We have isolated and characterized ␣-conotoxin EpI, a novel sulfated peptide from the venom of the molluscivorous snail, Conus episcopatus. The peptide was classified as an ␣-conotoxin based on sequence, disulfide connectivity, and pharmacological target. EpI has homology to sequences of previously described ␣-conotoxins, particularly PnIA, PnIB, and ImI. However, EpI differs from previously reported conotoxins in that it has a sulfotyrosine residue, identified by amino acid analysis and mass spectrometry. Native EpI was shown to coelute with synthetic EpI. The peptide sequence is consistent with most, but not all, recognized criteria for predicting tyrosine sulfation sites in proteins and peptides.

Although G protein-coupled receptors are often categorized in terms of their primary coupling to a given type of G␣ protein subunit, it is now well established that many show promiscuous coupling and activate multiple signaling pathways. Furthermore, some agonists selectively activate signaling pathways by promoting interaction between distinct receptor conformational states and particular G␣ subunits or alternative signaling proteins. We have tested the capacity of agonists to stimulate Ca 2ϩ release, cAMP accumulation, and changes in extracellular acidification rate (ECAR) at the human ␣ 1A -adrenoceptor. Signaling bias factors were determined by novel application of an operational model of agonism and compared with the reference endogenous agonist norepinephrine; values significantly different from 1.0 indicated an agonist that promoted receptor conformations distinct from that favored by norepinephrine. Oxymetazoline was a full agonist for ECAR and a partial agonist for Ca 2ϩ release (bias factor 8.2) but failed to stimulate cAMP production. Phenylephrine showed substantial bias toward ECAR versus Ca 2ϩ release or cAMP accumulation (bias factors 21 and 33, respectively) but did not display bias between Ca 2ϩ and cAMP pathways. Cirazoline and N-[5-(4,5-dihydro-1H-imidazol-2-yl)-2-hydroxy-5,6,7,8-tetrahydronaphthalen-1-yl]methanesulfonamide (A61603) displayed bias toward cAMP relative to Ca 2ϩ release (bias factors of 7.4 and 8.6). It is noteworthy that epinephrine, a second endogenous adrenoceptor agonist, did not display bias relative to norepinephrine. Our finding that phenylephrine displayed significant signaling bias, despite being highly similar in structure to epinephrine, indicates that subtle differences in agonist-receptor interaction can affect conformational changes in cytoplasmic domains and thereby modulate the repertoire of effector proteins that are activated.

BACKGROUND AND PURPOSEAgonists acting at GPCRs promote biased signalling via Gα or Gβγ subunits, GPCR kinases and β-arrestins. Since the demonstration of biased agonism has implications for drug discovery, it is essential to consider confounding factors contributing to bias. We have examined bias at human α 1A -adrenoceptors stably expressed at low levels in CHO-K1 cells, identifying off-target effects at endogenous receptors that contribute to ERK1/2 phosphorylation in response to the agonist oxymetazoline. EXPERIMENTAL APPROACH Intracellular Ca2+ mobilization was monitored in a Flexstation® using Fluo 4-AM. The accumulation of cAMP and ERK1/2 phosphorylation were measured using AlphaScreen® proximity assays, and mRNA expression was measured by RT-qPCR. Ligand bias was determined using the operational model of agonism. KEY RESULTSNoradrenaline, phenylephrine, methoxamine and A61603 increased Ca 2+ mobilization, cAMP accumulation and ERK1/2 phosphorylation. However, oxymetazoline showed low efficacy for Ca +2 mobilization, no effect on cAMP generation and high efficacy for ERK1/2 phosphorylation. The apparent functional selectivity of oxymetazoline towards ERK1/2 was related to offtarget effects at 5-HT 1B receptors endogenously expressed in CHO-K1 cells. Phenylephrine and methoxamine showed genuine bias towards ERK1/2 phosphorylation compared to Ca 2+ and cAMP pathways, whereas A61603 displayed bias towards cAMP accumulation compared to ERK1/2 phosphorylation. CONCLUSION AND IMPLICATIONSWe have shown that while adrenergic agonists display bias at human α 1A -adrenoceptors, the marked bias of oxymetazoline for ERK1/2 phosphorylation originates from off-target effects. Commonly used cell lines express a repertoire of endogenous GPCRs that may confound studies on biased agonism at recombinant receptors. AbbreviationsECAR, extracellular acidification rate; HEAT, 2-(β-4-hydroxyphenyl)ethylaminomethyltetralone; PTX, pertussis toxin

No abstract

Abstract:The activity of ␣-conotoxin (␣-CTX) ImI, from the vermivorous marine snail Conus imperialis, has been studied on mammalian nicotinic receptors on bovine chromaffin cells and at the rat neuromuscular junction. Synthetic ␣-CTX ImI was a potent inhibitor of the neuronal nicotinic response in bovine adrenal chromaffin cells (IC 50 ϭ 2.5 M, log IC 50 ϭ 0.4 Ϯ 0.07), showing competitive inhibition of nicotine-evoked catecholamine secretion. ␣-CTX ImI also inhibited nicotine-evoked 45 Ca 2ϩ uptake but not 45 Ca 2ϩ uptake stimulated by 56 mM K ϩ . In contrast, ␣-CTX ImI had no effect at the neuromuscular junction over the concentration range 1-20 M. Bovine chromaffin cells are known to contain the ␣3␤4, ␣7, and (possibly) ␣3␤4␣5 subtypes. However, the secretory response of bovine chromaffin cells is not inhibited by ␣-bungarotoxin, indicating that ␣7 nicotinic receptors are not involved. We propose that ␣-CTX ImI interacts selectively with the functional (␣3␤4 or ␣3␤4␣5) nicotinic acetylcholine receptor to inhibit the neuronal-type nicotinic response in bovine chromaffin cells. Key Words: Neuronal nicotinic receptor-␣-Conotoxin ImI-␣-Bungarotoxin-␣7 nicotinic receptors.