DNA immunization combined with scFv phage display identifies antagonistic GCGR specific antibodies and reveals new epitopes on the small extracellular loops.

The identification of functional monoclonal antibodies directed against G-protein coupled receptors (GPCRs) is challenging because of the membrane-embedded topology of these molecules. Here, we report the successful combination of llama DNA immunization with scFv-phage display and selections using virus-like particles (VLP) and the recombinant extracellular domain of the GPCR glucagon receptor (GCGR), resulting in glucagon receptor-specific antagonistic antibodies. By immunizing outbred llamas with plasmid DNA containing the human GCGR gene, we sought to provoke their immune system, which generated a high IgG1 response. Phage selections on VLPs allowed the identification of mAbs against the extracellular loop regions (ECL) of GCGR, in addition to multiple VH families interacting with the extracellular domain (ECD) of GCGR. Identifying mAbs binding to the ECL regions of GCGR is challenging because the large ECD covers the small ECLs in the energetically most favorable 'closed conformation' of GCGR. Comparison of Fab with scFv-phage display demonstrated that the multivalent nature of scFv display is essential for the identification of GCGR specific clones by selections on VLPs because of avid interaction. Ten different VH families that bound 5 different epitopes on the ECD of GCGR were derived from only 2 DNA-immunized llamas. Seven VH families demonstrated interference with glucagon-mediated cAMP increase. This combination of technologies proved applicable in identifying multiple functional binders in the class B GPCR context, suggesting it is a robust approach for tackling difficult membrane proteins.

Analysis of the glucagon receptor first extracellular loop by the substituted cysteine accessibility method.

Glucagon is an important hormone for the prevention of hypoglycemia, and contributes to the hyperglycemia observed in diabetic patients, yet very little is known about its receptor structure and the receptor-glucagon interaction. In related receptors, the first extracellular loop, ECL1, is highly variable in length and sequence, suggesting that it might participate in ligand recognition. We applied a variant of the SCAM (Substituted Cysteine Accessibility Method) to the glucagon receptor ECL1 and sequentially mutated positions 197 to 223 to cysteine. Most of the mutations (15/27) affected the glucagon potency, due either to a modification of the glucagon binding site, or to the destabilization of the active receptor conformation. We reasoned that side chains accessible to glucagon must also be accessible to large, hydrophilic cysteine reagents. We therefore evaluated the accessibility of the introduced cysteines to maleimide-PEO(2)-biotin ((+)-biotinyl-3-maleimido-propionamidyl-3,6-dioxa-octanediamine), and tested the effect of pretreatment of intact cells with a large cationic cysteine reagent, MTSET ([2-(trimethylammonium)ethyl]methanethiosulfonate bromide), on glucagon potency. Our results suggest that the second and third transmembrane helices (TM2 and TM3) are extended to position 202 and from position 215, respectively, and separated by a short ß stretch (positions 203-209). Glucagon binding induced a conformational change close to TM2: L198C was accessible to the biotin reagent only in the presence of glucagon. Most other mutations affected the receptor activation rather than glucagon recognition, but S217 and D218 (at the top of TM3) were good candidates for glucagon recognition and V221 was very close to the binding site.

Mutational and cysteine scanning analysis of the glucagon receptor N-terminal domain.

The glucagon receptor belongs to the B family of G-protein coupled receptors. Little structural information is available about this receptor and its association with glucagon. We used the substituted cysteine accessibility method and three-dimensional molecular modeling based on the gastrointestinal insulinotropic peptide and glucagon-like peptide 1 receptor structures to study the N-terminal domain of this receptor, a central element for ligand binding and specificity. Our results showed that Asp(63), Arg(116), and Lys(98) are essential for the receptor structure and/or ligand binding because mutations of these three residues completely disrupted or markedly impaired the receptor function. In agreement with these data, our models revealed that Asp(63) and Arg(116) form a salt bridge, whereas Lys(98) is engaged in cation-π interactions with the conserved tryptophans 68 and 106. The native receptor could not be labeled by hydrophilic cysteine biotinylation reagents, but treatment of intact cells with [2-(trimethylammonium)ethyl]methanethiosulfonate increased the glucagon binding site density. This result suggested that an unidentified protein with at least one free cysteine associated with the receptor prevented glucagon recognition and that [2-(trimethylammonium)ethyl]methanethiosulfonate treatment relieved this inhibition. The substituted cysteine accessibility method was also performed on 15 residues selected using the three-dimensional models. Several receptor mutants, despite a relatively high predicted cysteine accessibility, could not be labeled by specific reagents. The three-dimensional models show that these mutated residues are located on one face of the protein. This could be part of the interface between the receptor and the unidentified inhibitory protein, making these residues inaccessible to biotinylation compounds.

