Competitive Inhibition Under Changing Substrate Concentrations

Why Substrate Concentration Changes Inhibition

A competitive inhibitor and the substrate compete for access to the same free enzyme. Increasing inhibitor concentration therefore tends to reduce the fraction of enzyme processing substrate. But the prediction is not determined by drug concentration alone: at the same inhibitor exposure, more substrate makes the inhibitor less effective in this model. This matters in a mechanistic QSP model whenever substrate concentrations vary across tissues, disease states, or time.

The key quantity is an apparent inhibition constant that combines the inhibitor’s intrinsic binding affinity with the substrate concentration. It is “apparent” because it describes inhibition under a specified substrate condition; it is not a change in the inhibitor’s underlying affinity for the enzyme.

Model and Derivation

Consider the reaction scheme in which substrate binds free enzyme to form a productive enzyme–substrate complex, while inhibitor binds free enzyme to form an inactive enzyme–inhibitor complex. The substrate and inhibitor cannot occupy the enzyme simultaneously. Assume reversible binding, initial-rate conditions, constant substrate and inhibitor concentrations during the rate measurement, and no substantial depletion of either ligand by enzyme binding. The inhibitor is assumed to equilibrate with the enzyme, and the unbound concentration at the target is represented by \(I_u\). The usual Michaelis–Menten description is assumed to apply to substrate processing.

In this model, \(K_m\) characterizes the substrate dependence of enzyme processing, while \(K_i\) characterizes inhibitor binding to free enzyme. These constants have concentration units. The total enzyme is divided among free enzyme, enzyme–substrate complex, and enzyme–inhibitor complex. The fraction in the productive complex is proportional to \(S/K_m\), while inhibitor binding contributes a competing term \(I_u/K_i\).

With no inhibitor, the rate at the same substrate concentration is \(v_0=V_{\max}S/(K_m+S)\). Define fractional inhibition as the decrease in rate divided by the uninhibited rate at that same \(S\): \(F_I=1-v_I/v_0\). Substitution and simplification give the result below.

The algebra identifies the apparent constant \(K_{i,\mathrm{app}}=K_i(1+S/K_m)\). It follows that \(F_I=I_u/(I_u+K_{i,\mathrm{app}})\), a concentration-response form analogous to the unbound-concentration inhibition relation, but with substrate-dependent potency. At \(I_u=K_{i,\mathrm{app}}\), the predicted rate is reduced by one half relative to the uninhibited rate at that substrate concentration. When \(S=K_m\), \(K_{i,\mathrm{app}}=2K_i\); when \(S\) is much larger than \(K_m\), the apparent constant rises approximately in proportion to \(S/K_m\).

Worked Predictions at Fixed Drug Exposure

Interpretation, Diagnostics, and Limits

The substrate dependence follows directly from the model’s competition for free enzyme. Increasing \(S\) raises \(K_{i,\mathrm{app}}\), so the same \(I_u\) produces a smaller fractional reduction in the rate compared with the no-inhibitor rate at that same \(S\). This does not mean that substrate changes the inhibitor’s intrinsic binding constant \(K_i\). Nor does it imply that the absolute inhibited rate must fall when substrate rises: the uninhibited rate itself usually rises with substrate, and the absolute rate depends on both effects.

The classical competitive model predicts an unchanged \(V_{\max}\) and an increased apparent \(K_m\), specifically \(K_m(1+I_u/K_i)\). This gives a useful diagnostic: rates measured across substrate concentrations at more than one inhibitor concentration should be compatible with a shared \(V_{\max}\), with inhibitor-related shifts in the substrate dependence. A systematic change in the fitted maximum rate, inconsistent shifts, or inability to describe rates using one \(K_i\) may indicate that the competitive assumptions are inadequate, that concentrations are not at equilibrium, or that the measurements are affected by processes outside the model. Such patterns warrant testing a different mechanistic description rather than forcing a competitive fit.

The formula is also a practical comparison with a concentration-only inhibition calculation. A calculation based on \(I_u\) and \(K_i\) alone would omit the factor \(1+S/K_m\) and therefore could overpredict inhibition when substrate is appreciable. Conversely, an empirically measured half-inhibitory concentration obtained at one substrate level should not be transferred unchanged to a different substrate condition. Direct measurement of enzyme rates over relevant substrate and inhibitor ranges is useful when the substrate varies substantially or when the mechanism is uncertain.

Several limitations matter in QSP applications. The derivation describes initial enzyme rates, not automatically a time-varying pathway response. It assumes the target-site unbound inhibitor concentration is known and approximately constant over the measurement, that free substrate is represented by \(S\), and that enzyme binding does not appreciably deplete substrate or inhibitor. If substrate or inhibitor concentrations are transient, a dynamic model may be needed. If binding is slow, the equilibrium prediction may not hold at the time of interest. Finally, fractional enzyme-rate inhibition is not necessarily the same as fractional change in a downstream biomarker: downstream dynamics and pathway organization determine how a rate change propagates.

Common Pitfall

A common error is to calculate inhibition from \(I_u/(I_u+K_i)\) and treat the result as independent of substrate. That expression uses the intrinsic inhibitor constant as though substrate were absent. For the competitive model at a specified substrate concentration, the correct denominator is \(I_u+K_i(1+S/K_m)\). For example, if \(S=3K_m\), the apparent constant is \(4K_i\), not \(K_i\). The correction is to state the substrate condition, calculate \(K_{i,\mathrm{app}}\), and then compare \(I_u\) with that apparent constant.