Pharmacological onomastics: what's in a name?

Drugs are named for their primary receptor target and overt action (agonism, antagonism) but the observation of multiple or collateral efficacies emanating from drugs activating a single receptor target is posing a challenge for drug classification and nomenclature. With increasing abilities to detect alteration in cellular function has come the identification of efficacies that are not necessarily manifest in obvious changes in cell response. Specifically, some agonists selectively activate cellular pathways, demonstrate phenotypic behaviour associated with cell type and some antagonists actively induce receptor internalization without activation. In addition, the effects of allosteric modulators can be linked to the nature of the co-binding ligand posing a similar complication in classification and naming. Thus, accurate labels for this new generation of selective drugs may require identification of receptor partners (G-protein type, beta-arrestin) or pathway or, in the case of allosteric modulators, identification of co-binding ligands. The association of distinct phenotypic behaviours with molecules opens the opportunity to better associate clinical effects with distinct pharmacological properties.

Quantitation in receptor pharmacology.

The activity of drugs on receptors can be visualized in appropriate systems. However, for comparisons of activity to be made, these effects must be quantified. The common currency for this quantitation is the dose-response curve. Thus, the shape, location (along the concentration axis), and maximal asymptote of dose-response curves are used to quantify drug activity and comparisons to mathematical models of receptors are made to describe mechanism of action. This review cites examples of where these quantitative procedure yield information beyond what is readily apparent through observation of the data and thus support the use of quantitative methods to maximize the information gained from experiments.

Uncovering biases in high throughput screens of G-protein coupled receptors.

The ability of high throughput membrane binding assays to detect ligands for G-protein coupled receptors was examined using mathematical models. Membrane assay models were developed using the extended ternary complex model (Samama et al., 1993) as a basis. Ligand binding to whole cells was modeled by adding a G-protein activation step. Results show that inverse agonists bind more slowly and with a lower affinity to receptors in the membrane binding assay than to receptors in whole cells, causing the membrane assay to miss pharmaceutically important inverse agonists. Assay modifications to allow detection of inverse agonists are discussed. Finally, kinetic binding data are shown to provide information about ligand efficacy. This work demonstrates the utility of mathematical modeling in detecting biases in drug-screening assay, and also in suggesting techniques to correct those biases.

Protean effects of a natural peptide agonist of the G protein-coupled secretin receptor demonstrated by receptor mutagenesis.

G protein-coupled receptors initiate signaling cascades after associating with heterotrimeric G proteins. This is typically initiated by agonist binding, but can also occur spontaneously, particularly in receptors bearing distinct missense mutations. Two such mutations in the parathyroid hormone receptor are associated with constitutive activity, manifesting clinically as Jansen's metaphyseal chondroplasia. We introduce analogous mutations separately and together into the secretin receptor to explore their impact on another family member. Constructs were expressed transiently in COS cells, and had binding and signaling (cAMP generation) studied. Each construct was processed appropriately to lead to cell surface expression and signaling. Secretin bound to the wild-type receptor with two affinity states recognized, 1% of sites in the high affinity state (Ki = 0.5 +/- 0.1 nM) and 99% in the low affinity state (Ki = 23 +/- 3 nM). Mutant receptor binding best fit a single affinity state, having values for Ki of 5 +/- 1 nM (H156R), 8 +/- 1 nM (T322P) and 6 +/- 1 nM (H156R/T322P), with each of these demonstrating a shift to higher affinity than the predominent low affinity state of the wild-type receptor. Each mutant receptor expressed small to moderate constitutive activity, with basal levels of cAMP activity greater than control (P < .01): H156R, 1.4-fold; T322P, 4.5-fold and H156R/T322P, 6.8-fold. The level of basal activity of even the most active construct was only 15% of the maximal response of wild-type receptor. Although each of the single site mutants responded to secretin by increasing their cAMP levels in a concentration-dependent manner, the dual mutant decreased its cAMP in response to hormone (EC50 = 13 nM). Thus, a natural agonist had become an inverse agonist at this unique construct. Because this could reflect reduced normal coupling with Gs or increased aberrant coupling with Gi, the mechanism was further explored using pertussis toxin and a stable analogue of GTP. Although ligand-binding determinants were retained in the dual receptor mutant, the conformation of this receptor upon secretin binding effected a reduction in its basal coupling with Gs, thereby resulting in inverse agonism.