An alternative presentation of metabolism: Enzyme-catalyzed reactions can be viewed as atoms flowing through sluice gates.

Why should we think of enzymes as sluice gates? We are used to talk about "metabolic flux": I should like to suggest that we should indeed envision metabolism as a "liquid" flowing in different channel segments, each of which contains a single metabolite. Enzymes facilitate chemical reactions without affecting the reaction equilibrium, creating, or destroying matter: they can be visualized as "sluice gates" (mobile metal or wooden plates that control water flow in channels and sluices: Fig. 1), or, rather, as openings in metal or wooden plates that block the channel. Metabolism control and homeostasis maintenance are equivalent to controlling the water flux while maintaining the water level in each channel segment approximately constant. The "gates" representing non cooperative (Michaelis-Menten) and cooperative enzymes have very different shapes, reflecting their different roles in metabolic regulation: the activity of noncooperative enzymes catalyzing irreversible reactions has the greater effect over the flux, but cooperative enzymes are necessary to maintain homeostasis when irreversible reactions are involved.

Propiverine and metabolites: differences in binding to muscarinic receptors and in functional models of detrusor contraction.

Propiverine is a commonly used antimuscarinic drug used as therapy for symptoms of an overactive bladder. Propiverine is extensively biotransformed into several metabolites that could contribute to its spasmolytic action. In fact, three propiverine metabolites (M-5, M-6 and M-14) have been shown to affect various detrusor functions, including contractile responses and L-type calcium-currents, in humans, pigs and mice, albeit with different potency. The aim of our study was to provide experimental evidence for the relationship between the binding of propiverine and its metabolites to human muscarinic receptor subtypes (hM(1)-hM(5)) expressed in chinese hamster ovary cells, and to examine the effects of these compounds on muscarinic receptor-mediated detrusor function. Propiverine, M-5, M-6 and M-14 bound to hM(1)-hM(5) receptors with the same order of affinity for all five subtypes: M-6 > propiverine > M-14 > M-5. In HEK-293 cells expressing hM(3), carbachol-induced release of intracellular Ca(2+) ([Ca(2+)](i)) was suppressed by propiverine and its metabolites; the respective concentration-response curves for carbachol-induced Ca(2+)-responses were shifted to the right. At higher concentrations, propiverine and M-14, but not M-5 and M-6, directly elevated [Ca(2+)](i). These results were confirmed for propiverine in human detrusor smooth muscle cells (hDSMC). Propiverine and the three metabolites decreased detrusor contractions evoked by electric field stimulation in a concentration-dependent manner, the order of potency being the same as the order of binding affinity. We conclude that, in comparison with the parent compound, loss of the aliphatic side chain in propiverine metabolites is associated with higher binding affinity to hM(1)-hM(5) receptors and higher functional potency. Change from a tertiary to a secondary amine (M-14) results in lower binding affinity and reduced potency. Oxidation of the nitrogen (M-5) further lowers binding affinity as well as functional potency.

Techniques: promiscuous Galpha proteins in basic research and drug discovery.

Assay technologies that measure the activation of heterotrimeric (alphabetagamma) G proteins by G-protein-coupled receptors (GPCRs) are well established within the pharmaceutical industry, either for pharmacological characterization or for the identification of natural or surrogate receptor ligands. Despite recent evidence indicating that GPCR-linked signalling events might not be mediated exclusively by G proteins, G-protein activation remains a common benchmark for assessing GPCR family members. Thus, assay systems that translate ligand-mediated modulation of GPCRs into G-protein-dependent intracellular responses still represent key components of both basic research and the drug discovery process. In this article, the current knowledge and recent progress of integrating Galpha subunits into assay systems for GPCR drug discovery will be reviewed. Emphasis is given to novel promiscuous and chimeric Galpha proteins. Because of their ability to interact with a wide range of GPCRs, such novel G proteins are likely to be incorporated rapidly into drug discovery programmes.

Motilin and erythromycin-A share a common binding site in the third transmembrane segment of the motilin receptor.

