Atrial fibrillation after coronary artery bypass surgery: association with changes in G protein levels in mononuclear leukocytes

Atrial fibrillation (AFib) is a frequent complication of coronary artery bypass grafting (CABG). Its cause, however, is unknown. As the adrenergic system is involved in some types of AFib, we hypothesized that a change in guanine nucleotide-binding protein (G protein) expression plays a role in the development of post-CABG AFib. In 28 patients undergoing CABG surgery, the G(s alpha)/G(i alpha) ratio (stimulatory/inhibitory G protein alpha) at the protein (Western blotting) and mRNA (reverse transcription polymerase chain reaction) levels was measured before and after surgery. As a suitable test system allowing multiple analysis mononuclear leukocytes (MNL) were chosen. The perioperative change of the G(s alpha)/G(i alpha) ratio of protein and mRNA was significantly different in patients who subsequently developed AFib (eight patients) and in patients who did not (20 patients; P<0.01 and <0.001, respectively). On average, the protein G(s alpha)/G(i alpha) ratio decreased from 1.79+/-1.13 (mean+/-SD) to 1.32+/-0.69 in patients without AFib (P=0.1, n.s.) whereas a significant increase from 0.86+/-0.44 to 1.62+/-0.65 (P<0.01) was observed in patients subsequently developing AFib. The mRNA G(s alpha)/G(i alpha) ratio decreased significantly from 0.53+/-0.24 to 0.36+/-0.11 in patients without AFib (P<0.01) whereas a significant increase from 0.31+/-0.14 to 0.47+/-0.13 was observed in those who subsequently developed AFib (P<0.05). These results indicate that an increase of the G(s alpha)/G(i alpha) ratio in MNL is associated with AFib after CABG surgery and possibly may be used as a prognostic indicator of this complication.

G proteins as drug targets

The structure and function of heterotrimeric G protein subunits is known in considerable detail. Upon stimulation of a heptahelical receptor by the appropriate agonists, the cognate G proteins undergo a cycle of activation and deactivation; the alpha-subunits and the beta gamma-dimers interact sequentially with several reaction partners (receptor, guanine nucleotides and effectors as well as regulatory proteins) by exposing appropriate binding sites. For most of these domains, low molecular weight ligands have been identified that either activate or inhibit signal transduction. These ligands include short peptides derived from receptors, G protein subunits and effectors, mastoparan and related insect venoms, modified guanine nucleotides, suramin analogues and amphiphilic cations. Because compounds that act on G proteins may be endowed with new forms of selectivity, we propose that G protein subunits may therefore be considered as potential drug targets.

A G protein alpha subunit from Cochliobolus heterostrophus involved in mating and appressorium formation

A Galpha subunit-encoding gene (CGA1) was cloned from Cochliobolus heterostrophus, a heterothallic foliar pathogen of corn. The deduced amino acid sequence showed similarity to Galpha proteins from other filamentous fungi and suggested that CGA1 is a member of the Galphai class. cga1 mutants had reduced ability to form appressoria on glass surfaces and on corn leaves; mutants nevertheless caused lesions on corn plants like those of wild type. cga1 mutants were female sterile; sexual development was completely abolished when the mutant allele was homozygous in a cross. Ascospores produced in crosses heterozygous at Cga1 were all wild type. The signal transduction pathway represented by CGA1 appears to be involved in developmental pathways leading to either appressorium formation or mating; in sexual development CGA1 is required for both fertility and ascospore viability.

Selective activation of G-protein subtypes by vertebrate and invertebrate rhodopsins

We have quantitatively investigated specificities in activating G-protein subtype by bovine and squid rhodopsins to examine whether or not the phototransduction cascade in each of the photoreceptor cells is determined by the colocalization of a large amount of G-protein subtype (Gt or Gq). In contrast to the efficient activation of respective Gt and Gq, bovine and squid rhodopsins scarcely activated G-protein counterparts. Exchange of alpha- and betagamma-subunits of Gt and Gq indicated the critical role of the alpha-subunit in specific binding to respective rhodopsins. Thus the specific recognition of G-protein subtype by each rhodopsin is a major mechanism in determining the phototransduction cascade.

