Physical Chemistry/Part III: Chemical Kinetics/Enzyme Kinetics

Enzyme Kinetics

Reading time: ~60 minutes | Topics: Michaelis-Menten equation, Lineweaver-Burk analysis, enzyme inhibition, allosteric regulation, Hill equation, catalytic efficiency, diffusion-limited reactions

1. Introduction: Enzymes as Biological Catalysts

Enzymes are remarkable molecular machines that accelerate the rates of biochemical reactions by factors of$10^6$ to $10^{17}$ while operating under mild conditions of temperature, pressure, and pH. Unlike inorganic catalysts, enzymes exhibit extraordinary specificity — each enzyme typically catalyzes a single reaction or a narrow class of reactions with exquisite selectivity for particular substrates.

The quantitative study of enzyme-catalyzed reaction rates, known as enzyme kinetics, provides essential insights into catalytic mechanisms, the effects of inhibitors and activators, and the regulation of metabolic pathways. Enzyme kinetics bridges physical chemistry and biochemistry by applying the principles of chemical kinetics — rate laws, steady-state approximations, and transition state theory — to biological catalysis.

Key Characteristics of Enzymes

• Rate enhancement: Enzymes lower the activation energy $\Delta G^\ddagger$ without altering the equilibrium constant $K_{\text{eq}}$.
• Saturation kinetics: At high substrate concentration, the rate approaches a maximum value $V_{\max}$ because all enzyme active sites are occupied.
• Specificity: The active site geometry and chemical environment selectively bind particular substrates (lock-and-key or induced-fit models).
• Regulation: Enzyme activity can be modulated by inhibitors, activators, covalent modifications, and allosteric effectors.
• Turnover number: The catalytic constant $k_{\text{cat}}$ measures how many substrate molecules one enzyme molecule converts per second.

The central equation of enzyme kinetics — the Michaelis-Menten equation — was first derived in 1913 and remains the foundation upon which all more complex kinetic models are built. In this chapter, we develop the complete theoretical framework from first principles, including the treatment of inhibition, cooperativity, and the physical limits on catalytic efficiency.

Central question: How do we mathematically describe the rate of an enzyme-catalyzed reaction as a function of substrate concentration, and what do the kinetic parameters reveal about catalytic mechanism?

2. Derivation 1: The Michaelis-Menten Equation

The simplest enzymatic mechanism involves a single substrate binding reversibly to the enzyme to form an enzyme-substrate complex, which then irreversibly converts to product:

$$\text{E} + \text{S} \underset{k_{-1}}{\overset{k_1}{\rightleftharpoons}} \text{ES} \overset{k_2}{\longrightarrow} \text{E} + \text{P}$$

Here $k_1$ is the rate constant for substrate binding, $k_{-1}$ is the rate constant for substrate dissociation from ES, and $k_2$ (often written $k_{\text{cat}}$) is the catalytic rate constant for product formation.

Step 1: Rate of Change of [ES]

The enzyme-substrate complex ES is formed by the binding of E and S (rate $k_1[\text{E}][\text{S}]$) and consumed by two processes: dissociation back to E + S (rate $k_{-1}[\text{ES}]$) and conversion to product (rate $k_2[\text{ES}]$). The rate of change is:

$$\frac{d[\text{ES}]}{dt} = k_1[\text{E}][\text{S}] - k_{-1}[\text{ES}] - k_2[\text{ES}]$$

Step 2: The Steady-State Approximation

After a brief transient period (typically microseconds), the concentration of ES reaches a steady state where it is being formed and consumed at equal rates. This is the Briggs-Haldane steady-state assumption:

$$\frac{d[\text{ES}]}{dt} = 0 \quad \Longrightarrow \quad k_1[\text{E}][\text{S}] = (k_{-1} + k_2)[\text{ES}]$$

Step 3: Define the Michaelis Constant

We define the Michaelis constant $K_M$ as:

$$K_M = \frac{k_{-1} + k_2}{k_1}$$

Note that $K_M$ has units of concentration (M). When $k_2 \ll k_{-1}$,$K_M \approx k_{-1}/k_1 = K_d$, the dissociation constant for the ES complex. In general, however,$K_M$ is a kinetic parameter, not a true equilibrium constant. From the steady-state condition:

$$[\text{ES}] = \frac{[\text{E}][\text{S}]}{K_M}$$

Step 4: Enzyme Conservation

The total enzyme concentration is conserved: $[\text{E}]_0 = [\text{E}] + [\text{ES}]$. Substituting$[\text{E}] = [\text{E}]_0 - [\text{ES}]$ into the expression for [ES]:

$$[\text{ES}] = \frac{([\text{E}]_0 - [\text{ES}])[\text{S}]}{K_M}$$

Solving for [ES]:

$$[\text{ES}] \cdot K_M = [\text{E}]_0[\text{S}] - [\text{ES}][\text{S}]$$

$$[\text{ES}](K_M + [\text{S}]) = [\text{E}]_0[\text{S}]$$

$$[\text{ES}] = \frac{[\text{E}]_0[\text{S}]}{K_M + [\text{S}]}$$

Step 5: The Michaelis-Menten Equation

The initial rate of the reaction is the rate of product formation:

$$v = k_2[\text{ES}] = \frac{k_2[\text{E}]_0[\text{S}]}{K_M + [\text{S}]}$$

Defining the maximum velocity as $V_{\max} = k_2[\text{E}]_0 = k_{\text{cat}}[\text{E}]_0$, we obtain the Michaelis-Menten equation:

$$\boxed{v = \frac{V_{\max}[\text{S}]}{K_M + [\text{S}]}}$$

Physical Interpretation of $K_M$

• When $[\text{S}] = K_M$: $v = V_{\max}/2$. Thus $K_M$ is the substrate concentration at which the reaction rate is half-maximal.
• When $[\text{S}] \ll K_M$: $v \approx (V_{\max}/K_M)[\text{S}]$ (first-order in [S]).
• When $[\text{S}] \gg K_M$: $v \approx V_{\max}$ (zero-order in [S]; enzyme is saturated).
• A small $K_M$ indicates high substrate affinity (enzyme reaches half-saturation at low [S]).

