Biochemistry
Michaelis-Menten Kinetics
v = (V_max·[S])/(K_m + [S]) — the canonical hyperbolic enzyme rate equation
Michaelis-Menten kinetics is the rate equation v = V_max·[S]/(K_m + [S]) that describes how a single-substrate enzyme responds to substrate concentration. K_m is the [S] at half-V_max; k_cat = V_max/[E_total] is the turnover number, and k_cat/K_m is the catalytic efficiency, bounded above by the diffusion limit ~10^9 M^-1 s^-1. Leonor Michaelis and Maud Menten published the rapid-equilibrium derivation in 1913; George Briggs and J.B.S. Haldane gave the more general steady-state derivation in 1925. The hyperbolic curve covers ~95% of one-substrate enzymes and is the substrate from which competitive, noncompetitive, and uncompetitive inhibition kinetics are derived.
- Equationv = V_max·[S]/(K_m + [S])
- K_m[S] at half-V_max (units: concentration)
- k_cat range~10^-2 to 10^7 s^-1
- Diffusion limitk_cat/K_m ~10^8–10^9 M^-1 s^-1
- OriginMichaelis & Menten 1913, Briggs-Haldane 1925
- Curve shapeHyperbolic (vs sigmoidal cooperative)
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
Why Michaelis-Menten matters
- It quantifies enzyme behavior. Two parameters — K_m and V_max (equivalently k_cat once you know [E]) — summarize the catalytic behavior of any one-substrate enzyme over the entire substrate range. Compare two enzymes' K_m and k_cat and you can predict which one dominates flux at any [S].
- Foundation of pharmacology. Almost every drug-target affinity number you see — IC50, Ki, EC50 — descends from Michaelis-Menten algebra. Competitive, noncompetitive, and uncompetitive inhibition kinetics are derived by adding an inhibitor binding term to the basic equation. Pharmacology is applied enzyme kinetics.
- K_m matched to physiology. Hexokinase K_m for glucose is ~0.1 mM — far below the 5 mM blood glucose, so it is saturated and runs at V_max. Glucokinase (liver) K_m ~10 mM — near blood glucose, so it titrates with [S] and acts as a glucose sensor. Same reaction, different K_m, different physiological role. K_m is evolutionarily tuned, not arbitrary.
- k_cat/K_m identifies catalytic perfection. A few enzymes — triose phosphate isomerase (~2.4×10^8 M^-1 s^-1), catalase (~4×10^8), fumarase (~1.5×10^8), acetylcholinesterase (~1.5×10^8) — operate at the diffusion limit. They cannot evolve faster because they are limited by how often substrate hits the active site. Their existence implies that natural selection has explored the entire kinetic landscape.
- Quick filter for cooperativity. Plot v vs [S]. If the curve is hyperbolic, Michaelis-Menten applies. If sigmoidal, the enzyme is cooperative (allosteric, multi-site) and needs the Hill or MWC framework instead. Hemoglobin's O2-binding curve is sigmoidal because it is a tetramer; myoglobin's is hyperbolic because it is a monomer.
- Identifies inhibitor mechanism. The way K_m and V_max change in response to an inhibitor distinguishes competitive (K_m up, V_max same), noncompetitive (V_max down, K_m same), and uncompetitive (both down by the same factor) inhibition. This is one of the few clean experimental discriminations available in pharmacology.
- Key parameter for metabolic modeling. Whole-cell flux balance analysis uses Michaelis-Menten parameters to build kinetic models of metabolism. The BRENDA database catalogues ~140,000 K_m and k_cat values across enzymes, substrates, and organisms. Any quantitative cell biology builds on this corpus.
Common misconceptions
- K_m is the dissociation constant. Only when k_cat << k_-1. The general Briggs-Haldane definition is K_m = (k_-1 + k_cat)/k_1, which equals K_d only in the rapid-equilibrium limit Michaelis and Menten originally assumed. For most modern enzymes k_cat is comparable to k_-1, so K_m and K_d differ.
- Higher k_cat means a better enzyme. Not necessarily. At physiological [S] far below K_m, the rate is governed by k_cat/K_m. An enzyme with k_cat=100 and K_m=0.01 outpaces one with k_cat=10000 and K_m=10 at [S]=0.1, despite the second's apparent k_cat advantage.
- V_max is a property of the enzyme. V_max = k_cat × [E_total]. Doubling enzyme concentration doubles V_max. The intrinsic property is k_cat. Always report kinetics with [E] specified, or quote k_cat directly.
- The plot is hyperbolic so the mechanism is single-step. Michaelis-Menten kinetics emerge from any kinetic scheme where a steady-state [ES] exists. Multi-step mechanisms with the rate-limiting step at any single point also fit. The hyperbola tells you nothing about the number of intermediates.
- Lineweaver-Burk is the standard analysis. Not since the 1980s. Nonlinear regression directly on v vs [S] is the correct method. Lineweaver-Burk distorts errors and biases estimates. Use it only as a teaching aid for visualizing inhibitor mechanisms.