Molecular cloning of a novel GTP-binding protein induced in fish cells by rhabdovirus infection

We have cloned and sequenced a cDNA encoding GTP-binding protein from a fish cell, CHSE-214. The clone was 1493 bp long and contained an open reading frame encoding 364 amino acids. It has the five sequence motifs G1-G5 that are conserved in all GTP-binding proteins. Its amino acid sequences are strikingly different from those of the well-characterized G-proteins. However, sequences closely related to this protein are found in various kinds of species including human, Arabidopsis, Drosophila and archaebacteria, suggesting a novel subfamily within the superfamily of the GTP-binding proteins. Northern analysis indicates that this gene is constitutively expressed at a low level in normal cells but is induced by fish rhabdovirus infection at about 24 h post infection and disappears thereafter. Based on these observations, we propose that this protein represents an evolutionarily conserved novel subfamily of GTP-binding proteins which may play an important role in fish rhabdovirus infection.

Gustducin and its role in taste

The mechanisms responsible for taste signal transductions are very complex. A key molecule, alpha-gustducin, a primarily taste-specific G protein alpha-subunit, was discovered in 1992 and was later found to be involved in both bitter and sweet taste transduction. A proposed mechanism for alpha-gustducin involves coupling specific cell-surface receptors with a cyclic nucleotide phosphodiesterase which would open a cyclic nucleotide-suppressible cation channel leading to influx of calcium, and ultimately leading to release of neurotransmitter. Although "knock-out" animals deficient in the alpha-gustducin gene clearly demonstrate that gustducin is an essential molecule for tasting certain bitter and sweet compounds, the precise role of alpha-gustducin in bitter and sweet taste is presently unclear. Indeed, there are several other signaling mechanisms in sweet and bitter taste, apparently unrelated to alpha-gustducin, that increase cyclic AMP or inositol 1,4,5 trisphosphate. Thus, proposed models for alpha-gustducin and those found by other laboratories may be parallel and interdependent.

Gq-protein alpha subunit expression and distribution in pregnant rat myometrial tissues

OBJECTIVE

Multiple G-protein isoforms play an integral role in signal transduction; the Gq subtype of G-protein alpha subunits is involved in the activation of the phosphatidylinositol signaling pathway. The studies described herein evaluate the expression of Gq, along with Gs and Gi, in pregnant and nonpregnant rat myometrial tissues.

METHODS

Myometrium and other tissues were obtained from nonpregnant and timed-pregnant Sprague-Dawley rats. Western blot studies were performed using polyclonal G-protein isoform-specific antibodies. Immunohistochemical studies were performed using the same antibodies with specimens of myometrium, intestine, and skeletal muscle.

RESULTS

The Western blot studies confirmed differential expression of all types of G-protein alpha subunit subtypes in rat myometrial tissues. In pregnant rat myometrium, the expression of Gq and Gs was sustained through day 22, whereas, Gi expression decreased on day 20 and remained low through the remainder of gestation. The immunohistochemical studies revealed significant staining for Gq and Gs in the myometrial layers of the pregnant and nonpregnant rat uterus; in contrast, immunostaining for Gi was minimal in nonpregnant myometrium, and even lower in myometrium from pregnant uteri.

CONCLUSIONS

These studies have confirmed expression of the Gq, Gi, and Gs alpha subunits in rat myometrial tissue. Immunohistochemistry confirmed that Gq was expressed at high levels in the myometrial layer of the pregnant and nonpregnant uterus. These observations support the hypothesis that Gq expression is critically important for the transduction of hormone signals, such as those responsible for the generation of phasic myometrial contractions.