A mathematical model for analysis of pharmacologically induced changes in the kinetics of cardiac muscle.

A mathematical model of the isometric contraction of cardiac muscle is developed and utilized to characterize the inotropic and lusitropic effects of cardioactive compounds in isolated guinea pig left atria. In contrast to metrics that are based on minima and maxima of an isometric twitch and its derivative function, the entire time course of the twitch is used to quantify the kinetics of the contraction-relaxation cycle. The model relates observed tension to a time-dependent activation function that describes generation of internal force and a coupling function that determines mechanical response to the activation function. The model is structured so that it is suitable for nonlinear curve fitting to observed data. Results obtained using the model for fitting experimental data from tissues treated with different classes of cardioactive compounds agree with more qualitative results presented by other authors. Experiments using the model to fit data over an extended (90 min) time course revealed differences in the kinetic profiles of milrinone and forskolin. Computer simulations that demonstrate the effect of each model parameter on twitch kinetics are presented, and the relationships between the model and other theoretical and empirical models of cardiac muscle are discussed. The mathematical model is useful to enable a more quantitative understanding of the kinetics of cardiac muscle contraction and relaxation and identify compounds that may be selective for inotropic or lusitropic effects.

The cubic ternary complex receptor-occupancy model. III. resurrecting efficacy.

Early work in pharmacology characterized the interaction of receptors and ligands in terms of two parameters, affinity and efficacy, an approach we term the bipartite view. A precise formulation of efficacy only exists for very simple pharmacological models. Here we extend the notion of efficacy to models that incorporate receptor activation and G-protein coupling. Using the cubic ternary complex model, we show that efficacy is not purely a property of the ligand-receptor interaction; it also depends upon the distributional details of the receptor species in the native receptor ensemble. This suggests a distinction between what we call potential efficacy (a vector) and realized efficacy (a scalar). To each receptor species in the native receptor ensemble we assign a part-worth utility; taken together these utilities comprise the potential efficacy vector. Realized efficacy is the expectation of these part-worth utilities with respect to the frequency distribution of receptor species in the native receptor ensemble. In the parlance of statistical decision theory, the binding of a ligand to a receptor ensemble is a random prospect and realized efficacy is the utility of this prospect. We explore the implications that our definition of efficacy has for understanding agonism and in assessing the legitimacy of the bipartite view in pharmacology.

Use of resampling techniques to estimate the variance of parameters in pharmacological assays when experimental protocols preclude independent replication: an example using Schild regressions.

Estimates of variance in pharmacological assays are usually made by repeating the experiment with different tissues. Biological factors, such as the inability to wash a drug from tissue, may preclude the type of replication that is appropriate for the statistics of interest. For example, in Schild regressions, replication is usually done at each concentration of antagonist. In some test systems, replication of dose-response curves is not possible. For example, some persistent agonists cannot be removed from tissues after exposure, while in other systems, rapid desensitization severely alters tissue sensitivity to repeated challenge with agonist. In this paper, we demonstrate how a statistical resampling method, bootstrapping, can be used to derive estimates of the confidence intervals for pA2, pKB, and slope from Schild plots. This method utilizes the speed of the computer to estimate variance by repeatedly resampling the data. The advantage to this method is that it can be used for many different experimental designs. For a data set obtained from a Schild regression of atenolol antagonism of isoproterenol in the guinea pig left atrium, bootstrap estimates of confidence limits were calculated for cases where dose ratios were derived from the same tissue and randomly paired tissues. These estimates showed good agreement with estimates obtained using conventional analytical methods, thus suggesting that this method may be useful in practice.

Physiological effects of inverse agonists in transgenic mice with myocardial overexpression of the beta 2-adrenoceptor.