- Enzymes always obey Michaelis-Menten. Cooperative enzymes (hemoglobin, ATCase, PFK) give sigmoidal curves. Multi-substrate enzymes need Cleland's classification. Substrate inhibition causes the curve to fall after a peak. Pre-steady-state kinetics on sub-millisecond timescales follows different equations entirely.
How the rate equation is derived
Michaelis and Menten in 1913 used a rapid-equilibrium assumption: substrate binding is fast and reaches equilibrium before the slow chemical step. So [ES]/([E][S]) = 1/K_d, and v = k_cat[ES]. This works when k_cat << k_-1. Briggs and Haldane in 1925 generalized to a steady-state approximation: d[ES]/dt = 0 across most of the reaction, even when k_cat is comparable to k_-1. The steady-state assumption gives [ES] = [E_total][S]/(K_m + [S]) where K_m = (k_-1 + k_cat)/k_1, and v = k_cat[ES] = V_max[S]/(K_m + [S]) where V_max = k_cat[E_total]. This is the form modern textbooks use.
Two regimes. At [S] << K_m, v ≈ (V_max/K_m)[S] = (k_cat/K_m)[E][S] — pseudo-first-order in substrate, with k_cat/K_m the rate constant. At [S] >> K_m, v ≈ V_max — zero-order in substrate. The crossover between these regimes is at [S] = K_m, where v = V_max/2. Plotting v vs [S] gives a rectangular hyperbola; plotting log v vs log [S] gives a sigmoid in log space whose midpoint is K_m. The diffusion limit on k_cat/K_m comes from the Smoluchowski-Debye theory of reaction encounter rates: roughly 10^9 M^-1 s^-1 for small molecules in water.
Inhibition adds new equilibria. Competitive: I binds free E with constant K_i, raising apparent K_m to K_m(1 + [I]/K_i) without changing V_max. Noncompetitive: I binds both E and ES equally, lowering apparent V_max to V_max/(1 + [I]/K_i) with K_m unchanged. Uncompetitive: I binds only ES, lowering both V_max and K_m by 1/(1 + [I]/K_i) — substrate-induced inhibition. Mixed inhibition (different K_i for E and ES) is the more general form. Pharmacologically, methotrexate (DHFR), captopril (ACE), and statins (HMG-CoA reductase) are competitive; cyanide on cytochrome c oxidase is noncompetitive; lithium on inositol monophosphatase is uncompetitive.
Competitive vs Noncompetitive vs Uncompetitive inhibition
| Property | Competitive | Noncompetitive | Uncompetitive |
|---|---|---|---|
| Inhibitor binds | Free E (active site) | Both E and ES (allosteric) | Only ES complex |
| Apparent K_m | Increases | Unchanged | Decreases |
| Apparent V_max | Unchanged | Decreases | Decreases |
| Lineweaver-Burk pattern | Same y-intercept, different slope | Same x-intercept, different slope | Parallel lines |
| k_cat/K_m | Decreases | Decreases | Unchanged |
| Overcome by high [S]? | Yes | No | No (worsens) |
| Drug example | Methotrexate (DHFR), captopril (ACE) | Cyanide on cyt c oxidase | Lithium on IMPase |
| Mechanism | Substrate analog | Conformational change | Stabilizes ES complex |
Famous enzymes and their numbers
- Carbonic anhydrase II. k_cat ~10^6 s^-1 — among the fastest known enzymes. K_m for CO2 ~12 mM. Catalyzes CO2 + H2O <-> HCO3- + H+ in red blood cells, essential for CO2 transport. The Zn2+ in the active site activates a water molecule for nucleophilic attack.
- Hexokinase vs glucokinase. Hexokinase (most tissues) K_m ~0.1 mM glucose; glucokinase (liver, beta cells) K_m ~10 mM. Same reaction, different K_m, different role. Glucokinase's glucose-sensing K_m underpins insulin secretion in beta cells.
- Acetylcholinesterase. k_cat ~1.4×10^4 s^-1, K_m ~95 µM, k_cat/K_m ~1.5×10^8 M^-1 s^-1 — a diffusion-limited 'perfect' enzyme. Its speed is necessary because synaptic acetylcholine must be cleared in <1 ms for firing rates to track. Sarin nerve agents covalently inhibit it.
- HIV protease. K_m for natural Gag/Gag-Pol substrates ~1 to 100 µM; k_cat ~1 to 10 s^-1. Saquinavir, ritonavir, lopinavir are all transition-state mimics with Ki in the nanomolar range — competitive inhibitors that revolutionized HIV treatment in the mid-1990s.
- Triose phosphate isomerase. k_cat/K_m ~2.4×10^8 M^-1 s^-1. Catalyzes the interconversion of dihydroxyacetone phosphate and glyceraldehyde-3-phosphate in glycolysis. Operates near the diffusion limit and is the textbook example of catalytic perfection.
