Competitive Inhibition — Apparent K_m

A competitive inhibitor is a molecule that fits the active site. It cannot be turned into product, but while it is sitting there the substrate cannot get in, so the two compete for the same seat. The consequence is a clean one: the enzyme needs more substrate to reach any given rate, so $K_m$ appears to rise by the factor $\alpha = 1 + [I]/K_i$, while $V_{max}$ does not move at all.

The unchanged $V_{max}$ is the diagnosis and it is worth understanding rather than memorising. Because inhibitor and substrate compete for the same site, the competition can always be won by weight of numbers: flood the system with substrate and essentially every enzyme molecule ends up holding substrate rather than inhibitor, and the ceiling is exactly where it was. Competitive inhibition can be OUTCOMPETED. That is not true of the other mechanisms, and it is why the distinction matters practically as well as diagnostically.

The other two classic patterns make the contrast. A pure non-competitive inhibitor binds somewhere other than the active site and disables the enzyme regardless of whether substrate is bound; adding substrate does not help, so $V_{max}$ falls and $K_m$ is unchanged. An uncompetitive inhibitor binds only the enzyme–substrate complex, so more substrate makes matters worse rather than better, and both constants fall by the same factor. On a double-reciprocal plot those three give three visibly different families of lines — pivoting about the vertical intercept, pivoting about the horizontal intercept, and parallel — which is the single best reason that plot is still drawn.

$K_i$ is a concentration. It is the dissociation constant of the enzyme–inhibitor complex, in the same units as $[I]$, and it is the inhibitor concentration at which the apparent $K_m$ is exactly doubled. A small $K_i$ means a potent inhibitor. Note that it is a property of the enzyme–inhibitor pair alone, whereas an IC₅₀ from a screening assay depends on the substrate concentration that assay happened to use — the Cheng–Prusoff relation converts between them, and quoting an IC₅₀ as though it were a $K_i$ is a common and entirely avoidable error.

Where this matters in an industrial process is usually product inhibition, which is very often competitive: the product resembles the substrate it came from closely enough to fit the same site. Glucose inhibiting a cellulase, or a sugar inhibiting the glycosidase that released it, are ordinary cases. The consequence is that conversion slows as it approaches completion, and the standard answers are all about removing the product — membrane reactors that pull it out continuously, coupled reactions that consume it, or simply accepting a lower conversion and recycling. The same arithmetic explains why substrate analogues are used deliberately as process control agents, and why an inhibitor carried over from an upstream step can quietly halve the throughput of a downstream one.

Run the relation backwards and it becomes an assay. An enzyme whose $K_m$ and $K_i$ are known reports the concentration of an inhibitor in an unknown sample by how much its apparent $K_m$ has shifted; that is the operating principle of a good many enzyme-based sensors used in environmental and process monitoring. Precision is best when $[I]$ is within a factor of a few of $K_i$ — far below, the shift disappears into the scatter; far above, the apparent $K_m$ grows until the assay runs out of substrate range.