Endothelin

Abstract

The endothelins comprise three structurally similar 21-amino acid peptides. Endothelin-1 and -2 activate two G-protein coupled receptors, ETA and ETB, with equal affinity, whereas endothelin-3 has a lower affinity for the ETA subtype. Genes encoding the peptides are present only among vertebrates. The ligand-receptor signaling pathway is a vertebrate innovation and may reflect the evolution of endothelin-1 as the most potent vasoconstrictor in the human cardiovascular system with remarkably long lasting action. Highly selective peptide ETA and ETB antagonists and ETB agonists together with radiolabeled analogs have accurately delineated endothelin pharmacology in humans and animal models, although surprisingly no ETA agonist has been discovered. ET antagonists (bosentan, ambrisentan) have revolutionized the treatment of pulmonary arterial hypertension, with the next generation of antagonists exhibiting improved efficacy (macitentan). Clinical trials continue to explore new applications, particularly in renal failure and for reducing proteinuria in diabetic nephropathy. Translational studies suggest a potential benefit of ETB agonists in chemotherapy and neuroprotection. However, demonstrating clinical efficacy of combined inhibitors of the endothelin converting enzyme and neutral endopeptidase has proved elusive. Over 28 genetic modifications have been made to the ET system in mice through global or cell-specific knockouts, knock ins, or alterations in gene expression of endothelin ligands or their target receptors. These studies have identified key roles for the endothelin isoforms and new therapeutic targets in development, fluid-electrolyte homeostasis, and cardiovascular and neuronal function. For the future, novel pharmacological strategies are emerging via small molecule epigenetic modulators, biologicals such as ETB monoclonal antibodies and the potential of signaling pathway biased agonists and antagonists.

Fig. 1.

Long lasting vasoconstrictor response to 10 nM ET-1 in human mammary artery is maintained for over 2 hours but can be reversed by the physiologic antagonist nitric oxide derived from a nitric oxide donor (A) or by the ET_A antagonist PD156707 but not the ET_B antagonist BQ788 (B).

Fig. 2.

Schematic diagram showing the sequences of ET, Big ET precursors and sarafotoxin peptides. Amino acids that differ between ET-1 and mature peptides or between Big ET-1 and precursors are shown in blue. Labeling of [¹²⁵I] radioligands at Tyr residues are shown by a filled red circle. Residues that are thought from X-ray crystallography to be aligned as a stripe as a result of a secondary helical structure secondary helical structure are indicated with a star (Janes et al., 1994; Orry and Wallace, 2000). These residues have also been shown experimentally to be crucial for binding (Huggins et al., 1993). Arrows indicate the site for cleavage of Big ETs to mature peptides: between Trp²¹-Val²² amino acids for Big ET-1 and ET-2 and between Trp²¹-Ile²² for ET-1 and for Big ET-3.

Fig. 3.

Schematic diagram illustrating synthesis of ET peptides and interaction with receptors. ET isoforms are synthesized by a three-step process: Messenger RNA encodes a prepro-peptide that after a proteolytic cleavage initial of the signal peptidase to yield the propeptide is further cleaved by furin to Big ET precursors. Synthesis of ET-1 has been studied in the most detail. Transformation to the mature, biologically active ET-1 is mainly by the action of ECE-1 at pH 7 but also by ECE-2 at pH 5.5 within endothelial cells. On release from endothelial cells, about one in five molecules of Big ET-1 escape conversion but further processing to ET-1 may occur by smooth muscle ECE or via alternative pathways catalyzed by chymase for ET-1. It is inferred that ET-2 from endothelial cells is synthesized by a similar pathway. ET-3 is not released from human endothelial cells but is also synthesized in other cells by ECE-1 with evidence for an additional pathway, KELL. ET-1 is metabolized by NEP to inactive metabolites. ETs mediate their actions via two GPCRs ET_A or ET_B and the recommended agonists and antagonists that are most widely used are indicated together with ECE inhibitors.

Fig. 4.