3. Derivation 2: Lineweaver-Burk and Other Linear Plots

Before nonlinear regression became routine, kinetic parameters were extracted by linearizing the Michaelis-Menten equation. Three classical linearization methods remain widely used for visualization and diagnostic purposes.

3.1 Lineweaver-Burk (Double Reciprocal) Plot

Taking the reciprocal of both sides of the Michaelis-Menten equation:

$$\frac{1}{v} = \frac{K_M + [\text{S}]}{V_{\max}[\text{S}]} = \frac{K_M}{V_{\max}} \cdot \frac{1}{[\text{S}]} + \frac{1}{V_{\max}}$$

This has the form $y = mx + b$ where:

$$\boxed{\frac{1}{v} = \frac{K_M}{V_{\max}} \cdot \frac{1}{[\text{S}]} + \frac{1}{V_{\max}}}$$

• Slope = $K_M / V_{\max}$
• y-intercept = $1 / V_{\max}$
• x-intercept = $-1 / K_M$

While widely used, the Lineweaver-Burk plot compresses data at high [S] and amplifies errors at low [S], where the reciprocal transformation distorts experimental uncertainty.

3.2 Eadie-Hofstee Plot

Rearranging the Michaelis-Menten equation by multiplying both sides by $(K_M + [\text{S}])$:

$$v(K_M + [\text{S}]) = V_{\max}[\text{S}]$$

$$v \cdot K_M + v[\text{S}] = V_{\max}[\text{S}]$$

Dividing through by $[\text{S}]$ and rearranging:

$$\boxed{v = V_{\max} - K_M \cdot \frac{v}{[\text{S}]}}$$

• Plot of $v$ vs. $v/[\text{S}]$ gives slope = $-K_M$ and y-intercept = $V_{\max}$.
• This plot distributes data more evenly but has the disadvantage that both axes depend on $v$, introducing correlated errors.

3.3 Hanes-Woolf Plot

Multiplying the Lineweaver-Burk equation by $[\text{S}]$:

$$\frac{[\text{S}]}{v} = \frac{[\text{S}]}{V_{\max}} + \frac{K_M}{V_{\max}}$$

$$\boxed{\frac{[\text{S}]}{v} = \frac{1}{V_{\max}}[\text{S}] + \frac{K_M}{V_{\max}}}$$

• Slope = $1 / V_{\max}$
• y-intercept = $K_M / V_{\max}$
• x-intercept = $-K_M$

The Hanes-Woolf plot generally provides the most uniform distribution of experimental error among the three linearization methods.

4. Derivation 3: Enzyme Inhibition

Enzyme inhibitors are molecules that reduce the catalytic activity of an enzyme. Understanding inhibition mechanisms is critical for drug design, toxicology, and metabolic regulation. There are three classical reversible inhibition types, each with a distinct effect on the kinetic parameters.

4.1 Competitive Inhibition

A competitive inhibitor (I) binds to the free enzyme at the active site, competing directly with the substrate. The inhibitor does not bind to ES:

$$\text{E} + \text{I} \rightleftharpoons \text{EI} \quad \text{with} \quad K_i = \frac{[\text{E}][\text{I}]}{[\text{EI}]}$$

The enzyme conservation becomes $[\text{E}]_0 = [\text{E}] + [\text{ES}] + [\text{EI}]$. Since $[\text{EI}] = [\text{E}][\text{I}]/K_i$, the free enzyme concentration is reduced. Defining $\alpha = 1 + [\text{I}]/K_i$, the modified rate equation becomes:

$$\boxed{v = \frac{V_{\max}[\text{S}]}{\alpha K_M + [\text{S}]} = \frac{V_{\max}[\text{S}]}{K_M^{\text{app}} + [\text{S}]}}$$

• Apparent $K_M$ increases: $K_M^{\text{app}} = \alpha K_M = K_M(1 + [\text{I}]/K_i)$
• $V_{\max}$ is unchanged — at sufficiently high [S], the substrate outcompetes the inhibitor.
• Lineweaver-Burk: Lines intersect at the y-axis ($1/V_{\max}$ unchanged, slopes increase).

4.2 Uncompetitive Inhibition

An uncompetitive inhibitor binds only to the ES complex, not to the free enzyme:

$$\text{ES} + \text{I} \rightleftharpoons \text{ESI} \quad \text{with} \quad K_i' = \frac{[\text{ES}][\text{I}]}{[\text{ESI}]}$$

Defining $\alpha' = 1 + [\text{I}]/K_i'$ and proceeding through the steady-state derivation with $[\text{E}]_0 = [\text{E}] + [\text{ES}] + [\text{ESI}]$:

$$\boxed{v = \frac{V_{\max}/\alpha' \cdot [\text{S}]}{K_M/\alpha' + [\text{S}]} = \frac{V_{\max}^{\text{app}}[\text{S}]}{K_M^{\text{app}} + [\text{S}]}}$$

• Both $K_M$ and $V_{\max}$ decrease by the same factor $\alpha'$.
• The ratio $V_{\max}/K_M$ remains unchanged.
• Lineweaver-Burk: Parallel lines (same slope, different y-intercepts).

