A mu-delta opioid receptor brain atlas reveals neuronal co-occurrence in subcortical networks

Opioid receptors are G protein-coupled receptors (GPCRs) that modulate brain function at all levels of neural integration, including autonomic, sensory, emotional and cognitive processing. Mu (MOR) and delta (DOR) opioid receptors functionally interact in vivo, but whether interactions occur at circuitry, cellular or molecular levels remains unsolved. To challenge the hypothesis of MOR/DOR heteromerization in the brain, we generated redMOR/greenDOR double knock-in mice and report dual receptor mapping throughout the nervous system. Data are organized as an interactive database offering an opioid receptor atlas with concomitant MOR/DOR visualization at subcellular resolution, accessible online. We also provide co-immunoprecipitation-based evidence for receptor heteromerization in these mice. In the forebrain, MOR and DOR are mainly detected in separate neurons, suggesting system-level interactions in high-order processing. In contrast, neuronal co-localization is detected in subcortical networks essential for survival involved in eating and sexual behaviors or perception and response to aversive stimuli. In addition, potential MOR/DOR intracellular interactions within the nociceptive pathway offer novel therapeutic perspectives.

Delta opioid receptors regulate temporoammonic-activated feedforward inhibition to the mouse CA1 hippocampus

The opioid system influences learning and memory processes. However, neural mechanisms underlying the modulation of hippocampal activity by opioid receptors remain largely unknown. Here, we compared how mu and delta receptors operate within the mouse CA1 network, and used knock-in mice expressing functional delta opioid receptors fused to the green fluorescent protein (DOR-eGFP) to determine how delta opioid receptor-expressing interneurons integrate within the hippocampal circuitry. Through whole cell patch-clamp recording of CA1 pyramidal neurons from wild-type and DOR-eGFP mice, we found that mu and delta receptors both modulate spontaneous GABAergic inhibition received by these cells. Interestingly, mu but not delta receptor activation decreased the feed-forward inhibitory input evoked by Schaffer collateral stimulation. However, mu and delta agonists modulated GABAergic feed-forward inhibition when evoked upon stimulation of the temporoammonic pathway. In addition, anterograde tracing using biotinylated dextran amine injected into the entorhinal cortex of DOR-eGFP mice suggests the existence of synaptic contacts between temporoammonic afferents and delta receptor-expressing interneurons processes in CA1. Altogether, our data demonstrate a distinct modulatory role of the hippocampal network activity by mu and delta opioid receptors, and show for the first time that delta receptor-expressing interneurons in the CA1 are recruited by the temporoammonic pathway rather than the Schaffer collateral.

Evaluation of cre recombinase delivery in mammalian cells using baculovirus infection

In vivo conditional knock-out of a protein is a method of choice to decipher its biological function. It can be achieved by encoding the cre-recombinase on a recombinant virus to exert spatio-temporal control of its expression and enzymatic activity and, subsequently, of the target gene deletion. Recombinant baculoviruses have been successfully used to express a wide range of proteins in insect cells. More recently, their potential to infect mammalian cells has been addressed but, so far, their ability to yield a conditional knock-out as a result of efficient in vivo cre-recombinase gene delivery has not been examined. Cre-recombinase fused to the green fluorescent protein was cloned under the control of the CAG promoter in a recombinant Autographa californica baculovirus expressing the vesicular stomatitis virus envelope G protein for increased mammalian cell infection. Gene delivery was evaluated in vitro in mammalian cells, neuroblastoma and mouse primary neuronal cultures as well as in vivo in the mouse brain. Infection with adeno-associated viruses encoding the cre-recombinase fused to the green fluorescent protein was performed as a positive control. Our results indicate that baculovirus infection leads to functional cre-recombinase expression in non-neuronal and neuroblastoma cell lines but not in mouse primary neuronal cultures or brain.

Distribution of delta opioid receptor-expressing neurons in the mouse hippocampus

Delta opioid receptors participate to the control of chronic pain and emotional responses. Recent data also identified their implication in spatial memory and drug-context associations pointing to a critical role of hippocampal delta receptors. We examined the distribution of delta receptor-expressing cells in the hippocampus using fluorescent knock-in mice that express a functional delta receptor fused at its carboxyterminus with the green fluorescent protein in place of the native receptor. Colocalization with markers for different neuronal populations was performed by immunohistochemical detection. Fine mapping in the dorsal hippocampus confirmed that delta opioid receptors are mainly present in GABAergic neurons. Indeed, they are mostly expressed in parvalbumin-immunopositive neurons both in the Ammon's horn and dentate gyrus. These receptors, therefore, most likely participate in the dynamic regulation of hippocampal activity.

