Intracellular trafficking and assembly of specific Kir3 channel/G protein complexes
Introduction
The classical paradigm for G protein activation of effectors (at least in the mammalian visual system) involves free diffusion of the relevant proteins at the plasma membrane. In humans, there are at least 20 different Gα subunits, 5 Gβ subunits, 13 Gγ subunits, multiple effectors and associated proteins and several hundred 7 transmembrane domain-containing receptors (7TM-Rs). Considering the fact that some 7TM-R effectors such as ion channels can be fully activated within 1 s following addition of an agonist [1], [2], this model encounters difficulties in accounting for the specificity and the rapidity of G protein signalling in other tissues (see [3] for review). A number of studies have demonstrated the existence of stable interactions, independent of receptor activation between the receptor and G protein [4], [5], G protein and the effector [6] and even between the receptor and the effector [7], [8]. If the signalling partners interact before receptor activation, the question of where these proteins initially interact together becomes critical. Recent studies have shown that many of these proteins interact initially in the endoplasmic reticulum (ER), including receptor equivalents in receptor dimers [9], receptor and Gβγ subunits [10] and effectors such as Kir3 channels and adenylyl cyclase with nascent Gβγ [6], [11], [12]. If these complexes are pre-formed during protein biosynthesis and maturation, they would need to be trafficked inside the cells as a complex and not necessarily as individual proteins.
The Kir3 family of inwardly rectifying potassium channels was among the first effectors demonstrated to be directly modulated by Gβγ [13]. To date, little specificity has been shown for the different Gβγ subunits with respect to Kir3 channel modulation in vitro, with the exception that combinations containing Gγ1 (found in the mammalian visual system) show reduced efficacy [14]. A recent study has demonstrated both constitutive and agonist-induced interactions between Kir3 channels and various Gβγ combinations in living cells using fluorescence resonance energy transfer (FRET) and total internal reflectance fluorescence (TIRF) microscopy [15]. Direct activation of Kir3 channels by Gβγ is therefore mostly independent of the subunit composition of the heterodimer with the exception of Gβ5 and Gγ1 [14], [15]. This is also generally true for other Gβγ effectors such as adenylyl cyclase [16] although there are exceptions. For example, P-Rex1 is a GEF for Rac which is activated by combinations of Gβ1-4 but not Gβ5 and these interactions have differing affinities depending on the Gγ isoform present [17]. However, most of these studies have focused naturally on interactions important for Gβγ-dependent signalling events. Here, we focus on specificity in precocious (i.e. agonist-independent) interactions between Kir3.1 or Kir3.2 channel subunits or adenylyl cyclase II (ACII) with different Gβγ combinations. Recently, the development of bioluminescence resonance energy transfer (BRET) and bimolecular fluorescence complementation (BiFC) approaches has allowed us to study the different types of interactions important in GPCR signalling systems in living cells [18], [19]. We are now also able to link these interactions to particular trafficking steps in the anterograde trafficking of these complexes. We show here that although the trafficking of Kir3 channels is sensitive to dominant negative Sar 1 and Rab 1 GTPases, interactions between the channel and Gβγ are not, confirming our previous observations that Kir3.1 and Kir3.2 subunits and Gβγ initially interact in the ER.
Section snippets
Constructs
Constructs encoding Kir3.1-RLuc, Kir3.2-Rluc, CD4-Rluc, eGFP-tagged Gβ1-5, AC2-Rluc, β2AR-RLuc, β2AR-GFP10, Kir3.4, YFP (1–158)-Gβ1, YFP (158–239)-Gγ1, YFP (158–239)-Gγ2, YFP (158–239)-Gγ3, YFP (158–239)-Gγ7, YFP (158–239)-Gγ11, GFP10-Gγ2, soluble βARK-CT, WT and DN Sar 1 and Rab constructs were used as previously described [6], [10], [11], [12], [20], [21]. Flag-Gβ1-5, HA-Gγ1, HA-Gγ2, HA-Gγ3, HA-Gγ7, and HA-Gγ11 were obtained from UMR cDNA Resource Center (www.cdna.org). An extracellularly
Kir3.1/3.4 trafficking to the cell surface
To be transported to the cell surface, the Kir3.1 subunit must be assembled with either Kir3.2 or Kir3.4 subunits, as it lacks an ER export signal [25]. The use of dominant negative forms of the Sar 1 and Rab GTPases is now a common way to modulate trafficking itineraries of proteins during their maturation and subsequent insertion into the plasma membrane. These small GTPases regulate multiple anterograde protein trafficking events including the budding of transport vesicles from donor
Discussion
We have demonstrated that cell surface expression of Kir3.1 channels depends on Rab 1- and Sar 1-dependent anterograde protein trafficking. We confirm that channel assembly per se was not affected as even though trafficking was blocked, Kir3.1 and Kir3.4 still become associated as demonstrated using co-immunoprecipitation (Fig. 5). We and others [10], [30], [31] have previously demonstrated that the β2AR also uses these pathways. Curiously, after exit from the ER, the receptor uses a pathway
Acknowledgements
This work was supported by grants from the Canadian Institutes of Health Research and the Heart and Stroke Foundation of Quebec to T.E.H. T.E.H. is a Chercheur National of the Fonds de la Recherche en Santé du Québec. M.R. holds a doctoral award from the Fonds de la Recherche en Santé du Québec. NR is a recipient of a McGill Faculty of Medicine Internal Studentship Award. We thank Dr. Denis J. Dupré (Dalhousie University, Halifax, Canada) for generating a number of the reagents (listed in
References (68)
- et al.
