Design and construction of conformational biosensors to monitor ion channel activation: A prototype FlAsH/BRET-approach to Kir3 channels.

Ion channels play a vital role in numerous physiological functions and drugs that target them are actively pursued for development of novel therapeutic agents. Here we report a means for monitoring in real time the conformational changes undergone by channel proteins upon exposure to pharmacological stimuli. The approach relies on tracking structural rearrangements by monitoring changes in bioluminescence energy transfer (BRET). To provide proof of principle we have worked with Kir3 neuronal channels producing 10 different constructs which were combined into 17 donor-acceptor BRET pairs. Among these combinations, pairs bearing the donor Nano-Luc (NLuc) at the C-terminal end of Kir3.2 subunits and the FlAsH acceptor at the N-terminal end (NT) or the interfacial helix (N70) of Kir3.1 subunits were identified as potential tools. These pairs displayed significant changes in energy transfer upon activation with direct channel ligands or via stimulation of G protein-coupled receptors. Conformational changes associated with channel activation followed similar kinetics as channel currents. Dose response curves generated by different agonists in FlAsH-BRET assays displayed similar rank order of potency as those obtained with conventional BRET readouts of G protein activation and ion flux assays. Conformational biosensors as the ones reported herein should prove a valuable complement to other methodologies currently used in channel drug discovery.

Endocytic profiles of δ-opioid receptor ligands determine the duration of rapid but not sustained cAMP responses.

Traditional assays that monitor cAMP inhibition by opioid receptor ligands require second-messenger accumulation over periods of 10-20 minutes. Since receptor regulation occurs within a similar time frame, such assays do not discriminate the actual signal from its modulation. Here we used bioluminescence resonance energy transfer to monitor inhibition of cAMP production by δ-opioid receptor (DOR) agonists in real time. cAMP inhibition elicited by different agonists over a period of 15 minutes was biphasic, with response buildup during the first 6 to 7 minutes, followed by a second phase of response decay or of no further increment. The rate at which the cAMP response disappeared was correlated with operational parameters describing ligand efficiency [log(τ/KA)] to promote Gαi activation, as well as with ligand ability to promote internalization during the time course of the assay. Thus, ligands that displayed low signaling efficiency and poor sequestration(SB235863 ([8R-(4bS*,8aα,8aß,12bß)]7,10-dimethyl-1-methoxy-11-(2-ethylpropyl)oxycarbonyl 5,6,7,8,12,12b-hexahydro-(9H)-4,8-methanobenzofuro[3,2-e]pyrrolo[2,3-g]isoquinoline hydrochloride), morphine) had minimal or no response decay. On the other hand, the decay rate was pronounced for deltorphin II, [d-Pen(2), d-pen(5)]-enkephalin, met-enkephalin, and SNC-80 ((+)-4-[(αR)-α-((2S,5R)-4-allyl-2,5-dimethyl-1-piperazinyl)-3-methoxybenzyl]-N,N-diethylbenzamide), which displayed high signaling efficiency and internalization. Moreover, inhibition of internalization by dynasore reduced or abolished response decay by internalizing ligands. Unlike acute responses, endocytic profiles were not predictive of whether an agonist would induce prolonged cAMP inhibition over sustained (30-120 minutes) DOR stimulation. Taken together, the data indicate that ligand ability to evoke G-protein activation or promote endocytosis was predictive of response duration over short, but not over sustained periods of cAMP inhibition.

5-HT2A receptor signalling through phospholipase D1 associated with its C-terminal tail.

The 5-HT2AR (5-hydroxytryptamine-2A receptor) is a GPCR (G-protein-coupled receptor) that is implicated in the actions of hallucinogens and represents a major target of atypical antipsychotic agents. In addition to its classical signalling though PLC (phospholipase C), the receptor can activate several other pathways, including ARF (ADP-ribosylation factor)-dependent activation of PLD (phospholipase D), which appears to be achieved through a mechanism independent of heterotrimeric G-proteins. In the present study we show that wild-type and inactive constructs of PLD1 (but not PLD2) respectively facilitate and inhibit ARF-dependent PLD signalling by the 5-HT2AR. Furthermore we demonstrate that PLD1 specifically co-immunoprecipitates with the receptor and binds to a distal site in GST (glutathione transferase) fusion protein constructs of its C-terminal tail which is distinct from the ARF-interaction site, thereby suggesting the existence of a functional ARF-PLD signalling complex directly associated with this receptor. This reveals the spatial co-ordination of an important GPCR, transducer and effector into a physical complex that is likely to reinforce the impact of receptor activation on a heterotrimeric G-protein-independent signalling pathway. Signalling of this receptor through such non-canonical pathways may be important to its role in particular disorders.

