Cooperative and Allosteric Receptor Binding

Why Use Cooperative and Allosteric Binding Models?

The one-site equilibrium equation describes a receptor population with a simple relationship between free ligand concentration and fractional occupancy. Some measured binding curves, however, rise more steeply or more gradually than that equation predicts. A Hill curve can summarize this shape with a coefficient that adjusts the steepness. A distinct situation arises when a second ligand binds at another site and changes the first ligand’s binding. A simple allosteric model represents that interaction explicitly.

These models describe binding occupancy, not the functional effect of binding. In the equations below, \(L\) is the orthosteric ligand of interest, \(M\) is an allosteric modulator, and \([L]\) and \([M]\) denote their free concentrations. Fractional occupancy is the fraction of the modeled receptor population in a specified bound state or set of states.

Definition: A Hill binding curve is an empirical equilibrium relationship between free ligand concentration and fractional occupancy, parameterized by a half-occupancy concentration and a Hill coefficient. A simple allosteric binding model assigns equilibrium statistical weights to receptor states in which an orthosteric ligand and a modulator may bind, with a cooperativity factor describing how binding of either ligand changes binding of the other.

The Hill Binding Curve

A common form of the Hill curve is

\theta_L=\frac{[L]^{n_H}}{K_{0.5}^{n_H}+[L]^{n_H}}

Here, \(\theta_L\) is the fraction of receptors occupied by \(L\), \(K_{0.5}\) is the free ligand concentration at which the modeled occupancy is one-half, and \(n_H\) is the Hill coefficient. The concentrations in the numerator and denominator must use the same units, so that the ratio is dimensionless. The standard binding-curve interpretation assumes a stable receptor population and conditions under which the measured occupancy is at equilibrium with the stated free ligand concentration.

The half-occupancy interpretation follows directly: at \([L]=K_{0.5}\), the two terms in the denominator are equal. Therefore,

\theta_L=\frac{K_{0.5}^{n_H}}{K_{0.5}^{n_H}+K_{0.5}^{n_H}}=\frac{1}{2}

When \(n_H=1\), the Hill equation reduces to the one-site equilibrium occupancy equation. Values above one produce a steeper transition around \(K_{0.5}\); values between zero and one produce a shallower transition. The coefficient is a measure of curve shape, not automatically the number of ligand molecules that bind or a direct count of interacting sites. A Hill curve can describe observed behavior without identifying the molecular mechanism that produced it.

A useful transformation, when \(0<\theta_L<1\), makes the meaning of \(n_H\) explicit:

\frac{\theta_L}{1-\theta_L}=\left(\frac{[L]}{K_{0.5}}\right)^{n_H} \qquad\Longrightarrow\qquad \log\left(\frac{\theta_L}{1-\theta_L}\right) =n_H\log([L])-n_H\log(K_{0.5})

Thus, on a plot of log occupancy odds against log concentration, the Hill model has slope \(n_H\). This is a property of the specified curve; it does not establish that a particular microscopic binding mechanism is present.

Worked Example: Interpreting a Steep Hill Curve

Suppose an equilibrium binding curve is summarized by \(K_{0.5}=3.00\) nM and \(n_H=2.00\). Calculate occupancy at \([L]=6.00\) nM and interpret the parameters.

The assumed Hill model is applicable if the receptor population and free concentration are appropriately represented and the measurement is at equilibrium. Taking the reported parameters as a hypothetical curve summary, substitute the values:

\theta_L=\frac{(6.00\ \mathrm{nM})^{2.00}} {(3.00\ \mathrm{nM})^{2.00}+(6.00\ \mathrm{nM})^{2.00}} =\frac{36.0}{9.00+36.0}=0.800

The predicted fractional occupancy is 0.800, or 80.0%. The concentration \(K_{0.5}=3.00\) nM gives 50% occupancy for this curve. The coefficient \(n_H=2.00\) means that the curve is steeper near half occupancy than a one-site curve; it does not, by itself, prove that two ligand molecules bind cooperatively.

A Two-Site Allosteric Model

The Hill curve adjusts the shape of the concentration-occupancy relationship but does not represent a modulator explicitly. A simple allosteric model instead considers four receptor states: unbound \(R\), ligand-bound \(RL\), modulator-bound \(RM\), and doubly bound \(RLM\). The ligands bind at distinct modeled sites, and the doubly bound state is allowed. At equilibrium, state weights can be written relative to the unbound receptor as \(1\), \([L]/K_L\), \([M]/K_M\), and \(\alpha[L][M]/(K_LK_M)\).

