNP promote renal sodium excretion and relax vascular smooth muscles. To date, ANP and NPR-A gene variants and mutations have been reported in patients with hypertension and cardiac hypertrophy [4, 5*, 6*, 7–9, 10*]. ANP and BNP also play a role in regulating energy metabolism by stimulating fat oxidation in skeletal muscles and lipolysis in adipocytes [11*, 12*, 13]. Under pathological conditions, such as heart failure, ANP and BNP expression is highly up-regulated, which serves as a compensatory mechanism to lower blood pressure and volume. Currently, ANP, BNP and their related peptides are used as biomarkers in assessing heart failure [14].

CNP differs from ANP and BNP in many aspects including tissue distribution and biological function. CNP is made mainly in vascular endothelial cells, smooth muscles and chondrocytes, where it acts through its receptor, natriuretic peptide receptor-B (NPR-B), to regulate cell growth, vascular remodeling and bone differentiation [15, 16]. Defects in CNP and NPR-B cause skeletal abnormalities [17–20]. CNP also functions in the reproductive system. In ovaries, for example, CNP has been shown to promote follicle development and regulate oocyte maturation [21, 22]. The expression of CNP in peripheral neurons also has been reported [23]. Recent studies show that CNP may function in the gastrointestinal track to stimulate intestinal motility [24, 25].

Peptide hormones commonly are synthesized in precursor forms, which are processed to mature forms by proteolytic enzymes either in specific intracellular compartments or extracellularly. This step is essential for activating the peptide hormones. Similarly, the natriuretic peptides are made as prepropeptides [19, 26]. Signal peptidase moves the prepeptide in the ER to generate pro-natriuretic peptides, which remain inactive. In recent years, corin, a serine protease identified in the heart [27], has been shown to play a critical role in processing natriuretic peptides. In this review, we will focus on recent findings of corin function and discuss the role of corin in hypertension and kidney disease.

Corin and natriuretic peptide processing

Corin is a trypsin-like protease highly expressed in cardiomyocytes [27–29]. Unlike trypsin, which is a secreted soluble protein, corin has an N-terminal transmembrane domain tethering corin on the cell surface [27]. The catalytic protease domain of corin is located at the C-terminus. Between these two domains, there are two frizzled-like domains, eight LDL receptor-like repeats and a scavenger receptor-like domain [27]. The overall corin domain structure and membrane topology are similar to those of type II transmembranse serine proteases [30]. Corin, however, is the only trypsin-like serine protease known to contain frizzled-like domains. The molecular biology and biochemical properties of corin have been described in previous reviews [26, 31].

The primary function of corin in the heart is to activate pro-ANP [32–34]. When pro-ANP is released from the dense granules of cardiomyocytes, corin cleaves pro-ANP at residue Arg-98, producing a 26-amino-acid C-terminal active ANP. Corin also participates in the conversion of pro-BNP to BNP [33, 35–38]. This activity, however, is not corin-specific. The proprotein convertase furin also activates pro-BNP [37–40]. Remarkably, O-glycans in the pro-BNP propeptide inhibit corin- and furin-mediated pro-BNP processing [37, 41]. It appears that O-glycans, which are absent in pro-ANP and pro-CNP [42], may have a specific role in regulating BNP production. To date, there is no evidence to indicate a role of corin in pro-CNP processing. Furin appears to be the primary enzyme responsible for pro-CNP processing [43]. Thus, the enzymes responsible for processing pro-ANP, pro-BNP and pro-CNP differ considerably, even though the genes encoding these peptides were derived from a common origin. Probably, the divergence reflects specific regulatory requirements, as the peptides evolve to perform separate functions in different cell types.

Mouse models of corin deficiency

The importance of corin in controlling natriuretic peptide production and blood pressure has been studied in knockout (ko) mice [44]. Despite corin mRNA is expressed in embryonic hearts as early as day E7.5 [27], corin ko mice had no apparent developmental defects [44]. In this regard, corin appears similar to several other type II transmembrane serine proteases, such as hepsin [45] and HAT [46], which are dispensable for embryonic development and postnatal survival.

