Coupling of endothelin receptors to the ERK/MAP kinase pathway. Roles of palmitoylation and G(alpha)q.

Endothelins are potent mitogens that stimulate extracellular signal-regulated kinases (ERK/MAP kinases) through their cognate G-protein-coupled receptors, ET(A) and ET(B). To address the role of post-translational ET receptor modifications such as acylation on ERK activation and to identify relevant downstream effectors coupling the ET receptor to the ERK signaling cascades we have constructed a panel of palmitoylation-deficient ET receptor mutants with differential G(alpha) protein binding capacity. Endothelin-1 stimulation of wild-type ET(A) or ET(B) induced a fivefold to sixfold increase in ERK in COS-7 and CHO cells whereas full-length nonpalmitoylated ET(A) and ET(B) mutants failed to stimulate ERK. A truncated ET(B) lacking the C-terminal tail domain including putative phosphorylation and arrestin binding site(s) but retaining the critical palmitoylation site(s) was still able to fully stimulate ERK activation. Using mutated ET receptors with selective G-protein-coupling we found that endothelin-induced stimulation of G(alpha)q, but not of G(alpha)i or G(alpha)s, is essential for endothelin-mediated ERK activation. Inhibition of protein kinases A and C or epidermal growth factor receptor kinase failed to prevent ET(A)- and ET(B)-mediated ERK activation whereas blockage of phospholipase C-beta completely abrogated endothelin-promoted ERK activation through ET(A) and ET(B) in recombinant COS-7 and native C6 cells. Complex formation of Ca2+ or inhibition of Src family tyrosine kinases prevented ET-1-induced ERK-2 activation in C6-cells. Our results indicate that endothelin-promoted ERK/MAPK activation criticially depends on palmitoylation but not on phosphorylation of ET receptors, and that the G(alpha)q/phospholipase C-beta/Ca2+/Src signaling cascade is necessary for efficient coupling of ET receptors to the ERK/MAPK pathway.

Evidence for downregulation of the endothelin-B-receptor by the use of fluorescent endothelin-1 and a fusion protein consisting of the endothelin-B-receptor and the green fluorescent protein.

We generated fusion proteins consisting of the endothelin-B (ET(B))-receptor and the enhanced green fluorescent protein (EGFP) to visualize receptor internalization. In Madin Darby canine kidney (MDCK) clones expressing ET(B)/EGFP fusion proteins, single class high affinity binding sites for [125I]endothelin-1 (ET-1) were found (for two different clones apparent K(D) values were 31 +/- 15 pM and 30 +/- 7 pM). Pretreatment of membranes with GTPgammaS prior to saturation analysis did not alter these values. We also labelled ET-1 with cyanine-dyes (Cy3/ET-1, Cy5/ET-1). In displacement analyses with membranes of MDCK ET(B)/EGFP clones using [125I]ET-1, we found reduced affinity for Cy3/ET-1 and Cy5/ET-1 (about 5- to 10-fold, respectively), but normal efficacy when compared to unlabelled ET-1. Both fluorescent ligands and the ET(B)/EGFP fusion protein were suitable for analysis of receptor trafficking in living cells and cells fixed at different timepoints. Laser scanning microscopy of MDCK ET(B)/EGFP clones incubated with Cy3/ET-1 or Cy5/ET-1 revealed rapid internalization of ligand/receptor complexes, which clustered in large, perinuclear structures (most probably late endosomes). Our data argue against recycling of the ET(B) receptor and favour its targeting to the lysosomal pathway.

Late endosomal/lysosomal targeting and lack of recycling of the ligand-occupied endothelin B receptor.

