Selective targeting of G-protein-coupled receptor subtypes with venom peptides.

The G-protein-coupled receptor (GPCR) family is one of the largest gene superfamilies with approx. 370 members responding to endogenous ligands in humans and a roughly equal amount of receptors sensitive to external stimuli from the surrounding. A number of receptors from this superfamily are well recognized targets for medical treatment of various disease conditions, whereas for many others the potential medical benefit of interference is still obscure. A general problem associated with GPCR research and therapeutics is the insufficient specificity of available ligands to differentiate between closely homologous receptor subtypes. In this context, venom peptides could make a significant contribution to the development of more specific drugs. Venoms from certain animals specialized in biochemical hunting contain a mixture of molecules that are directed towards a variety of membrane proteins. Peptide toxins isolated from these mixtures usually exhibit high specificity for their targets. Muscarinic toxins found from mamba snakes attracted much attention during the 1990s. These are 65-66 amino acid long peptides with a structural three-finger folding similar to the α-neurotoxins and they target the muscarinic acetylcholine receptors in a subtype-selective manner. Recently, several members of the three-finger toxins from mamba snakes as well as conotoxins from marine cone snails have been shown to selectively interact with subtypes of adrenergic receptors. In this review, we will discuss the GPCR-directed peptide toxins found from different venoms and how some of these can be useful in exploring specific roles of receptor subtypes.

Adrenoceptor activity of muscarinic toxins identified from mamba venoms.

BACKGROUND AND PURPOSE: Muscarinic toxins (MTs) are snake venom peptides named for their ability to interfere with ligand binding to muscarinic acetylcholine receptors (mAChRs). Recent data infer that these toxins may have other G-protein-coupled receptor targets than the mAChRs. The purpose of this study was to systematically investigate the interactions of MTs with the adrenoceptor family members. EXPERIMENTAL APPROACH: We studied the interaction of four common MTs, MT1, MT3, MT7 and MTα, with cloned receptors expressed in insect cells by radioligand binding. Toxins showing modest to high-affinity interactions with adrenoceptors were additionally tested for effects on functional receptor responses by way of inhibition of agonist-induced Ca²âº increases. KEY RESULTS: All MTs behaved non-competitively in radioligand displacement binding. MT1 displayed higher binding affinity for the human α(2B)-adrenoceptor (IC50 = 2.3 nM) as compared with muscarinic receptors (IC50 ≥ 100 nM). MT3 appeared to have a broad spectrum of targets showing high-affinity binding (IC50 = 1-10 nM) to M4 mAChR, α(1A)-, α(1D)- and α(2A)-adrenoceptors and lower affinity binding (IC50 ≥ 25 nM) to α(1B)- and α(2C)-adrenoceptors and M1 mAChR. MT7 did not detectably bind to other receptors than M1, and MTα was specific for the α(2B)-adrenoceptor. None of the toxins showed effects on ß1- or ß2-adrenoceptors. CONCLUSIONS AND IMPLICATIONS: Some of the MTs previously found to interact predominantly with mAChRs were shown to bind with high affinity to selected adrenoceptor subtypes. This renders these peptide toxins useful for engineering selective ligands to target various adrenoceptors.

The STC-1 cells express functional orexin-A receptors coupled to CCK release.

Orexins are newly discovered neuropeptides regulating feeding and vigilance and have been detected in neuroendocrine cells of the gut. Potential neuroendocrine functions of orexin are unknown. Therefore, the effects of orexin-A on the intestinal neuroendocrine cell line, STC-1, were investigated as a model system. RT-PCR demonstrated the presence of both OX(1) and OX(2) receptors. Stimulation with orexin-A produced a dose-dependent release of cholecystokinin (CCK), which was abolished by removal of extracellular Ca(2+) or the presence of the voltage-gated L-type Ca(2+)-channel blocker diltiazem (10 microM). Orexin-A (Ox-A) elevated intracellular Ca(2+), which was dependent on extracellular Ca(2+). Furthermore, orexin-A caused a membrane depolarization in the STC-1 cells. Ox-A neither elevated cAMP levels nor stimulated phosphoinositide turnover in these cells. These data demonstrate a functional orexin receptor in the STC-1 cell line. Ox-A produces CCK release in these cells, by a mechanism involving membrane depolarization and subsequently activation of L-type voltage-gated Ca(2+)-channels.