G-protein-coupled receptors are thought to have an inactive conformation (R), requiring an agonist-induced conformational change for receptor/G-protein coupling. But new evidence suggests a two-state model in which receptors are in equilibrium between the inactive conformation (R), and a spontaneously active conformation (R*) that can couple to G protein in the absence of ligand (Fig. 1). Classic agonists have a high affinity for R* and increase the concentration of R*, whereas inverse agonists have a high affinity for R and decrease the concentration of R*. Neutral competitive antagonists have equal affinity for R and R* and do not displace the equilibrium, but can competitively antagonize the effects both of agonists and of inverse agonists. The lack of suitable in vivo model systems has restricted the evidence for the existence of inverse agonists to computer simulations and in vitro systems. We have used a transgenic mouse model in which there is such marked myocardial overexpression of beta 2-adrenoceptors that a significant population of spontaneously activated receptor (R*) is present, inducing a maximal response without agonist. We show that the beta 2-adrenoceptor ligand ICI-118,551 functions as an inverse agonist, providing evidence supporting the existence of inverse agonists and validating the two-state model of G-protein-coupled receptor activation.

Tissue response as a functional discriminator of receptor heterogeneity: effects of mixed receptor populations on Schild regressions.

A model is described that predicts the behavior of competitive antagonists in tissues with more than one receptor mediating response. The receptor stimuli for two receptor types are summed and processed, via a cellular stimulus-response mechanism, into tissue response. The primary receptor is described by a standard Langmurian isotherm, and a secondary receptor input with a variable maximal strength, for which the agonist has variable sensitivity, is added. The prediction of drug effects in this system does not depend on the way in which the two stimuli are combined or on the absolute magnitudes of the parameters used to make the calculations. The model is maximally flexible, in that no pharmacological significance is put on the magnitudes of the inputs from the secondary receptor system (i.e., they can vary with either agonist intrinsic efficacy, receptor number, or efficiency of stimulus-response coupling). The theoretical Schild regressions for selective antagonists in two-receptor systems are calculated for various secondary receptor inputs. These regressions generally are curvilinear whenever the secondary receptor significantly contributes to agonist response. These calculated data also indicate that minor variations in biological input from secondary receptor systems would obscure curvature in the Schild regression and result in a seemingly linear regression with a slope of less than unity. However, further calculations indicate three possible ways to use Schild analysis to detect receptor heterogeneity in tissues. One indicator of receptor heterogeneity is a change in the slope of the dose-response curve for the agonist in the presence of a selective antagonist. A second indicator would be a marked heteroscedasticity of errors in the Schild regression, i.e., the magnitude of the standard errors in the ordinate values would depend upon the concentration of the antagonist. A third, and most experimentally accessible, aspect of heterogeneous receptor systems predicts that changes in the overall sensitivity of organ response mechanisms will differentially alter the relative strength of two receptor inputs. This would be observed as a change in the potency of an antagonist. Under these circumstances, differences in the stimulus-response characteristics of a tissue would result in a change in the Schild regression for a selective antagonist. These concepts are discussed in terms of the use of Schild analysis in functional systems for the detection of physiologically relevant mixtures of receptors and the possible advantages over biochemical binding data.

Biphasic dose-response curves to arecoline in rat atria-mediation by a single promiscuous receptor or two receptor subtypes?

Arecoline produces a biphasic response in rat left atria, i.e., a depression of basal inotropy at low doses and a positive inotropic effect at higher doses. These present studies were designed to determine whether it can be shown that the two separate responses to arecoline are mediated by two distinct cell surface muscarinic receptors. The antagonists scopolamine, 4-DAMP and AF-DX 116 produced apparent simple competitive antagonism of the negative responses to arecoline. Schild analysis was used to measure the equilibrium dissociation constant of the antagonist-receptor complex for antagonism of this response to arecoline by these antagonists. In atria from rats treated with pertussis toxin, the negative inotropy to arecoline was abolished and only the positive inotropic effects were observed. The antagonism of the positive inotropic response to arecoline by these antagonists was studied separately in atria from rats treated with pertussis toxin by the Schild technique. The pKB estimates made from the Schild regressions indicated no evidence to suggest that the two responses to arecoline (negative and positive inotropy) were mediated by two separate receptors in rat left atria. These data are discussed in terms of a single muscarinic receptor in this tissue mediating these two responses by interaction with two G-proteins in the same cell membrane. These data also are discussed in terms of the use of agonist potency ratios for the classification of receptors.