In biochemical analysis, heart tissues from corin ko mice were found to contain high levels of unprocessed pro-ANP but no detectable amounts of mature ANP, reflecting a defect in pro-ANP processing [44]. Intravenous injection of a soluble active corin into the ko mice restored pro-ANP processing and elevated plasma cGMP levels [44]. The results show that corin is essential for pro-ANP processing and that its activity cannot be compensated by other proteases in vivo.

Natriuretic peptides are critical for normal blood pressure. In mice, lacking ANP or NPR-A leads to hypertension [47, 48]. A hypertensive phenotype also was found in corin ko mice [44, 49]. When the ko mice were challenged with high-salt diets, their blood pressure elevated further, indicating that the hypertension in corin ko mice was salt-sensitive [49]. Thus, the hypertensive phenotypes in corin, ANP and NPR-A ko mice were similar, supporting a critical role of corin in activating the ANP pathway in vivo.

In addition to the ko mice, corin deficiency has been reported in a naturally-occurring mutant mouse strain, C57BL/6-Kit^W-sh (W^sh), known for mast cell-deficiency due to an abnormal Kit locus [50]. Genomic sequencing revealed a genetic inversion disrupting the transcriptional regulatory region upstream of the c-kit gene, which accounts for the hematopoietic defects in these mice [51]. Interestingly, the generic inversion also disrupted the corin gene between exons 5 and 6. As a result, W^sh mice lacked corin mRNA expression and had high levels of unprocessed pro-ANP in the heart [51, 52]. The mice developed cardiac hypertrophy and had poor cardiac function [51, 52]. Similar pathological findings also were reported in corin ko mice [44, 53].

Unexpectedly, corin expression was detected in the dermal papilla of hair follicles in mice and humans [54, 55]. The biological significance of corin in the skin is not completely understood. In mice, corin appears involved in coat color regulation, as corin ko mice had a lighter yellowish color than that of wild-type mice [54]. This skin phenotype, however, occurred only when mice had a functional agouti allele [54], suggesting that corin may interact with a molecule in the agouti pathway to regulate hair pigmentation. To date, the specific corin substrate in the mouse skin has not been defined. It remains to be determined if corin has a similar role in human skin biology.

Human corin variants and mutations

The human corin gene is on chromosome 4p12–13, consisting of 22 exons [56]. To date, human corin gene variants have been identified. Dries et al. reported a corin variant allele (T555I/Q568P) in African Americans with hypertension and heart disease [57]. Individuals with this variant allele had severe cardiac hypertrophy and high levels of unprocessed natriuretic peptides in blood [58, 59], suggesting that corin protein encoded by this variant allele may be defective. This hypothesis was supported by biochemical studies, in which recombinant corin variant T555I/Q568P was shown to process pro-ANP and pro-BNP poorly [60]. Apparently, the amino acid changes caused by the gene variant in the propeptide region altered corin protein conformation and inhibited corin zymogen activation [60].

Wang et al. further examined the effect of this corin variant on blood pressure by generating a transgenic mouse model that expressed T555I/Q568P variant in a corin null background [53*]. In these mice, corin activity was significantly reduced, resulting in high levels of pro-ANP in the heart. The results confirmed that the corin variant is defective in vivo. More importantly, the transgenic mice developed hypertension and cardiac hypertrophy, which were exacerbated upon high salt-diet challenge [53*]. The overall hypertensive phenotype of the transgenic mice mimics the clinical features in the African Americans carrying the variant allele [57, 58]. These data suggest that the corin variant allele, which is present in ~10–12% of African Americans [57], may contribute to hypertension and heart disease in this high-risk population.

The corin T555I/Q568P variant appears to originate in Africa, as this allele is present mostly in African Americans but not in other ethnic groups [57]. Recently, Dong et al. reported a C-to-T mutation in exon 12 of the corin gene in a Chinese patient family of hypertension [61]. The mutation resulted in an Arg-to-Cys change at residue 539 in corin frizzled-2 domain. Within this family, individuals carrying the mutantion had high systolic and/or diastolic blood pressure. In functional studies, the R539C mutant had reduced pro-ANP processing activity and exhibited a dominant-negative effect on wild-type corin. It appeared that the mutant Cys formed an alternative disulfide bond [61], altering the conformation of corin frizzled-2 domain, which is required for interacting with pro-ANP [62]. The R539C mutation also caused corin self-cleavage, producing an alternative inactive fragment [61]. These results indicate that genetic mutations reducing corin activity may represent a molecular mechanism underlying hypertension. As more genetic studies are conducted, additional corin mutations are expected to be identified in hypertensive patients.