4.3 Mixed (Noncompetitive) Inhibition

A mixed inhibitor binds to both the free enzyme and the ES complex, but with different affinities ($K_i \neq K_i'$):

$$\text{E} + \text{I} \rightleftharpoons \text{EI} \quad (K_i) \qquad \text{ES} + \text{I} \rightleftharpoons \text{ESI} \quad (K_i')$$

The total enzyme is $[\text{E}]_0 = [\text{E}] + [\text{ES}] + [\text{EI}] + [\text{ESI}]$. With $\alpha = 1 + [\text{I}]/K_i$ and $\alpha' = 1 + [\text{I}]/K_i'$:

$$\boxed{v = \frac{V_{\max}/\alpha' \cdot [\text{S}]}{\alpha K_M / \alpha' + [\text{S}]}}$$

• Apparent $V_{\max}$ decreases to $V_{\max}/\alpha'$.
• Apparent $K_M$ changes to $\alpha K_M / \alpha'$.
• Special case $K_i = K_i'$ ($\alpha = \alpha'$): Pure noncompetitive inhibition, where $K_M$ is unchanged but $V_{\max}$ decreases. Lineweaver-Burk lines intersect on the x-axis.
• General mixed case: Lineweaver-Burk lines intersect to the left of the y-axis (above or below the x-axis depending on whether $K_i > K_i'$ or $K_i < K_i'$).

Summary: Lineweaver-Burk Diagnostics

Inhibition Type	$K_M^{\text{app}}$	$V_{\max}^{\text{app}}$	L-B Pattern
Competitive	$\alpha K_M$ (increases)	Unchanged	Intersect on y-axis
Uncompetitive	$K_M/\alpha'$ (decreases)	$V_{\max}/\alpha'$ (decreases)	Parallel lines
Mixed	$\alpha K_M/\alpha'$	$V_{\max}/\alpha'$ (decreases)	Intersect left of y-axis

5. Derivation 4: Allosteric Enzymes and the Hill Equation

Many enzymes, particularly those with multiple subunits, do not follow simple Michaelis-Menten kinetics. Instead, they display sigmoidal (S-shaped) saturation curves, indicative of cooperative substrate binding. The Hill equation provides an empirical description of cooperativity.

5.1 The Hill Equation

Consider an enzyme with $n$ binding sites that bind substrate with perfect cooperativity (all-or-none model):

$$\text{E} + n\text{S} \rightleftharpoons \text{ES}_n \longrightarrow \text{E} + n\text{P}$$

If the binding is infinitely cooperative (all sites fill simultaneously), the fractional saturation$Y$ is:

$$Y = \frac{[\text{S}]^n}{K_{0.5}^n + [\text{S}]^n}$$

Since $v = V_{\max} \cdot Y$, we obtain the Hill equation:

$$\boxed{v = \frac{V_{\max}[\text{S}]^n}{K_{0.5}^n + [\text{S}]^n}}$$

Here $K_{0.5}$ is the substrate concentration at half-maximal velocity and $n$ is the Hill coefficient:

• $n = 1$: No cooperativity (reduces to Michaelis-Menten).
• $n > 1$: Positive cooperativity — binding of one substrate molecule enhances binding of subsequent molecules. Sigmoidal curve.
• $n < 1$: Negative cooperativity — binding of one substrate molecule impedes further binding.
• $n$ is always less than or equal to the actual number of binding sites.

5.2 The Hill Plot

Rearranging the Hill equation in logarithmic form:

$$\log\left(\frac{v}{V_{\max} - v}\right) = n \log[\text{S}] - n \log K_{0.5}$$

A plot of $\log(v/(V_{\max} - v))$ vs. $\log[\text{S}]$ gives a straight line with slope $n$ (the Hill coefficient) near $Y = 0.5$.

5.3 The MWC (Monod-Wyman-Changeux) Concerted Model

The MWC model (1965) provides a mechanistic basis for cooperativity. It assumes an oligomeric enzyme exists in two conformational states in equilibrium:

• T state (tense): Low affinity for substrate, dissociation constant $K_T$.
• R state (relaxed): High affinity for substrate, dissociation constant $K_R$.

Define $L = [\text{T}_0]/[\text{R}_0]$ (the allosteric constant), $c = K_R/K_T$ (the ratio of dissociation constants), and the normalized substrate $\bar{\alpha} = [\text{S}]/K_R$. For an enzyme with$n$ identical subunits, the fractional saturation is:

$$\boxed{Y = \frac{\bar{\alpha}(1 + \bar{\alpha})^{n-1} + Lc\bar{\alpha}(1 + c\bar{\alpha})^{n-1}}{(1 + \bar{\alpha})^n + L(1 + c\bar{\alpha})^n}}$$

Key features of the MWC model:

• All subunits switch conformation simultaneously (concerted transition).
• When $L \gg 1$ (T state strongly favored) and $c \ll 1$ (S binds much more tightly to R), strong positive cooperativity emerges.
• Allosteric activators shift the equilibrium toward R; inhibitors shift it toward T.
• In the limit of extreme cooperativity, the MWC model approximates the Hill equation.

6. Derivation 5: Catalytic Efficiency and the Diffusion Limit

6.1 Catalytic Efficiency: $k_{\text{cat}}/K_M$

When $[\text{S}] \ll K_M$, the Michaelis-Menten equation simplifies to:

$$v \approx \frac{V_{\max}}{K_M}[\text{S}] = \frac{k_{\text{cat}}}{K_M}[\text{E}]_0[\text{S}]$$

The ratio $k_{\text{cat}}/K_M$ is the catalytic efficiency or specificity constant. It is a second-order rate constant (units: $\text{M}^{-1}\text{s}^{-1}$) that captures both the enzyme's catalytic prowess and its affinity for the substrate in a single parameter:

$$\frac{k_{\text{cat}}}{K_M} = \frac{k_1 k_2}{k_{-1} + k_2}$$

This quantity determines the enzyme's ability to discriminate between competing substrates: the substrate with the larger $k_{\text{cat}}/K_M$ will be preferentially converted.