Modelling of promiscuous receptor-Gi/Gs-protein coupling and effector response.

A single G-protein-coupled receptor might activate multiple G-protein species. This multiplex coupling ability can be used by tissues to regulate signalling; for the pharmacologist, such multiplex coupling might cause difficulties in the interpretation of experimental data. In this article, we present mathematical models for the activation of two separate G-protein species by a single receptor. Issues addressed concern mutual antagonism between the G proteins and the availability of an already activated receptor for interaction with a new G protein (receptor-G-protein-effector complexing versus free diffusion of G proteins) in addition to receptor-G-protein precoupling at different G-protein and receptor expression levels. The output from the receptor models uses, as readout, a new model for adenylyl cyclase regulation by two allosteric regulators (i.e. G(s) and G(i)).

Role of G-protein availability in differential signaling by alpha 2-adrenoceptors.

The impact of G-protein expression on the coupling specificity of the human alpha(2B)-adrenergic receptor (alpha(2B)-AR) was studied in Sf9 cells. The alpha(2B)-AR was shown to activate both coexpressed G(s)- and G(i)-proteins in a [(35)S]GTPgammaS binding assay. Noradrenaline and the synthetic agonist UK14,304 were equally potent and efficacious in stimulating G(i) activation. At the effector level (adenylyl cyclase), both ligands stimulated cAMP production. In the presence of forskolin, the effects of the agonists were more complex. Noradrenaline stimulated cAMP production, while UK14,304 showed a biphasic concentration-response curve with inhibition of stimulated cAMP production at low agonist concentrations and further stimulation at high agonist concentrations. G(s) coexpression caused a monophasic stimulatory response with both ligands. Coexpression with G(i) resulted in a biphasic concentration-response curve for noradrenaline and a monophasic inhibition with UK14,304. Experiments with a panel of agonists demonstrated that the more efficacious an agonist is in stimulating cAMP production, the weaker is its ability to couple to inhibition of cAMP accumulation via exogenous G(i). To be able to explain the mechanistic consequences of dual G-protein coupling described above, we developed a mathematical model based on the hypothesis that an agonist induces different conformations of the receptor having different affinity for different G-proteins. The model reproduced the profiles seen in the concentration-response curves with G(s) and G(i) coexpression. The model predicts that the affinity of the receptor conformation for G-proteins as well as the availability of G-proteins will determine the ultimate response of the receptor.

Recombinant expression of a selective blocker of M(1) muscarinic receptors.

Mamba venoms contain peptides with high selectivity for muscarinic receptors. Due to the limited availability of the M(1) muscarinic receptor-selective MT7 or m1-toxin 1, the peptide was expressed in Sf9 cells using a synthetic cDNA and purified. The isolated peptide had over four orders of magnitude higher affinity for the M(1) compared to M(2)-M(5) muscarinic receptors. The peptide strongly inhibited Ca(2+) mobilisation through recombinant and endogenously expressed M(1) receptors, having no effect on the function of the other subtypes. The MT7 peptide provides a unique tool for identification and functional characterisation of M(1) receptors in cells and tissues.

Role of third intracellular loop of galanin receptor type 1 in signal transduction.

To determine the domains essential for G-protein coupling of the human galanin receptor type 1 (GalR1), we have used both GalR1 mutants and synthetic receptor-derived peptides in(125)I-galanin and [(35)S]-GTPgammaS binding studies. Replacement of potential phosphorylation sites by Leu in the third intracellular loop (IC3) of GalR1 did not affect K(D)values for the receptor. Peptides derived form the IC3 loop, and especially the N-terminal part of it were able to increase the rate of [(35)S]-GTPgammaS binding to the trimeric Gialpha1beta1gamma2, but not to Gsalphabeta1gamma2, whereas the peptides corresponding to the IC1 and IC2 loops had no such effect. IC3 loop peptides also inhibited the binding of(125)I-galanin to GalR1 in membranes from Rin m5F cells. Our results suggest that the IC3 loop of GalR1, especially its N-terminal part, defines the coupling of the receptor to the Gialpha1beta1gamma2 protein and consequently, to the signal transduction cascade.