Mouse δ opioid receptors are located on presynaptic afferents to hippocampal pyramidal cells

Delta opioid receptors participate in the control of chronic pain and emotional responses. Recent data have also identified their implication in drug-context associations pointing to a modulatory role on hippocampal activity. We used fluorescent knock-in mice that express a functional delta opioid receptor fused at its carboxy terminus with the green fluorescent protein in place of the native receptor to investigate the receptor neuroanatomical distribution in this structure. Fine mapping of the pyramidal layer was performed in hippocampal acute brain slices and organotypic cultures using fluorescence confocal imaging, co-localization with pre- and postsynaptic markers and correlative light-electron microscopy. The different approaches concurred to identify delta opioid receptors on presynaptic afferents to glutamatergic principal cells. In the latter, only scarce receptors were detected that were confined within the Golgi or vesicular intracellular compartments with no receptor present at the cell surface. In the mouse hippocampus, expression of functional delta opioid receptors is therefore mostly associated with interneurons emphasizing a presynaptic modulatory effect on the pyramidal cell firing rate.

Permeabilization of lipid membranes and cells by a light-responsive copolymer

Membrane permeabilization is achieved via numerous techniques involving the use of molecular agents such as peptides used in antimicrobial therapy. Although high efficiency is reached, the permeabilization mechanism remains global with a noticeable lack of control. To achieve localized control and more gradual increase in membrane perturbation, we have developed hydrophobically modified poly(acrylic acid) amphiphilic copolymers with light-responsive azobenzene hydrophobic moieties. We present evidence for light triggered membrane permeabilization in the presence azobenzene-modified polymers (AMPs). Exposure to UV or blue light reversibly switches the polarity of the azobenzene (cis-trans isomerization) in AMPs, hence controlling AMP-loaded lipid vesicles permeabilization via in situ activation. Release of encapsulated probes was studied by microscopy on isolated AMP-loaded giant unilamellar vesicles (pol-GUVs). We show that in pH and ionic strength conditions that are biologically relevant pol-GUVs are kept impermeable when they contain predominantly cis-AMPs but become leaky with no membrane breakage upon exposure to blue light due to AMPs switch to a trans-apolar state. In addition, we show that AMPs induce destabilization of plasma membranes when added to mammal cells in their trans-apolar state, with no loss of cell viability. These features make AMPs promising tools for remote control of cell membrane permeabilization in mild conditions.

In vivo delta opioid receptor internalization controls behavioral effects of agonists

Background

GPCRs regulate a remarkable diversity of biological functions, and are thus often targeted for drug therapies. Stimulation of a GPCR by an extracellular ligand triggers receptor signaling via G proteins, and this process is highly regulated. Receptor activation is typically accompanied by desensitization of receptor signaling, a complex feedback regulatory process of which receptor internalization is postulated as a key event. The in vivo significance of GPCR internalization is poorly understood. In fact, the majority of studies have been performed in transfected cell systems, which do not adequately model physiological environments and the complexity of integrated responses observed in the whole animal.

Methods and Findings

In this study, we used knock-in mice expressing functional fluorescent delta opioid receptors (DOR-eGFP) in place of the native receptor to correlate receptor localization in neurons with behavioral responses. We analyzed the pain-relieving effects of two delta receptor agonists with similar signaling potencies and efficacies, but distinct internalizing properties. An initial treatment with the high (SNC80) or low (AR-M100390) internalizing agonist equally reduced CFA-induced inflammatory pain. However, subsequent drug treatment produced highly distinct responses. Animals initially treated with SNC80 showed no analgesic response to a second dose of either delta receptor agonist. Concomitant receptor internalization and G-protein uncoupling were observed throughout the nervous system. This loss of function was temporary, since full DOR-eGFP receptor responses were restored 24 hours after SNC80 administration. In contrast, treatment with AR-M100390 resulted in retained analgesic response to a subsequent agonist injection, and ex vivo analysis showed that DOR-eGFP receptor remained G protein-coupled on the cell surface. Finally SNC80 but not AR-M100390 produced DOR-eGFP phosphorylation, suggesting that the two agonists produce distinct active receptor conformations in vivo which likely lead to differential receptor trafficking.