6.2 The Diffusion Limit (Smoluchowski Equation)

There is a fundamental physical upper bound on $k_{\text{cat}}/K_M$: the enzyme cannot catalyze a reaction faster than substrate molecules can diffuse to its active site. The Smoluchowski equation gives the diffusion-limited bimolecular rate constant:

$$k_{\text{diff}} = 4\pi (D_E + D_S) R_{ES} N_A$$

where $D_E$ and $D_S$ are the diffusion coefficients of enzyme and substrate,$R_{ES}$ is the encounter distance, and $N_A$ is Avogadro's number. For typical small substrates diffusing to enzyme active sites in aqueous solution:

$$\boxed{k_{\text{diff}} \approx 10^8 - 10^9 \; \text{M}^{-1}\text{s}^{-1}}$$

6.3 Catalytic Perfection

An enzyme whose $k_{\text{cat}}/K_M$ approaches the diffusion limit is said to have achieved catalytic perfection — its rate is limited only by how fast substrate can arrive at the active site. Notable examples include:

Enzyme	$k_{\text{cat}}$ ($\text{s}^{-1}$)	$K_M$ (M)	$k_{\text{cat}}/K_M$ ($\text{M}^{-1}\text{s}^{-1}$)
Acetylcholinesterase	$1.4 \times 10^4$	$9 \times 10^{-5}$	$1.6 \times 10^8$
Carbonic anhydrase	$1.0 \times 10^6$	$1.2 \times 10^{-2}$	$8.3 \times 10^7$
Catalase	$4.0 \times 10^7$	$1.1 \times 10^{0}$	$4.0 \times 10^7$
Superoxide dismutase	$2.0 \times 10^9$	$3.6 \times 10^{-4}$	$7.0 \times 10^9$

Superoxide dismutase actually exceeds the simple Smoluchowski estimate due to electrostatic guidance — the charged substrate is steered toward the active site by the enzyme's electric field, effectively increasing the encounter radius.

7. Applications

Drug Design: Enzyme Inhibitors

Most modern drugs are enzyme inhibitors. HIV protease inhibitors (e.g., ritonavir) are competitive inhibitors designed to mimic the transition state of peptide bond hydrolysis. Statins inhibit HMG-CoA reductase competitively to lower cholesterol. Kinetic analysis of inhibition type ($K_i$, mechanism) guides lead compound optimization through structure-activity relationships.

Metabolic Engineering

Understanding kinetic parameters of enzymes in metabolic pathways enables rational design of engineered organisms. By modifying enzymes to alter $K_M$ or $k_{\text{cat}}$, metabolic flux can be redirected toward desired products. Allosteric regulation points are key targets for pathway optimization in industrial biotechnology.

Biosensors

Enzyme-based biosensors exploit the specificity and kinetics of enzyme-substrate interactions for analytical detection. Glucose oxidase sensors (used in diabetes management) operate in the first-order regime ($[\text{S}] \ll K_M$) where the signal is proportional to analyte concentration, providing a linear calibration range.

Clinical Diagnostics

Serum enzyme levels serve as biomarkers for tissue damage. Elevated serum levels of lactate dehydrogenase (LDH), creatine kinase (CK), and cardiac troponin indicate myocardial infarction. Isoenzyme analysis (different$K_M$ and electrophoretic mobility) identifies the tissue of origin, enabling differential diagnosis.

8. Historical Context

Michaelis and Menten (1913)

Leonor Michaelis and Maud Menten published their landmark paper on the kinetics of invertase (sucrase), introducing the concept of enzyme-substrate complex formation and deriving the hyperbolic rate equation. Their original derivation used the rapid equilibrium assumption($K_M = K_d = k_{-1}/k_1$), treating the ES formation as a true equilibrium. Maud Menten, one of the first women to earn a medical doctorate in Canada, made fundamental contributions to biochemistry despite facing significant barriers in the academic establishment.

Briggs and Haldane (1925)

George Edward Briggs and J.B.S. Haldane generalized the Michaelis-Menten analysis by replacing the equilibrium assumption with the more broadly applicable steady-state approximation. This gave the same mathematical form but with $K_M = (k_{-1} + k_2)/k_1$, which equals$K_d$ only when $k_2 \ll k_{-1}$. The steady-state treatment is the standard derivation used today and applies even when the catalytic step is fast relative to substrate dissociation.

Monod, Wyman, and Changeux (1965)

Jacques Monod, Jeffries Wyman, and Jean-Pierre Changeux proposed the concerted (MWC) model for allosteric regulation, explaining the sigmoidal kinetics observed in regulatory enzymes such as aspartate transcarbamoylase and phosphofructokinase. Their model introduced the concept of symmetry-conserving conformational transitions in oligomeric proteins, providing a framework that unifies cooperative binding in enzymes and oxygen transport (hemoglobin). Monod received the Nobel Prize in 1965 (shared with Jacob and Lwoff) for work on gene regulation.

9. Python Simulation: Michaelis-Menten and Inhibition

This simulation generates four plots: (1) Michaelis-Menten curves for uninhibited and three inhibition types, (2) corresponding Lineweaver-Burk double-reciprocal plots showing the diagnostic patterns, (3) Eadie-Hofstee plots, and (4) Hill equation curves demonstrating the effect of cooperativity on the saturation profile. The numerical output compares apparent kinetic parameters under each inhibition condition.