The relative efficiency of beta adrenoceptor coupling to myocardial inotropy and diastolic relaxation: organ-selective treatment for diastolic dysfunction.

The relative effects of drugs which elevate cytosolic cyclic AMP on inotropy and diastolic relaxation (lusitropy) of guinea pig atria were quantified in vitro. There was a temporal difference between these responses in that inotropy reached peak response considerably faster than lusitropy. Also, although the relaxation response was sustained to an elevated steady state, the inotropic responses to beta adrenoceptor agonists were transient and returned to base line over 90 min. However, the inotropic responses to forskolin and dibutyryl cyclic AMP (cAMP) were sustained. For all of the drugs tested, the lusitropic response was at least 4 times more sensitive than the inotropic response (i.e., the concentration response curve for relaxation was shifted to the left of the curve for inotropy). In the case of beta adrenoceptor agonists, these differences were greater, presumably because of the fading inotropic response over 90 min. It was found that although high efficacy beta adrenoceptor agonists such as isoproterenol (and the direct activator of adenylate cyclase forskolin) produced both inotropy and lusitropy, lower efficacy agonists produced predominant lusitropy. The low efficacy agonist prenalterol produced insignificant inotropy but 60% maximal lusitropy. These data were modeled mathematically by a "differential coupling model" which assumed that a uniform cytosolic level of elevated cAMP activated two biochemical processes of differing sensitivity. Thus, the lusitropic response (phosphorylation of phospholamban) was coupled more efficiently to the cAMP response than the inotropic response (phosphorylation of calcium channels). A second model ("differential messenger concentration model") which calculated the effects of a compartmentalization of cAMP concentration within the cardiac cell by restricted diffusion and/or selective degradation by phosphodiesterases also was used.(ABSTRACT TRUNCATED AT 250 WORDS)

Promiscuous or heterogeneous muscarinic receptors in rat atria? I. Schild analysis with simple competitive antagonists.

Carbachol has been shown to produce a biphasic response in rat left atria. At low concentrations, carbachol depresses basal inotropy, while at high doses a positive inotropic effect is observed. The negative inotropic response can be selectively eliminated by pretreatment of rats with pertussis toxin. The aim of these studies was to determine whether or not evidence could be obtained to show that different muscarinic receptors produced these different biochemical responses to the agonist carbachol. Schild analysis was used to measure the equilibrium dissociation constant of the antagonist-receptor complex for antagonism of the negative inotropy to carbachol by atropine, scopolamine 4-diphenylacetoxy-N-methylpiperidine methiodide (4-DAMP) and AF-DX 116. The antagonism of the positive inotropic response to carbachol by these antagonists was studied in atria from rats pretreated with pertussis toxin where the negative inotropy was nearly completely abolished. In general, it was found that the antagonists did not produce simple competitive blockade of the positive inotropy but rather a nominal shift to the right of the dose-response curves followed by a depression of maximal responses. However, it was found that when pA2 or pKb values could be calculated, they coincided with those determined for the antagonism of the negative inotropy to carbachol. The conclusion drawn from these experiments was that no evidence was obtained to disprove the null hypothesis that a common receptor, interacting with two G-proteins, mediates these two effects of carbachol in rat left atria. The implications of these data for the classification of drug receptors with agonists is discussed.

Promiscuous or heterogeneous muscarinic receptors in rat atria? II. Antagonism of responses to carbachol by pirenzepine.