Corin in patients with heart disease

Hypertension is a major risk factor for heart disease. In patients with heart failure, plasma levels of unprocessed natriuretic peptides are highly elevated [63–65], suggesting that corin activity could not be compensated adequately to meet the demand under the pathological condition [66–68]. Presently, how corin expression and activity are regulated in the heart is not well understood. Studies indicate that a number of regulatory mechanisms may be involved, including transcriptional control [56], cell surface targeting [69], N-glycosylation [70, 71], and zymogen activation [60, 72].

In cell-based studies, active corin molecules on the cell surface were shown to undergo proteolytic shedding [73], a process that has been found in many membrane-bound proteases [30, 74]. Understandably, excessive proteolytic activities in tissues could be detrimental. It is possible that the shedding process may serve as a mechanism to control corin activity in the heart. In principle, shed corin fragments could be degraded in the heart, which is known containing many proteolytic enzymes [75]. It is also possible that some of the corin fragments may enter the circulation.

Indeed, corin antigen and activity have been detected in human blood [36, 76, 77]. Dong et al. showed that plasma corin levels decreased progressively in patients with late stages of heart failure [78]. Similar findings of reduced corin antigen and activity levels in heart failure patients were reported in another study [79]. Most recently, low serum corin levels were found to be an independent predictor for poor clinical outcomes in patients with coronary disease [80*]. These results are consistent with the elevated levels of unprocessed natriuretic peptides in patients with heart disease, suggesting that corin deficiency may be a contributing factor in failing hearts [81]. In supporting this hypothesis, a recent study showed that corin overexpression improved cardiac function, reduced pulmonary edema, and increased survival in a mouse model of dilated cardiomyopathy [82*]. The data also support a possible therapeutic strategy to enhance corin activity to treat patients with heart disease.

Corin in pre-eclampsia

Pre-eclampsia is a serious complication in pregnancy, afflicting millions of women worldwide [83]. It has been long suspected that a defective uteroplacental interface is a primary factor underlying the disease [84, 85]. Among pathologic findings in pre-eclamptic patients, delayed trophoblast invasion and poorly remodeled spiral arteries in the uterus are common [84, 85]. Possibly, the maternal hypertension reflects a compensatory response attempting to increase blood flow to the ischemic placenta caused by narrow uterine spiral arteries. To date, the mechanism underlying such uteroplacental defects in pre-eclamptic patients remains poorly understood.

Interestingly, corin expression was found in the pregnant uterus in mice and humans [27, 86, 87], suggesting a possible corin function in pregnancy. Studies have shown that ANP and NPR-A are expressed in the uterus, where ANP antagonized uteroplacental vessel contraction and stimulated myometrial relaxation [88]. ANP also is known to promote vessel wall remodeling in angiogenic processes [89, 90]. In a matrigel-based invasion study, ANP was shown to enhance human trophoblast transmigration [86**]. These data suggest that corin and ANP may participate in tissue and vascular remodeling in the pregnant uterus, which is necessary for developing a healthy maternal-fetal interface.

Consistent with this hypothesis, pregnant ko mice lacking either corin or ANP had delayed trophoblast invasion and poorly remodeled spiral arteries in the uterus [86**]. The mice also developed late gestational hypertension and proteinuria, which was alleviated once pups were delivered [86**]. The overall phenotype of the pregnant corin and ANP ko mice resembled pathological features in pre-eclamptic patients [86, 91]. Importantly, low levels of uterine corin expression were found in pre-eclamptic patients [86**]. Moreover, missense mutations in the corin gene were identified in patients with pre-eclampsia. In functional studies, the corin mutants were shown to have markedly reduced activity in processing pro-ANP [86**]. Paradoxically, plasma corin levels were increased in patients with pre-eclampsia [86, 92]. The results suggest that the observed role of corin in the uterus was likely mediated by locally produced, but not heart-derived, corin and ANP. Together, these findings suggest that defects in the corin and ANP pathway may be a novel mechanism underlying pre-eclampsia.