Influence of ionic strength on drug adsorption onto and release from a poly(acrylic acid) grafted poly(vinylidene fluoride) membrane.

Ion exchange resins have several applications in pharmacy for controlled or sustained release of drugs. In the present study, effects of the ionic strengths of adsorption medium and dissolution medium on drug adsorption onto and release from a acrylic acid grafted poly(vinylidene fluoride) (PAA-PVDF) were studied. Despite their porosity, PAA-PVDF membranes act reasonable well as cation exchange membranes. It was observed, that ionic strength of adsorption medium, degree of grafting and concentration of propranolol-HCl in adsorption medium affect propranolol-HCl adsorption onto the membrane. The fluxes of smaller molecules (MW < 500) across the membrane decreased with ionic strength of buffer solution, whereas the fluxes of the large molecules (FITC-dextran, MW 4400) increased with ionic strength. Release rate of adsorbed propranolol-HCl from the membrane into phosphate buffer was greatly affected by ionic strength of adsorption medium. These results can be explained by a cation exchange process between membrane and cations present in the buffer solution and swelling behavior of the grafted PAA chains.

Endogenous extracellular purine nucleotides redirect alpha2-adrenoceptor signaling.

Many receptors coupled to inhibitory Go/Gi-type G proteins often also produce stimulatory signals like Ca2+ mobilisation. When expressed in CHO cells the alpha2-adrenoceptor subtypes alpha2A, alpha2B and alpkha2C mobilised Ca2+. These responses were strongly reduced by the P2Y-purinoceptor antagonist suramin. A large proportion of the total pool of purine nucleotides was found extracellularly. Removal of extracellular nucleotides with apyrase or by constant perfusion had a similar effect as suramin. These treatments did not affect the alpha2-adrenoceptor-mediated inhibition of cAMP production. This indicates that cells may be primed or their signaling pathways redirected towards Ca2+ mobilisation by 'autocrine' release of nucleotides.

Transport of drugs across porous ion exchange membranes.

Porous ion exchange membranes have potential applications for drug delivery systems. Permeability of these membranes can be controlled by environmental factors like pH and ionic strength but also the drug properties have an important role in the permeation process. In this paper the influence of the drug charge, lipophilicity and molecular weight on the diffusional drug flux is demonstrated. The membranes under study were poly(acrylic acid) (PAA) grafted porous poly(vinylidene fluoride) (PVDF) membranes which are cation selective due to the partial ionization of carboxyl groups in grafted PAA chains. At low pH the membrane pores are open and the drugs can diffuse through the membrane quite easily. However, at pH 7 the grafted chains partially block the pores and the diffusional flux of bigger drug molecules (Mw9400) decreases five orders of magnitude and also the flux of smaller molecules is clearly reduced. When the influence of the drug charge on the diffusion of the drugs across the membranes was studied, it turned out that the PAA-PVDF membranes facilitate the transport of cationic drugs and repel anionic ones. The presented mathematical model, based on Donnan drugs equilibrium and measured transport number data, predicted the observed trends reasonably well.

Protean agonism at alpha2A-adrenoceptors.