In this model, \(K_L\) is the dissociation constant for \(L\) when \(M\) is absent, and \(K_M\) is the dissociation constant for \(M\) when \(L\) is absent. Smaller dissociation constants indicate tighter binding. The dimensionless cooperativity factor \(\alpha\) adjusts the weight of the doubly bound state relative to independent binding. The model assumes equilibrium, a common receptor population with these four states, and free ligand concentrations that are known or adequately approximated. It also assumes that binding is represented by these states rather than by additional receptor forms or binding sites.

The sum of all state weights is the binding polynomial, or normalizing denominator. The fraction of receptors with \(L\) bound includes both \(RL\) and \(RLM\):

Z=1+\frac{[L]}{K_L}+\frac{[M]}{K_M} +\alpha\frac{[L][M]}{K_LK_M}

\theta_L= \frac{\dfrac{[L]}{K_L}+ \alpha\dfrac{[L][M]}{K_LK_M}}{Z}

For compactness, define \(x=[L]/K_L\) and \(y=[M]/K_M\). Then

\theta_L=\frac{x+\alpha xy}{1+x+y+\alpha xy} =\frac{x(1+\alpha y)}{1+y+x(1+\alpha y)}

This form also gives the modulator-dependent apparent dissociation constant for \(L\). Dividing the numerator and denominator by \(1+\alpha y\) gives

\theta_L=\frac{[L]}{[L]+K_{L,\mathrm{app}}} \qquad K_{L,\mathrm{app}}=K_L\frac{1+[M]/K_M}{1+\alpha[M]/K_M}

This apparent constant is conditional on the specified modulator concentration; it is not a replacement for the intrinsic \(K_L\) defined when \(M\) is absent. If \(\alpha=1\), the ligands bind independently and \(K_{L,\mathrm{app}}=K_L\). If \(\alpha>1\), modulator binding favors the doubly bound state and lowers \(K_{L,\mathrm{app}}\). If \(0\leq\alpha<1\), it disfavors that state and raises \(K_{L,\mathrm{app}}\). At \(\alpha=0\), the model gives zero weight to the doubly bound state, an extreme negative interaction in this simple representation.

The modulator-bound fraction can be obtained by summing the weights of \(RM\) and \(RLM\):

\theta_M=\frac{y+\alpha xy}{1+x+y+\alpha xy}

A receptor may therefore contribute to both \(\theta_L\) and \(\theta_M\) when both ligands are bound. These fractions are not mutually exclusive categories. This is why the two occupancies need not sum to one.

Worked Example: Positive Allosteric Cooperativity

Consider a hypothetical receptor with \(K_L=10.0\) nM, \(K_M=20.0\) nM, and \(\alpha=5.00\). At equilibrium, let \([L]=10.0\) nM and \([M]=20.0\) nM. Assume these are free concentrations and that the four-state model applies.

The dimensionless concentrations are \(x=10.0/10.0=1.00\) and \(y=20.0/20.0=1.00\). The state-weight denominator is

$$ Z=1+1.00+1.00+(5.00)(1.00)(1.00)=8.00 $$

The ligand-bound fraction includes the \(RL\) and \(RLM\) weights:

\theta_L=\frac{1.00+(5.00)(1.00)(1.00)}{8.00} =\frac{6.00}{8.00}=0.750

Thus, 75.0% of receptors are predicted to have \(L\) bound. Without modulator, the one-site occupancy at \([L]=K_L\) would be \(10.0/(10.0+10.0)=0.500\). The same result follows from the conditional constant: \(K_{L,\mathrm{app}}=10.0(1+1.00)/(1+5.00)=3.33\) nM, so \(\theta_L=10.0/(10.0+3.33)=0.750\). The calculation describes a change in occupancy, not a functional response.

Worked Example: Negative Allosteric Modulation

For a second hypothetical case, let \(K_L=4.00\) nM, \(K_M=8.00\) nM, \(\alpha=0.200\), and \([L]=4.00\) nM and \([M]=8.00\) nM. Assume equilibrium and known free concentrations as in the model.

Here \(x=y=1.00\), so the denominator and the ligand-bound fraction are

$$ Z=1+1.00+1.00+(0.200)(1.00)(1.00)=3.20 $$

\theta_L=\frac{1.00+0.200}{3.20} =\frac{1.20}{3.20}=0.375

The predicted \(L\)-bound fraction is 37.5%. For comparison, with \(M\) absent and \([L]=K_L\), occupancy is 50.0%. The apparent constant at \([M]=K_M\) is \(4.00(1+1.00)/(1+0.200)=6.67\) nM; substituting gives \(4.00/(4.00+6.67)=0.375\), consistent with the state-weight calculation.

Choosing and Checking a Model

The Hill model is a compact description when the main goal is to summarize a concentration-occupancy curve, especially when there is no explicit modulator variable in the experiment. The allosteric model is more informative when two ligands bind at distinct sites and the question concerns how one ligand changes the other’s occupancy. Its additional parameters have mechanistic interpretations within the four-state assumptions, but that does not guarantee the model is uniquely identified by a given data set.

