Why Occupancy Alone Can Be Ambiguous
A measured receptor occupancy is often interpreted alongside a ligand concentration to assess target engagement. The interpretation depends on which concentration is available. The ligand concentration that drives binding is the free concentration at the receptor site; a plasma concentration or total tissue concentration need not equal it. If affinity and site concentration are both unknown, even a precise occupancy measurement may not distinguish them.
This tutorial focuses on the ambiguity that remains under a simple equilibrium binding model. It does not require occupancy to be a functional response: occupancy describes the fraction of available receptors bound by ligand. The central question is what the occupancy observation can identify, and what additional measurement is needed to identify affinity separately from concentration.
Here, \(C\) and \(K_D\) must use the same concentration units, such as nM. The equation assumes a single reversible binding class, equilibrium, and a concentration \(C\) that is effectively constant during binding. It also assumes that the measured occupancy corresponds to the fraction of available receptors bound by this ligand. Competition, multiple receptor classes, cooperativity, receptor turnover, and time-varying binding can require different models; the simple equation should not be applied to those settings without justification.
What One Occupancy Measurement Identifies
Rearranging the equilibrium equation makes the ambiguity explicit. For \(0<\theta<1\),
If \(C\) is known, the last expression gives \(K_D\). If \(K_D\) is known, the same equation gives \(C=K_D\theta/(1-\theta)\). But if both are unknown, one occupancy value supplies only their ratio \(C/K_D\). For every positive choice of \(C\), there is a corresponding \(K_D=C(1-\theta)/\theta\) that produces the same occupancy. Thus, exact measurement of occupancy alone does not resolve the ambiguity.
Worked Example: Two Concentration–Affinity Pairs, One Occupancy
Suppose a hypothetical direct occupancy assay reports \(\theta=0.60\) at equilibrium. Find two different free-concentration and affinity pairs consistent with that result.
The identifiable ratio is \(\theta/(1-\theta)=0.60/0.40=1.5\). Therefore, \(C/K_D=1.5\).
For \(C=3.00\) nM, the implied affinity constant is \(K_D=3.00(1-0.60)/0.60=3.00(0.40)/0.60=2.00\) nM.
For \(C=12.0\) nM, it is \(K_D=12.0(0.40)/0.60=8.00\) nM. Checking the original model, \(3.00/(2.00+3.00)=0.600\) and \(12.0/(8.00+12.0)=0.600\).
Both pairs give the same occupancy. The observation supports the ratio \(C/K_D=1.5\), but cannot tell which pair is the actual one.
Why the Ambiguity Occurs
The equilibrium relation follows from balancing binding and unbinding. For the reaction \(L+R \rightleftharpoons B\), the forward binding rate is \(k_{\mathrm{on}}CR\) and the reverse unbinding rate is \(k_{\mathrm{off}}B\). At equilibrium these rates are equal:
Because \(\theta=B/(R+B)\), substitute \(B/R=C/K_D\), or equivalently \(B=(C/K_D)R\):
Only the ratio \(C/K_D\) enters this expression. Multiplying both \(C\) and \(K_D\) by the same positive factor leaves occupancy unchanged. This symmetry is the structural source of the confounding, not merely a consequence of noisy data or a weak assay. Better precision on the same single occupancy measurement cannot break it.
Measurements That Separate Concentration and Affinity
The direct remedy is to measure one of the two unknown quantities independently or to collect occupancy observations under conditions that provide known, varying concentrations. The measurement must match the model: a concentration used in the equation should represent free ligand at the binding site, not a convenient but unvalidated surrogate.
- Measure free site concentration. With \(C\) measured at the relevant site and time, calculate \(K_D=C(1-\theta)/\theta\), subject to the equilibrium and single-site assumptions.
- Measure affinity independently. An independent binding experiment that estimates \(K_D\) allows occupancy to be used to infer \(C\). The independent estimate must apply to the same ligand, receptor state, and relevant conditions.
- Use a controlled concentration series. Measure occupancy at several known free concentrations while the receptor system is at equilibrium. If the concentrations span the binding transition, the location of the transition informs \(K_D\).
- Measure bound and free ligand together. A validated paired measurement can supply the concentration information missing from occupancy alone, provided the measured free pool represents concentration at the receptor site.
Worked Example: Using a Measured Free Concentration
In a hypothetical equilibrium experiment, a direct occupancy assay gives \(\theta=0.80\), and a separate measurement estimates free ligand at the receptor site as \(C=4.00\) nM. Estimate \(K_D\).
The model assumptions are that the assay reports fractional occupancy of one available receptor class and that the measured site concentration represents the free concentration driving binding at equilibrium. Apply \(K_D=C(1-\theta)/\theta\):
\(K_D=4.00\text{ nM}\times(1-0.80)/0.80 =4.00\text{ nM}\times0.20/0.80 =1.00\text{ nM}\).
The estimated dissociation constant is \(1.00\) nM. Substitution verifies the result: \(4.00/(1.00+4.00)=0.800\). If the 4.00 nM value were instead a plasma total concentration, using it here would not be justified without evidence that it represents free ligand at the receptor site.