A fusion protein consisting of the endothelin B (ET(B)) receptor and the enhanced green fluorescent protein (EGFP) in conjunction with Cyanin3- or fluorescein-conjugated endothelin 1 (Cy3-ET1, Fluo-ET1) was used to investigate the ligand-mediated internalization of the ET(B) receptor. The ET(B) receptor and the ET(B)/EGFP fusion protein displayed very similar pharmacological properties when expressed in Chinese hamster ovary cells. The integrity of the fusion protein was verified by low temperature PAGE analysis of the (125)I-ET1-bound ET(B) receptor and the (125)I-ET1-bound ET(B)/EGFP fusion protein. Fluorescence microscopy of Chinese hamster ovary cells expressing the ET(B)/EGFP fusion protein demonstrated strong signals at the plasma membrane. On addition of Cy3-ET1, internalization of ligand and receptor occurred within 5 min via a sucrose-sensitive (i.e., clathrin-mediated) pathway. On further incubation, ET(B)/EGFP and Cy3-ET1 fluorescences were found in the perinuclear region, colocalized with fluorescent low density lipoproteins, a marker of the late endosomal/lysosomal pathway, but not with fluorescent transferrin, a marker of the recycling pathway. No dissociation of Cy3-ET1 from the receptor was seen within 4 h. Using (125)I-ET1 or Cy3-ET1, binding sites were again demonstrable at the cell surface within 2 h. The reappearance of binding sites was abolished by prior treatment of the cells with cycloheximide, an inhibitor of protein synthesis. The data demonstrate that the ligand-occupied ET(B) receptor is internalized; however, it does not recycle like most of the G protein-coupled receptors but is sorted to the late endosomal/lysosomal pathway in a manner similar to that of the family of protease-activated receptors.

Palmitoylation of endothelin receptor A. Differential modulation of signal transduction activity by post-translational modification.

Post-translational modifications such as phosphorylation and palmitoylation play important roles for the function and regulation of receptors coupled to heterotrimeric guanyl nucleotide-binding proteins. Here we demonstrate that the human endothelin receptor A (ETA) incorporates [3H]palmitate. Mutation of a cluster of five cysteine residues present in the cytoplasmic tail of ETA into serine or alanine residues completely prevented palmitoylation of the receptor. The ligand binding affinity of the non-palmitoylated ETA mutants was essentially unchanged as compared to the palmitoylated wild type ETA suggesting that the replacement of the cysteine residues did not alter the overall structure of the receptor. Furthermore, the ligand-induced stimulation of adenylyl cyclase by the mutant ETA was unaffected by the mutation. In contrast, the mutated non-palmitoylated receptors but not the wild type receptor failed to stimulate phosphatidylinositol hydrolysis by phospholipase C activation upon challenge by endothelin-1. Furthermore, the mutant receptors failed to stimulate the ligand-induced transient increase in the cytoplasmic calcium seen with the wild type ETA. Endothelin-1 induced mitogenic stimuli via the wild type receptors but not through the mutated receptors suggesting an important role for phospholipase C in this signal transduction pathway. The differential regulation of distinct signal transduction pathways by post-translational modification suggests that palmitoylation of the ETA provides a novel mechanism of modulating ETA receptor activity.

The CMS-associated 16 kDa protein encoded by orfH522 in the PET1 cytoplasm is also present in other male-sterile cytoplasms of sunflower.

In sunflower plants carrying the PET1 cytoplasm male sterility (CMS) is associated with a new open reading frame (orfH522) in the 3'-flanking region of the atpA gene and an additional 16 kDa protein. Twenty-seven male-sterile cytoplasms of different origin were studied for the expression of the 16 kDa protein. In addition to the PET1 cytoplasm nine other male-sterile cytoplasms express the CMS-associated protein. These CMS sources originate from different interspecific crosses, from spontaneously occurring male-sterile plants in wild sunflower and from induced mutagenesis. Polyclonal antisera were raised against fusion proteins which contain 421 bp of the 3'-coding region of orfH522 to verify by immunological methods the identity of the other CMS cytoplasms. The anti-ORFH522 antiserum showed a positive reaction in the immunoblot with all CMS cytoplasms which expressing the 16 kDa protein. Investigations of the mitochondrial DNA demonstrated that all ten CMS cytoplasms which express the 16 kDa protein have the same organization at the atpA locus. OrfH522 as probes gave the same transcript pattern for the investigated CMS cytoplasms, just as for PET1. The MAX1 cytoplasm has an orfH522-related sequence but does not synthesize the 16 kDa protein. Using the sodium carbonate treatment the 16 kDa protein proved to be membrane-bound. Computer analyses predict that the hydrophobic N-terminal region of ORFH522 may form a transmembrane helix functioning as membrane anchor.