Michaelis-Menten Kinetics & Inhibition Analysis

Python

Plots Michaelis-Menten curves, Lineweaver-Burk, Eadie-Hofstee, and Hill equation. Computes apparent kinetic parameters for competitive, uncompetitive, and mixed inhibition.

enzyme_kinetics.py195 lines

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# =====================================================
# Enzyme Kinetics: Michaelis-Menten and Inhibition
# =====================================================

def michaelis_menten(S, Vmax, Km):
    """Standard Michaelis-Menten rate equation."""
    return Vmax * S / (Km + S)

def competitive_inhibition(S, Vmax, Km, Ki, I):
    """Competitive inhibition: apparent Km increases."""
    Km_app = Km * (1 + I / Ki)
    return Vmax * S / (Km_app + S)

def uncompetitive_inhibition(S, Vmax, Km, Ki, I):
    """Uncompetitive inhibition: both Km and Vmax decrease."""
    alpha_prime = 1 + I / Ki
    return (Vmax / alpha_prime) * S / ((Km / alpha_prime) + S)

def mixed_inhibition(S, Vmax, Km, Ki, Ki_prime, I):
    """Mixed (noncompetitive) inhibition."""
    alpha = 1 + I / Ki
    alpha_prime = 1 + I / Ki_prime
    return (Vmax / alpha_prime) * S / ((Km * alpha / alpha_prime) + S)

# Parameters
Vmax = 100.0   # uM/s
Km = 20.0      # uM
Ki = 10.0      # uM (inhibitor dissociation constant)
Ki_prime = 15.0
I_conc = 15.0  # uM inhibitor concentration

S = np.linspace(0.5, 200, 500)
S_lb = np.linspace(2, 200, 200)  # for Lineweaver-Burk

# Compute rates
v_no_inhib = michaelis_menten(S, Vmax, Km)
v_competitive = competitive_inhibition(S, Vmax, Km, Ki, I_conc)
v_uncompetitive = uncompetitive_inhibition(S, Vmax, Km, Ki, I_conc)
v_mixed = mixed_inhibition(S, Vmax, Km, Ki, Ki_prime, I_conc)

# Lineweaver-Burk data
v_no_lb = michaelis_menten(S_lb, Vmax, Km)
v_comp_lb = competitive_inhibition(S_lb, Vmax, Km, Ki, I_conc)
v_uncomp_lb = uncompetitive_inhibition(S_lb, Vmax, Km, Ki, I_conc)
v_mixed_lb = mixed_inhibition(S_lb, Vmax, Km, Ki, Ki_prime, I_conc)

fig, axes = plt.subplots(2, 2, figsize=(14, 11))
fig.patch.set_facecolor('#0a0a0a')
fig.suptitle('Enzyme Kinetics: Michaelis-Menten & Inhibition Analysis',
             fontsize=16, color='white', y=0.98)

colors = ['#f87171', '#60a5fa', '#34d399', '#fbbf24']
labels = ['No Inhibitor', 'Competitive', 'Uncompetitive', 'Mixed']

# --- Plot 1: v vs [S] ---
ax = axes[0, 0]
for v_data, color, label in zip(
    [v_no_inhib, v_competitive, v_uncompetitive, v_mixed],
    colors, labels):
    ax.plot(S, v_data, linewidth=2, color=color, label=label)

ax.axhline(y=Vmax, color='gray', linestyle='--', alpha=0.5, linewidth=1)
ax.text(170, Vmax + 2, 'Vmax', color='gray', fontsize=10)
ax.axhline(y=Vmax/2, color='gray', linestyle=':', alpha=0.4)
ax.set_xlabel('[S] (\u03bcM)', fontsize=12, color='white')
ax.set_ylabel('v (\u03bcM/s)', fontsize=12, color='white')
ax.set_title('Michaelis-Menten Curves', fontsize=14, color='#f87171')
ax.legend(fontsize=9, facecolor='#1a1a2e', edgecolor='#f87171', labelcolor='white')
ax.grid(True, alpha=0.2, color='gray')
ax.set_facecolor('#0f172a')
ax.tick_params(colors='white')
for spine in ax.spines.values():
    spine.set_color('#f87171')

# --- Plot 2: Lineweaver-Burk ---
ax = axes[0, 1]
for v_data, color, label in zip(
    [v_no_lb, v_comp_lb, v_uncomp_lb, v_mixed_lb],
    colors, labels):
    ax.plot(1.0/S_lb, 1.0/v_data, linewidth=2, color=color, label=label)

ax.set_xlabel('1/[S] (1/\u03bcM)', fontsize=12, color='white')
ax.set_ylabel('1/v (s/\u03bcM)', fontsize=12, color='white')
ax.set_title('Lineweaver-Burk Plots', fontsize=14, color='#f87171')
ax.legend(fontsize=9, facecolor='#1a1a2e', edgecolor='#f87171', labelcolor='white')
ax.grid(True, alpha=0.2, color='gray')
ax.set_facecolor('#0f172a')
ax.tick_params(colors='white')
for spine in ax.spines.values():
    spine.set_color('#f87171')

# --- Plot 3: Eadie-Hofstee ---
ax = axes[1, 0]
for v_full, color, label in zip(
    [v_no_inhib, v_competitive, v_uncompetitive, v_mixed],
    colors, labels):
    ratio = v_full / S
    ax.plot(ratio, v_full, linewidth=2, color=color, label=label)

ax.set_xlabel('v/[S] (1/s)', fontsize=12, color='white')
ax.set_ylabel('v (\u03bcM/s)', fontsize=12, color='white')
ax.set_title('Eadie-Hofstee Plots', fontsize=14, color='#f87171')
ax.legend(fontsize=9, facecolor='#1a1a2e', edgecolor='#f87171', labelcolor='white')
ax.grid(True, alpha=0.2, color='gray')
ax.set_facecolor('#0f172a')
ax.tick_params(colors='white')
for spine in ax.spines.values():
    spine.set_color('#f87171')

