Isothermal titration calorimetry
Measure affinity, stoichiometry and binding thermodynamics directly.
Our ruleSimulations prepare you for the bench but don't replace hands-on practice.
1Measuring binding as heat
Almost every binding event releases or absorbs heat. An isothermal titration calorimeter holds a sample cell and a reference cell at the same constant temperature and adds the ligand to the sample cell from a motor-driven syringe, a few microlitres at a time, while a paddle stirs.
2The c-value decides what you can measure
The shape of the isotherm depends on one number, c = n·[protein in the cell]/KD. Between roughly 10 and 100 the curve is a clear sigmoid and KD, n and ΔH are all well determined.
ΔH
Heat per mole of ligand when every injected molecule binds; measured directly, not inferred from temperature dependence.
n
Molar ratio at the inflection. It reports active protein as well as stoichiometry, so it carries every concentration error.
KD
From the curvature of the transition; reliable only when c is roughly 10–100.
ΔG, −TΔS
ΔG from KD, and the entropic term by difference: the enthalpy–entropy balance of the interaction.
3How it goes wrong
The calorimeter measures every heat in the cell, so anything that differs between syringe and cell is measured as well. Ligand dissolved in water, a different pH, a few percent of DMSO in the syringe but not in the cell, or a reducing agent that oxidises during the run all give heats of mixing that can be larger than the binding heat and drift through the titration.
Key terms
Practise these as flashcards →The difference in heater power between sample and reference cell needed to keep them at the same temperature; integrated over a peak it gives the heat of one injection.
n times the cell concentration divided by KD. It sets the shape of the isotherm; about 10–100 is the useful window for measuring KD.
Moles of ligand bound per mole of protein at saturation, taken from the midpoint of the isotherm; it also reports the active fraction of the protein.
The heat of injecting ligand into buffer without a binding partner, measured in a control titration and subtracted or fitted as an offset.
Total ligand over total protein in the cell after each injection, the x-axis of the isotherm.
Titrating a tight ligand into protein pre-bound to a weaker one, which lowers the apparent affinity into the measurable range.
The common observation that changes in ΔH between related ligands are partly offset by changes in −TΔS, leaving ΔG similar.
When to use it
- Measuring affinity, stoichiometry and binding enthalpy in one label-free experiment in solution, for protein–small molecule, protein–protein or protein–nucleic acid interactions.
- Explaining why a ligand binds: separating enthalpic from entropic contributions, for example when comparing analogues in a medicinal chemistry series.
- Checking the stoichiometry and active fraction of a purified protein preparation before other binding assays.
- Interactions where a label or a surface could change the binding, and where enough protein (tens to hundreds of micrograms per titration) is available.
Compared with alternatives
Gives on- and off-rates as well as KD with far less material, but needs one partner on a surface and gives no enthalpy or stoichiometry directly.
KD in solution from microlitres and nanomolar labelled protein; prefer it when protein is scarce, but it gives no thermodynamics and needs a label.
Much higher throughput for a small labelled ligand and good for competition screens; no ΔH or n.
The calorimetry of unfolding rather than binding: Tm, ΔH and ΔCp of the protein, and ligand-induced stabilisation.