Conclusions

Together our data show that delta agonists retain full analgesic efficacy when receptors remain on the cell surface. In contrast, delta agonist-induced analgesia is abolished following receptor internalization, and complete behavioral desensitization is observed. Overall these results establish that, in the context of pain control, receptor localization fully controls receptor function in vivo. This finding has both fundamental and therapeutic implications for slow-recycling GPCRs.

Co-expression of mu and delta opioid receptors as receptor-G protein fusions enhances both mu and delta signalling via distinct mechanisms

Mu and delta opioid receptors (MORs and DORs) were co-expressed as fusion proteins between a receptor and a pertussis insensitive mutant Galpha(i/o) protein in human embryonic kidney 293 cells. Signalling efficiency was then monitored following inactivation of endogenous Galpha(i/o) proteins by pertussis toxin. Co-expression resulted in increased delta opioid signalling which was insensitive to the mu specific antagonist d-Phe-Cys-Tyr-d-Trp-Arg-Thr-Pen-Thr-NH2. Under these conditions, mu opioid signalling was also increased and insensitive to the delta specific antagonist Tic-deltorphin. In this latter case, however, no G protein activation was observed in the presence of the delta specific inverse agonist N,N(CH3)2-Dmt-Tic-NH2. When a MOR fused to a non-functional Galpha subunit was co-expressed with the DOR-Galpha protein fusion, delta opioid signalling was not affected whereas mu opioid signalling was restored. Altogether our results suggest that increased delta opioid signalling is due to enhanced DOR coupling to its tethered Galpha subunit. On the other hand, our data indicate that increased mu opioid signalling requires an active conformation of the DOR and also results in activation of the Galpha subunit fused the DOR.

Two-photon excitation fluorescence cross-correlation assay for ligand-receptor binding: cell membrane nanopatches containing the human micro-opioid receptor

Current ligand-receptor binding assays for G-protein coupled receptors cannot directly measure the system's dissociation constant, Kd, without purification of the receptor protein. Accurately measured Kd's are essential in the development of a molecular level understanding of ligand-receptor interactions critical in rational drug design. Here we report the introduction of two-photon excitation fluorescence cross-correlation spectroscopy (TPE-FCCS) to the direct analysis of ligand-receptor interactions of the human micro opioid receptor (hMOR) for both agonists and antagonists. We have developed the use of fluorescently distinct, dye-labeled hMOR-containing cell membrane nanopatches ( approximately 100-nm radius) and ligands, respectively, for this assay. We show that the output from TPE-FCCS data sets can be converted to the conventional Hill format, which provides Kd and the number of active receptors per nanopatch. When ligands are labeled with quantum dots, this assay can detect binding with ligand concentrations in the subnanomolar regime. Interestingly, conjugation to a bulky quantum dot did not adversely affect the binding propensity of the hMOR pentapeptide ligand, Leu-enkephalin.

Mu-delta opioid receptor functional interaction: Insight using receptor-G protein fusions

Fusion proteins between a receptor and a pertussis toxin-insensitive G(i)alpha subunit were used to gain insight into the molecular interactions that take place upon mu and delta opioid receptor heterodimerization. When mu opioid receptor-G(i1)alpha fusions were coexpressed with nonfused delta opioid receptors in human embryonic kidney 293 cells, or vice versa, receptor heterodimers were detected by coimmunoprecipitation. In pertussis toxin-treated cells, receptor coexpression decreased the amount of guanosine 5'-O-(3-[35S]thio)triphosphate ([35S]GTPgammaS) incorporated in the fused G alpha protein after the addition of agonists specific for the receptor-G(i1)alpha fusion. In addition, activation of the G alpha protein occurred in heterodimers upon addition of an agonist specific for the nonfused receptor. It remained unaffected by an inverse agonist specific for the receptor-G(i1)alpha fusion. These data suggest that signaling through the receptor-G(i1)alpha fusion protein is impaired in heterodimers and support a mechanism in which activation of the G alpha subunit is promoted by a direct interaction with the nonfused receptor. Alternatively, receptor coexpression did not modify the ligand binding properties for the high-affinity state of the receptor-G(i1)alpha fusion nor the EC50 values for agonist-induced [35S]GTPgammaS incorporation in the G(i1)alpha subunit. In addition, no binding competition was observed between delta and mu ligands. Together, the data point to mu-delta opioid receptor heterodimers formed by contact interactions between monomers that retain their structural integrity.

