Direct Receptor Occupancy Assays

From a Receptor Signal to Fractional Occupancy

A receptor assay produces a signal: counts from a radioligand, fluorescence from a probe, or label associated with a drug. The biological quantity of interest is usually not that signal itself, but the fraction of receptors occupied by the drug of interest at a specified time. Translating one into the other requires identifying what the assay detects and what receptor population provides the denominator.

Direct occupancy assays either measure drug bound to receptors or use a labeled probe to report how much receptor remains available to bind. These are not interchangeable measurements. A probe signal can fall because fewer receptors are available, but it can also fall because receptor abundance changed, the probe did not reach the target, or the assay signal was affected by background or handling. Interpretation therefore begins with a model of the measured signal, not with a percentage calculation.

Definition: Let $B_D$ be the amount of receptor bound by the drug of interest, and let $R_{\mathrm{avail}}$ be the amount of receptor available for binding by that drug. If these are the only receptor states being counted, total receptor is $R_T=B_D+R_{\mathrm{avail}}$. The fractional occupancy by the drug is $$ \theta_D=\frac{B_D}{R_T} =\frac{B_D}{B_D+R_{\mathrm{avail}}}. $$ Occupancy is dimensionless and ranges from 0 to 1; multiplying by 100 expresses it as a percentage.

The definition concerns receptor amounts, not raw detector readings. If a measurement has been converted into receptor-equivalent amounts, bound and available amounts can be used directly. If measurements are in counts or fluorescence units, a ratio is valid only when the signal is proportional to the relevant receptor amount and the compared measurements share an appropriate scale. The later material on calibrating occupancy assay signals addresses that conversion in detail; here the focus is how the assay's measured bound or available signal maps to occupancy.

What the Main Assay Formats Measure

In a labeled-drug assay, the drug of interest carries a detectable label. After accounting for label that is not specifically associated with the target, the signal is interpreted as drug bound to receptor. To calculate fractional occupancy, this bound amount must be related to the total receptor amount in the same relevant sample or compartment. Bound drug alone is not a fraction: a value of 50 pmol could represent low occupancy in a sample containing 500 pmol of receptor, or high occupancy in one containing 60 pmol.

In a radioligand or fluorescent-ligand availability assay, a labeled probe binds receptor sites that remain accessible to it. If drug and probe bind mutually exclusively to the same site, less probe binding after drug exposure is evidence that fewer sites are available. A matched control without the drug provides a reference for the probe signal when the target is available. With unchanged receptor abundance and assay conditions, the fraction of reference probe binding lost estimates the fraction of sites occupied by the drug.

That inference depends on the probe signal being specific and proportional to available receptor under the compared conditions. A no-target or excess-unlabeled-ligand control can help identify nonspecific probe binding. Background should be handled consistently before comparing signals. In addition, samples should be compared at a defined time: drug may dissociate during collection or processing, so an ex vivo measurement may not preserve occupancy at the original sampling time.

A labeled probe is not necessarily the drug of interest. In a displacement design, the probe reports availability indirectly through competition. By contrast, a labeled version of the therapeutic compound can report its own receptor-bound amount more directly, but still requires separation of specific from nonspecific association and a measure of total receptor for fractional occupancy.

Why a Displacement Ratio Can Represent Occupancy

Consider a simple assay in which a labeled probe binds available receptors. Assume a single accessible receptor population, mutually exclusive binding of drug and probe, the same probe concentration and binding conditions in treated and control samples, and a probe signal proportional to probe-bound receptor. Also assume the control contains no drug occupying the target and that receptor abundance is unchanged between the samples. At equilibrium, the probe-bound amount is proportional to the amount of receptor available to the probe. Thus a treated-to-control probe-binding ratio estimates the fraction of receptor that remains available.

Let $A_0$ denote the specific probe-binding signal in the drug-free control and $A_D$ the specific probe-binding signal after drug exposure. If the signal is proportional to available receptor and the control represents the full receptor population, then $A_D/A_0$ estimates the unoccupied fraction. The occupied fraction is its complement:

\widehat{\theta}_D=1-\frac{A_D}{A_0}.

