Ligand Depletion and the Binding Mass Balance

Why Total Ligand May Not Be Free Ligand

The familiar equilibrium occupancy equation is written in terms of free ligand concentration. In a binding experiment or a target-rich tissue compartment, however, some ligand molecules are bound to target. If this bound amount is not negligible, the supplied total ligand concentration is not the concentration available to bind. Using total ligand as though it were free can then overstate occupancy, sometimes substantially.

The distinction is important in PK/PD models because dose or measured total concentration is not automatically the concentration driving target binding. This tutorial considers a single ligand binding reversibly to one class of identical target sites at equilibrium. It derives the mass-balance solution and gives practical checks for when the simpler equation is an adequate approximation.

Definition: Let \(L\) denote free ligand concentration, \(R\) free target-site concentration, and \(C\) ligand–target complex concentration. The corresponding total concentrations are \(L_T=L+C\) and \(R_T=R+C\). For reversible 1:1 binding at equilibrium, the dissociation constant is \(K_D=LR/C\). Fractional target occupancy is \(\theta=C/R_T\), provided \(R_T>0\).

All concentrations in these equations must use compatible units and refer to the same binding compartment. The mass balance assumes that each complex contains one ligand molecule and one target site, and that the totals include all free and specifically bound molecules or sites in that system.

Deriving the Binding Mass-Balance Solution

Substitute the mass balances \(L=L_T-C\) and \(R=R_T-C\) into the equilibrium definition. This gives one equation in the unknown complex concentration:

K_D=\frac{(L_T-C)(R_T-C)}{C}.

Multiplying through by \(C\), expanding, and collecting terms yields a quadratic:

$$ C^2-(L_T+R_T+K_D)C+L_TR_T=0. $$

The quadratic formula gives two roots. The physically admissible root is the smaller one: binding cannot consume more ligand or target than is present, so \(0\le C\le\min(L_T,R_T)\). The larger root is outside that feasible range (except in degenerate boundary cases). The smaller root is therefore

C=\frac{L_T+R_T+K_D-\sqrt{(L_T+R_T+K_D)^2-4L_TR_T}}{2}.

This expression involves subtracting two similar numbers when binding is weak or one total concentration is small. For more reliable numerical calculation in such cases, rationalize the numerator to obtain an equivalent form:

C=\frac{2L_TR_T}{L_T+R_T+K_D+\sqrt{(L_T+R_T+K_D)^2-4L_TR_T}}.

Both forms give the same physical solution. Once \(C\) is found, calculate \(L=L_T-C\), \(R=R_T-C\), and \(\theta=C/R_T\). The free ligand is not an extra fitted quantity in this closed, equilibrium system; it follows from the same mass balance.

Key result: For a single 1:1 binding system at equilibrium, with conserved totals and \(K_D>0\), the complex concentration is the smaller quadratic root above. The physically consistent solution satisfies \(0\le C\le\min(L_T,R_T)\); occupancy is \(C/R_T\), not \(C/L_T\).

Worked Binding Calculations

Worked Example: Substantial Ligand Depletion

Suppose a hypothetical equilibrium system has \(L_T=10.0\) nM, \(R_T=10.0\) nM, and \(K_D=3.617\) nM. Calculate complex concentration, free ligand, and target occupancy. Assume a homogeneous population of single binding sites, 1:1 reversible binding, conserved totals, and that equilibrium has been reached.

The assumptions allow the equilibrium constant and both mass balances to describe the same system. The discriminant is

(10.0+10.0+3.617)^2-4(10.0)(10.0) =23.617^2-400 =157.762689\ {\rm nM}^2.

Its square root is approximately \(12.5604\) nM. Applying the smaller-root expression,

C=\frac{23.617-12.5604}{2} =5.5283\ {\rm nM}.

Thus free ligand is \(L=L_T-C=10.0-5.5283=4.4717\) nM, and free target is \(R=10.0-5.5283=4.4717\) nM. Occupancy is

\theta=\frac{C}{R_T} =\frac{5.5283\ {\rm nM}}{10.0\ {\rm nM}} =0.5528.

About 55.28% of the supplied ligand is bound, so treating all 10.0 nM as free is not justified for an accurate calculation. The mass balance also checks: \(L+C=4.4717+5.5283=10.0\) nM and \(R+C=10.0\) nM.

Worked Example: When Depletion Is Small

Consider a hypothetical system with \(L_T=100\) nM, \(R_T=1.00\) nM, and \(K_D=10.0\) nM. Under the same assumptions of one-site 1:1 equilibrium binding and conserved totals, compare exact occupancy with the calculation that substitutes total ligand for free ligand.

The discriminant is \( (111)^2-4(100)(1)=11{,}921\ {\rm nM}^2\), whose square root is approximately \(109.1833\) nM. Using the numerically stable form,

C=\frac{2(100)(1)}{111+109.1833} =\frac{200}{220.1833} =0.9083\ {\rm nM}.

Exact occupancy is \(0.9083/1.00=0.9083\), and free ligand is \(100-0.9083=99.0917\) nM. The simple equation, if total ligand is substituted for free ligand, gives

\theta_{\rm simple}=\frac{L_T}{K_D+L_T} =\frac{100}{10.0+100} =0.9091.

The absolute occupancy difference is about \(0.0008\), or 0.08 percentage points. Here the approximation is close because at most 1.00 nM can be bound, small relative to the 100 nM ligand total. Whether this difference is acceptable depends on the precision needed for the modeling task; the comparison does not make the approximation exact.

