ITC for Antibody-Antigen Interactions: A Practical Guide to Kd Determination, pH Conditions, and Sample Requirements

Why ITC Matters for Antibody-Antigen Characterization — and Where Most Experiments Go Wrong

Isothermal titration calorimetry (ITC) is the only biophysical technique that delivers binding affinity (Kd), enthalpy (ΔH), entropy (ΔS), and stoichiometry (n) from a single label-free experiment. For antibody-antigen characterization, ITC provides quantitative thermodynamic profiles that ELISA and SPR alone cannot supply—revealing not just whether binding occurs, but how and why it changes with pH, buffer composition, or formulation conditions.

Yet despite its power, ITC experiments for antibody-antigen interactions fail more often than necessary. The most common causes—incorrect concentration ranges, buffer mismatch, and insufficient sample quality—are entirely preventable. This guide walks through the critical parameters—cell concentration, buffer matching, pH optimization, sample preparation, and data quality—so that researchers can design ITC experiments that yield publication-ready antibody binding data on the first attempt.

For a general introduction to ITC principles, refer to our Isothermal Titration Calorimetry overview. An overview of the technique and its applications is available on the ITC analysis service page.

Cell Concentration and Stoichiometry: Getting the Numbers Right

The most common reason for failed antibody-antigen ITC experiments is incorrect concentration choice. ITC measures heat released or absorbed as discrete injections of one binding partner (typically the antigen in the syringe) are titrated into the other (the antibody in the cell). For the binding isotherm to be well-defined, the concentrations of both partners must be chosen relative to the expected Kd.

General concentration guidelines

For example, if the expected Kd is 50 nM (as in a typical high-affinity antibody-antigen interaction), the antibody concentration in the cell should be 1–5 µM, and the antigen concentration in the syringe should be 15–75 µM.

ITC thermogram (top) and integrated binding isotherm (bottom) for a typical antibody-antigen titration, showing heat response per injection and the sigmoidal saturation curve.

Stoichiometry considerations

Antibodies are bivalent (IgG) or multivalent (IgM), which affects the expected binding stoichiometry (n). For a monoclonal IgG:

• Fab–antigen: If using the full IgG in the cell and monovalent antigen in the syringe, the stoichiometry (n) should approach 2—two antigen binding sites per antibody molecule.
• Fab fragment: If using Fab fragments, n should approach 1.
• Actual n values: Slight deviations from theoretical n (e.g., n = 1.7–1.9) are normal and may reflect partial activity, aggregation, or imprecise concentration estimates.

What to do when Kd is unknown

If the approximate Kd is not known, run a scouting experiment with the antibody at 10–20 µM in the cell and antigen at 150–300 µM in the syringe. A well-defined sigmoidal isotherm confirms that the concentrations are in the right range. A flat or nearly linear trace indicates either that the concentration is too low (Kd is higher than expected) or that binding is too weak to detect. In the latter case, increase both concentrations proportionally.

Buffer Matching and pH Optimization

Buffer composition is the second most common source of ITC failure. Because ITC measures heat directly, any difference in buffer composition between the cell and syringe solutions produces a heat of dilution that can obscure or distort the binding signal.

Buffer matching protocol

• Dialyze or buffer-exchange both the antibody and antigen into the identical buffer batch. Use at least two buffer exchanges (≥1,000× volume each).
• Filter both solutions through 0.22 µm filters to remove particulates that cause baseline noise.
• Degas both solutions (or use vacuum degassing if available) to prevent bubble formation during the injection.
• Confirm pH of both solutions is identical (±0.02 pH units) at the experimental temperature.

pH titration experiments

For experiments requiring Kd comparison at multiple pH conditions (e.g., pH 7.4 vs. pH 5.5):

• Prepare separate matched buffer sets for each pH condition. Each pH requires its own complete buffer exchange.
• Run pH 7.4 first: physiological pH typically yields the highest binding signal for most antibody-antigen pairs.
• Run pH 5.5 second: acidic pH may mimic endosomal conditions or test pH-dependent release in ADC and bispecific antibody design.
• Between pH runs: clean the ITC cell thoroughly with 1% SDS followed by extensive water and buffer washes.

How pH affects antibody-antigen binding

pH changes the ionization state of residues on both the antibody paratope and the antigen epitope. Histidine (pKa ~6.0), glutamate (pKa ~4.3), and aspartate (pKa ~3.9) are the most pH-sensitive residues in the physiological-to-acidic range. A Kd shift from nM at pH 7.4 to µM at pH 5.5 is common and may be biologically relevant—for example, in recycling antibodies or pH-switchable therapeutics. ITC has been used to characterize pH-dependent antibody variants, with published studies demonstrating that substituting histidine residues at the heavy-light chain interface can produce Kd ratios (pH 7.4 / pH 5.5) exceeding 100-fold (Schröter et al., 2024). Comparing ΔH and −TΔS across pH values reveals whether binding is driven by specific polar interactions (enthalpy) or hydrophobic burial (entropy) and how these contributions shift.

pH-dependent antibody-antigen binding: physiological pH 7.4 (left) maintains electrostatic interactions, while acidic pH 5.5 (right) disrupts key ionic contacts via histidine protonation.