This equation is a relative estimate, not a universal conversion from any two detector readings. It is justified only when the control and treated signals are comparable measures of receptor availability. If receptor abundance differs, the ratio confounds occupancy with abundance. If the drug does not compete with the probe, a change in probe binding cannot be attributed to drug occupancy by this argument.

The derivation is direct. Under the stated conditions, write $A_D=kR_{\mathrm{avail}}$ and $A_0=kR_T$, where $k$ is the common signal per available receptor. Then

1-\frac{A_D}{A_0} =1-\frac{kR_{\mathrm{avail}}}{kR_T} =1-\frac{R_{\mathrm{avail}}}{R_T} =\frac{R_T-R_{\mathrm{avail}}}{R_T} =\frac{B_D}{R_T} =\theta_D.

The cancellation of $k$ explains why a matched ratio can be useful without converting every signal to receptor amount. It also shows exactly where the argument fails: if the signal scale differs between conditions, if the control is not a drug-free reference, or if the counted receptor pool changes, the cancellation or the interpretation of the ratio is no longer justified.

Worked Example: Radioligand Displacement

In a hypothetical assay, a radioligand is used to measure receptor availability. After nonspecific binding has been subtracted, a drug-free control gives 120,000 counts per minute and a drug-treated sample gives 45,000 counts per minute. Assume both samples have the same receptor abundance, probe concentration, incubation conditions, and counting efficiency; the control has no drug occupancy; and the drug and probe compete for the same sites.

The estimated remaining available fraction is $$ \frac{A_D}{A_0}=\frac{45{,}000}{120{,}000}=0.375. $$

The estimated drug occupancy is therefore $$ \widehat{\theta}_D=1-0.375=0.625. $$

The assay estimates 62.5% occupancy at the sampled condition and time. The 45,000 counts per minute are not themselves an occupancy percentage; the interpretation depends on the matched control and the assumptions stated above.

Worked Interpretations Across Assay Types

Worked Example: Fluorescent Probe and Receptor Availability

A hypothetical cell assay uses a fluorescent ligand to report receptor availability. After background subtraction, the drug-free control has a signal of 800 fluorescence units per $10^5$ viable cells. The treated sample has 280 units per $10^5$ viable cells. A separate receptor-abundance check indicates comparable total receptor per viable cell in the two samples. Assume probe binding is specific, proportional to available receptor, and measured under identical conditions; also assume the drug competes with the probe at the target.

Because the signals are already expressed per the same number of viable cells, their ratio is $$ \frac{A_D}{A_0}=\frac{280}{800}=0.350. $$

The estimated occupancy is $$ \widehat{\theta}_D=1-0.350=0.650. $$

The result is an estimated 65.0% occupancy. The receptor-abundance check matters: without it, a 65% reduction in probe signal could reflect fewer receptors per cell rather than drug occupying the same receptor population. Equal viable-cell normalization alone would not resolve that ambiguity.

Worked Example: Labeled Drug and a Receptor-Amount Denominator

In a hypothetical labeled-drug assay, specific target-bound drug is estimated as 72 pmol in a tissue sample. A matched measurement gives 120 pmol of total receptor in the same sampled compartment. Assume the label does not materially alter binding, the measured bound drug is specifically receptor-associated, and the receptor amount includes the relevant drug-bound and available states.

Here the direct bound amount is the numerator and total receptor is the denominator: $$ \widehat{\theta}_D=\frac{72\ \mathrm{pmol}}{120\ \mathrm{pmol}}=0.600. $$

The estimated occupancy is 60.0%. Under the stated two-state accounting, the available amount is $$ R_{\mathrm{avail}}=120\ \mathrm{pmol}-72\ \mathrm{pmol}=48\ \mathrm{pmol}. $$

Checking the alternative form gives $72/(72+48)=0.600$. If the 72 pmol included nonspecific tissue association, or if the total-receptor measurement covered a different compartment, this calculation would not estimate receptor occupancy as intended.

Assumptions, Diagnostics, and Failure Modes

A direct occupancy interpretation is strongest when the assay measures the relevant receptor population at a defined time, the label reports the intended molecular interaction, and the denominator represents the same target pool. For a displacement assay, the key comparisons are treated versus control probe binding. For a labeled-drug assay, they are specific target-bound drug versus total relevant receptor. In either case, the binding state must be preserved well enough during sampling and measurement to answer the biological question.

