Expression of two human beta-adrenergic receptors in Escherichia coli: functional interaction with two forms of the stimulatory G protein

When expressed in Escherichia coli, the human beta 1- and beta 2-adrenergic receptors retain their ligand binding specificity. Their functional integrity was investigated by analyzing receptor-guanine nucleotide-binding regulatory (G) protein coupling by using two splice variants of the alpha subunit of the stimulatory G protein Gs synthesized in E. coli (rGs alpha-S and rGs alpha-L) and the beta gamma subunits of G protein purified from bovine brain. In competition binding experiments with (-)-[125I]iodocyanopindolol and (-)-isoproterenol, rGs alpha-S.beta gamma and rGs alpha-L.beta gamma reconstituted guanine nucleotide-sensitive high-affinity agonist binding with comparable affinities, whereas rGs alpha PT, a mutant of rGs alpha-L with an altered carboxyl terminus, and a recombinant subtype of the alpha subunit of the inhibitory G protein, rGi alpha-1, were approximately 20- and approximately 200-fold less potent, respectively. A comparison of the beta 1- and beta 2-adrenergic receptor expressed in E. coli with the beta 2-receptor in S49 murine lymphoma cyc- cell membranes revealed a similar affinity of rGs alpha-S and rGs alpha-L for the recombinant and native receptors. After stable incorporation of rGs alpha-S.beta gamma into E. coli membranes, receptor-G protein coupling was also verified by determining the isoproterenol-mediated acceleration of the rate for guanine 5'-[gamma-[35S]thio]triphosphate binding. These results show that (i) receptor-G protein coupling can be reconstituted in E. coli using recombinant components and that (ii) such an approach may be more generally used to evaluate coupling preferences between defined molecular species of receptors and G-protein subunits.

Are several G proteins involved in the different effects of endothelin-1 in mouse striatal astrocytes?

High-affinity specific receptors of endothelin (ET-1) were identified on primary cultures of mouse embryo striatal astrocytes by binding experiments performed with 125I-ET-1. Stimulation of production of inositol phosphates, a biphasic increase of the intracellular calcium concentration, and inhibition of cyclic AMP accumulation were observed in the same cells under ET-1 stimulation. Pretreatment of these cells with Bordetella pertussis toxin affected these effects to different extends, an observation suggesting that they are mediated by multiple transduction pathways, possibly involving several guanine nucleotide-binding proteins.

Functional endothelin-1 receptors in rat astrocytoma C6

Rat astrocytoma C6 cells have been recently identified as target cells for ET-1, which stimulates inositol lipid turnover in these cells. It is shown here that binding of ET-1 to high-affinity receptors on C6 cells leads to 40-45% inhibition of isoproterenol-induced intracellular cyclic AMP accumulation, as well as to stimulation of inositol lipid turnover, both effects characterized by an absolute requirement of extracellular calcium. Moreover, ET-1, which has been generally reported to have a mitogenic effect on a variety of target cells including primary rat astrocytes, is shown here to stimulate or, alternatively, inhibit DNA synthesis in C6 cells, depending on the subclone considered.

Expression of three human beta-adrenergic-receptor subtypes in transfected Chinese hamster ovary cells

The genes coding for three pharmacologically distinct subtypes of human beta-adrenergic receptors (beta 1 AR, beta 2 AR and beta 3 AR) were transfected for expression in Chinese hamster ovary (CHO) cells. Stable cell lines expressing each receptor were analyzed by ligand binding, adenylate cyclase activation and photoaffinity labeling, and compared to beta AR subtypes observed in previously described tissues, primary cultures and transfected cell lines. Each of the three receptor subtypes displayed saturable [125I]iodocyanopindolol-binding activity. They showed the characteristic rank order of potencies for five agonists, determined by measuring the accumulation of intracellular cAMP. These recombinant cell lines express a homogeneous population of receptors and display the known pharmacological properties of beta 1 AR and beta 2 AR, in human tissues. It is therefore likely that the pattern of ligand binding and adenylate cyclase activation, mediated by the new beta 3 AR in CHO cells, also reflects the yet-undetermined pharmacological profile in humans.