Frequently asked questions
What does K_m actually mean?
K_m is the substrate concentration at which the enzyme operates at half its maximum velocity. Mathematically K_m = (k_-1 + k_cat)/k_1 in the Briggs-Haldane derivation, where k_1 is the on-rate of substrate to enzyme, k_-1 is the off-rate, and k_cat is the catalytic step. K_m has units of concentration (typically µM or mM). It is not a binding constant unless k_cat << k_-1, in which case it reduces to the dissociation constant K_d. Hexokinase has K_m ~0.1 mM for glucose, exquisitely tuned to physiological blood glucose ~5 mM (well above K_m, so the enzyme is saturated). Glucokinase, the liver isoform, has K_m ~10 mM, near physiological glucose, allowing the liver to titrate uptake against blood glucose levels.
What is k_cat and why does k_cat/K_m matter?
k_cat is the turnover number — the maximum number of substrate molecules each enzyme active site converts to product per second when fully saturated. Carbonic anhydrase has the fastest known k_cat at ~10^6 s^-1; catalase (~10^7 s^-1) is similar. Most enzymes have k_cat in the 1 to 1000 s^-1 range. The ratio k_cat/K_m is the catalytic efficiency — the apparent second-order rate constant when [S] is far below K_m, where v ≈ (k_cat/K_m)[E][S]. It is bounded above by the diffusion limit, roughly 10^8 to 10^9 M^-1 s^-1, set by how fast enzyme and substrate can collide in solution. Several enzymes — triose phosphate isomerase, fumarase, acetylcholinesterase — operate near this catalytic perfection, meaning they have evolved to the physical maximum.
How is K_m measured?
Run the enzyme at fixed [E_total] across a range of substrate concentrations spanning roughly 0.1·K_m to 10·K_m, measure initial velocity v0 (within the first 5 to 10 percent of substrate consumption), and plot v0 vs [S]. Fit the hyperbola v = V_max·[S]/(K_m + [S]) by nonlinear regression. Historically people used linearized plots — Lineweaver-Burk (1/v vs 1/[S]), Eadie-Hofstee (v vs v/[S]), or Hanes-Woolf ([S]/v vs [S]) — but those distort the error structure; Lineweaver-Burk in particular weights low-[S] points heavily and is now considered statistically inferior. Modern practice fits the hyperbolic equation directly. Reliable K_m measurements typically require 8 to 12 substrate concentrations and at least three technical replicates per point.
Why does Lineweaver-Burk distort the data?
The Lineweaver-Burk plot transforms v -> 1/v and [S] -> 1/[S]. Small errors at low v become enormous after inversion, so points at low [S] dominate the linear regression. Points at high [S] cluster near the y-intercept and contribute little leverage. The result is biased estimates of K_m and V_max with poorly characterized confidence intervals. James Eadie's 1942 plot (v vs v/[S]) and Charles Hanes's 1932 plot ([S]/v vs [S]) have less severe error distortion but still violate the assumption of homoscedastic residuals. Nonlinear regression on the original v-[S] data is unambiguously the correct approach today; computers make it trivial. Lineweaver-Burk persists in textbooks because its slope-and-intercept geometry visualizes inhibitor mechanisms well — it is a teaching tool, not a fitting tool.
What is the difference between competitive and noncompetitive inhibition?
Competitive inhibitors bind the active site and compete with substrate. They raise the apparent K_m without changing V_max — at saturating [S], substrate outcompetes inhibitor. On Lineweaver-Burk the lines pivot around the same y-intercept (1/V_max). Methotrexate competing with dihydrofolate at DHFR is the classic example. Noncompetitive (more strictly mixed-type) inhibitors bind a separate site and reduce the effective [E_active], lowering V_max while leaving K_m roughly unchanged. Lines pivot around the same x-intercept (-1/K_m). Pure noncompetitive is rare; mixed inhibition is more common. Uncompetitive inhibitors bind only the ES complex and decrease both K_m and V_max by the same factor, giving parallel Lineweaver-Burk lines. Lithium on inositol monophosphatase is a textbook uncompetitive case.
When does Michaelis-Menten break down?
Five common cases. (1) Multi-substrate reactions need ordered/random/ping-pong kinetic schemes — Cleland's notation handles them. (2) Cooperative enzymes give sigmoidal curves, not hyperbolic; use Hill equation or MWC model. (3) Substrate inhibition at high [S] causes velocity to fall off after a peak — common in hexokinase at high glucose, alcohol dehydrogenase, monoamine oxidase. (4) Slow product release in the off step makes V_max lower than k_cat-limited would predict; use full chemical-mechanism kinetics. (5) Pre-steady-state (sub-millisecond) or single-turnover regimes need stopped-flow or quenched-flow methods because the Briggs-Haldane assumption that d[ES]/dt ≈ 0 fails. Real enzymes also show transient burst kinetics on the timescale of one catalytic cycle, which Michaelis-Menten averages away.