A modification of receptor theory

TLDR: An attempt has been made to determine the relation between log concentration and effect for acetylcholine and the frog rectus abdominis after blocking the cholinesterase activity of isolated rabbit auricles.

Abstract: It is usually supposed that a drug combines with some part of a tissue as a necessary step in producing its action on that tissue. This view was expressed by Langley in 1878 while discussing experiments with pilocarpine and atropine. He later referred to “receptive substance” (Langley, 1905) and subsequently the term “receptor” has been widely used to express the same concept. Clark (1937a) attempted to put the concept on a quantitative basis by applying the adsorption isotherms of Langmuir which were derived from the application of the mass laws to the adsorption of gases on metal surfaces. The derivation of the equations as they apply to drugs in solution, and receptors or active patches in tissue, may be put as follows: Drug molecules combine with receptors at a rate which is proportional to the concentration of molecules in solution and to the number of free receptors. The resulting complex breaks down at a rate proportional only to the number of complexes. Thus the rate of combination is k1 A (l – y) and the rate of dissociation is k2y, where k1 and k2 are constants, A the concentration of drug, and y the proportion of receptors occupied by the drug. When the two rates are equal we have, putting K for 1 When y = 0.5, then 1a Equation 1 therefore relates the concentration of drug and the proportion of receptors which are occupied by drug molecules at equilibrium. Clark, however, went a step further than this and, by taking y as the response, used equation 1 to relate the response of the tissue to the concentration of drug. Thus there was in Clark's treatment the implicit assumption that the percentage of receptors occupied is equal to the percentage response of the tissue. When 50% of the receptors are occupied there is 50% of the possible response; when the response is maximal all the receptors are occupied. Clark sought to show that the relation between concentration of drug and the effect produced did in fact agree with equation 1, and his apparent success was some justification for the assumption. The two situations he investigated were (a) the contraction of the isolated frog rectus abdominis produced by acetylcholine, and (b) the inhibition produced by acetylcholine of the contraction of strips of electrically stimulated frog ventricle. These tests have a common disadvantage in that both tissues contain enzymes which destroy acetylcholine, so that the concentration of drug in equilibrium with the receptors is less than that applied externally, and, though the applied concentration is known, that in the tissue is not. The receptors, indeed, may not all be in equilibrium with the same concentration of acetylcholine; a gradient is presumably established with a high concentration near the surface and a lower concentration at the centre of the tissue—because the further the acetylcholine penetrates the greater is its chance of being destroyed. Such a gradient would produce a log-concentration-effect curve flatter than the true curve. Furthermore, it cannot be assumed that the ratio of the applied concentration to the effective concentration (or concentrations) is constant for different applied concentrations. A block of cholinesterase activity would therefore be expected to alter the slope of log-concentration-effect curves; Webb (1950) has shown that this is so, and that the change of slope is considerable. Calculation from his graphs shows that, when the cholinesterase of isolated rabbit auricles is blocked with physostigmine, the slope of the acetylcholine log-concentration-effect curve is about 10 times steeper than on normal auricles (see Table I). An attempt has been made to determine the relation between log concentration and effect for acetylcholine and the frog rectus after blocking the cholinesterase with TEPP so that the size of the effect could be assessed in one of the systems used by Clark. This proved to be difficult, since the response of the rectus, always slow, became very much slower after TEPP, and the contractions continued to increase for an hour after acetylcholine was applied. The action of carbachol on normal recti was similarly slow. It was therefore impossible to get an accurate estimate of slope, but it was evident that the increase of effect with increased concentration was greater than required by equation 1. In any case it was clear that reliable evidence in support of any particular equation would be difficult to obtain from this tissue. Table I Slopes of Concentration-Effect Curves Clark discussed several other drug-tissue systems, but he was mainly concerned to show that a graded response was produced over a wide range of concentration and, in most of the examples he quoted, the slopes of the log concentration curves are not in accord with equation 1. A simple way of reducing data about concentration-effect curves to a single figure which is independent of any theory of drug action, and which can therefore be used to test any particular theory, is to use, as Clark did, the ratio of concentrations causing specified effects. Clark used the ratio for 84% and 16% of the maximum contraction; but a narrower range is easier to cover, and there is no need to restrict the ratio measured to any particular pair of responses. Evidence from which such ratios can be calculated is rare, because the maximum response is seldom recorded, but a few examples are shown in Table I. Some of these ratios represent new data for acetylcholine and histamine on guinea-pig ileum; details are given in the section on results. Furchgott and Bhadrakom commented on the theoretical implication of the slope, and considered that their results agreed with equation 1; the ratio given in Table I expresses this agreement. The other ratios in Table I have been calculated from published graphs: they are not in agreement with equation 1. A variety of slopes can be accounted for by modifications of equation 1 obtained by postulating that a molecule of drug combines with either less or more than one receptor. The latter possibility may be true in some instances, but the postulate is unlikely to be generally correct—particularly where fractions are involved. Equation 1 appears to be the most likely relation between concentration of drug and the percentage of receptors occupied, but Table I shows that there is no experimental justification for supposing that it represents a general relation between concentrations of drug and the response of the tissue. It therefore becomes necessary to consider possibilities other than the one assumed by Clark for the relation between the proportion of receptors occupied and the response, and we can broaden the application of the receptor theory by the following hypotheses: A maximum effect can be produced by an agonist when occupying only a small proportion of the receptors. The response is not linearly proportional to the number of receptors occupied. Different drugs may have varying capacities to initiate a response and consequently occupy different proportions of the receptors when producing equal responses. This property will be referred to as the efficacy of the drug. The concept of efficacy, which may vary from zero to a large positive value, is at variance with Clark's (1937c) view of the action of a drug at a receptor as all or none; an agonist (Reuse, 1948) causes an effect and an antagonist causes no effect so that the activity of a drug of either kind is simply a measure of its affinity for the receptors. According to hypothesis 3, however, this remains true only for antagonists (which have zero efficacy): the activity of agonists is the product of their affinity and their efficacy. A drug with very low efficacy may produce a response which is less than the maximum, even when it is occupying nearly all the receptors—and because it is occupying the receptors, it diminishes the action of a drug with high efficacy added simultaneously. Compounds with such a low efficacy that they possess properties intermediate between agonists and antagonists will be called “partial agonists.” (See p. 384.) Ariens and Groot (1954) have also observed substances with these intermediate properties, and they ascribe this to a low “intrinsic activity.” Ariens and Groot, however, continue to assume that all normal agonists have the same “intrinsic activity” and their concept is more limited in application than hypothesis 3. The above hypotheses account for the progressive variation in properties in a series of alkyltrimethylammonium salts. The lower homologues have acetylcholine-like activity whereas the higher members have an atropine-like blocking action (Raventos, 1937a and b; Clark and Raventos, 1937). It was expected that the members of such a simple homologous series would display some progressive change in their affinity for the receptors, and consequently, if Clark was correct, the reciprocals of the concentrations of the agonists causing a 50% response (see equation 1a) would form part of the same progression as the affinity constants of the antagonists. The demonstration of a sharp discontinuity between the two sets of constants, such that the agonists are effective in lower concentration than the antagonists, would, however, constitute good evidence for hypothesis 1. Raventos showed that the acetylcholine-like activity of the lower homologues varied rather irregularly, but calculation from his data suggested that the expected discontinuity between agonist and antagonist activity did occur. The nature of the evidence did not allow a firm conclusion. It was therefore decided to re-examine the compounds, with the intention of obtaining accurate values for the affinity constants of the first 4 or 5 antagonists, and to compare these with the reciprocals of the concentrations of the agonists which cause a 50% contraction.