Coexpression of beta 1- and beta 2-adrenergic receptors on bovine brain capillary endothelial cells in culture

The pharmacological study of the blood-brain barrier has often been hampered by the unavailability of a large number of pure and fully differentiated brain capillary endothelial cells. Here we describe a homogeneous culture of brain capillary endothelial cells isolated from bovine brain (BBECs), which retain at least some phenotypic characteristics of the functional blood-brain barrier: intracellular tight junctions and monoamine oxidase activity. These cells were subcultured in vitro, in the absence of any neuronal or glial influences, for greater than 100 doublings without any sign of senescence. The present study is focused on the expression of beta-adrenergic receptors on BBECs. By Northern blot hybridization, subtype-specific ligand binding, and cyclic AMP accumulation experiments, we demonstrate that beta 1- and beta 2-adrenergic receptors are coexpressed (in the respective proportions of 42 and 58%) on BBEC membranes and are functionally coupled to adenylate cyclase. This is the first report documenting a significant number of beta 1-adrenergic receptors on brain capillary endothelial cells. The results are discussed in light of the known noradrenergic innervation of brain capillaries.

Coexistence of muscarinic acetylcholine receptors and somatostatin in nonpyramidal neurons of the rat dorsal hippocampus

This study describes the colocalization of muscarinic acetylcholine receptors (mAChRs) and the neuropeptide somatostatin (SOM) in nonpyramidal neurons of the rat dorsal hippocampus. SOM and mAChRs were identified by immunocytochemistry employing antibody S309 and M35, respectively. Half of the SOMergic cell population is found to be immunoreactive for muscarinic receptor protein as obtained by fluorescent double-labeling techniques. These findings provide additional evidence for a direct cholinergic influence upon SOMergic, nonpyramidal neurons, and defines the anatomical distribution of SOMergic, cholinoceptive neurons in the dorsal hippocampus. Concerning the muscarinic cholinoceptive, nonpyramidal neuron population of the dorsal hippocampus, a considerable number (approximately one-third) was found to be colocalized with somatostatin. These results indicate that a significant part of the cholinergic influence upon hippocampal nonpyramidal neurons is relayed via SOMergic neurons.

Colocalization of muscarinic and nicotinic receptors in cholinoceptive neurons of the suprachiasmatic region in young and aged rats

In the present study muscarinic and nicotinic cholinergic receptors in the SCN region were demonstrated and analyzed, employing monoclonal antibodies to purified muscarinic and nicotinic cholinergic receptor proteins. A near-total colocalization of the two acetylcholine receptor subclasses in cholinoceptive neurons of the SCN area was found. The antibodies applied to aging rat brain (at 30-34 months) revealed a clear decrease in immunoreactivity in senescence albeit with a high level of individual variability. Furthermore, in 8 out of 10 aged animals examined a considerable increase of astrocytes possessing muscarinic cholinergic receptors was observed.

Structure/function relationship of proteins belonging to the family of receptors coupled to GTP-binding proteins

The structural properties of a number of proteins belonging to the family of receptors coupled to GTP binding proteins are discussed in relation to their function. The structure of the ligand binding site and of the regions involved in coupling to the G proteins are analyzed mainly for the adrenergic and muscarinic cholinergic receptors, for which site-directed mutagenesis and chimaeric constructions have been studied. The structure of the genes are compared and the presence of various regulatory elements is discussed in relation to control of expression. Mechanism of desensitization and internalization, while mostly studied for the beta 2-adrenergic receptor, are proposed to be generally applicable to all G-protein-coupled receptors.

Receptor-based assays

Receptor-based assays have benefitted from the newest advances in biotechnology and electronics in three main ways: genetically engineered cells expressing single receptor subtypes have been developed for many natural and synthetic ligands; assays have been designed which take advantage of a variety of signals triggered in cells by binding, or inhibition of binding, of ligands to surface-bound receptors; and radiolabelled ligand assays have been considerably improved and simplified by novel electronic devices.

Haplotype specific homology scanning algorithm to predict T-cell epitopes from protein sequences

We present a homology scanning microcomputer program to predict functional T-cell epitopes within proteins. By taking into account particular human or mouse restriction elements the predictions are made haplotype-specific. The generality of this approach is confirmed by (i) identification of well-characterized immunogenic T-cell determinants in lysozyme (ii) search for potential T epitopes on unanalysed proteins like the human beta 2-adrenoreceptor (iii) modification of non-immunogenic peptide sequences in order to generate T-cell determinants.