UNLABELLED: The motilin receptor (MTLR) represents a clinically useful pharmacological target, as agonists binding to the MTLR have gastroprokinetic properties. In order to compare the molecular basis for interaction of the MTLR with motilin and with the non-peptide motilin agonist, erythromycin-A (EM-A), the negatively charged E119 located in the third transmembrane (TM3) region was mutated to D (E119D) and Q (E119Q), respectively, and changes in activity of the mutant receptors were verified. METHODS: Each mutant receptor was stably transfected in CHO-cells containing the Ca2+ indicator apo-aequorin. Receptor activation in response to motilin, EM-A and their analogues was assessed by Ca2+-luminescense. RESULTS: In the E119Q mutant, the Ca2+ response to motilin and EM-A was abolished while in the E119D mutant it was reduced with 62% (motilin) and 81% (EM-A). The pEC50 values were shifted from 9.65+/-0.03 to 7.41+/-0.09 (motilin) and from 6.63+/-0.12 to 4.60+/-0.07 (EM-A). Acetylation of the N-terminal amine group as in [N-acetyl-Phe]1 mot (1-14), decreased the potency 6.3-fold (WT-MTLR) and 148-fold (E119D). Acetylation of EM-A enol ether induced a more pronounced shift in potency: 7943-fold (WT-MTLR) and 1413-fold (E119D). CONCLUSION: The comparable loss of affinity of the mutant receptors for motilin and EM-A indicate that these agonists both interact with the TM3 domain of the MTLR. The results with acetylated derivatives support an ionic interaction between E119 of the MTLR with the N+ of the desosamine sugar in EM-A, but not with the N+ of the free amine group in motilin.

G protein activation by G protein coupled receptors: ternary complex formation or catalyzed reaction?

G protein coupled receptors catalyze the GDP/GTP exchange on G proteins, thereby activating them. The ternary complex model, designed to describe agonist binding in the absence of GTP, is often extended to G protein activation. This is logically unsatisfactory as the ternary complex does not accumulate when G proteins are activated by GTP. Extended models taking into account nucleotide binding exist, but fail to explain catalytic G protein activation. This review puts forward an enzymatic model of G protein activation and compares its predictions with the ternary complex model and with observed receptor phenomenon. This alternative model does not merely provide a new set of formulae but leads to a new philosophical outlook and more readily accommodates experimental observations. The ternary complex model implies that, HRG being responsible for efficient G protein activation, it should be as stable as possible. In contrast, the enzyme model suggests that although a limited stabilization of HRG facilitates GDP release, HRG should not be "too stable" as this might trap the G protein in an inactive state and actually hinder G protein activation. The two models also differ completely in the definition of the receptor "active state": the ternary complex model implies that the active state corresponds to a single active receptor conformation (HRG); in contrast, the catalytic model predicts that the active receptor state is mobile, switching smoothly through various conformations with high and low affinities for agonists (HR, HRG, HRGGDP, HRGGTP, etc.).

Ghrelin degradation by serum and tissue homogenates: identification of the cleavage sites.

The endogenous ligand for the GH secretagogue receptor is ghrelin, a peptide recently purified from the stomach. Ghrelin is n-octanoylated on the Ser(3) residue, and this modification is essential for its interaction with the receptor. The degradation of ghrelin by rat and human serum, purified commercial enzymes, and tissues homogenates was analyzed by combining HPLC and mass spectrometry. In serum, ghrelin was desoctanoylated, without proteolysis. The desoctanoylation was significantly reduced by phenylmethylsulfonyl fluoride, a serine proteases and esterases inhibitor. In rat serum, the carboxylesterase inhibitor bis-p-nitrophenyl-phosphate totally inhibited ghrelin desoctanoylation, and a correlation was found between ghrelin desoctanoylation and carboxylesterase activity. Moreover, purified carboxylesterase degraded ghrelin. Thus, carboxylesterase could be responsible for ghrelin desoctanoylation in that species. In human serum, ghrelin desoctanoylation was partially inhibited by eserine salicylate and sodium fluoride, two butyrylcholinesterase inhibitors, but not by bis-p-nitrophenyl-phosphate and EDTA. Purified butyrylcholinesterase was able to degrade ghrelin, and there was a correlation between the butyrylcholinesterase and ghrelin desoctanoylation activities in human sera. This suggested that several esterases, including butyrylcholinesterase, contributed to ghrelin desoctanoylation in human serum. In contact with tissues homogenates, ghrelin was degraded by both desoctanoylation and N-terminal proteolysis. We identified five cleavage sites in ghrelin between residues -Ser(2)-(acyl)Ser(3)- (stomach and liver), -(acyl?)Ser(3)-Phe(4)- (stomach, liver, and kidney), -Phe(4)-Leu(5)- (stomach and kidney), -Leu(5)-Ser(6)- and -Pro(7)-Glu(8)- (kidney). In all cases, the resulting fragments were biologically inactive.