For complementary biophysical characterization, see our Co-Immunoprecipitation guide for orthogonal binding validation.

Sample Preparation and Quality Control

ITC is exquisitely sensitive to sample quality. A 50 nM Kd interaction requires clean, monomeric, well-characterized reagents.

Minimum sample requirements

Pre-experiment QC checklist

• Protein concentration confirmed by A280 using calculated extinction coefficients
• SEC chromatogram shows >95% monomer
• DLS indicates monodisperse sample (PdI < 0.2) — see our biophysical characterization services for QC options
• Buffer exchanged ≥2 times into final experimental buffer
• Samples centrifuged at 14,000 × g for 10 min before loading
• Blank buffer prepared identically for reference injections

Concentration determination

The single largest source of error in Kd determination is inaccurate protein concentration. For antibodies, use calculated extinction coefficients based on amino acid sequence rather than Bradford or BCA assays, which vary across antibodies. Confirm concentration by A280 on a UV-Vis spectrophotometer with the same buffer as blank. If the antibody contains post-translational modifications that affect absorbance, consider amino acid analysis for absolute quantification.

ITC vs. Other Antibody Characterization Methods

When to use each method: Use ITC when the complete thermodynamic profile (ΔH, ΔS), binding stoichiometry, or pH-dependent binding mechanisms are required. Our ITC platform is optimized for this application. Use SPR when high-throughput kinetic screening (ka and kd separately) is needed. Use ELISA for relative binding comparison across many samples. Use MST when sample volume is limited.

Complementary binding analysis options include Surface Plasmon Resonance (SPR) and Microscale Thermophoresis (MST).

Decision tree for selecting the appropriate antibody-antigen binding analysis method based on experimental requirements: thermodynamic profile, kinetic screening, throughput, or sample conservation.

Common Mistakes and How to Avoid Them

Designing an ITC Experiment for Antibody-Antigen Binding

When Kd is approximately known

• Set cell concentration to 20–100× expected Kd
• Set syringe concentration to 12–15× cell concentration
• Choose 20 injections of 1.5–2 µL each with 150 s spacing
• Run at 25°C (standard); test 37°C for therapeutically relevant conditions
• Fit with one-site binding model; verify n, Kd, ΔH all converge

When Kd is unknown

• Start with cell: 10–20 µM antibody; syringe: 150–300 µM antigen
• If isotherm is well-defined: proceed to optimized run
• If isotherm is too shallow (near-linear): increase concentrations 2–3×
• If isotherm is too sharp (

pH comparison experiment design

• Prepare two complete buffer sets (pH 7.4 and pH 5.5)
• Perform complete buffer exchange for antibody and antigen into each buffer
• Run pH 7.4 first; clean cell thoroughly; run pH 5.5
• Compare ΔΔG = −RT ln(Kd₂/Kd₁) between conditions
• Decompose ΔΔG into ΔΔH and −TΔΔS to understand the thermodynamic basis

For experimental design support, see our ITC analysis service overview.

FAQ

What is the minimum antibody concentration needed for an ITC experiment?

The minimum concentration depends on the expected Kd. For a 50 nM Kd interaction, the cell should contain 1–5 µM antibody (350–500 µL). For weaker interactions (µM Kd), concentrations in the 10–50 µM range are needed. The key rule is that the cell concentration must be 20–100× the Kd.

Can ITC distinguish between specific and non-specific binding?

Yes. Specific binding produces a saturable, sigmoidal isotherm that fits a defined binding model (typically 1:1). Non-specific binding produces a linear, non-saturating heat signal that continues beyond the expected stoichiometric point.

How much protein do I need to send for an ITC experiment?

For a single Kd measurement at 25°C, provide 400–600 µg of antibody and 200–300 µg of antigen (or the molar equivalent). For pH comparison studies, double these amounts.

Can ITC be performed at non-physiological pH?

Yes. ITC can be performed at pH 5.0–9.0 with most proteins. Below pH 5.0, antibody stability may be compromised. Verify antibody stability at the target pH by DLS or CD before the experiment.

What is the difference between a one-site and two-site binding model?

A one-site model assumes identical, independent binding sites (standard for monoclonal IgG with monovalent antigen). A two-site model assumes two classes of sites with different affinities. Start with a one-site model; only use a two-site model if residuals are clearly non-random.

How do I know if my ITC data are reliable?

Key QC metrics: Chi²

Can ITC detect very weak or very strong binding?

For weak binding (Kd > 100 µM), detectability is limited by max achievable concentration. For very strong binding (Kd

Why does my ITC experiment show no heat signal?

Possible causes: concentrations too low relative to Kd, binding has very low enthalpy (ΔH near zero), the antigen or antibody has aggregated, buffer mismatch obscures the signal, or pH is outside the optimal binding range.

Related Services