Comparison of binding characteristics of endothelin receptors on subpopulations of astrocytes

Astrocytes in primary culture originating from different brain areas of the mouse embryo (striatum, cerebral cortex and mesencephalon) were compared for their [125I]-Endothelin-1 (ET-1) binding characteristics, in terms of affinity, binding capacity and specificity. Our results indicate that astrocytes from mesencephalon express about twice as many receptors as astrocytes from striatum or cortex (149,000 +/- 9700 vs 63,700 +/- 5600 and 81,900 +/- 5300, respectively), with similar affinities. Specificity patterns for the various peptides of the endothelins/sarafotoxins family (ET-1, -2, -3; SRTXa, b, c) are comparable in the three subpopulations of astrocytes: ET-1, -2 and SRTXb exhibit higher affinities than SRTXa and SRTXc. In addition, ET-3 and SRTXc seem to discriminate between different subsets of [125I]-ET-1 binding sites in the three subpopulations.

Effector-dependent conformational changes in protein kinase C gamma through epitope mapping with inhibitory monoclonal antibodies

Three monoclonal antibodies (mAb) directed against the regulatory domain of the protein kinase C gamma (PKC gamma); 15G4, 5A2 and 36G9, were shown to display distinct properties with respect to PKC gamma kinase activity [Cazaubon, S., Marais, R., Parker, P. & Strosberg, A.D. (1989) Eur. J. Biochem. 182, 401-406]. The mAb 5A2 and 36G9, which act as potent inhibitors of the cofactor-dependent kinase activity, can no longer bind PKC gamma in the presence of phosphatidylserine and phosphatidylserine/phorbol ester, respectively; 15G4 binding is not influenced by effectors. Due to this functional relationship between the inhibitory mAb- and cofactor-binding sites, we sought to localize the mAb epitopes with respect to the functional sites of PKC gamma. For this purpose, several deletions were introduced at the 5' end of the PKC gamma cDNA and the mutant proteins were expressed in Escherichia coli. The determination of the immunoreactivity of the deleted PKC gamma proteins shows that the amino acid residues essential to the binding of 5A2 and 36G9 are directly adjacent to the second cysteine-rich motif: these are contained in the sequences at positions 151-163 and 164-197, respectively. In addition, various deletions around the C1 region of the regulatory domain allowed the identification of the second cysteine-rich motif as a functional binding site for phorbol dibutyrate. These deletion studies thus demonstrate that the epitopes recognized by the inhibitory mAbs 5A2 and 36G9 are distinct from the cofactor-binding sites. This suggests that the binding of phosphatidylserine and phorbol ester induce conformational changes in the regulatory domain of PKC, which are thus responsible for the loss of the 5A2 and 36G9 immunoreactivity of the native protein. In this conformational state, PKC gamma conserves its ability to interact with the non-inhibitory mAb 15G4. By using synthetic peptides, the 15G4 epitope was localized to the sequence 297-310 in the V3 variable region. This indicates that the flexibility of the V3 region, which delimits the C-terminus of the regulatory domain, may not be necessary for the allosteric activation of PKC. In view of these results, we propose that PKC activation by its cofactors results in intramolecular changes which allow the enzyme to bind exogenous substrates.

Isolation, purification and partial analysis of the lipopolysaccharide antigenic determinant recognized by a monoclonal antibody to Legionella pneumophila serogroup 1

Monoclonal antibody II-6-18 recognizes a serogroup-1-specific Legionella pneumophila antigenic determinant which has been shown to be virulence-associated. We previously reported the physicochemical characterization by means of a quantitative fluorometric assay of monoclonal antibody II-6-18 binding to L. pneumophila, and its implications concerning the nature of the antigen. We describe here the isolation and the purification of the antigen by chemical and immunological methods, followed by its partial chemical analysis. The results demonstrate that the epitope--an immunodominant carbohydrate which includes a fucosamine-like residue--is part of the cell wall lipopolysaccharide (LPS). It is localized in the polysaccharide moiety of the LPS which contains KDO, rhamnose, mannose, glucosamine and an unidentified aminodideoxyhexose X1, but no heptose. The aminodideoxyhexose X1 could be fucosamine and is probably the immunodominant residue in the epitope, localized, at least partially, at the end of the polysaccharide chain.

