c-value could be achieved by lowering the cell protein concentration. However, the total amount of ligand that can be bound and
therefore the total amount of heat that can be released or absorbed
during the titration is determined by this cell concentration. Therefore, there is a lower limit to the protein concentration defined by
both the smallest amount of heat detectable in a single injection
and the smallest number of injections required to define a binding
curve. As a consequence, ITC is currently able to provide optimal
characterization of binding interactions using a single standard
experiment where the K d is between 10’s of nM and 100’s of μM.
3.4 How Do
the Concentrations
Influence the Results?
Determining the concentration of protein and ligand in an ITC
experiment is essential for accurate quantitation of the binding.
UV–VIS absorbance spectroscopy is the easiest option (see Notes
9 and 10) while amino acid analysis can be used for proteins or
peptides lacking aromatic residues. These methods should be able
to reproducibly quantify concentration within a few percent.
Weighing lyophilized solid and/or colorimetric methods are also
options but are potentially much less accurate.
The effects of concentration errors on ITC data have been
examined in detail elsewhere [3, 4] but as an exercise in underlining
its importance it is informative to refit the test reaction data using
deliberately inaccurate values mimicking a Æ20% error in concentration as seen in Fig. 4.
It is evident that errors in the syringe concentration translate
into equivalent percentage errors in all of the directly fitted parameters (although not in ΔG because of the logarithmic relation to
K d ). Interestingly, the cell concentration only affects the value of
stoichiometry n, while K d and ΔH overlay and are unaltered. Since
the concentration of material used in the syringe for ITC titrations
is often quite high it may not be possible to measure directly its UV
absorbance. In this case a dilution should be prepared so that
absorbance is in a good range for measurement and this dilution
should be repeated a few times for independent measurement and
averaging.
Since both the cell and the syringe concentrations affect the
determination of stoichiometry, this value will potentially have an
error of at least 5% or more from the combined measurements.
These errors can compensate if one is an overestimate while the
other underestimates concentration, but stoichiometries within the
generous range 0.9–1.1 should probably be considered as consistent with standard 1:1 binding with n ¼ 1. Reflecting this it is also
possible to fit ITC data with a fixed stoichiometry of binding of
1 but with the concentration of either cell or syringe component as
a variable.
There can also be errors in concentration that are not the result
of inaccuracy of measurement. For example, the “active” or “binding competent” concentration of a sample can be lower than the
Isothermal Titration Calorimetry
145
therefore the total amount of heat that can be released or absorbed
during the titration is determined by this cell concentration. Therefore, there is a lower limit to the protein concentration defined by
both the smallest amount of heat detectable in a single injection
and the smallest number of injections required to define a binding
curve. As a consequence, ITC is currently able to provide optimal
characterization of binding interactions using a single standard
experiment where the K d is between 10’s of nM and 100’s of μM.
3.4 How Do
the Concentrations
Influence the Results?
Determining the concentration of protein and ligand in an ITC
experiment is essential for accurate quantitation of the binding.
UV–VIS absorbance spectroscopy is the easiest option (see Notes
9 and 10) while amino acid analysis can be used for proteins or
peptides lacking aromatic residues. These methods should be able
to reproducibly quantify concentration within a few percent.
Weighing lyophilized solid and/or colorimetric methods are also
options but are potentially much less accurate.
The effects of concentration errors on ITC data have been
examined in detail elsewhere [3, 4] but as an exercise in underlining
its importance it is informative to refit the test reaction data using
deliberately inaccurate values mimicking a Æ20% error in concentration as seen in Fig. 4.
It is evident that errors in the syringe concentration translate
into equivalent percentage errors in all of the directly fitted parameters (although not in ΔG because of the logarithmic relation to
K d ). Interestingly, the cell concentration only affects the value of
stoichiometry n, while K d and ΔH overlay and are unaltered. Since
the concentration of material used in the syringe for ITC titrations
is often quite high it may not be possible to measure directly its UV
absorbance. In this case a dilution should be prepared so that
absorbance is in a good range for measurement and this dilution
should be repeated a few times for independent measurement and
averaging.
Since both the cell and the syringe concentrations affect the
determination of stoichiometry, this value will potentially have an
error of at least 5% or more from the combined measurements.
These errors can compensate if one is an overestimate while the
other underestimates concentration, but stoichiometries within the
generous range 0.9–1.1 should probably be considered as consistent with standard 1:1 binding with n ¼ 1. Reflecting this it is also
possible to fit ITC data with a fixed stoichiometry of binding of
1 but with the concentration of either cell or syringe component as
a variable.
There can also be errors in concentration that are not the result
of inaccuracy of measurement. For example, the “active” or “binding competent” concentration of a sample can be lower than the
Isothermal Titration Calorimetry
145