Classification of protein families and detection of the determinant residues with an improved self-organizing map

Using a SOM (self-organizing map) we can classify sequences within a protein family into subgroups that generally correspond to biological subcategories. These maps tend to show sequence similarity as proximity in the map. Combining maps generated at different levels of resolution, the structure of relations in protein families can be captured that could not otherwise be represented in a single map. The underlying representation of maps enables us to retrieve characteristic sequence patterns for individual subgroups of sequences. Such patterns tend to correspond to functionally important regions. We present a modified SOM algorithm that includes a convergence test that dynamically controls the learning parameters to adapt them to the learning set instead of being fixed and externally optimized by trial and error. Given the variability of protein family size and distribution, the addition of this features is necessary. The method is successfully tested with a number of families. The rab family of small GTPases is used to illustrate the performance of the method.

Isolation of a novel human cDNA (rhoHP1) homologous to rho genes

A novel full-length cDNA showing homology with rho genes was isolated from a human placenta cDNA library. Sequencing of a total of 1086 nucleotides of this clone revealed an open reading frame of 630 nucleotides (210 amino acids). In view of its degree of homology to members of the Rho family of molecules (50-54% identical amino acids, 60-63% identical nucleotides within the coding region), the predicted product was designated RhoHP1(Rho-related protein HP1). Northern analysis indicated that a message about 1.2-kb long is expressed in human heart, placenta, liver, skeletal muscle, and pancreas and, with weaker intensity, in several other tissues.

Molecular cloning and characterization of rab27a and rab27b, novel human rab proteins shared by melanocytes and platelets

Rabs are prenylated, membrane-bound proteins involved in vesicular fusion and trafficking. We isolated the complete cDNAs of two rab isoforms, rab27a and rab27b, from human melanoma cells and melanocytes, Rab27a is the human homolog of a rat megakaryocyte rab called ram p25. Rab27b corresponds to a small GTP-binding protein, c25KG, which was previously purified from platelets but whose cDNA had not been cloned. Sequence comparisons with known rabs indicate that rab27a and rab27b comprise a melanocyte/platelet subfamily within the rab family. In addition, rab27a was expressed in a large variety of cell and tissue types, excluding brain, and rab27b manifested itself primarily in testis. Bacterially expressed and purified rab27a and rab27b exhibited GTP-binding activity and can now be used for antibody production and studies of the substrate specificities of geranylgeranyl transferase. In addition, the expression of rab27a and rab27b in both melanocytes and platelets makes them candidates for involvement in mouse and human disorders characterized by the combination of pigment dilution and a platelet storage pool defect.

Analysis of small GTP-binding proteins of the lens by GTP overlay assay reveals the presence of unique GTP-binding proteins associated with fiber cells

Low molecular weight GTP-binding proteins are molecular switches which are thought to play pivotal roles in cell growth, differentiation, cytoskeletal organization and vesicular trafficking. In this study, members of this family of proteins have been identified and characterized in the eye lens, for the first time. [alpha 33P]GTP blot overlay assays of monkey and human lens water soluble and membranous insoluble fractions revealed the presence of specific GTP-binding proteins in the range of 20-30 kDa (small GTPases) in both fractions, with much higher amounts in the membranous insoluble fraction. In the insoluble fraction, in addition to 20-30 kDa GTPases, there are three distinct GTP-binding proteins, ranging from 33-45 kDa. The small GTPases (20-30 kDa) were present throughout the lens in epithelium, cortex and nucleus, while the 33-45 kDa GTP-binding protein bands were exclusively associated with the cortex and nucleus (fiber cells). Analysis of lens fractions by two-dimensional electrophoresis, immunoprecipitation using monoclonal and sequence specific polyclonal antibodies and C3 exoenzyme mediated ADP-ribosylation demonstrated the presence of Ras, Rap, Rho, Rac, Rab and several other small GTPases. The 33-45 kDa GTP-binding proteins that are associated with lens fiber cells appear to be distinct from the small GTPases and from heterotrimeric GTPases, and were not detected in brain or heart tissue. The presence of different complements of GTP-binding proteins in lens fibers and epithelial, cells suggests their involvement in important regulatory functions, possibly related to cell growth, differentiation and organization of the cytoskeleton.