Mapping of a functional autoimmune epitope on the beta 1-adrenergic receptor in patients with idiopathic dilated cardiomyopathy

The presence and properties of serum autoantibodies against beta-adrenergic receptors in patients with idiopathic dilated cardiomyopathy were studied using synthetic peptides derived from the predicted sequences of the human beta-adrenergic receptors. Peptides corresponding to the sequences of the second extracellular loop of the human beta 1- and beta 2-adrenergic receptors were used as antigens in an enzyme immunoassay to screen sera from patients with dilated cardiomyopathy (n = 42), ischemic heart disease (n = 17), or healthy blood donors (n = 34). The sera of thirteen dilated cardiomyopathy patients, none of the ischemic heart disease patients, and four of the healthy controls monospecifically recognized the beta 1-peptide. Only affinity-purified antibodies of these patients had a inhibitory effect on radioligand binding to the beta 1 receptor of C6 rat glioma cells. They recognized the receptor protein by immunoblot and bound in situ to human myocardial tissue. We conclude that a subgroup of patients with idiopathic dilated cardiomyopathy have in their sera autoantibodies specifically directed against the second extracellular loop of the beta 1-adrenergic receptor. These antibodies could serve as a marker of an autoimmune response with physiological and/or pathological implications.

Differential regulation of beta 1- and beta 2-adrenergic receptor protein and mRNA levels by glucocorticoids during 3T3-F442A adipose differentiation

The regulation by dexamethasone of beta 1- and beta 2-adrenergic receptor expression during the adipose differentiation of 3T3-F442A cells was investigated at the receptor protein and mRNA level. Preadipocytes were poorly responsive to beta-adrenergic receptor (beta-AR) agonists and expressed few beta-ARs (approximately 3,000 sites/cell) solely of beta 1 subtype. Differentiation increased adrenergic sensitivity and total beta-AR number (approximately 16,000 sites/cell) with a beta 1/beta 2 ratio of approximately 90/10. Long term exposure of either differentiating cells or mature adipocytes to dexamethasone induced down-regulation of (-)-isoproterenol-sensitive adenylate cyclase activity which paralleled a 2- to 3.5-fold decrease in beta-ARs, while the beta 1/beta 2 ratio switched to approximately 20/80. The ratios of beta 1/beta 2 binding sites were always consistent with the rank order of potency of beta-adrenergic agonists in stimulating the adenylate cyclase system. The action of steroid agonists and antagonist suggested a glucocorticoid receptor-mediated mechanism. The beta 1-AR mRNA (3.2 kilobases) was stimulated 3-4.7 times in differentiated cells, as compared with preadipose cells; this beta 1-AR transcript was repressed in dexamethasone-treated cells. The beta 2-AR mRNA species (2.3 kilobases), absent in preadipocytes, was expressed at low levels in untreated adipocytes, but reached 11-fold this level in dexamethasone-exposed cells. The switch in receptor subtype protein and mRNA levels elicited by dexamethasone demonstrates the differential genetic control by glucocorticoids of beta-AR subtype expression in 3T3-F442A cells. We suggest that this regulation of beta-AR gene expression requires interactions of glucocorticoid receptors with specific DNA targets and with one (or several) transcription factor(s) that are cell- and differentiation state-dependent.

Molecular cloning of cDNAs encoding bovine and human lactoperoxidase

Peptide sequences obtained from cyanogen bromide fragments of bovine lactoperoxidase (bLPO) were used to design oligonucleotide probes for library screening. These probes were used to screen a cDNA library constructed from bovine mammary tissue. Three overlapping clones were obtained, the longest of which (T3) contained a reading frame of 712 amino acid residues. The encoded amino acid sequence was homologous to those recently reported for myelo-, thyro-, and eosinophil peroxidases. Two possible amino termini of the mature enzyme were identified, and the predicted mature protein matched previous molecular weight estimates of 78,500. Of eight bovine tissues tested, transcription of T3 sequences were detected in mammary tissue only. Using the bLPO cDNA as a probe, a single hybridizing clone was found in a human mammary gland cDNA library. This clone (M1) encoded the carboxy-terminal 324 residues of a peroxidase distinct from the other three known human peroxidases, and was closely related to bLPO. This result confirms the presence of at least one distinct lactoperoxidase in humans.