The coupling of the endogenously expressed alpha2A-adrenoceptors in human erythroleukemia cells (HEL 92.1.7) to Ca2+ mobilization and inhibition of forskolin-stimulated cAMP production was investigated. The two enantiomers of medetomidine [(+/-)-[4-(1-[2, 3-dimethylphenyl]ethyl)-1H-imidazole]HCl] produced opposite responses. Dexmedetomidine behaved as an agonist in both assays (i.e. , it caused Ca2+ mobilization and depressed forskolin-stimulated cAMP production). Levomedetomidine, which is a weak agonist in some test systems, reduced intracellular Ca2+ levels and further increased forskolin-stimulated cAMP production and therefore can be classified as an inverse agonist. A neutral ligand, MPV-2088, antagonized responses to both ligands. Several other, chemically diverse alpha2-adrenergic ligands also were tested. Ligands that could promote increases in Ca2+ levels and inhibition of cAMP production could be classified as full or partial agonists. Their effects could be blocked by the alpha2-adrenoceptor antagonist rauwolscine and by pertussis toxin treatment. Some typical antagonists such as rauwolscine, idazoxan, and atipamezole had inverse agonist activity like levomedetomidine. The results suggest that the alpha2A-adrenoceptors in HEL 92.1.7 cells exist in a precoupled state with pertussis toxin-sensitive G proteins, resulting in a constitutive mobilization of intracellular Ca2+ and inhibition of cAMP production in the absence of agonist. This constitutive activity can be antagonized by inverse agonists such as levomedetomidine and rauwolscine. Levomedetomidine can be termed a "protean agonist" because it is capable of activating uncoupled alpha2-adrenoceptors in other systems and inhibiting the constitutive activity of precoupled alpha2-adrenoceptors in HEL 92.1. 7 cells. With this class of compounds, the inherent receptor "tone" could be adjusted, which should provide a new therapeutic principle in receptor dysfunction.

[3H]RS79948-197 binding to human, rat, guinea pig and pig alpha2A-, alpha2B- and alpha2C-adrenoceptors. Comparison with MK912, RX821002, rauwolscine and yohimbine.

The Kd values of the recently introduced radioligand [3H]RS79948-197 ((8a R,12aS,13a-S)-5,8,8a,9,10,11,12,12a,13,13a-decahydro-3-metho xy-12-(ethylsulphonyl)-6H-isoquino[2,1-g][1,6]naphthyridine) were determined for the recombinant human and rat alpha2A-, alpha2B- and alpha2C- as well as guinea pig alpha2B- and alpha2c-adrenoceptors expressed in COS (CV-1 Origin, SV40) cells. In addition, the Kd values were also determined for [3H]RS79948-197 for the guinea pig spleen alpha2A-adrenoceptor and for pig alpha2A-, alpha2B- and alpha2C-adrenoceptors in membranes obtained from kidney and striatum. Available radioligands for alpha2-adrenoceptors, besides [3H]RS79948-197 are the tritiated forms of MK912 ((2S,12bS)1',3'-dimethylspiro(1,3,4,5',6,6',7,12b-octa hydro-2H-benzo[b]furo[2,3-a]quinazoline)-2,4'-pyrimidin-2'-one), RX821002 (2-methoxy-idazoxan), rauwolscine and yohimbine. In the present article the binding constants of all these substances for the alpha2A-, alpha2B- and alpha2C-adrenoceptor subtypes in human, pig, rat and guinea pig are reviewed. In all species tested MK912 was alpha2C-selective, RX821002 showed a minor alpha2A-selectivity, whereas [3H]RS79948-197 was non-selective among the alpha2-adrenoceptor subtypes, showing high affinity for all three subtypes. Rauwolscine and yohimbine showed relatively low affinities for nmost of the alpha2-adrenoceptor subtypes investigated, the exception being rauwolscine having high affinity for the human and porcine alpha2C-adrenoceptors.

Pseudo-noncompetitive antagonism of M1, M3, and M5 muscarinic receptor-mediated Ca2+ mobilization by muscarinic antagonists.

Muscarinic receptors M1, M3, and M5 were expressed in Sf9 cells. Three different patterns of inhibition of Ca2+ elevations could be resolved for the subtype nonselective muscarinic receptor antagonists: (i) a right shift of the agonist dose-response curve, (ii) a right shift of the agonist dose-response curve and a depression of the maximum signal, and (iii) an intermediate pattern where the antagonist apparently behaved more competitively at higher concentrations. A simulation performed assuming that these differences are due to differences in the dissociation rates of the antagonists reproduced all three different modes of inhibition; the novel intermediate pattern (iii) is suggested to be caused by an intermediate antagonist dissociation rate. A direct correlation between the type of inhibition and the measured dissociation rate of the antagonists was also observed. Functional selectivity between receptor subtypes based on the dissociation constants is suggested based on the results.