Enhanced spontaneous activity of the mu opioid receptor by cysteine mutations: characterization of a tool for inverse agonist screening

Background

The concept of spontaneous- or constitutive-activity has become widely accepted and verified for numerous G protein-coupled receptors and this ligand-independent activity is also acknowledged to play a role in some pathologies. Constitutive activity has been reported for the mu opioid receptor. In some cases the increase in receptor basal activity was induced by chronic morphine administration suggesting that constitutive activity may contribute to the development of drug tolerance and dependence. Constitutively active mutants represent excellent tools for gathering information about the mechanisms of receptor activation and the possible physiological relevance of spontaneous receptor activity. The high basal level of activity of these mutants also allows for easier identification of inverse agonists, defined as ligands able to suppress spontaneous receptor activity, and leads to a better comprehension of their modulatory effects as well as possible in vivo use.

Results

Cysteines 348 and 353 of the human mu opioid receptor (hMOR) were mutated into alanines and Ala^348,353 hMOR was stably expressed in HEK 293 cells. [³⁵S] GTPγS binding experiments revealed that Ala^348,353 hMOR basal activity was significantly higher when compared to hMOR, suggesting that the mutant receptor is constitutively active. [³⁵S] GTPγS binding was decreased by cyprodime or CTOP indicating that both ligands have inverse agonist properties. All tested agonists exhibited binding affinities higher for Ala^348,353 hMOR than for hMOR, with the exception of endogenous opioid peptides. Antagonist affinity remained virtually unchanged except for CTOP and cyprodime that bound the double mutant with higher affinities. The agonists DAMGO and morphine showed enhanced potency for the Ala^348,353 hMOR receptor in [³⁵S] GTPγS experiments. Finally, pretreatment with the antagonists naloxone, cyprodime or CTOP significantly increased Ala^348,353 hMOR expression.

Conclusion

Taken together our data indicate that the double C348/353A mutation results in a constitutively active conformation of hMOR that is still activated by agonists. This is the first report of a stable CAM of hMOR with the potential to screen for inverse agonists.

G protein-coupled receptor overexpression with the baculovirus-insect cell system: a tool for structural and functional studies

G protein-coupled receptors, whose topology shows seven transmembrane domains, form the largest known family of receptors involved in higher organism signal transduction. These receptors are generally of low natural abundance and overexpression is usually a prerequisite to their structural or functional characterisation. The baculovirus-insect cell system constitutes a versatile tool for the maximal production of receptors. This heterologous expression system also provides interesting alternatives for receptor functional studies in a well-controlled cellular context.

Agonists activate Gi1 alpha or Gi2 alpha fused to the human mu opioid receptor differently

As preferential coupling of opioid receptor to various inhibitory Galpha subunits is still under debate, we have investigated the selectivity of the human mu opioid receptor fused to a pertussis toxin insensitive C351I Gi1 alpha or C352I Gi2 alpha in stably transfected HEK 293 cells. Overall agonist binding affinities were increased for both fusion constructs when compared to the wild type receptor. [35 S]GTPgammaS binding was performed on pertussis toxin treated cells to monitor coupling efficiency of the fusion constructs. Upon agonist addition hMOR-C351I Gi1 a exhibited an activation profile similar to the non-fused receptor while hMOR-C352I Gi2 alpha was poorly activated. Interestingly no correlation could be drawn between agonist binding affinity and efficacy. Upon agonist addition, forskolin-stimulated cAMP production, as measured using a reporter gene assay, was inhibited by signals transduced via the fused Gi1 alpha and Gi2 alpha mainly. In contrast both fusion constructs were able to initiate ERK-MAPK phosphorylation via coupling to endogenous G proteins only. In conclusion our data indicate that hMOR couples more efficiently to Gi1 alpha than Gi2 alpha and that the coupling efficacy is clearly agonist-dependent.

Expression of the human mu opioid receptor in a stable Sf9 cell line

The cDNA that encodes the human mu opioid receptor (hMOR) has been cloned and expressed in Spodoptera frugiperda (Sf9) cells using a nonlytic vector system. The coding sequence fused to the cleavable glycoprotein signal peptide gp 64, and a C-terminal histidine tag was placed under the transcriptional control of the Orgyia pseudotsugata multicapsid nucleopolyhedrosis virus immediate-early 2 (OpIE2) promoter. Transfected cells were selected using Zeocin resistance and the receptor was constitutively expressed at approximately 12000 receptors per cell. Immunofluorescence images illustrated that more than 75% of the Sf9 cells expressed hMOR at the plasma membrane. This is the first report of the constitutive and heterologous expression of a G protein-coupled receptor in a stably transfected Sf9 cell line, under the control of the OpIE2 promoter.