Carbachol produces both negative and positive inotropy in rat left atria. It is not clear whether these two effects are mediated by two separate cell surface muscarinic receptors or a single receptor interacting with two coupling proteins in the cell membrane. Pirenzepine, known to selectively block some biochemical muscarinic responses, was used in this study to block the biphasic response to carbachol in rat left atria. The negative inotropy to carbachol was blocked by pirenzepine, and Schild analysis indicated a -log dissociation constant (pKb) for the pirenzepine-receptor complex of 6.2. However, the Schild analysis may have been complicated by positive inotropy observed with pirenzepine. This positive inotropic effect was sensitive to blockade by other muscarinic antagonists. In atria from rats pretreated with pertussis toxin, carbachol produced a positive inotropic effect. Schild analysis with pirenzepine for antagonism of this response indicated a -log equilibrium dissociation constant for the pirenzepine-receptor complex of 6.7, significantly different from that for antagonism of negative inotropy. This ostensibly suggested a difference in the receptors mediating these responses. In view of the possible complicating effects of the positive inotropic effects of pirenzepine in this assay, an alternative method for the measurement of pirenzepine affinity was utilized. Resultant analysis was used to measure the pKb for pirenzepine antagonism of negative inotropy to carbachol. This method had the advantage of cancelling the positive inotropy to pirenzepine. Under these circumstances, pirenzepine had a pKb of 6.9, a value not significantly different from for antagonism of the positive inotropy to carbachol. The relevance of these findings is discussed in terms of a single promiscuous muscarinic receptor or heterogeneous receptors in this tissue. These data do not support the hypothesis that two separate receptors mediate these two effects.

Theoretical effects of single and multiple transducer receptor coupling proteins on estimates of the relative potency of agonists.

A mathematical model is presented that simulates the steady state kinetics of agonists interacting with a promiscuous receptor. The model system consists of a single receptor that forms a ternary complex with either of two transducer proteins (G proteins). At a given agonist concentration, the concentrations of the two ternary complexes are determined by the relative quantities of the two G proteins and the ratio of the dissociation constants for the two ternary complexes. Accordingly, the potency of an agonist is dependent upon the relative quantities of the G proteins. If receptors are truly promiscuous and if the distribution of G proteins varies with tissue type, then the agonist potency ratio would be tissue dependent as well as receptor dependent. Experimental data from literature studies are reviewed in the context of the promiscuous receptor model, and implications of the model regarding pharmacologic classification of receptors are discussed.

Challenges for receptor theory as a tool for drug and drug receptor classification.

Increased understanding in the field of receptor pharmacology, born of the sophisticated techniques now available to us, has confounded rather than simplified the problem of receptor classification. The International Union of Pharmacology (IUPHAR) is currently sponsoring a Receptor Nomenclature Committee whose aims are to recommend a rational system of classification, a formidable task given the complexity and volume of data in the literature (agonist/antagonist potencies, coupling mechanisms, primary structures, etc.) that will need to be incorporated. The Committee's chairman, Terry Kenakin, outlines here the limitations of classical receptor theory for drug receptor classification and suggests that any functional classification system must take into account not only affinity and intrinsic efficacy but also, at the very least, parameters relating to the transducing properties of receptors. If this is not done, then receptor classification data obtained from studies with agonists and antagonists may be different and lead to confusion.

Classification of phenoxybenzamine/prazosin-resistant contractions of rat spleen to norepinephrine by Schild analysis: similarities and differences to postsynaptic alpha-2 adrenoceptors.

The striking resistance of norepinephrine contractions of rat splenic strips to antagonism by the selective alpha-1 adrenoceptor antagonist prazosin was examined by Schild analysis. Prazosin was a simple competitive antagonist of contractions to phenylephrine indicating that this tissue possesses alpha-1 adrenoceptors. In contrast, the Schild regression for prazosin, with norepinephrine as the agonist, was nonlinear and had an overall slope of 0.24. These data indicated that norepinephrine activated a prazosin-resistant adrenoceptor in this tissue. As a working hypothesis, it was assumed that the prazosin-resistant receptor was an alpha-2 adrenoceptor; the concomitant addition of yohimbine, in concentrations below those required to block alpha-1 adrenoceptors, converted the atypical Schild regression for prazosin (norepinephrine as agonist) to a linear regression identical with that found for antagonism of phenylephrine responses. Selective alkylation of alpha-1 adrenoceptors with phenoxybenzamine (POB) eliminated responses to phenylephrine but not those to norepinephrine. After POB-alkylation and in the presence of a concentration of prazosin that was sufficient to produce a profound blockade of alpha-1 adrenoceptors, a response to norepinephrine remained. It was determined that the POB/prazosin-resistant response most likely was mediated by a homogeneous population of receptors by the finding that the Schild regressions for both yohimbine and idazoxan were identical with respect to slope and elevation when either norepinephrine or cobefrin were utilized as agonists, i.e., a difference in the regressions for these antagonists would be expected if the two agonists activated a heterogeneous receptor population.(ABSTRACT TRUNCATED AT 250 WORDS)