Alpha2B-adrenoceptors couple to Ca2+ increase in both endogenous and recombinant expression systems.

The ability of cloned human alpha2B-adrenoceptors heterologously expressed in Sf9 cells and endogenous alpha2B-adrenoceptors in NG 108-15 neuroblastoma x glioma cells to couple to increase of intracellular Ca2+ was studied. Ca2+ increases in NG 108-15 cells were detectable but slight, whereas those in alpha2B-adrenoceptor-expressing Sf9 cells were greater. In the latter, the maximum Ca2+ increase correlated positively, and the EC50-value of noradrenaline negatively, with the receptor expression density. The order of potency of the agonists was D-medetomidine ([D]-4-[5]-[1-(2,3-dimethylphenyl)ethyl]-1H-imidazole) > noradrenaline approximately = clonidine > oxymetazoline, with clonidine and UK14,304 (5-bromo-N-[4,5-dihydro-1H-imidazole-2-yl]-6-quinoxalinamine) being weak partial agonists. In Sf9 cells Ca2+ increases consisted of concomitant mobilization from an intracellular store and influx of extracellular Ca2+. In these cells alpha2B-adrenoceptor stimulation also increased the inositol 1,4,5-trisphosphate mass. We conclude that alpha2B-adrenoceptors can couple to intracellular Ca2+ increases which may involve prior activation of phospholipase C.

The second intracellular loop of the alpha2-adrenergic receptors determines subtype-specific coupling to cAMP production.

The alpha2-adrenergic receptors (alpha2-ARs), which primarily couple to inhibition of cAMP production, have been reported to have a stimulating effect on adenylyl cyclase activity in certain cases. When expressed in Spodoptera frugiperda Sf9 cells the alpha2A subtype showed only inhibition of forskolin-stimulated cAMP production when activated by norepinephrine (NE), whereas the alpha2B subtype displayed a biphasic dose-response curve with inhibition at low concentrations of NE and a potentiation at higher concentrations. To further investigate the subtype-specific coupling, we expressed a set of chimeric alpha2A-/alpha2B-ARs at similar expression levels in Sf9 cells to determine the structural domain responsible for the difference between the two subtypes. When the third intracellular loops were interchanged between alpha2A and alpha2B subtypes, the coupling specificity remained unchanged, indicating that this loop does not confer selectivity toward a stimulating response. A biphasic dose-response curve, typical for the alpha2B subtype, could be seen when the second intracellular loop of the alpha2B subtype was inserted into the alpha2A subtype, suggesting that this loop is important for determining the subtype-specific coupling of alpha2-ARs to cAMP production. Site-directed mutagenesis of non-conserved amino acids in the second intracellular loop of the alpha2A subtype indicated that several residues are involved in the coupling specificity.

Functional properties of muscarinic receptor subtypes Hm1, Hm3 and Hm5 expressed in Sf9 cells using the baculovirus expression system.

The human muscarinic ACh receptor subtypes m1, m3 and m5 have been expressed in Sf9 cells using the baculovirus expression system. Stimulation of all three subtypes with CCh caused an increase in inositol-1,4,5-trisphosphate and intracellular Ca++. The increase in cytosolic free Ca++ was to a large extent due to influx. The levels of receptors (< 0.1-1 pmol/mg protein) increased with infection time in a narrow time span (24-36 h). The changes in the receptor densities did not significantly affect the EC50 values of CCh-mediated Ca++ mobilization with the m3 or the m5 subtype. The EC50 value was higher with the m1 receptor at low expression levels (approximately 100 fmol/mg protein), and it decreased with an increase in receptor density. The receptor subtypes displayed no gross differences in their response to oxotremorine-M, which behaved as a full agonist, or to oxotremorine and pilocarpine, which were less active. With the m3 subtype, there was an increase in the maximal response to oxotremorine with longer infection times. The results demonstrate that the recombinant muscarinic receptors, expressed in Sf9 cells, show many of the characteristics of endogenously expressed receptors when studied at low expression levels and that the receptor density may significantly affect the receptor pharmacology.