Mutational analysis of ligand binding activity of beta 2 adrenergic receptor expressed in Escherichia coli

A novel in situ screening procedure was used to identify neutral mutations in the human beta 2 adrenergic receptor (beta 2AR). The coding region of the human beta 2AR gene was subcloned under transcriptional control of an inducible T7 promoter and used to transform Escherichia coli. Colonies expressing the beta 2AR bound the radiolabeled antagonist [125I]iodocyanopindolol and could be identified by autoradiography after transfer to nitrocellulose filters. The region of the beta 2AR between residues 76 and 83, in the second transmembrane helix, was mutagenized by a saturation mutagenesis technique, so that virtually all of the beta 2AR genes contained at least one mutation. Colonies retaining ligand binding activity were isolated using the in situ screen. Sequence analysis of the active mutant receptor genes allowed the identification of individual amino acid side chains which are essential for either ligand binding or structural integrity of the beta 2AR receptor.

Targeted killing of yeast expressing a HIV-1 peptide by antibody-conjugated glucose oxidase and horseradish peroxidase

The epitope recognized by monoclonal antibody directed against the HIV-1 recombinant gp160 protein was precisely delineated by using a number of peptides comprising amino acid positions 302-330 of the protein. Two different enzymes, glucose oxidase and horseradish peroxidase, were then coupled to distinct antibody molecules and the efficacy of the immunoenzymes in killing yeast cells which express the recognized peptide was evaluated by flow cytometry analysis. The antibody-glucose oxidase conjugate alone was cytotoxic only at large doses (over 35 micrograms/ml) while in the presence of the antibody-horseradish peroxidase conjugate, killing was observed at nine times lower concentrations (4 micrograms/ml). The procedure described here may provide a new immunotherapy tool for microbial infection.

Biotechnology of beta-adrenergic receptors

The emergence of Biotechnology has provided pharmacologists with a variety of methods for investigating the structure, the function, and the regulation of membrane-bound receptors with a precision that was not imagined even five years ago. These new tools have been developed and used to analyze the known catecholamine beta 1- and beta 2-adrenergic receptors and to discover and study a new subtype, the beta 3 receptor. We review here the salient features of each of these three receptors, compare their structural and functional properties, and propose models to explain their differential regulation in time and space. A whole family of proteins has now been found to share with the beta-adrenergic receptors their most prominent features, including seven transmembrane domains and coupling with GTP-binding "G" proteins. We therefore propose that the biotechnology-based procedures developed for the beta-adrenergic receptors will be well applicable to the other members of this "R7G" family of receptors.

Selective binding of ligands to beta 1, beta 2 or chimeric beta 1/beta 2-adrenergic receptors involves multiple subsites

The molecular basis of ligand binding selectivity to beta-adrenergic receptor subtypes was investigated by designing chimeric beta 1/beta 2-adrenergic receptors. These molecules consisted of a set of reciprocal constructions, obtained by the exchange between the wild-type receptor genes of one to three unmodified transmembrane regions, together with their extracellular flanking regions. Eight different chimeras were expressed in Escherichia coli and studied with selective beta-adrenergic ligands. The evaluation of the relative effect of each chimeric exchange on ligand binding affinity was based on the analysis of delta G values, calculated from the equilibrium binding constants, as a function of the number of substituted beta 2-adrenergic receptor transmembrane domains. The data showed that the contribution of each exchanged region to subtype selectivity varies with each ligand; moreover, while several regions are critical for the pharmacological selectivity of all ligands, others are involved in the selectivity of only some compounds. The selectivity displayed by beta-adrenergic compounds towards beta 1 or beta 2 receptor subtypes thus results from a particular combination of interactions between each ligand and each of the subsites, variably distributed over the seven transmembrane regions of the receptor; these subsites are presumably defined by the individual structural properties of the ligands.

Immunocytochemical visualization of muscarinic cholinoceptors in the human cerebral cortex

The monoclonal antibody M 35 was used to study the cellular and subcellular distribution of muscarinic acetylcholine receptors in the human cerebral cortex. M35, raised against purified muscarinic receptor protein, exerts muscarinic agonist-like effects at cholinergic synapses. Applied to human cortical tissue immediately fixed upon neurosurgical removal, light microscopically M 35 revealed immunoreactive perikarya predominantly of the pyramidal cell type in layers II/III and V, their labeled apical dendrites extending into the superficial layers. Furthermore, immunoprecipitate decorated distinct regions of the perikaryal cytoplasm, numerous dendritic profiles and synapses of both symmetric and asymmetric appearance. In the perikarya immunoprecipitate was associated with ribosomes, the endoplasmic reticulum and the Golgi apparatus. In dendrites the microtubular system, in synaptic complexes the postsynaptic membranes were decorated. The results suggest that novel informations about the cell type specific and subcellular distribution of human muscarinic cholinoceptors can be obtained by M 35 immunocytochemistry.