Identification of G-protein alpha-subunits during distinct stages of lens cell differentiation

PURPOSE

The purpose of this study was to identify alpha-subunits of heterotrimeric guanosine triphosphate-binding proteins in lens cell populations at various stages of terminal differentiation.

METHODS

Crude cell membranes were isolated from the annular pad, cortical fibers, and nuclear fibers of adult chickens and subjected to cholera and pertussis toxin-mediated ribosylation reactions. Specific labeling of toxin substrates was visualized after SDS-PAGE and radioautography. In complementary experiments, cell membranes were first separated by SDS-PAGE, transferred to a nitrocellulose support membrane, and probed with a panel of commercially available antibodies that recognize various classes of G-protein alpha-subunits.

RESULTS

A cholera toxin substrate was identified in cortical fibers whose labeling was dependent on a soluble factor. No cholera toxin substrates were labeled in annular pad cells. Two pertussis toxin substrates were seen in the relatively undifferentiated annular pad and the more differentiated cortical fibers. Relative abundance of the pertussis toxin substrates differed between the two cell types. Antibody staining revealed the presence of alpha-subunits belonging to the Gs, G(o), and G(i), families throughout the lens. Molecular weight differences of G alpha 2 polypeptides were noted between annular pad and cortical fiber cells.

CONCLUSIONS

These results suggest that G-protein-based signal transduction pathways continue to function in lens cells during various stages of terminal differentiation. Differences noted between annular pad and cortical fibers provide additional evidence that dynamic alterations in receptor-mediated processes may be responsible for the accumulation of differentiated characteristics during fiber formation. This also indicates that ocular pharmacologic intervention could affect various aspects of lens physiology throughout the process of fiber formation.

Simple purification and functional reconstitution of octopus photoreceptor Gq, which couples rhodopsin to phospholipase C

In invertebrate photoreceptors, illuminated rhodopsin activates multiple G proteins, which are assumed to initiate multiple phototransduction cascades. In this paper, we focused on one of the phototransduction cascades, which utilizes rhodopsin, a Gq-like G protein, and phospholipase C (PLC). A Gq-like G protein from octopus photoreceptors was successfully purified to apparent homogeneity as an active form by simple two-step chromatography. The purified G protein had an alpha beta gamma-trimeric structure consisting of 44-kDa alpha, 37-kDa beta, and 9-kDa gamma subunits. The 44-kDa alpha subunit was assigned to the Gq class by western blot with antiserum against mammalian Gq alpha and by partial amino acid sequencing of its proteolytic fragments. Light-dependent binding of GTP gamma S was observed when the purified octopus Gq was reconstituted with octopus rhodopsin that had been integrated into phospholipid vesicles. Octopus Gq activated PLC beta 1 purified from bovine brain dose-dependently in the presence of A1F4-. Finally, light- and GTP-dependent activation of PLC beta 1 was observed in a reconstitution system consisting of octopus rhodopsin, Gq, and bovine PLC beta 1.

How Ras-related proteins talk to their effectors

More and more effectors for the Ras-related protein superfamily are being discovered and it is emerging that these GTP-binding proteins interact with more than one effector to generate more than one cellular signal. Atomic details for the interaction of Rap/Ras with one of the effectors, the protein kinase c-Raf-1, have recently become available by X-ray structure analysis. The implications for the specificity of the signal transduction pathway, and how the GTP-dependent switch mechanism modulates the interaction with effectors will be discussed here, using Ras as a paradigm.

The SpoIIAA protein of Bacillus subtilis has GTP-binding properties

SpoIIAA is the first protein of the spoIIA operon. Here we show that SpoIIAA can bind and hydrolyze GTP. The protein also accepts ATP, but with lower affinity. GDP competes poorly for binding of GTP. The GTPase activity of SpoIIAA is within the range found for other GTP-binding proteins.