# --- Plot 4: Hill equation ---
ax = axes[1, 1]
K_half = 20.0
for n_hill, color, lstyle in zip([0.5, 1.0, 2.0, 4.0],
    ['#60a5fa', '#f87171', '#34d399', '#fbbf24'],
    ['-', '--', '-', '-.']):
    v_hill = Vmax * S**n_hill / (K_half**n_hill + S**n_hill)
    ax.plot(S, v_hill, linewidth=2, color=color, linestyle=lstyle,
            label=f'n = {n_hill}')

ax.axhline(y=Vmax/2, color='gray', linestyle=':', alpha=0.4)
ax.axvline(x=K_half, color='gray', linestyle=':', alpha=0.4)
ax.set_xlabel('[S] (\u03bcM)', fontsize=12, color='white')
ax.set_ylabel('v (\u03bcM/s)', fontsize=12, color='white')
ax.set_title('Hill Equation (Cooperativity)', fontsize=14, color='#f87171')
ax.legend(fontsize=9, facecolor='#1a1a2e', edgecolor='#f87171', labelcolor='white')
ax.grid(True, alpha=0.2, color='gray')
ax.set_facecolor('#0f172a')
ax.tick_params(colors='white')
for spine in ax.spines.values():
    spine.set_color('#f87171')

plt.tight_layout(rect=[0, 0, 1, 0.96])
plt.savefig('output.png', dpi=150, bbox_inches='tight', facecolor='#0a0a0a')
plt.close()

# =====================================================
# Numerical output
# =====================================================
print("=" * 60)
print("ENZYME KINETICS ANALYSIS")
print("=" * 60)

print(f"\nParameters:")
print(f"  Vmax = {Vmax} \u03bcM/s")
print(f"  Km   = {Km} \u03bcM")
print(f"  Ki   = {Ki} \u03bcM")
print(f"  [I]  = {I_conc} \u03bcM")

print(f"\n--- Uninhibited Enzyme ---")
print(f"  At [S] = Km = {Km}: v = Vmax/2 = {Vmax/2:.1f} \u03bcM/s")
print(f"  Catalytic efficiency kcat/Km determines specificity")

print(f"\n--- Competitive Inhibition ---")
Km_app_comp = Km * (1 + I_conc / Ki)
print(f"  Apparent Km = Km(1 + [I]/Ki) = {Km_app_comp:.1f} \u03bcM")
print(f"  Vmax unchanged = {Vmax:.1f} \u03bcM/s")
print(f"  v at [S]=50: {competitive_inhibition(50, Vmax, Km, Ki, I_conc):.1f} \u03bcM/s")

print(f"\n--- Uncompetitive Inhibition ---")
alpha_p = 1 + I_conc / Ki
Km_app_unc = Km / alpha_p
Vmax_app_unc = Vmax / alpha_p
print(f"  Apparent Km  = Km/(1+[I]/Ki) = {Km_app_unc:.1f} \u03bcM")
print(f"  Apparent Vmax = Vmax/(1+[I]/Ki) = {Vmax_app_unc:.1f} \u03bcM/s")
print(f"  v at [S]=50: {uncompetitive_inhibition(50, Vmax, Km, Ki, I_conc):.1f} \u03bcM/s")

print(f"\n--- Mixed Inhibition ---")
alpha_m = 1 + I_conc / Ki
alpha_p_m = 1 + I_conc / Ki_prime
print(f"  alpha = 1 + [I]/Ki = {alpha_m:.2f}")
print(f"  alpha\' = 1 + [I]/Ki\' = {alpha_p_m:.2f}")
print(f"  Apparent Km  = Km*alpha/alpha\' = {Km*alpha_m/alpha_p_m:.1f} \u03bcM")
print(f"  Apparent Vmax = Vmax/alpha\' = {Vmax/alpha_p_m:.1f} \u03bcM/s")

print(f"\n--- Hill Equation ---")
for n in [0.5, 1.0, 2.0, 4.0]:
    v_50 = Vmax * 50**n / (K_half**n + 50**n)
    print(f"  n={n}: v at [S]=50 = {v_50:.1f} \u03bcM/s ({v_50/Vmax*100:.1f}% Vmax)")

print(f"\n--- Catalytic Efficiency ---")
kcat = 500  # s^-1 (example)
eff = kcat / (Km * 1e-6)  # M^-1 s^-1
print(f"  kcat = {kcat} s^-1, Km = {Km} \u03bcM")
print(f"  kcat/Km = {eff:.2e} M^-1 s^-1")
print(f"  Diffusion limit: ~10^8 - 10^9 M^-1 s^-1")
if eff > 1e8:
    print(f"  This enzyme approaches catalytic perfection!")
else:
    print(f"  Below diffusion limit by factor of {1e9/eff:.0f}x")

Click Run to execute the Python code

Code will be executed with Python 3 on the server

10. Fortran Simulation: Kinetic Parameter Estimation

This Fortran program simulates experimental enzyme kinetics data (with realistic noise) and extracts$V_{\max}$ and $K_M$ using three independent linearization methods: Lineweaver-Burk, Eadie-Hofstee, and Hanes-Woolf. By comparing the parameter estimates from each method, we can assess their relative accuracy and sensitivity to experimental error — a classical exercise in physical chemistry.