Localization and characterization of three different beta-adrenergic receptors expressed in Escherichia coli

After fusion with the N-proximal portion of the outer membrane protein LamB, three beta-adrenergic receptors, the human beta 1- and beta 2- and turkey beta 1-adrenergic receptor, were expressed in Escherichia coli with retention of their own specific pharmacological properties. Molecular characterization and localization of the three receptors in bacteria and comparison of the behaviour of each hybrid protein are reported. The bacteria were lysed and fractionated on a sucrose gradient. Saturable [125I]iodocyanopindolol binding activity was found associated mainly with the inner membrane fraction, suggesting that the receptor is correctly folded in this membrane. Binding activity was also found in the outer membrane fraction but varied according to the receptor type. Photoaffinity labeling experiments revealed that the receptors exhibit binding activity only after proteolytic removal of the LamB moiety from the fusion protein. The three hybrid proteins, detected in immunoblots by anti-peptide antibodies, were found mainly in the outer membrane fraction. Each of them exhibited different susceptibility to intrinsic bacterial proteolytic enzymes; sites of proteolytic cleavage were localized by the use of anti-peptide antibodies. The functional expression in E. coli of three beta-adrenergic receptors with similar structure but different amino acid sequences suggests that this expression system may be a general feature among similar receptors of the family of G-protein-coupled receptors. The level of expressed binding activity of a given receptor will be within the control of proteolytic degradation processes, depending on the primary sequence of the receptor. Constructions of new hybrid proteins, in combination with expression in protease mutants of E. coli, should help in controlling such processes.

Amino acid sequence of the variable domains of a human anti-Rh(c) antibody: presence of an unusually long CDR3 in the lambda chain

The complete amino acid sequence of the lambda light chain and the variable domain of the heavy chain of an anti-Rh(c) human monoclonal antibody were determined. The lambda chain presents a long third complementarity-determining region sequence with unusual amino acid insertions at the C-terminus. The proposed sequence indicates that this lambda chain may be assigned to the variable region subgroup I. The J segment is identical to that of J lambda 2 except for the first amino acid residue. Positions 152 (serine) and 190 (arginine) from this sequence correspond to the Kern-Oz- isotype, respectively. The VH segment can be classified as a VHIII subgroup member. The CDR1 segment of the anti-Rh(c) VH region has the same sequence as the VH of human BRO protein except for the first residue of the CDR1. The amino acid sequence of the anti-Rh(c) D segment does not match any published D segment. The JH segment used in this protein can be classified as a JH3 with a single amino acid difference at the fourth residue.

Internalization of beta-adrenergic receptor in A431 cells involves non-coated vesicles

To characterize the mechanism of internalization of beta-adrenergic catecholamine receptors on human epidermoid A431 carcinoma cells, their distribution was analyzed by immunocytochemistry using the monoclonal anti-receptor antibody BRK2. In preconfluent cultures, the receptors appeared to be randomly distributed on the cell surface. Exposure to the agonist isoproterenol induced an overall decrease in the number of cell surface receptors as determined by binding experiments and visualized by immunofluorescence. When cells were incubated at 4 degrees C with BRK2 and anti-mouse IgG-gold and then transferred at 37 degrees C, non-coated invaginations and vesicles were labeled. The addition of isoproterenol resulted in an increased rate of internalization of the receptor-BRK2-anti-IgG-gold complex. When incubation with the two antibody reagents was prolonged (with or without isoproterenol), non-coated vesicles fused in the endosomal compartment, and receptors were transferred to multivesicular bodies and lysosomes. At no stage in this process was there any indication that clathrin-coated pits or vesicles participated. Furthermore, we found that an intracellular potassium depletion treatment known to inhibit endocytosis, did not affect the normal pattern of desensitization of beta-adrenergic receptors.