Expression in human endothelial cells of ADP-ribosylation factors, 20-kDa guanine nucleotide-binding proteins involved in the initiation of vesicular transport

ADP-ribosylation factors (ARFs) are approximately 20-kDa, guanine nucleotide-binding proteins, initially discovered as stimulators of cholera toxin ADP-ribosyltransferase activity and subsequently shown to participate in vesicular trafficking. Five of the six mammalian ARFs have been identified in human tissues by molecular cloning. They fall into three classes (class I: ARFs 1-3; class II: ARFs 4, 5; class III: ARF 6) based on deduced amino acid sequence, size, phylogenetic analysis, and gene structure. Similar to the rab family of approximately 20 kDa guanine nucleotide-binding proteins, the ARFs appear to function in specific trafficking pathways. The presence of a specific ARF might serve as a marker for that pathway. To verify expression of ARF mRNA and protein in human umbilical vein endothelial cells, immunoreactivity using antibodies specific for each ARF class, quantitative polymerase chain reaction (PCR) using ARF-specific, internal cRNA standards containing unique restriction enzyme cleavage sites introduced by point mutations, and Northern analysis with probes specific for ARFs 1, and 3-6, were utilized. PCR and Northern analysis were in agreement in showing that amounts of mRNA for ARF 1 and ARF 4 were similar and higher than those of ARF 3 and ARF 5 which were greater than ARF 6. Primarily, Class 1 ARF proteins were detected by immunoreactivity, with the majority in the supernatant fraction. The relative expression of ARFs in endothelial cells thus differs from that in neuronal tissues where it had been found that ARF3 is the predominant species.

Identification and isoprenylation of plant GTP-binding proteins

To identify isoprenylated plant GTP-binding proteins, Arabidopsis thaliana and Nicotiana tabacum cDNA expression libraries were screened for cDNA-encoded proteins capable of binding [32P]GTP in vitro. ATGB2, an Arabidopsis homologue of the GTP-binding protein Rab2, was found to bind GTP in vitro and to be a substrate for a geranylgeranyl:protein transferase (GGTase) present in plant extracts. The carboxyl terminus of this protein contains a -GCCG sequence, which has not previously been shown to be recognized by any prenyl:protein transferase (PTase), but which most closely resembles that isoprenylated by the type II GGTase (-XXCC, -XCXC, or -CCXX). In vitro geranylgeranylation of an Arabidopsis Rab1 protein containing a carboxyl-terminal-CCGQ sequence confirmed the presence of a type II GGTase-like activity in plant extracts. Several other proteins were also identified by in vitro GTP binding, including Arabidopsis and tobacco homologues of Rab11, ARF (ADP-ribosylation factor) and Sar proteins, as well as a novel 22 kDa Arabidopsis protein (ATG81). This 22 kDa protein had consensus GTP-binding motifs and bound GTP with high specificity, but its structure was not closely related to that of any known GTP-binding protein (it most resembled proteins within the ARF/Sar and G protein alpha-subunit superfamilies).

Rho family GTPases: the cytoskeleton and beyond

Until recently, members of the Rho family of GTPases were considered primarily to be regulators of the distinct structures making up the actin cytoskeleton. Several Rho GTPases have now also been shown to play an important role in cell transformation. In addition, Cdc42, Rac and Rho activate transcription, providing a possible mechanism for their role in growth control.

Sensory transduction in vomeronasal neurons: evidence for G alpha o, G alpha i2, and adenylyl cyclase II as major components of a pheromone signaling cascade