Role of the conserved NPxxY motif of the 5-HT2A receptor in determining selective interaction with isoforms of ADP-ribosylation factor (ARF).

In this study we have shown that N376 to D mutation in the conserved NPxxY motif within the carboxy terminal tail domain (CT) of the 5-HT2A receptor alters the binding preference of GST-fusion protein constructs of the CT domain from ARF1 to an alternative isoform, ARF6. These findings were corroborated by experiments investigating co-immunoprecipitation of the wild type (WT) and N376D mutant of the 5-HT2A receptor with ARF1 or 6 or dominant negative ARF1/6 constructs co-expressed in COS7 cells. In functional assays of 5-HT-induced phospholipase D (PLD) activation responses of the WT receptor were inhibited by a dominant negative mutant of ARF1 but not ARF6, whereas responses of the N376D mutant were strongly inhibited by negative mutant ARF6. No equivalent effect of the ARF mutants was seen on phospholipase C activation. In experiments assaying 5-HT-induced increases in [35S]GTPgammaS binding to ARF 1/6 immunoprecipitates as a measure of ARF activation, increased ARF6 activation was seen only with the mutant receptor. When cellular PLD responses of other NPxxY- or a DPxxY-containing GPCRs were measured in the presence of dominant negative ARF1/6 constructs, the majority, but not all, fitted the pattern exemplified by the 5-HT2A receptor and its N376D mutant. These data suggest that the presence of the N or a D in this highly conserved motif is an important, but not exclusive, determinant of which ARF isoform interacts with the GPCR.

ADP-ribosylation factor-dependent phospholipase D activation by the M3 muscarinic receptor.

G protein-coupled receptors can potentially activate phospholipase D (PLD) by a number of routes. We show here that the native M3 muscarinic receptor in 1321N1 cells and an epitope-tagged M3 receptor expressed in COS7 cells substantially utilize an ADP-ribosylation factor (ARF)-dependent route of PLD activation. This pathway is activated at the plasma membrane but appears to be largely independent of G, phospholipase C, Ca2+ q/11, protein kinase C, tyrosine kinases, and phosphatidyl inositol 3-kinase. We report instead that it involves physical association of ARF with the M3 receptor as demonstrated by co-immunoprecipitation and by in vitro interaction with a glutathione S-transferase fusion protein of the receptor's third intracellular loop domain. Experiments with mutant constructs of ARF1/6 and PLD1/2 indicate that the M3 receptor displays a major ARF1-dependent route of PLD1 activation with an additional ARF6-dependent pathway to PLD1 or PLD2. Examples of other G protein-coupled receptors assessed in comparison display alternative pathways of protein kinase C- or ARF6-dependent activation of PLD2.

Selective interaction of ARF1 with the carboxy-terminal tail domain of the 5-HT2A receptor.

The 5-hydroxytryptamine 2A receptor (5-HT2AR) is a member of the class I family of rhodopsin-related G protein-coupled receptors. The receptor is known to activate phospholipase C via the heterotrimeric G proteins Gq/11, but we showed previously that it can also signal through the phospholipase D (PLD) pathway in an ADP-ribosylation factor (ARF)-dependent manner that seems to be independent of Gq/11 (Mitchell et al., 1998). Both coimmunoprecipitation experiments and the effects of negative mutant ARF constructs on 5-HT2AR-induced PLD activation here suggested that ARF1 may play a greater role than ARF6 in the function of this receptor. Furthermore, we demonstrated using glutathione S-transferase (GST)-fusion proteins of receptor domains that ARF1 and ARF6 bind to the third intracellular loop (i3) and the carboxy terminal tail (ct) of the 5-HT2AR. The association of ARF1 with the ct domain of the receptor was stronger than its interaction with i3, or the interactions of ARF6 with either construct. Experiments using ARF mutants that are deficient in GTP loading, and the in vitro addition of GTPgammaS suggested that GTP loading enhances ARF1 binding to the receptor. The N376PxxY motif in the transmembrane 7 domain of the receptor (rather than a N376DPxxY mutant form) was shown to be essential for ARF-dependent PLD signaling and ARF1 coimmunoprecipitation. In GST-fusion proteins of the 5-HT2AR ct domain, mutation of Asn376 to Asp also markedly reduced ARF1-HA binding, although additional motifs in the Asn376-Asn384 sequence and to a lesser extent elsewhere, seem also to contribute to the interaction.