A method to assess concomitant cardiac phosphodiesterase inhibition and positive inotropy.

The ability of positive inotropic drugs to inhibit phosphodiesterase was assessed by observance of the potentiation of inotropic responses to the low intrinsic efficacy beta-adrenoceptor partial agonist prenalterol. Previous studies have shown that an increase in tissue stimulus response capability, as is produced for beta-adrenoceptors by blockade of phosphodiesterase, produces an increase in the maximal response to beta-adrenoceptor partial agonists. Using this principle, we tested cardiotonic drugs on normal and prenalterol-pretreated guinea pig left atria and compared the resulting inotropic responses statistically. Prenalterol pretreatment did not potentiate inotropic responses to methoxamine and CaCl2 and blocked (in accordance with beta-adrenoceptor occupation by a low-efficacy partial agonist) responses to norepinephrine (NE), tyramine, and to a certain extent ouabain. This latter finding was attributed to a low-level catecholamine-release by ouabain in this tissue. The inotropic responses to the phosphodiesterase inhibitors isobutylmethylxanthine (IBMX), theophylline, and enprophylline were greatly potentiated. Similarly, the responses to the cardiotonic drugs (known also to be inhibitors of phosphodiesterase) milrinone, amrinone, and fenoximone were potentiated. Positive inotropy to the cardiotonic drug sulmazole was not significantly potentiated by this procedure, indicating that in this tissue sulmazole may produce inotropic responses by other mechanisms as well. In general, this simple assay may be useful to detect blockade of cardiac phosphodiesterase concomitant with positive inotropy. Although a causal relationship between the PDE inhibition and positive inotropy may not be made from these data, knowledge of PDE blockade may still be useful in the assessment of inotropic mechanism and propensity for toxic side effects.

Measurement of antagonist affinity for purine receptors of drugs producing concomitant phosphodiesterase blockade: the use of pharmacologic resultant analysis.

It was observed experimentally that both theophylline and isobutylmethylxanthine (IBMX) produced surmountable antagonism of the agonist effects of 2-chloroadenosine on purine receptors in rat vas deferens and guinea pig atria. In the case of IBMX, there was a statistically significant difference between the Schild regressions in the two tissues, ostensibly indicating a possible purine receptor heterogeneity with respect to the binding of this antagonist. However, the analysis was complicated by the fact that both theophylline and IBMX produced positive inotropic responses in guinea pig atria (presumably by inhibition of cardiac phosphodiesterase), making the calculation of dose ratios subjective and ambiguous. To determine whether the phosphodiesterase blocking property of theophylline and IBMX was interfering with the observation of purine receptor antagonism in guinea pig atria, a new technique described as pharmacologic resultant analysis was utilized to measure the purine receptor blocking properties of theophylline and IBMX in guinea pig atria in the presence of phosphodiesterase blockade. Based on the principle of additive dose ratios for two antagonists competing for one receptor, pharmacologic resultant analysis measures the effects of a test antagonist (in this case theophylline or IBMX) on the competitive blockade of a reference antagonist; for these studies, the reference antagonist was 8-sulfophenyltheophylline. Under these circumstances, the direct effects of the test antagonist (positive inotropy) are expressed throughout the experiment and cancel with the normal null methods used for measurement of competitive antagonism. Using this technique, it was found that the potencies of both theophylline and IBMX on the purine receptors in rat vas deferens and guinea pig atria were identical. The use of this method in the delineation of multiple drug activities is discussed.