Allosteric drugs acting at muscarinic acetylcholine receptors.

The binding properties of muscarinic acetylcholine receptors are affected by various drugs acting at a second (allosteric) binding site, usually (but not always) at supratherapeutic concentrations. Allosteric drugs acting at GABA receptors present advantages over competitive drugs; this explains the interest raised by allosteric effects on muscarinic receptors. A theoretical and practicable definition of allosteric drugs acting at muscarinic receptors will be given in this work, together with a summary of recent data concerning the number, position, and structural requirements of their binding sites.

Evidence for a direct interaction between the Thr11 residue of vasoactive intestinal polypeptide and Tyr184 located in the first extracellular loop of the VPAC2 receptor.

We developed previously VPAC(1) [vasoactive intestinal peptide (VIP)/pituitary adenylate cyclase-activating peptide (PACAP) receptor]>VPAC(2) receptor selective ligands. Replacement of the VIP-Thr(11) by an Arg(11) in these ligands contributed to their selectivity: Arg(11)-VIP had a 200-fold lower affinity when compared with VIP at VPAC(2) receptors as opposed to 3- to 5-fold higher affinity at VPAC(1) receptors. Comparison of the binding and functional properties of related VIP analogues suggested that the VPAC(1) selectivity of Arg(11)-VIP was due to the loss of a hydrogen bond between the hydroxy group of Thr residue and the VPAC(2) receptor, steric hindrance between the Arg side chain and the VPAC(2) receptor and charge attraction by the VPAC(1) receptor. Comparison of the ability of VIP analogues to activate adenylate cyclase through chimaeric VPAC(1)/VPAC(2) and VPAC(2)/VPAC(1) receptors indicated that the first extracellular receptor loop carried most of the VPAC(2) receptors' ability to discriminate VIP from Arg(11)-VIP. Based on results obtained for a truncated VPAC(2) receptor and the closely related PACAP-preferring receptor (PAC(1)) and secretin receptors, we hypothesized that Thr(11) interacted with the VPAC(2) receptor Tyr(184) (similar to the VPAC(1) receptor Phe(200) residue). The Y184F (Tyr(184)-->Phe) VPAC(2) mutant lost the ability to discriminate VIP from Val(11)-VIP, and the F200Y VPAC(1) mutant acquired the ability to discriminate the natural peptide from Val(11)-VIP. These results support the hypothesis that the hydroxy group of the native VIP-Thr(11) side chain can indeed form a hydrogen bond with the Tyr side chain in the VPAC(2) receptor.

Synthesis and affinity studies of himbacine derived muscarinic receptor antagonists.

A series of himbacine (1)-related analogues has been prepared featuring three different isomeric configurations with respect to the B-ring (a, b and natural c) and three different interconnecting two-carbon unsaturated units [natural (E)-ene, (Z)-ene, and yne]. The study of the binding affinities of the nine resulting compounds, including synthetic (+)-himbacine (3c), towards the M(1)-M(4) muscarine receptor subtypes revealed that analogues 3a and 5c display a promising 10-fold selectivity for the M(2) receptor as compared to the M(1) receptor.

Transglutaminase-mediated polyamination of vasoactive intestinal peptide (VIP) Gln16 residue modulates VIP/PACAP receptor activity.

Previous data showing an increase of receptor binding activity of [R16]VIP, a vasoactive intestinal peptide (VIP) structural analogue containing arginine at the position 16 of its amino acid sequence, have pointed out the importance of a positive charge at this site. Here, the functional characterization of three VIP polyaminated adducts (VIPDap, VIPSpd, and VIPSpm), obtained by a transglutaminase-catalysed reaction between the VIP Gln16 residue and 1,3-diaminopropane (Dap), spermidine (Spd), or spermine (Spm), is reported. Appropriate binding assays and adenylate cyclase enzymatic determinations have shown that these VIP adducts act as structural VIP agonists, both in vitro and in vivo. In particular, their IC50 and EC50 values of human and rat VIP/pituitary adenylate cyclase activating peptide (PACAP)1 and VIP/PACAP2 receptors indicate that VIPDap is a VIP agonist, with an affinity and a potency higher than that of VIP, while VIPSpd and VIPSpm are also agonists but with affinities lower than that of VIP. These findings suggest that the difference in adduct agonist activity reflects the differences in the positive charge and carbon chain length of the polyamine covalently linked with the VIP Gln16 residue. In addition, the data obtained strongly suggest that the length of polyamine carbon chain could be critical for the interaction of the agonist with its receptor, even though possible hydrophobic interaction cannot be ruled out. In vivo experiments on murine J774 macrophage cell cultures have shown the ability of these compounds to stimulate the inducible nitric oxide synthase activity at the transcriptional level.

