Inverse Agonism and Constitutive Receptor Activity

Basal Signaling Changes What “Antagonist” Means

An antagonist is often described as a ligand that blocks an agonist’s effect. That description does not by itself say what the ligand does when no agonist is present. If receptors signal spontaneously in the absence of an added ligand, a ligand can occupy those receptors and reduce their basal signaling. Such a ligand is an inverse agonist, not a neutral antagonist. Distinguishing the two requires a reference response measured in the ligand-free system, rather than a response measured only in the presence of an agonist.

Here, constitutive activity means receptor-mediated signaling that occurs without an added agonist. It is distinct from any response due to background processes unrelated to the receptor. The model below represents ligand binding as changing a receptor stimulus, which is then converted to a measured response. It is a deliberately compact model for predicting the direction and size of a response change; it does not imply that measured response is a direct count of active receptors.

For one-site reversible binding at equilibrium, let \([L]\) be the free ligand concentration and \(K\) its equilibrium dissociation constant, expressed in the same concentration units. The fractional receptor occupancy is \(f=[L]/(K+[L])\). Assume the stimulus is the occupancy-weighted average of the unbound and ligand-bound receptor states. The resulting response is transformed from stimulus by a saturable function:

Here \(E([L])\) is the measured response, \(E_{\mathrm{sys}}>0\) is the system maximum response, and \(K_E>0\) is the stimulus producing half of that maximum in this response function. \(S_0\) and \(S_L\) are nonnegative, dimensionless stimuli; \(K_E\) has stimulus units. The model assumes equilibrium occupancy, a single binding site, a fixed response system, and a stimulus that varies linearly with the fraction of receptors occupied. It also assumes that the same response function applies with and without ligand. If the system has no constitutive stimulus, \(S_0=0\), this model does not predict a response decrease below the ligand-free response through inverse agonism, because there is no positive basal stimulus to reduce.

Why the Ligand’s Effect Has a Defined Direction

The ligand-free response is \(E(0)=E_{\mathrm{sys}}S_0/(K_E+S_0)\). To determine whether ligand changes this response, first consider how concentration changes stimulus. For \(K>0\),

For nonnegative \([L]\), the denominator is positive. Thus, if \(S_L-S_0\) is positive, negative, or zero, stimulus changes in that same direction as concentration increases. The response transformation has derivative

When \(E_{\mathrm{sys}}>0\), \(K_E>0\), and \(S\geq 0\), this derivative is positive. The response therefore changes in the same direction as stimulus. Combining the derivatives gives

Under the stated conditions, including \(K>0\), this derivative has the sign of \(S_L-S_0\). An inverse agonist lowers response as its concentration increases; a neutral ligand leaves response unchanged; and an agonist raises response. This direction result relies on the conditions: for example, if \(K_E=0\), then every positive stimulus gives \(E=E_{\mathrm{sys}}\), so differences in stimulus need not produce differences in response.

The size of the response change can also be seen directly. Subtracting the ligand-free response gives

For positive \(E_{\mathrm{sys}}\) and \(K_E\), and nonnegative stimuli, the denominator and the other factors outside the stimulus difference are positive. Hence the response difference has the sign of \(S([L])-S_0\). As \([L]\) tends to infinity, occupancy tends to one, stimulus tends to \(S_L\), and response tends to \(E_{\mathrm{sys}}S_L/(K_E+S_L)\). High occupancy does not necessarily mean a large measured response change: the response can be compressed by saturation.

Neutral Antagonism and Inverse Agonism Compared

The distinguishing feature is efficacy relative to the ligand-free receptor state, not merely binding affinity or the fact that a ligand blocks another ligand’s action. When \(S_L=S_0\), the stimulus equation reduces to \(S([L])=S_0\) at every concentration. Thus the neutral ligand leaves basal response unchanged in this model, even when occupancy is nearly complete. If \(S_L<S_0\), occupancy moves stimulus toward a lower value and the ligand is an inverse agonist. A ligand with \(S_L>S_0\) increases stimulus and is an agonist.

Estimating Efficacy From a Response

When the response function and occupancy are known, a measured response can be translated back to stimulus. This is useful for checking whether the implied ligand-bound stimulus is below the basal stimulus. The inversion requires \(E<E_{\mathrm{sys}}\), as expected for finite nonnegative stimulus when \(K_E>0\):

Assumptions, Diagnostics, and Common Errors

A useful interpretation requires evidence that the ligand-free reference is stable and that the assay can register changes around that reference. Constitutive signaling may be weak or absent in a particular tissue or assay. In that case an inverse agonist may show little measurable decrease, or may be indistinguishable from a neutral antagonist, even if it has lower efficacy at the receptor. Conversely, changes in background signaling over time can be mistaken for ligand effects. The earlier material on baseline and background response components is relevant to defining a reliable ligand-free reference; it does not replace the receptor-state comparison made here.

Inspect the response across a concentration range, verify that concentrations plausibly reach appreciable occupancy, and check whether the ligand-free response is reproducible. A response that declines only when assay conditions drift, or that changes similarly in a receptor-free control, does not by itself support receptor-mediated inverse agonism. If measured responses plateau before the predicted range, consider whether system saturation, a response floor, or an incorrect value of \(E_{\mathrm{sys}}\) or \(K_E\) is masking the stimulus change. The model also becomes questionable when binding is not at equilibrium, multiple receptor populations contribute, the ligand changes the response machinery independently of the receptor, or occupancy does not translate to stimulus by the assumed linear relation.

A common mistake is to call every ligand that blocks an agonist an inverse agonist. That reasoning confuses blocking another ligand’s action with reducing basal activity. The correction is to compare the ligand’s response with the ligand-free basal response: \(S_L=S_0\) is neutral in this model, whereas \(S_L<S_0\) predicts a decrease. Another important pitfall is to claim that the response must change whenever \(S_L\ne S_0\). If \(K_E=0\), every positive stimulus maps to the same maximal response; more generally, a nearly saturated response function can make a real stimulus change difficult to observe. The direction result above therefore requires strictly positive \(E_{\mathrm{sys}}\), \(K_E\), and \(K\), along with a defined response and nonnegative stimulus.

This model is not a substitute for the full operational model of agonism or for explicit analysis of mixtures of ligands. Its value here is narrower: it makes the basal reference, ligand-bound stimulus, and response transformation explicit so their roles are not conflated. Binding affinity \(K\) governs how concentration controls occupancy, while \(S_L-S_0\) determines the direction of the stimulus change. Neither alone determines the observed response change; the response system’s saturation also matters.