Worked Example: A Known-Concentration Series
Consider hypothetical equilibrium measurements at three controlled free concentrations: \(1.00\), \(3.00\), and \(9.00\) nM. The measured occupancies are \(0.250\), \(0.500\), and \(0.750\), respectively. Assume one shared \(K_D\) across conditions and that the concentrations are known and measured without material error.
For the first observation, \(K_D=1.00(1-0.250)/0.250=1.00(0.750)/0.250=3.00\) nM. For the second, \(K_D=3.00(1-0.500)/0.500=3.00\) nM. For the third, \(K_D=9.00(1-0.750)/0.750=3.00\) nM.
All three observations are consistent with \(K_D=3.00\) nM. The half-occupancy observation occurs at \(C=K_D\), since \(\theta=0.500\) implies \(C/(K_D+C)=0.500\) and therefore \(C=K_D\). In real data, measurement noise means the individual calculations may differ; a model fit to the full series is generally preferable to treating each point as an exact estimate.
Information, Diagnostics, and Limitations
A concentration series resolves the particular concentration–affinity ambiguity only when its concentration scale is known and its data contain information about the binding transition. If all concentrations are unknown up to a common multiplicative scale, multiplying both the concentrations and \(K_D\) by that scale preserves the predicted occupancies. Likewise, a series that samples only a narrow range or lies almost entirely at saturation may estimate \(K_D\) imprecisely, even when the model is structurally identifiable.
The model gives \(\theta=0.5\) at \(C=K_D\). Observations near this transition are generally more informative about \(K_D\) than observations very close to zero or one. The inversion \(K_D=C(1-\theta)/\theta\) also shows why endpoint measurements are fragile: it is undefined at \(\theta=0\), and as \(\theta\) approaches 1, small errors in occupancy can cause large changes in the inferred \(K_D\). These cases should not be treated as exact affinity estimates from the inversion.
Useful checks include confirming that the occupancy assay is calibrated to a fraction, confirming that concentration is free rather than total, checking that observations are consistent with a monotone, saturating response, and assessing whether the assumed equilibrium and single-site model are plausible. A systematic departure from that shape may indicate model inadequacy, heterogeneity, time dependence, or measurement problems; it does not by itself identify which explanation is responsible.
| Approach | What it adds | Main condition or limitation |
|---|---|---|
| One occupancy measurement | Identifies \(C/K_D\) under the simple equilibrium model | Cannot estimate both unknowns separately |
| Independent free-concentration measurement | Allows \(K_D\) to be calculated from occupancy | Must represent free ligand at the receptor site |
| Known-concentration series | Can estimate a shared \(K_D\) from the transition in occupancy | Needs a known concentration scale and informative range |
| Independent affinity experiment | Provides \(K_D\), allowing occupancy to inform concentration | Affinity must transfer to the conditions of interest |
These approaches address the same ambiguity with different information. An independent concentration measurement is direct for an in vivo concentration question but may be technically difficult at the target site. An in vitro affinity estimate can be easier to obtain but may not represent the receptor environment in vivo. A concentration series uses the occupancy relation itself, but depends on controlled, known concentrations and a shared equilibrium model. No approach makes an incorrect model correct.
The simple relationship is also limited when equilibrium is not reached, when free concentration changes materially during binding, or when several receptor populations or binding mechanisms contribute to the measured occupancy. In such cases, a richer model and additional observations may be needed. The key diagnostic remains: before interpreting a fitted affinity, ask what independent information fixes the concentration scale and whether the occupancy model describes the system being measured.
Check Your Understanding
Answer each question using the assumptions of the single-site equilibrium model unless the question says otherwise. Brief solutions follow the questions.
- An occupancy of \(0.40\) is measured, but neither free concentration nor \(K_D\) is known. What combination of parameters is identified?
- At measured free concentration \(6.00\) nM, occupancy is \(0.75\). Calculate \(K_D\), with units.
- Why does increasing the precision of one occupancy measurement not solve the ambiguity when both \(C\) and \(K_D\) are unknown?
- A concentration series has known free concentrations, but every occupancy is near \(0.99\). What concern remains for estimating \(K_D\)?
- Explain why a measured plasma total concentration cannot automatically be substituted for \(C\) in the occupancy equation.
Solutions. 1. The identified ratio is \(C/K_D=\theta/(1-\theta)=0.40/0.60=0.667\) (approximately); the separate values are not identified. 2. \(K_D=6.00(1-0.75)/0.75=6.00(0.25)/0.75=2.00\) nM. 3. The equation depends only on \(C/K_D\), so infinitely many pairs produce the same exact occupancy; precision does not add a second independent constraint. 4. The series may contain little information about the transition and may estimate \(K_D\) imprecisely; high saturation makes small occupancy errors consequential. 5. The equation uses free concentration at the receptor site, whereas plasma total concentration may include bound ligand and may not match the site concentration.