Enzyme Kinetics: Parameter Estimation from Simulated Data

Fortran

Generates noisy kinetic data and estimates Vmax, Km via Lineweaver-Burk, Eadie-Hofstee, and Hanes-Woolf regression. Computes catalytic efficiency.

enzyme_kinetics.f90195 lines

program enzyme_kinetics
  implicit none
  ! ==========================================================
  ! Enzyme Kinetics Parameter Estimation from Simulated Data
  ! Computes Km, Vmax via Lineweaver-Burk linear regression
  ! and compares with Eadie-Hofstee and Hanes-Woolf methods
  ! ==========================================================

integer, parameter :: dp = selected_real_kind(15, 307)
  integer, parameter :: n_data = 12

real(dp) :: S(n_data), v_obs(n_data), v_calc(n_data)
  real(dp) :: inv_S(n_data), inv_v(n_data)
  real(dp) :: v_over_S(n_data), S_over_v(n_data)
  real(dp) :: Vmax_true, Km_true, kcat, E0
  real(dp) :: Vmax_lb, Km_lb, Vmax_eh, Km_eh, Vmax_hw, Km_hw
  real(dp) :: slope, intercept, r_squared
  real(dp) :: sum_x, sum_y, sum_xy, sum_x2, sum_y2, n_real
  real(dp) :: ss_res, ss_tot, y_mean, residual
  real(dp) :: noise, kcat_est, efficiency
  integer  :: i, seed

! True parameters
  Vmax_true = 150.0_dp   ! uM/s
  Km_true   = 25.0_dp    ! uM
  kcat      = 600.0_dp   ! s^-1
  E0        = Vmax_true / kcat  ! total enzyme conc

! Substrate concentrations (uM)
  S = (/ 1.0_dp, 2.5_dp, 5.0_dp, 10.0_dp, 15.0_dp, 20.0_dp, &
         30.0_dp, 50.0_dp, 75.0_dp, 100.0_dp, 150.0_dp, 250.0_dp /)

! Generate simulated data with small noise
  seed = 42
  write(*,'(A)') '# Enzyme Kinetics Parameter Estimation'
  write(*,'(A)') '# ======================================'
  write(*,'(A,F8.2,A)') '# True Vmax = ', Vmax_true, ' uM/s'
  write(*,'(A,F8.2,A)') '# True Km   = ', Km_true, ' uM'
  write(*,'(A,F8.2,A)') '# kcat      = ', kcat, ' s^-1'
  write(*,'(A,ES12.4,A)') '# [E]_0     = ', E0, ' uM'
  write(*,'(A)') '#'

write(*,'(A)') '# Simulated Experimental Data:'
  write(*,'(A)') '#   [S](uM)     v_obs(uM/s)   v_exact(uM/s)'

do i = 1, n_data
    v_calc(i) = Vmax_true * S(i) / (Km_true + S(i))
    ! Simple pseudo-random noise (linear congruential)
    seed = mod(seed * 1103515245 + 12345, 2147483647)
    noise = (real(mod(seed, 1000), dp) / 1000.0_dp - 0.5_dp) * 0.04_dp
    v_obs(i) = v_calc(i) * (1.0_dp + noise)
    write(*,'(A,F10.2,F14.3,F14.3)') '#  ', S(i), v_obs(i), v_calc(i)
  end do

! ============================
  ! Method 1: Lineweaver-Burk
  ! ============================
  n_real = real(n_data, dp)
  do i = 1, n_data
    inv_S(i) = 1.0_dp / S(i)
    inv_v(i) = 1.0_dp / v_obs(i)
  end do

sum_x  = 0.0_dp; sum_y  = 0.0_dp
  sum_xy = 0.0_dp; sum_x2 = 0.0_dp; sum_y2 = 0.0_dp
  do i = 1, n_data
    sum_x  = sum_x  + inv_S(i)
    sum_y  = sum_y  + inv_v(i)
    sum_xy = sum_xy + inv_S(i) * inv_v(i)
    sum_x2 = sum_x2 + inv_S(i)**2
    sum_y2 = sum_y2 + inv_v(i)**2
  end do

slope     = (n_real * sum_xy - sum_x * sum_y) / (n_real * sum_x2 - sum_x**2)
  intercept = (sum_y - slope * sum_x) / n_real

Vmax_lb = 1.0_dp / intercept
  Km_lb   = slope * Vmax_lb

! R-squared
  y_mean = sum_y / n_real
  ss_res = 0.0_dp; ss_tot = 0.0_dp
  do i = 1, n_data
    residual = inv_v(i) - (slope * inv_S(i) + intercept)
    ss_res = ss_res + residual**2
    ss_tot = ss_tot + (inv_v(i) - y_mean)**2
  end do
  r_squared = 1.0_dp - ss_res / ss_tot

write(*,'(A)') ''
  write(*,'(A)') '=== Method 1: Lineweaver-Burk (1/v vs 1/[S]) ==='
  write(*,'(A,F12.6)') '  Slope (Km/Vmax)     = ', slope
  write(*,'(A,F12.6)') '  Intercept (1/Vmax)  = ', intercept
  write(*,'(A,F10.3,A)') '  Estimated Vmax      = ', Vmax_lb, ' uM/s'
  write(*,'(A,F10.3,A)') '  Estimated Km        = ', Km_lb, ' uM'
  write(*,'(A,F10.6)')   '  R-squared           = ', r_squared
  write(*,'(A,F8.2,A)') '  Vmax error          = ', &
    abs(Vmax_lb - Vmax_true)/Vmax_true * 100.0_dp, ' %'
  write(*,'(A,F8.2,A)') '  Km error            = ', &
    abs(Km_lb - Km_true)/Km_true * 100.0_dp, ' %'

! ============================
  ! Method 2: Eadie-Hofstee
  ! v = Vmax - Km * (v/[S])
  ! ============================
  do i = 1, n_data
    v_over_S(i) = v_obs(i) / S(i)
  end do

sum_x  = 0.0_dp; sum_y  = 0.0_dp
  sum_xy = 0.0_dp; sum_x2 = 0.0_dp
  do i = 1, n_data
    sum_x  = sum_x  + v_over_S(i)
    sum_y  = sum_y  + v_obs(i)
    sum_xy = sum_xy + v_over_S(i) * v_obs(i)
    sum_x2 = sum_x2 + v_over_S(i)**2
  end do

slope     = (n_real * sum_xy - sum_x * sum_y) / (n_real * sum_x2 - sum_x**2)
  intercept = (sum_y - slope * sum_x) / n_real