Feature	Hill curve	Simple allosteric model
Primary purpose	Summarize occupancy-curve shape	Represent modulator-dependent binding
Key parameters	\(K_{0.5}\), \(n_H\)	\(K_L\), \(K_M\), \(\alpha\)
What it describes directly	Occupancy versus free ligand concentration	Fractions of receptor states with \(L\), \(M\), or both bound
Main limitation	Does not specify the binding mechanism	Depends on the assumed states, equilibrium, and free concentrations

Several checks help identify when a fitted model may be inadequate or its parameters difficult to interpret:

Check equilibrium and concentration assumptions. If concentrations change during measurement, binding has not equilibrated, or ligand depletion makes free concentration differ substantially from the value used, these equilibrium formulas may not describe the data as written.
Check the physical range. Predicted fractions should be between zero and one. In the four-state model with nonnegative concentrations and \(\alpha\geq0\), all state weights are nonnegative, so the occupancy fractions are valid probabilities.
Check whether the curve shape is supported by data. A steep or shallow fitted Hill curve should not be treated as proof of a particular molecular mechanism. Sparse observations around the transition can also make \(K_{0.5}\) and \(n_H\) imprecise.
Check parameter identifiability. In the allosteric model, measurements only at one modulator concentration may not distinguish intrinsic \(K_L\) from the change represented by \(\alpha\). Data across ligand and modulator concentrations are more informative, although adequate coverage and a valid state model are still needed.
Check for model mismatch. Receptor subpopulations, additional binding states, or nonequilibrium behavior can produce systematic departures from either simple curve. A numerical fit alone does not resolve such structural mismatches.

The allosteric model can be reduced to the one-site occupancy equation when \([M]=0\): then \(y=0\), and \(\theta_L=x/(1+x)=[L]/(K_L+[L])\). When \(\alpha=1\), the denominator factors as \((1+x)(1+y)\), and the ligand occupancy is again \(x/(1+x)\), unaffected by \(M\). These limiting cases are useful checks on both the algebra and an implementation.

Common Mistake and Practical Interpretation

A common mistake is to interpret \(n_H=2\) as proof that exactly two ligand molecules bind each receptor. The Hill coefficient describes the steepness of the specified occupancy curve; the same apparent steepness can arise from different underlying processes, and other mechanisms can produce non-unit values. The correction is to report \(n_H\) as a curve-shape parameter unless independent mechanistic evidence supports a more specific interpretation.

A second mistake is to call \(\alpha>1\) an increase in ligand efficacy. In this model, \(\alpha\) changes the relative weight of the doubly bound state and hence the predicted occupancy relationship. It says nothing by itself about the response produced by an occupied receptor. The model’s scope is binding and target occupancy only.

Key takeaway: Use \(K_{0.5}\) and \(n_H\) to summarize the position and steepness of a Hill binding curve. Use \(K_L\), \(K_M\), and \(\alpha\) to describe conditional binding in a specified two-ligand equilibrium model. In both cases, interpret parameters within the model assumptions rather than as proof of an unobserved mechanism.

Check Your Understanding

Work through the questions, then compare your reasoning with the solutions below.

For a Hill curve with \(K_{0.5}=5.00\) nM and \(n_H=1.00\), what is the occupancy at \([L]=5.00\) nM?
For \(K_{0.5}=2.00\) nM, \(n_H=2.00\), and \([L]=4.00\) nM, calculate the Hill occupancy.
In the allosteric model, what does \(\alpha=1\) mean for ligand occupancy as modulator concentration changes?
Let \(K_L=6.00\) nM, \(K_M=12.0\) nM, \(\alpha=3.00\), and \([L]=6.00\) nM, \([M]=12.0\) nM. Calculate \(\theta_L\).
Why is \(n_H=2.00\) not sufficient evidence that two ligand molecules bind each receptor?

Solutions. (1) At \([L]=K_{0.5}\), the two denominator terms are equal, so occupancy is \(0.500\), or 50.0%. (2) \(\theta_L=4.00^2/(2.00^2+4.00^2)=16.0/20.0=0.800\), or 80.0%. (3) At \(\alpha=1\), binding is independent in this model: the denominator factors into \((1+x)(1+y)\), and \(\theta_L=x/(1+x)\), so changing \([M]\) does not change ligand occupancy. (4) Here \(x=y=1.00\), \(Z=1+1+1+3=6.00\), and \(\theta_L=(1+3)/6=0.667\), or 66.7%. (5) The Hill coefficient describes curve steepness, not a direct count of bound molecules; different mechanisms can produce similar curve shapes.