Regioselective photolabeling of glycoprotein A in membranes

We have developed a chemical method for directly identifying the amino acid residues of the transmembrane domain of a protein that are located right in the center of the membrane. Glycophorin A (GPA), the major sialoglycoprotein of human erythrocytes, was the first membrane protein whose primary sequence was elucidated, but its three-dimensional structure is still not known. GPA has been reconstituted into liposomes formed from dimyristoylphosphatidylcholine, dimyristoylphosphatidylserine, cholesterol, and a bola-amphiphilic phospholipidic photoactivatable probe (radioactive probe 1) by a detergent-mediated method. Electron microscopy confirmed the formation of spherical vesicular structures, and sucrose-density gradients revealed that the proteoliposomes comprised only one membrane fraction. Proteinase-K digestion of GPA in the proteoliposomes suggested that the orientation of GPA in reconstituted proteoliposomes was virtually identical to that observed in natural erythrocyte membranes. After photo-irradiation of the reconstituted proteoliposomes and in situ tryptic digestion, the photolabeled amino acid residues were analyzed by Edman degradation and their radioactivity was measured. Val80 and Met81, which had been assumed to be located near the center of the transmembrane domain of GPA, were indeed highly selectively photolabeled by probe 1. The new method might be applied to analyze the three-dimensional arrangement of the transmembrane domain of protein complexes that are made up from several subunits.

The human delta opioid receptor activates G(i1)alpha more efficiently than G(o1)alpha

To assess the relative capacity of the human delta opioid receptor to activate closely related G proteins, fusion proteins were constructed in which the alpha-subunits of either G(i1) or G(o1), containing point mutations to render them insensitive to the actions of pertussis toxin, were linked in-frame with the C-terminus of the receptor. Following transient and stable expression in HEK 293 cells, both constructs bound the antagonist [(3)H]naltrindole with high affinity. D-ala(2),D-leu(5) Enkephalin effectively inhibited forskolin-stimulated adenylyl cyclase activity in intact cells in a concentration-dependent, but pertussis toxin-insensitive, manner. The high-affinity GTPase activity of both constructs was also stimulated by D-ala(2),D-leu(5) enkephalin with similar potency. However, enzyme kinetic analysis of agonist stimulation of GTPase activity demonstrated that the GTP turnover number produced in response to D-ala(2),D-leu(5) enkephalin was more than three times greater for G(i1)alpha than for G(o1)alpha. As the effect of agonist in both cases was to increase V:(max) without increasing the observed K:(m) for GTP, this is consistent with receptor promoting greater guanine nucleotide exchange, and thus activation, of G(i1)alpha compared with G(o1)alpha. An equivalent fusion protein between the human mu opioid receptor-1 and G(i1)alpha produced a similar D-ala(2),D-leu(5) enkephalin-induced GTP turnover number as the delta opioid receptor-G(i1)alpha fusion construct, consistent with agonist occupation of these two opioid receptor subtypes being equally efficiently coupled to activation of G(i1)alpha.

Internalization and recycling of delta-opioid receptor are dependent on a phosphorylation-dephosphorylation mechanism

Internalization, recycling, and resensitization of the human delta-opioid receptor (hDOR) were studied in the neuroblastoma cell line SK-N-BE, endogenously expressing this receptor. Conventional and confocal fluorescence microscopy observations, corroborated by Scatchard analysis, indicated that after a 100 nM Eto treatment, 60 to 70% of hDOR were rapidly internalized (t(1/2) < 15 min). This agonist-triggered internalization was reversible for a treatment not exceeding 1 h and became irreversible for prolonged treatment (4 h), leading probably to the degradation and/or down-regulation of the receptor. The rapid internalization of hDOR was totally blocked in the presence of heparin, known as an inhibitor of G protein-coupled receptor kinases (Benovic et al., 1989), a result indicating that phosphorylation by these kinases is a critical step in desensitization (Hasbi et al, 1998) and internalization of hDOR (present study) in SK-N-BE cell line. Blockade of internalization by agents not interferring with phosphorylation, as hypertonic sucrose or concanavalin A, also blocked the resensitization (receptor functional recovering) process. Furthermore, blockade of dephosphorylation of the internalized hDOR by okadaic acid totally suppressed its recycling to the plasma membrane and its subsequent resensitization. These results indicate that regulatory events leading to desensitization, internalization, and recycling in a functional state of hDOR involve phosphorylation by a G protein-coupled receptor kinase, internalization via clathrin-coated vesicles, and dephosphorylation by acid phosphatases.