The recommendations that follow have been updated from the proposals of a Technical Subcommittee set up by the International Union of Pharmacology Committee on Receptor Nomenclature and Drug Classification (Jenkinson DH, Barnard EA, Hoyer D, Humphrey PPA, Leff P, and Shankley NP (1995) International Union of Pharmacology Committee on Receptor Nomenclature and Drug Classification. IX. Recommendations on terms and symbols in quantitative pharmacology. Pharmacol Rev 47:255–266).
This update was undertaken to incorporate new information about multiple receptor conformational states and the recognition that multiple distinct agonist responses may result that have different pharmacological properties (Kenakin, 1995). Nomenclature concerning the actions of allosteric (allotopic) ligands is presented based on recent literature (Christopoulos and Kenakin, 2002). The implications of high receptor numbers in heterologous expression systems for interpretation of agonist function are discussed. Additional changes address the fact that many receptors are not single macromolecules but are made up of multiple subunits. Finally, there are new recommendations regarding nomenclature for equilibrium constants.
II. Working Definition of a Receptor
A cellular macromolecule, or an assembly of macromolecules, that is concerned directly and specifically in chemical signaling between and within cells. Combination of a hormone, neurotransmitter, drug, or intracellular messenger with its receptor(s) initiates a change in cell function. Thus NC-IUPHAR does not classify simple binding sites, without function (although truncated proteins without signaling function may be designated as such, to avoid confusion). Furthermore, a receptor may consist of several proteins, called subunits. In some cases the large number of combinatorial possibilities for assembly of multiple subunits may require NC-IUPHAR to use an interim nomenclature based on the individual subunits (Spedding et al., 2002). Nevertheless, the ultimate goal is to define the multi-subunit assemblies that occur in vivo.
The regions of the receptor macromolecule to which ligands bind are referred to collectively as the recognition site(s) of the receptor. Those at which the endogenous agonist binds are termed primary or orthosteric sites whereas other ligands may act through allosteric sites (see Table 1).
III. Use of Drugs in Definition of Receptors or of Signaling Pathways
When using drugs to define receptors or signaling pathways, it would be desirable to use a drug that acts only on the receptor or biological site of interest at all concentrations and doesn't interact with others at any achievable concentration. Unfortunately, there are very few or no drugs with this ideal property. Fortunately, there are numerous drugs with a detectable potency difference (in exceptional cases >103-fold but usually much less) between their primary target and other related receptors. Because these differences are not absolute, claims for the involvement of a particular receptor, or signaling protein, based on the use of such agents should be backed up by testing with multiple agents, and wherever possible, full concentration-response curves should be obtained for the definition of responses in in vitro experiments. Full dose-response curves should also be obtained in in vivo experiments, if ethical considerations allow.
A. The Expression of Amount of Drug: Concentration and Dose
Concentration. It is recommended that the molar concentration of substance X be denoted by either [X] or cx, with the former preferred. Decimal multipliers should be indicated by the use of either Le Système International d'Unités (International System of Units) prefixes (e.g., μM, nM) or by powers of ten (e.g., 3 × 10-8 M), with the former preferred.
Dose. In some circumstances (e.g., in therapeutics and clinical pharmacology, in in vivo experiments, and when tissues are perfused in vitro and exposed to a bolus application of drug), absolute drug concentrations are uncertain, and it becomes more appropriate to specify the quantity of drug administered. This may be done in terms of either mass or molar quantity. Units and routes of administration should be specified. In the case of in vivo experiments with animals, the quantity of drug is to be expressed per unit of animal mass (e.g., mol/kg, mg/kg). In therapeutics, milligrams per kilogram will normally be appropriate. Negative indices should be used where confusion otherwise arises (e.g., mg min-1 kg-1).