Vmax_eh = intercept
  Km_eh   = -slope

write(*,'(A)') ''
  write(*,'(A)') '=== Method 2: Eadie-Hofstee (v vs v/[S]) ==='
  write(*,'(A,F12.6)') '  Slope (-Km)         = ', slope
  write(*,'(A,F12.6)') '  Intercept (Vmax)    = ', intercept
  write(*,'(A,F10.3,A)') '  Estimated Vmax      = ', Vmax_eh, ' uM/s'
  write(*,'(A,F10.3,A)') '  Estimated Km        = ', Km_eh, ' uM'
  write(*,'(A,F8.2,A)') '  Vmax error          = ', &
    abs(Vmax_eh - Vmax_true)/Vmax_true * 100.0_dp, ' %'
  write(*,'(A,F8.2,A)') '  Km error            = ', &
    abs(Km_eh - Km_true)/Km_true * 100.0_dp, ' %'

! ============================
  ! Method 3: Hanes-Woolf
  ! [S]/v = [S]/Vmax + Km/Vmax
  ! ============================
  do i = 1, n_data
    S_over_v(i) = S(i) / v_obs(i)
  end do

sum_x  = 0.0_dp; sum_y  = 0.0_dp
  sum_xy = 0.0_dp; sum_x2 = 0.0_dp
  do i = 1, n_data
    sum_x  = sum_x  + S(i)
    sum_y  = sum_y  + S_over_v(i)
    sum_xy = sum_xy + S(i) * S_over_v(i)
    sum_x2 = sum_x2 + S(i)**2
  end do

slope     = (n_real * sum_xy - sum_x * sum_y) / (n_real * sum_x2 - sum_x**2)
  intercept = (sum_y - slope * sum_x) / n_real

Vmax_hw = 1.0_dp / slope
  Km_hw   = intercept * Vmax_hw

write(*,'(A)') ''
  write(*,'(A)') '=== Method 3: Hanes-Woolf ([S]/v vs [S]) ==='
  write(*,'(A,F12.6)') '  Slope (1/Vmax)      = ', slope
  write(*,'(A,F12.6)') '  Intercept (Km/Vmax) = ', intercept
  write(*,'(A,F10.3,A)') '  Estimated Vmax      = ', Vmax_hw, ' uM/s'
  write(*,'(A,F10.3,A)') '  Estimated Km        = ', Km_hw, ' uM'
  write(*,'(A,F8.2,A)') '  Vmax error          = ', &
    abs(Vmax_hw - Vmax_true)/Vmax_true * 100.0_dp, ' %'
  write(*,'(A,F8.2,A)') '  Km error            = ', &
    abs(Km_hw - Km_true)/Km_true * 100.0_dp, ' %'

! ============================
  ! Catalytic efficiency
  ! ============================
  write(*,'(A)') ''
  write(*,'(A)') '=== Catalytic Efficiency ==='
  kcat_est = Vmax_lb / E0
  efficiency = kcat_est / (Km_lb * 1.0e-6_dp)  ! M^-1 s^-1
  write(*,'(A,F10.2,A)') '  kcat (from LB Vmax) = ', kcat_est, ' s^-1'
  write(*,'(A,ES12.4,A)') '  kcat/Km             = ', efficiency, ' M^-1 s^-1'
  write(*,'(A)') '  Diffusion limit     ~ 1.0E+08 to 1.0E+09 M^-1 s^-1'
  if (efficiency > 1.0e8_dp) then
    write(*,'(A)') '  ** Near catalytic perfection! **'
  else
    write(*,'(A,F8.1,A)') '  Factor below limit  = ', 1.0e9_dp / efficiency, 'x'
  end if

write(*,'(A)') ''
  write(*,'(A)') '=== Comparison of Methods ==='
  write(*,'(A20,A12,A12)') 'Method', 'Vmax', 'Km'
  write(*,'(A20,A12,A12)') '------', '----', '--'
  write(*,'(A20,F12.3,F12.3)') 'True values', Vmax_true, Km_true
  write(*,'(A20,F12.3,F12.3)') 'Lineweaver-Burk', Vmax_lb, Km_lb
  write(*,'(A20,F12.3,F12.3)') 'Eadie-Hofstee', Vmax_eh, Km_eh
  write(*,'(A20,F12.3,F12.3)') 'Hanes-Woolf', Vmax_hw, Km_hw

end program enzyme_kinetics

Click Run to execute the Fortran code

Code will be compiled with gfortran and executed on the server

Key Equations Summary

Michaelis-Menten Equation

$$v = \frac{V_{\max}[\text{S}]}{K_M + [\text{S}]} \qquad K_M = \frac{k_{-1} + k_2}{k_1} \qquad V_{\max} = k_{\text{cat}}[\text{E}]_0$$

Lineweaver-Burk

$$\frac{1}{v} = \frac{K_M}{V_{\max}} \cdot \frac{1}{[\text{S}]} + \frac{1}{V_{\max}}$$

Hill Equation

$$v = \frac{V_{\max}[\text{S}]^n}{K_{0.5}^n + [\text{S}]^n}$$

Catalytic Efficiency & Diffusion Limit

$$\frac{k_{\text{cat}}}{K_M} \leq k_{\text{diff}} \approx 10^8 - 10^9 \; \text{M}^{-1}\text{s}^{-1}$$

MWC Concerted Model

$$Y = \frac{\bar{\alpha}(1 + \bar{\alpha})^{n-1} + Lc\bar{\alpha}(1 + c\bar{\alpha})^{n-1}}{(1 + \bar{\alpha})^n + L(1 + c\bar{\alpha})^n}$$

← Part III Overview