The mammalian vomeronasal organ (VNO) is an accessory olfactory structure implicated in the sensing of pheromones. Although virtually nothing is known about sensory transduction in the mammalian VNO, recent findings have raised the possibility that it proceeds via a G-protein-coupled mechanism and involves a cyclic nucleotide-gated ion channel as in the nasal olfactory epithelium. To investigate this possibility, we cloned G-protein alpha subunits, adenylyl cyclases, and guanylyl cyclases that are expressed in the VNO and examined their patterns of expression. Of seven G alpha subunits identified as being expressed in the VNO, we found that mRNAs encoding only two, G alpha o and G alpha i2, are highly expressed in VNO neurons. Moreover, G alpha o and G alpha i2 are highly expressed by separate subsets of neurons that are located in different regions of the VNO neuroepithelium. Immunohistochemical studies show that both G alpha o and G alpha i2 are enriched in VNO microvilli, suggesting that G-proteins containing both of these alpha subunits may be involved in VNO sensory transduction. Of the adenylyl and guanylyl cyclases that we cloned, we found that only one, adenylyl cyclase type II, is highly expressed in VNO neurons; furthermore, it is expressed by both G alpha o+ and G alpha i2+ subsets. Our findings suggest that spatially segregated subsets of VNO neurons may use different, but related, sensory transduction pathways in which G-proteins and an adenylyl cyclase play major roles.

Isolation of neonatal cardiomyocytes reduces the expression of the GTP-binding protein, Gh

alpha 1-Adrenoceptors in most tissues couple with the heterotrimeric GTP-binding protein Gq, the alpha subunit of which activates the beta-isoforms of phospholipase C. However, in heart (and in liver) alpha 1-adrenoceptors have been reported to couple to a high molecular weight GTP-binding protein. Gh, which functions both as a type II transglutaminase and as a receptor coupling protein. Gh activates a phospholipase isoform distinct from phospholipase C-beta. Here we report that isolation and culture of neonatal cardiomyocytes decreased the expression of Gh without reducing the content of Gq or Gi. Gh was readily detected in extracts from intact neonatal and adult heart tissues. The expression of Gh thus appears to be a feature of intact cardiac tissue.

Identification of heterotrimeric GTP-binding proteins in human megakaryoblastic leukemia cell line, MEG-01, and their alteration during cellular differentiation

Various heterotrimeric GTP-binding proteins (G proteins) are possible to have important functions in hematopoietic cells. However, there has been no information regarding their expression in magakaryoblasts and/or megakaryocytes. In the present study, protein contents of seven G protein alpha subunits (Gs alpha, Gi2 alpha, Gi3 alpha, Gz alpha, G11 alpha, Gq alpha and G12 alpha) and beta subunit in a human megakaryoblastic leukemia cell line, MEG-01, were analyzed by immunoblotting. Immature MEG-01 cells expressed the alpha subunits of Gs, Gi2, Gi3, Gz, G11 and G12 at protein molecule level. During the 12-O-tetradecanoyl-phorbol-13-acetate (TPA)-induced differentiation process, the contents of Gi2 alpha and Gi3 alpha increased, whereas the protein levels of Gz alpha, Gs alpha, G11 alpha and G12 alpha were observed to hardly change. beta subunit was also observed to be present in immature MEG-01 cells and to increase continuously throughout the differentiation process. For the expression of Gi2 alpha and beta subunits, chronic TPA-treatment was required although Rac2, a low M(r) GTP-binding protein, was expressed abundantly by only 30 min-TPA-treatment followed by 3 day-culture.

Differential coupling of m2 and m4 muscarinic receptors to inhibition of adenylyl cyclase by Gi alpha and G(o)alpha subunits

We compared the G-protein requirements for coupling of human and chicken m2 and m4 muscarinic acetylcholine receptors (mAChRs) to inhibition of adenylyl cyclase, using a luciferase reporter gene under the transcriptional control of a cAMP response element as a sensitive monitor of intracellular cAMP levels. Previously, we used this system to demonstrate that the chick m4 receptor preferentially coupled to Gi alpha-2 and G(o)alpha over Gi alpha-1 and Gi alpha-3. We found that both the chick and human m2 mAChRs can couple to Gi alpha-1, Gi alpha-2, Gi alpha-3, and G(o)alpha, while the human m4 mAChR preferentially couples to Gi alpha-2 and G(o)alpha. Both the G(o)1 and G(o)2 forms of the G(o)alpha subunit were effective in reconstituting coupling of the m2 and m4 mAChRs to inhibit adenylyl cyclase activity. The m2 and m4 mAChRs thus couple to inhibition of adenylyl cyclase by overlapping but different sets of G-protein alpha subunits.