J. Biol. Chem.
(2003) - et al.
Cell. Signal.
(1999) - et al.
J. Biol. Chem.
(2003) - et al.
J. Biol. Chem.
(2002) - et al.
J. Biol. Chem.
(2004) - et al.
J. Biol. Chem.
(2006) - et al.
Cell. Signal.
(2006) - et al.
Cell. Signal.
(2007) - et al.
Neuron
(2006) - et al.
J. Biol. Chem.
(1998)
J. Biol. Chem.
J. Biol. Chem.
Neuron
J. Biol. Chem.
J. Biol. Chem.
Cell. Signal.
J. Biol. Chem.
J. Biol. Chem.
J. Biol. Chem.
J. Biol. Chem.
J. Biol. Chem.
Neuron
Mol. Cell. Neurosci.
Cell. Signal.
FEBS Lett.
Cell
Biophys. J.
J. Biol. Chem.
Trends Pharmacol. Sci.
Cell. Signal.
J. Biol. Chem.
J. Biol. Chem.
J. Biol. Chem.
Curr. Opin. Cell Biol.
Cited by (27)
Shaky ground - The nature of metastable GPCR signalling complexes
2019, NeuropharmacologyCitation Excerpt :Stable or meta-stable (defined below) interactions of GPCRs with effector partners would easily contribute to rapid and specific signalling. Such signalling complexes containing either β1AR or β2AR (Lavine et al., 2002) assemble with effector partners such as adenylyl cyclase (Dupré et al., 2007; Baragli et al., 2008) and Kir3 channels (David et al., 2006; Rebois et al., 2006; Robitaille et al., 2009). Further, these complexes formed during receptor biosynthesis were insensitive to dominant negative versions of Rab1 or Sar1 (but not Rabs 2, 6 or 11) (Dupré et al., 2006, 2007), which regulate anterograde receptor trafficking (reviewed in (Dupré and Hébert, 2006; Dong et al., 2007)).
The Dynamics of GPCR Oligomerization and Their Functional Consequences
2018, International Review of Cell and Molecular BiologyCitation Excerpt :Stable or metastable (more on that later) interactions of GPCRs with effector partners can be viewed as a mechanism to assure rapid and specific signaling. Our own studies of the ontogeny of such signaling complexes suggest that complexes of either β1AR or β2AR (Lavine et al., 2002) assemble with effector partners such as adenylyl cyclase (Baragli et al., 2008; Dupré et al., 2007) and Kir3 channels (David et al., 2006; Rebois et al., 2006; Robitaille et al., 2009). Similar preassembled signaling complexes have recently been identified for M3 muscarinic receptors and Gαq (Qin et al., 2011).
Gβγ subunits—Different spaces, different faces
2016, Pharmacological ResearchCitation Excerpt :Biosynthesis of Gβ and Gγ subunits is a tightly regulated process that involves the functions of various chaperones that include phosducin-like protein 1 (PhLP1), chaperonin containing TCP-1 (CCT) and dopamine receptor-interacting protein 78 (DRiP78) at the endoplasmic reticulum [38–42]. We have previously described that upon their biosynthesis, Gβγ subunits interact with components of nascent signalling complexes, i.e. GPCRs, Gα subunits and various effectors, at the ER and are trafficked to the cellular membrane as a complex ([43–45], reviewed in [2]). These studies describe one instance of Gβγ-mediated sorting of signalling complexes in the ER prior to trafficking to the plasma membrane, however, there are several instances where Gβγ subunits have been implicated beyond such biosynthesis-dependent, intracellular translocation events.
Localization and Targeting of GIRK Channels in Mammalian Central Neurons
2015, International Review of NeurobiologyCytoskeletal roles in cardiac ion channel expression
2014, Biochimica et Biophysica Acta - BiomembranesCitation Excerpt :Two studies have examined the forward trafficking pathways that carry newly synthesized cardiac ion channels to the plasmalemma. Included in the first study were tests for the involvement of several Rab GTPases and Sar1 in the trafficking of Kir3.1/Kir3.4 heterotetramers in HEK293 cells [134]. This heterotetramer underlies the cardiac IKAch current.
Organizational Complexity of β-adrenergic Receptor Signaling Systems
2011, Current Topics in MembranesCitation Excerpt :In fact, significant evidence is accumulating that the assembly of GPCR signaling complexes occurs during their biosynthetic journey, rather than in response to agonist stimulation at the plasma membrane. We have studied the ontogeny of R/G/E complexes, initially focusing on β1AR and β2AR (Lavine et al., 2002) as well as AC (Dupré et al., 2007; Baragli, Grieco, Trieu, Villeneuve, & Hébert, 2008) and Kir3 channels (David, Richer, Mamarbachi, Villeneuve, Dupré, & Hébert, 2006; Rebois et al., 2006; Robitaille, Ramakrishnan, Baragli, & Hébert, 2009). Our data suggest that these complexes are formed during biosynthesis rather than through random, agonist-induced interactions at the plasma membrane.