A small sequence in the third intracellular loop of the VPAC(1) receptor is responsible for its efficient coupling to the calcium effector.

The stimulatory effect of VIP on intracellular calcium concentration ([Ca(2+)](i)) has been investigated in Chinese hamster ovary cells stably transfected with the reporter gene aequorin, and expressing human VPAC(1), VPAC(2), chimeric VPAC(1)/VPAC(2), or mutated receptors. The VIP-induced [Ca(2+)](i) increase was linearly correlated with receptor density and was higher in cells expressing VPAC(1) receptors than in cells expressing a similar VPAC(2) receptor density. The study was performed to establish the receptor sequence responsible for that difference. VPAC(1)/VPAC(2) chimeric receptors were first used for a broad positioning: those having the third intracellular loop (IC(3)) of the VPAC(1) or of the VPAC(2) receptor behaved, in that respect, phenotypically like VPAC(1) and VPAC(2) receptor, respectively. Replacement in the VPAC(2) receptor of the sequence 315-318 (VGGN) within the IC(3) by its VPAC(1) receptor counterpart 328-331 (IRKS) and the introduction of VGGN in state of IRKS in VPAC(1) was sufficient to mimic the VPAC(1) and VPAC(2) receptor characteristics, respectively. Thus, a small sequence in the IC(3) of the VPAC(1) receptor, probably through interaction with G(alphai) and G(alphaq) proteins, is responsible for the efficient agonist-stimulated [Ca(2+)](i) increase.

VPAC(1) receptors have different agonist efficacy profiles on membrane and intact cells.

The vasoactive intestinal peptide receptor VPAC(1) is preferentially coupled to G(alpha s) protein but also increases [Ca(2+)](i) through interaction with G(alpha i)/G(alpha q) protein. We evaluated a panel of full, partial and null agonists for their capability to stimulate adenylate cyclase activity in both intact cells and membrane and [Ca(2+)](i) in intact cells transfected with the reporter gene aequorin. In intact cells, the agonists efficacy for cAMP and calcium increase were well, but not linearly correlated: VPAC(1) receptors activated G(alpha s) protein more efficiently but with the same pharmacological profile as the other G proteins. In contrast, there was a difference between cAMP increase in intact and broken cell membranes: EC(50) values were generally lower in intact cells whereas the efficacy was higher. There was, however, no correlation between the shift in the EC(50) value and the intrinsic activity. Of interest, the (4-28) fragment, a reported antagonist on cell membrane, was a full agonist in intact cells. We concluded that the active states of the VPAC(1) receptor resulting from the coupling to different effector are undistinguishable by the VIP analogs tested but that receptor properties are different when evaluated in intact cells or cell membranes.

Mutational analysis of the glucagon receptor: similarities with the vasoactive intestinal peptide (VIP)/pituitary adenylate cyclase-activating peptide (PACAP)/secretin receptors for recognition of the ligand's third residue.

Receptor recognition by the Asp(3) residues of vasoactive intestinal peptide and secretin requires the presence of a lysine residue close to the second transmembrane helix (TM2)/first extracellular loop junction and an ionic bond with an arginine residue in TM2. We tested whether the glucagon Gln(3) residue recognizes the equivalent positions in its receptor. Our data revealed that the binding and functional properties of the wild-type glucagon receptor and the K188R mutant were not significantly different, whereas all agonists had markedly lower potencies and affinities at the I195K mutated receptor. In contrast, glucagon was less potent and the Asp(3)-, Asn(3)- and Glu(3)-glucagon mutants were more potent and efficient at the double-mutated K188R/I195K receptor. Furthermore, these alterations were selective for position 3 of glucagon, as shown by the functional properties of the mutant Glu(9)- and Lys(15)-glucagon. Our results suggest that although the Gln(3) residue of glucagon did not interact with the equivalent binding pocket as the Asp(3) residue of vasoactive intestinal peptide or secretin, the Asp(3)-glucagon analogue was able to interact with position 188 of the K188R/I195K glucagon receptor. Nevertheless, the Gln(3) side chain of glucagon probably binds very close to this region in the wild-type receptor.