Glass-ionomer cements based on poly(acrylic acid-co-vinyl alcohol) in drug release model formulations.

The release of methylene blue from glass-ionomer cements based on a commercial glass powder and partly hydrolysed poly(acrylic acid-co-vinyl alcohol) has been studied. A series of random copolymers were synthesized from acrylic acid and vinyl acetate monomers, using feed ratios of 10-90 mol% acrylic acid. The acetate group was hydrolysed using sodium hydroxide. According to proton nuclear magnetic resonance analysis, the degree of hydrolysis was about 40%. The residual Na+ content in the samples after hydrolysis was in the vicinity of 3 mmol g-1 polymeric sample. The viscosity average molecular weight of the copolymers was the region of 60,000 g mol-1 for samples with low acrylic acid content and in the region of 300,000 g mol-1 for samples with a larger fraction of acrylic acid. Cement sample discs were prepared and the porosity of the cement was determined using N2 adsorption-desorption isotherms. The specific surface area was 10 m2 g-1, and the volume of the pores (< 20 nm) 0.035 cm3 g-1. The swelling indices and the release rates in two different buffers, pH 2.0 and pH 7.4, were determined for fresh and dried samples. An acidic environment did not bring about significant swelling in the samples and release of the model agent at pH 2.0 was limited. At a physiological pH value the cements swelled in a hydrogel manner and the fractional release of model agent increased.

Two human alpha 2-adrenoceptor subtypes alpha 2A-C10 and alpha 2B-C2 expressed in Sf9 cells couple to transduction pathway resulting in opposite effects on cAMP production.

The baculovirus expression vector system utilizing the strong polyhedrin gene promoter of the Autographa californica nuclear polyhedrosis virus (AcNPV) was used for high level expression of the two alpha 2-adrenoceptor subtypes alpha 2A-C10 and alpha 2B-C2 in Spodoptera frugiperda (Sf-9) insect cells. For rapid screening of recombinant viruses the luciferase gene was expressed under the early ETL-promoter (early transcript large) in the same plasmid. Both receptor subtypes showed the same rank order of binding affinity for four agonists tested: dexmedetomidine > l-medetomidine = clonidine > noradrenaline. For the alpha 2A-C10 subtype, these agonists inhibited forskolin stimulated cAMP production through pertussis toxin sensitive G-proteins. In contrast, for the alpha 2B-C2 subtype the agonists stimulated both basal and forskolin stimulated cAMP production.

Muscarinic receptor subtypes in human neuroblastoma cell lines SH-SY5Y and IMR-32 as determined by receptor binding, Ca++ mobilization and northern blotting.

Muscarinic receptor subtypes in neuroblastoma cell lines IMR-32 and SH-SY5Y were determined with receptor binding, Ca++ mobilization and Northern blotting. Displacement of [3H]NMS with pirenzepine in IMR-32 cells revealed apparent binding sites with Kd values of 5 (41%) and 237 nM (59%). With 4-diphenylacetoxy-N-metylpiperidine metiodid, a similar proportion of apparent high- and low-affinity binding was obtained: 36 (Kd = 0.26 nM) and 64% (Kd = 6.3 nM), respectively. In SH-SY5Y cells, two different affinities with apparent Kd of 40 (24%) and 460 nM (76%) could be distinguished with pirenzepine, even though the Kd of the apparent high-affinity site varied markedly (variation = 8.7-96.8 nM). Inhibition of carbachol-induced Ca++ mobilization displayed high sensitivity to 4-diphenylacetoxy-N-methylpiperidine metiodid in both cell lines. IMR-32 cells displayed high sensitivity to pirenzepine, whereas the sensitivity varied between different batches of SH-SY5Y cells. DNA fragments (approximately 1000 base pairs) from SH-SY5Y DNA amplified with polymerase chain reaction were used as probes for muscarinic receptor mRNA. Northern blotting with the Hm1-specific probe gave a stronger signal for SH-SY5Y than for IMR-32, whereas the result obtained with the Hm2-probe was the opposite. Also, the Hm3 mRNA was detected in SH-SY5Y cells. The Hm4 and Hm5 transcripts were not detected in either of these cell lines.