Worked Example: Total Ligand Substitution Can Overstate Occupancy

For a hypothetical system with \(L_T=10.0\) nM, \(R_T=10.0\) nM, and \(K_D=2.00\) nM, calculate exact occupancy and compare it with the simple equation. Assume equilibrium, conserved totals, and homogeneous 1:1 binding sites.

The discriminant is \(22.0^2-4(10.0)(10.0)=84.0\ {\rm nM}^2\), with square root \(9.1652\) nM. Therefore,

C=\frac{22.0-9.1652}{2} =6.4174\ {\rm nM}, \qquad \theta=\frac{6.4174}{10.0}=0.6417.

Free ligand is \(10.0-6.4174=3.5826\) nM. In contrast, substituting \(L_T\) gives \(\theta_{\rm simple}=10.0/(2.00+10.0)=0.8333\). That is higher than the mass-balance occupancy because it treats the bound ligand as if it remained free. The exact result is internally consistent: \(LR/C\) is approximately \((3.5826)(3.5826)/6.4174=2.00\) nM, matching the stated \(K_D\).

When the Simple Occupancy Equation Works

If free ligand concentration is known, the equilibrium relation directly gives the usual occupancy equation. Since \(R=R_T-C\), rearranging \(K_D=LR/C\) gives \(C/R_T=L/(K_D+L)\). Thus

\theta=\frac{L}{K_D+L}.

This formula does not require negligible depletion when \(L\) truly is the free concentration. The potential error arises when \(L_T\), a total concentration, is substituted for \(L\). The exact relation is instead \(\theta=(L_T-C)/(K_D+L_T-C)\), with \(C\) determined by the mass balance.

A useful preliminary check is that the bound amount cannot exceed \(R_T\). If \(R_T\) is much smaller than \(L_T\), depletion is bounded by a small fraction of the ligand total. In the low-target limit, binding is approximately \(C\approx R_T L_T/(K_D+L_T)\), so the fraction of total ligand bound is approximately \(R_T/(K_D+L_T)\). This is a screening approximation, not a replacement for the exact solution when precision matters.

Practical takeaway: Use \(\theta=L/(K_D+L)\) when the input is free ligand. If the available input is total ligand and target binding may appreciably reduce free ligand, solve the mass balance first. A small estimated depletion relative to \(L_T\) supports—but does not prove—the adequacy of the total-ligand approximation.

Assumptions, Diagnostics, and Alternatives

The quadratic solution is exact for the stated equilibrium model, not for every biological binding system. Its assumptions are that binding is reversible and at equilibrium; one ligand binds one site per complex; sites are identical and independent; the system is sufficiently well mixed for concentrations to be meaningful; and the stated totals include the relevant free and bound pools. Other sinks, nonspecific binding, multiple site classes, changing target abundance, or incomplete equilibration can make the model inappropriate.

Several checks can reveal calculation or model problems:

Mass-balance check: Verify \(L_T=L+C\) and \(R_T=R+C\), with all concentrations nonnegative.
Feasibility check: The computed complex must satisfy \(C\le\min(L_T,R_T)\), and occupancy must lie between 0 and 1.
Equilibrium check: For positive \(C\), calculate \(LR/C\); it should equal the specified \(K_D\), up to rounding.
Input check: Confirm whether an assay or model input represents free or total ligand. The label “concentration” alone does not resolve this.
Model check: Persistent disagreement between predicted and observed binding across concentrations may indicate that the single-site equilibrium assumptions do not describe the system, rather than a need to alter the quadratic root.

For a model with only one unknown complex concentration, the closed-form quadratic is usually simplest. A numerical solver can also enforce the mass balances and equilibrium equation, which is useful when the model is extended or when the analytic expression is numerically inconvenient; the solver must still be constrained to the physical range. The simple hyperbolic equation is a useful alternative only when free ligand is known or when a justified approximation makes total and free ligand effectively interchangeable. A time-dependent binding model is needed if equilibrium is not an adequate description, but the equilibrium quadratic alone does not describe how quickly binding occurs.

A common mistake is to compute occupancy as \(L_T/(K_D+L_T)\) merely because total ligand is the reported concentration. The equation is defined using free ligand. The correction is to calculate \(C\) from the mass balance, obtain \(L=L_T-C\), and then evaluate \(L/(K_D+L)\), or equivalently use \(C/R_T\). Confusing these steps can lead to systematically inflated predicted occupancy when target binding appreciably depletes ligand.

Check Your Understanding

Answer each question, then compare with the solutions.

In a 1:1 system, what two mass balances connect free concentrations, complex, and total concentrations?
For \(L_T=8.0\) nM, \(R_T=2.0\) nM, and \(K_D=4.0\) nM, use the quadratic to find \(C\) and occupancy.
Why is the larger root of the quadratic not the physical binding solution?
When is \(\theta=L/(K_D+L)\) valid, and what is the issue with substituting \(L_T\) for \(L\)?

Solutions. (1) The balances are \(L_T=L+C\) and \(R_T=R+C\). (2) The coefficient sum is \(8+2+4=14\), and the discriminant is \(14^2-4(8)(2)=132\ {\rm nM}^2\). Thus \(C=(14-\sqrt{132})/2=(14-11.4891)/2=1.2554\) nM. Occupancy is \(1.2554/2.0=0.6277\), or 62.77%. This is feasible because \(C\) is below both totals. (3) The larger root exceeds the physically available amount of ligand or target, violating \(C\le\min(L_T,R_T)\). (4) The equation uses free ligand concentration. Substituting total ligand is justified only when total and free concentrations are sufficiently close for the intended accuracy; otherwise solve the mass balance first.