Selective coupling of beta 2-adrenergic receptor to hematopoietic-specific G proteins

The coupling of the beta 2-adrenergic receptor (AR) to the alpha subunits of the Gq class of G proteins was investigated in a cotransfection system. COS-7 cells cotransfected with the beta 2-AR cDNA and the G alpha 15 or G alpha 16 cDNA showed marked norepinephrine-induced increases in accumulation of inositol phosphates in a concentration-dependent manner. However, cells cotransfected with the cDNA encoding G alpha q, G alpha 11, or G alpha 14 instead of G alpha 16 gave no ligand-dependent activation of phospholipase C (PLC). The facts that the beta-AR agonist isoprenaline can also induce activation of PLC in cells coexpressing beta 2-AR and G alpha 16 and that the beta 2-AR-specific antagonist propranolol can block norepinephrine-induced activation of PLC in these cotransfected cells further indicate that it is the beta 2-AR that mediates the activation of phospholipase C in these cotransfected cells. To test the possibility of involvement of G beta gamma, a G beta gamma antagonist, G gamma 3 mutant with substitution of a Ser residue for the C-terminal Cys residue, was used because this protein, when expressed in COS-7 cells, can inhibit only G beta gamma-mediated but not G alpha-mediated activation of PLC. The result that the G gamma 3 mutant could not inhibit beta 2-adrenergic receptor-mediated activation of PLC in cells cotransfected with the G alpha 16 cDNA suggests that G beta gamma is unlikely to be a major mediator of beta 2-adrenergic receptor-induced activation of PLC. Thus, we conclude that the beta 2-adrenergic receptor can specifically couple to G alpha 15 and G alpha 16, but not to G alpha q, G alpha 11, or G alpha 14 to activate PLC.

Expression of the mu-opioid receptor in CHO cells: ability of mu-opioid ligands to promote alpha-azidoanilido[32P]GTP labeling of multiple G protein alpha subunits

The identities of heterotrimeric G proteins that can interact with the mu-opioid receptor were investigated by alpha-azidoanilido[32P]GTP labeling of alpha subunits in the presence of opioid agonists in Chinese hamster ovary (CHO)-MORIVA3 cells, a CHO clone that stably expressed mu-opioid receptor cDNA (MOR-1). This clone expressed 1.01 x 10(6) mu-opioid receptors per cell and had higher binding affinity and potency to inhibit adenylyl cyclase for the mu-opioid-selective ligands [D-Ala2,N-MePhe4, Gly-ol]-enkephalin and [N-MePhe3,D-Pro4]-morphiceptin, relative to the delta-selective opioid agonist [D-Pen2,D-Pen5]-enkephalin or the kappa-selective opioid agonist U-50,488H. mu-Opioid ligands induced an increase in alpha-azidoanilido[32P]GTP photoaffinity labeling of four G alpha subunits in this clone, three of which were identified as Gi3 alpha, Gi2 alpha, and Go2 alpha. The same pattern of simultaneous interaction of the mu-opioid receptor with multiple G alpha subunits was also observed in two other clones, one expressing about three times more and the other 10-fold fewer receptors as those expressed in CHO-MORIVA3 cells. The opioid-induced increase of labeling of these G proteins was agonist specific, concentration dependent, and blocked by naloxone and by pretreatment of these cells with pertussis toxin. A greater agonist-induced increase of alpha-azidoanilido[32P]GTP incorporation into Gi2 alpha (160-280%) and Go2 alpha (110-220%) than for an unknown G alpha (G? alpha) (60%) or Gi3 alpha (40%) was produced by three different mu-opioid ligands tested. In addition, slight differences were also found between the ability of various mu-opioid agonists to produce half-maximal labeling (ED50) of any given G alpha subunit, with a rank order of Gi3 alpha > Go2 alpha > Gi2 alpha = G? alpha. In any case, these results suggest that the activated mu-opioid receptor couples to four distinct G protein alpha subunits simultaneously.