ET-1 is synthesized in human endothelial cells by a dual secretory pathway. ET-1 is continuously released from the small vesicles of the constitutive pathway to interact with ET receptors to contribute to vasomotor tone. ET-1 is also released from the regulated pathway in response to external stimuli from Weibel-Palade bodies that are unique to endothelial cells. In the human vasculature, ET-1 released abluminally from endothelial cells, interacts mainly with ET_A receptors on the underlying smooth muscle, with a small population of ET_B receptors also mediating constriction in some but not all vessels. In animals, vasoconstriction can be via ET_B or a mixture of both depending on the vascular bed. The ET-1/ET_A complex undergoes internalization to the endosome before recycling of the receptor to the cell surface and provides a mechanism whereby ET_A antagonist can reverse an ET-1 response. Some ET-1 may also interact with endothelial ET_B receptors in an autocrine manner and limit the constrictor response by the release of vasodilators such as nitric oxide. ET-1/ET_B complex is internalized and degraded to the lysosome; as a result ET_B antagonists are unable to displace receptor bound ligand. Low levels of ET-1 and Big ET-1 that has escaped conversion can also be detected in the plasma. Big ET-1 can also undergo further conversion by smooth muscle ECE to the mature peptide.

Fig. 5.

Schematic diagram of the human ET_A receptor. Experimental mutations altering receptor function are shown in red. Potential sites for translational modifications are shown as phosphorylation (yellow), glycosylation (gray), and palmitoylation (purple). Diagrams generated using http://tools.gpcr.org/ (Isberg et al., 2014).

Fig. 6.

Schematic diagram of the human ET_B receptor. Naturally occurring mutation reported in patients with Hirschsprung disease (blue), experimental mutations altering receptor function (red). Potential sites for translational modifications as shown as phosphorylation (yellow), glycosylation (gray), and palmitoylation (purple). Green indicates both a phosphoraylation and site experimentally mutated. Diagrams generated using http://tools.gpcr.org/ (Isberg et al., 2014).

Fig. 7.

Relative expression of mRNA encoding ET_A (Ednra) or ET_B (Ednrb) receptors in 41 adult tissues. (Graphs constructed using data from Regard et al., 2008).

Fig. 8.

Ratio of ET_A to ET_B densities measured by saturation binding assays in the human brain, kidney, lung, liver, and heart and smooth muscle layer of the vessels from each organ.

Fig. 9.

Comparison of the selectivity of ET ligands for native ET_A and ET_B receptors measured in the same competition binding assay against [¹²⁵I]-ET-1. This assay uses human heart, which expresses both subtypes, resulting in a biphasic competition curve. Measurement of the affinity (the equilibrium dissociation constant or K_D) of each compound at the two receptor sub-types can be accurately determined using non-linear iterative curve fitting. Comparison of the two affinities provides a measure of the selectivity for each subtype (Maguire and Davenport, 2012).

Fig. 10.

Spectrum of selectivity of antagonists for ET_A versus ET_B receptors as reported by the companies that discovered the compounds mainly based on measuring affinity constants in separate competition assays against [¹²⁵I]-ET-1 using human recombinant ET_A versus ET_B receptors. Antagonists are classified as either selective for one subtype or alternatively as mixed antagonists that block both receptors. The classification is usually made by the manufacturer, but there is no agreed definition and there are anomalies. We proposed that antagonists that are ET_A-selective should display more than 100-fold selectivity for the ET_A subtype and those that block both ET_A and ET_B (mixed antagonists) should demonstrate less than 100-fold ET_A selectivity. The rational for this classification is based on receptor occupancy where a concentration can be calculated to block virtually all ET_A receptors but not ET_B for in vitro studies. The threshold is indicated by the lower dashed line. It is difficult to achieve exact plasma concentration in vivo and to achieve selectivity; compounds with at least two orders of magnitude are more useful. Bosentan, ambrisentan, and macitentan are currently approved for clinical use and are highlighted. Note that the manufacturers of macitentan classify the antagonist as a mixed antagonist although it displays ET_A selectivity. Sparsentan is a dual ET_A/angiotensin II receptor type 1 receptor antagonist (modified from Maguire and Davenport, 2014).

Fig. 11.

Pipeline of ET receptor antagonists measured by papers in PubMed (blue bars), papers with ChEMBL entries in Europe PubMed Central (red bars) and patents (green bars) in PATENTSCOPE with ET receptor antagonists combined with patent classification C07D (code for medicinal chemistry) (Bento et al. 2014).

Fig. 12.

Three-dimensional alignments shown as a top view (A) and the side view (B) of the 11-molecule cluster of ET antagonists using ChemAxon Marvin tool at default settings to perform a flexible overlap of atoms via on-the-fly adjustments. The following 11 were selected as having high affinity: peptide antagonist FR139317 and nonpeptide antagonists, A127722, ambrisentan, avosentan, darusentan, macitentan, PD156707, PD164333, SB234551, sitaxentan, and zibotentan. Top (C) and side view (D) of a three-molecule cluster of ET_B peptide antagonist, BQ788, and nonpeptide antagonists A192621 and RO468443.