Corin in kidney disease

In mice, corin mRNA expression was detected in the medulla of developing kidneys [27]. In rat and human kidneys, corin protein was found in the proximal tubule, thick ascending limb, connecting tubule, and collecting duct [36, 93]. In rat models of proteinuric kidney disease that were associated with sodium retention, renal corin mRNA and protein expression was found to be markedly reduced and the reduction was associated with increased renal -ENaC expression [93]. These results are intriguing because ENaC is known to play a critical role in renal sodium reabsorption. The findings from these rat model studies suggest that corin may function in the kidney to promote sodium excretion in an ENaC-dependent manner and that impaired corin expression and/or function may contribute to salt and water retention in kidney disease [94].

Consistent with these findings, corin ko mice were shown to have an impaired response to promote sodium excretion when dietary salt contents increased, resulting in sodium and water retention and exacerbated hypertension [49]. When corin ko mice on high-salt diet were treated with an ENaC inhibitor, amiloride, renal sodium excretion was significantly improved, leading to lowered blood pressure and reduced body weight [49]. The results point to an important role of corin and ANP in regulating renal sodium excretion and body fluid balance in response to fluctuating dietary salt levels [49, 95]. In migratory fish species such as eel and salmon that live in both salty and fresh water environments, natriuretic peptides are essential for maintaining electrolyte balance [1]. Apparently, such a physiological mechanism is well preserved in mammals, even though their living environments are drastically different from that of fish.

At this time, it is unknown if renal and cardiac corin expression is controlled under similar transcriptional mechanism. Interestingly, corin protein expressed on renal epithelial cells also undergoes a shedding process. Fang et al. reported soluble corin in human urine samples [96]. In patients with chronic kidney disease (CKD), urinary corin levels were significantly lower than that in normal controls [96]. The low levels of urinary corin were associated with hypertension and poor renal function in these patients. By immunostaining, reduced corin protein levels also were found in kidney biopsies in CKD patients [96]. The results suggest that reduced renal corin protein expression and/or activity may be a contributing factor in the pathogenesis of CKD. These findings are consistent with previous reports of fibrosis, tubular dilation and enhanced local inflammation in the kidney of NPR-A ko mice [97].

Conclusions

Hypertension is a major cardiovascular disease. As proposed by Irvine H. Page in his "mosaic theory of hypertension" more than half a century ago [98], a steady state exists in the circulation in which the important regulatory factors are in equilibrium to maintain blood pressure. To date, it is well accepted that environmental, behavioral and genetic factors disrupting such equilibrium can cause hypertension.

The natriuretic peptides act as an important endocrine system, connecting the heart and kidney in maintaining salt-water balance and blood pressure. Most recent studies show that ANP may also participate in a gut-heart cross-talk in regulating blood pressure and energy metabolism [99, 100**]. Corin was discovered as a novel cardiac protease [27]. Identification of corin as the long-sought pro-ANP convertase provides important insights into the biochemical mechanism underlying natriuretic peptide processing [26] (Fig. 1). As more studies are being conducted, we now know that corin acts not only in the heart, but also in many other tissues including the kidney, uterus and skin. Recent animal models and human genetic studies have shown that corin plays a critical role in major cardiovascular diseases, including hypertension, heart disease, pre-eclampsia and kidney disease (Fig. 1). These findings are expected to stimulate more studies to understand the biology of corin and its role in cardiovascular disease. Such studies may also help to translate basic discoveries in corin and natriuretic peptide research into novel therapies to treat hypertension and heart disease in patients.

Fig. 1.

Corin and ANP in cardiovascular biology and disease. Corin converts pro-ANP to ANP, which in return enhances cardiac function, renal sodium excretion, vasodilation, and uterine spiral artery remodeling. Defects in the corin and ANP pathway may lead to major diseases, such as cardiac hypertrophy (CH), heart failure (HF), hypertension (HP), chronic kidney disease (CKD), and pre-eclampsia (PE).