← All simulations

Binding Curves and Titrant Depletion

A protein P binds a single site on a DNA D (or any other ligand), and θ is the fraction of the DNA that is bound, as total protein PT is added. When the DNA concentration DT is well below Kd, binding hardly changes the free protein concentration, the curve is the simple hyperbola, and it is half-saturated at PT = Kd. When DT is comparable to Kd or larger, the DNA soaks up protein, the curve moves to the right and sharpens, and its midpoint is Kd + DT/2. Measuring Kd then requires the full quadratic equation, and when DT ≫ Kd the curve hardly depends on Kd at all.

The top two plots also show why protein concentrations in a binding experiment should be a geometric series: use the switch above either plot to compare evenly spaced and geometric points. Concentrations are in nM, but any unit works if they are all the same. Move the pointer over a plot to read off values.

Settings
nM
nM
tonM
nM
nM

Binding curve, linear axis

Evenly spaced points pile up on the plateau; a geometric series spreads them across the transition

Evenly spacedGeometric series

Binding curve, logarithmic axis

Evenly spaced points pile up on the plateau; a geometric series spreads them across the transition

Evenly spacedGeometric series

Effect of the DNA concentration DT

LinearLogarithmic

Effect of Kd

LinearLogarithmic
How this is calculated

This page is the web version of the Matlab script Binding_DT_script.m. For P + D ⇌ PD, with mass balances for the protein and the DNA,

Kd = [P][D]/[PD], PT = [P] + [PD], DT = [D] + [PD], theta = [PD]/DT

In terms of the free protein concentration the binding curve is always a hyperbola:

theta = [P]/(Kd + [P])

but experiments control the total protein. Eliminating [P] and [D] gives a quadratic equation for [PD], whose physically meaningful root is

theta = ((PT + DT + Kd) - sqrt((PT + DT + Kd)^2 - 4 PT DT))/(2 DT)

(the page evaluates it in an equivalent form that avoids round-off when DT is tiny). Only when DT ≪ Kd is the bound protein a negligible part of the total, so that [P] ≈ PT:

theta approx PT/(Kd + PT) when DT much less than Kd

Solving the quadratic for PT gives the protein needed for any fraction bound, including the midpoint:

PT = theta/(1 - theta) Kd + theta DT; theta = 1/2 at PT = Kd + DT/2

At the midpoint half the DNA, DT/2, has taken protein out of solution, and the free protein is exactly Kd. When DT ≪ Kd the midpoint gives Kd; when DT ≫ Kd it gives only DT/2, and the experiment becomes a stoichiometric titration, which can measure the concentration of active protein but not Kd.

Spacing the protein concentrations. The evenly spaced series runs from 0 to the highest protein concentration and the geometric series from the lowest to the highest, with the same number of points (21 in the Matlab script); the switch above the top plots chooses which one is plotted and listed in the table. The “informative points” in the summary are those between 10% and 90% bound, where the data actually constrain Kd. The logarithmic plots leave out PT = 0.

This page was checked by its author on desktop browsers in light mode. No human testing was done in dark mode or on phones.

Jason D. Kahn, Dept. of Chemistry and Biochemistry, University of Maryland College Park, using Claude

Complete table of values

To use these numbers elsewhere, click Download .csv: the file opens directly in Excel, Numbers or Google Sheets, and in Matlab T = readtable('filename.csv') reads it into a table. Or click Copy table and paste into a spreadsheet; each value lands in its own cell. The export has full precision.