Several checks help detect a failing interpretation:

Check specificity. If a substantial signal remains when target binding is prevented, the raw signal includes non-target association. Occupancy should not be calculated from that total without accounting for the nonspecific component.
Check receptor abundance. A lower availability signal can result from receptor loss, internalization, or reduced expression, as well as occupancy. Compare treated and reference samples for receptor amount or use a design that measures it independently.
Check assay comparability. Different probe concentrations, incubation conditions, sample handling, or detector response can change the signal independently of availability. A matched reference is meaningful only if these conditions are controlled.
Check the competition premise. If the drug and probe bind distinct sites, or if binding is allosterically coupled rather than mutually exclusive, a simple lost-signal fraction need not equal drug occupancy.
Check the time point. If occupancy changes during collection or incubation, the assay may report the state at measurement rather than the state at sampling. State which time the estimate is intended to represent.
Check plausibility. Under the simple two-state definition, occupancy lies between 0 and 1. A negative estimate or one above 1 indicates measurement variability, background or normalization problems, or a violated model assumption; it is not a meaningful physical occupancy.

A common mistake is to say, “The probe signal fell by 40%, so 40% of receptors are occupied,” without checking receptor abundance or whether the drug and probe compete. The correction is to identify what the signal measures, establish the required comparison and denominator, and state the assumptions that make signal loss correspond to occupied receptor. A signal change alone is not proof of occupancy.

Choosing an Occupancy Readout

A labeled-drug assay is conceptually direct because it tracks the drug bound to the target, but it can be difficult to distinguish target-bound label from nonspecific association and to measure total receptor in the same compartment. A labeled-probe displacement assay can avoid labeling the therapeutic drug and can use a matched reference, but it infers drug occupancy from receptor availability and therefore relies on competition and stable receptor abundance.

Both approaches differ from a functional readout, such as a downstream response, and from an indirect target-engagement biomarker. Functional response can be affected by receptor reserve, efficacy, pathway amplification, and other biological processes; it is not a direct fraction of occupied receptors. An indirect biomarker may be useful when the target cannot be measured directly, but its relationship to occupancy needs separate justification. Direct binding assays are valuable precisely because they interrogate target binding more directly, not because they eliminate measurement assumptions.

The earlier material on equilibrium occupancy and time-varying concentrations helps determine whether an equilibrium interpretation is appropriate. An assay can report occupancy at a time point without establishing that the system was at equilibrium. Likewise, receptor turnover and internalization can change the target population, so a probe signal should not automatically be interpreted as competition alone. Report the assay format, comparison, time point, denominator, and assumptions alongside any occupancy estimate.

Key takeaway: Occupancy is a fraction of a defined receptor pool. A bound-drug assay estimates it as specific bound drug divided by total receptor; a matched displacement assay estimates it as one minus the fraction of probe binding that remains. Either interpretation depends on valid specificity, normalization, and target-pool assumptions.

Check Your Understanding

For each question, identify the relevant quantity and show the calculation or reasoning.

A matched displacement assay gives specific probe signals of 150 units in control and 90 units after drug exposure. Under the assumptions of the displacement model, what occupancy is estimated?
A labeled-drug assay measures 30 pmol of specific receptor-bound drug and 50 pmol of total receptor. Calculate occupancy and the available receptor amount.
A treated sample has half the probe signal of control, but receptor abundance was not measured and may have fallen. Can the signal ratio alone establish 50% drug occupancy? Explain.
Why is a functional response reduction not automatically the same as fractional receptor occupancy?

Solutions. (1) The available fraction is $90/150=0.600$, so estimated occupancy is $1-0.600=0.400$, or 40.0%. (2) Occupancy is $30/50=0.600$, or 60.0%; available receptor is $50-30=20$ pmol. (3) No. Reduced receptor abundance could also reduce probe binding, so the matched-signal assumption that the receptor pool is unchanged is not established. (4) Functional response also depends on efficacy, receptor reserve, and downstream pathway processes; a response change does not directly count the fraction of receptors bound.