B. General Terms Used to Describe Drug Action
C. Experimental Measures of Drug Action
A. Microscopic and Macroscopic Equilibrium Constants
Microscopic and macroscopic equilibrium constants should be distinguished when describing complex equilibria, which occur with all agonists. The latter refers to a single constant describing the overall equilibrium (i.e., the value that would be obtained in a ligand binding experiment), whereas the former refers to each individual constant that describes each reaction step within the equilibrium. For the scheme the macroscopic equilibrium dissociation constant (denoted here as Kapp for “Kapparent”) is given by Here K1 and K2 are the microscopic equilibrium constants for the first and second reactions, respectively. Note that in this scheme, saturation radioligand binding assays and Furchgott's (1966) irreversible antagonist method for determining the equilibrium dissociation constant for an agonist would each provide an estimate of Kapp rather than K1.
This distinction is also important when considering those receptors (e.g., ligand-gated ion channels) that have more than one binding site for the agonist.
B. Schild Equation and Plot—Further Detail
The Schild equation is based on the assumptions that (a) agonist and antagonist combine with the receptor macromolecule in a freely reversible but mutually exclusive manner, (b) equilibrium has been reached and that the law of mass action can be applied, (c) a particular level of response is associated with a unique degree of occupancy or activation of the receptors by the agonist, (d) the response observed is mediated by a uniform population of receptors, and (e) the antagonist has no other relevant actions, e.g., on the relationship between receptor and response. Under these circumstances, the slope of the Schild plot should be 1 and the resulting estimate of the pA2 should be equal to the pK (negative logarithm of the antagonist equilibrium dissociation constant).
For an antagonist to be classified as reversible and competitive on the basis of experiments in which a biological response is measured, the following criteria must hold:
In the presence of the antagonist, the log agonist concentration-effect curve should be shifted to the right in a parallel fashion.
The relationship between the extent of the shift (as measured by the concentration ratio) and the concentration of the antagonist should follow the Schild equation over as wide a range of antagonist concentrations as practicable. Usually, the data are presented in the form of the Schild plot, and adherence to the Schild equation is judged by the finding of a linear plot with unit slope (see Note 2 below). Nonlinearity and slopes other than unity can result from many causes. For example, a slope greater than 1 may reflect incomplete equilibration with the antagonist or depletion of a potent antagonist from the medium, as a consequence either of binding to receptors or to other structures. A slope that is significantly less than 1 may indicate removal of agonist by a saturable uptake process, or it may arise because the agonist is acting at more than one receptor (the Schild plot may then be nonlinear). See Kenakin (1997) for a detailed account.
Note 1: The finding that the Schild equation is obeyed over a wide range of concentrations does not prove that the agonist and antagonist act at the same site. All that may be concluded is that the results are in keeping with the hypothesis of mutually exclusive binding, which may of course result from competition for the same site but can also arise in other ways (see Allosteric Modulators in Table 1 and Competitive Antagonism in Table 7).
Note 2: Traditional Schild analysis is based on the use of linear regression. Nowadays, with the almost ubiquitous availability of computers in most research environments, a more accurate approach to performing Schild analysis is to use computerized nonlinear regression to directly fit agonist/antagonist concentration-response data to the Gaddum/Schild equations. The advantages of this approach over traditional Schild analysis are described elsewhere (Waud, 1975; Black et al., 1985; Lew and Angus, 1995). One simple method is to fit agonist EC50 data, determined in the absence and presence of antagonist, to the following equation: where pEC50 and pA2 are as defined previously in Tables 3 and 4, respectively, [B] denotes the antagonist concentration, S is a logistic slope factor analogous to the Schild slope and log c is a fitting constant (Motulsky and Christopoulos, 2003). This equation is based on a modification of the original Gaddum/Schild equations that results in more statistically reliable parameter estimates than those obtained using the original equations for nonlinear regression (Waud et al., 1978; Lazareno and Birdsall, 1993). If S is not significantly different from 1, then it should be constrained as such and the equation re-fitted to the data.
C. The Relationship between the Hill and Logistic Equation
The logistic function is defined by the equation where a and b are constants. If a is redefined as -loge(Kb), and x as loge z, then which has the same form as the Hill equation.
We acknowledge helpful comments from Tom Bonner, Steven Foord, Steve Watson, and Sir James Black.
- The American Society for Pharmacology and Experimental Therapeutics