established in the control experiments described above, a linear
term may be added to the exponential equation used to analyze
the transient, to account for this effect.
It is common to observe artifacts resulting from (bad) mixing
in the early parts of the trace. This is another reason to be careful
when interpreting rate constants in the range 300–400 s
À1 or
higher, in particular if the rate constants do not increase linearly
with increasing protein concentration under pseudo-first-order
conditions. Thus, a hyperbolic dependence of k obs as expected
from a three-state binding model (Schemes 2 and 3) may be due
to an instrumental mixing limitation rather than any conformational change in the protein(s). While such artifact-based changes
do not appear perfectly hyperbolic, they can be hard to distinguish
from a genuine three-state binding. Decreasing the temperature or
adding a cosolvent might reduce k obs values into a range where data
is reliable. Alternatively, the hyperbolic kinetics can be confirmed by
another kinetic method such as temperature jump that does not use
mixing as the source of equilibrium perturbation.
We have also noted an unusual artifact on a long timescale with
a “rate constant” approximately 0.2 s
À1
, i.e., in a range that could
interfere with slow conformational changes in kinetic experiments.
This artifact is not related to the more common photobleaching
effect. Interestingly, the amplitude changes with temperature and
reverses sign around room temperature while k obs for the phase
remains unaltered. The basis for the artifact is possibly related to
small differences in temperature between flow lines and the
observation cell.
3.6 Example Studies
and What They Tell Us
About IDP Binding
3.6.1 Case Study 1:
p53TAD and MDM2,
an Apparent One-Step
Binding
p53 is a central transcription factor in cell cycle regulation. The
transactivation domain of p53 (p53TAD) contains a conserved Trp
residue, which is directly involved in binding to the TAD-binding
domain of MDM2, a negative regulator of p53. p53TAD is intrinsically disordered in the free state but forms an α-helix upon binding to MDM2. Stopped-flow kinetic traces were obtained using a
fixed concentration of MDM2 (1 μM) and a range of p53TAD
concentrations (2–10 μM) and were well described by a single
exponential fit (Fig. 2).
Observed rate constants were fitted using Eq. 2 (pseudo-first
order) and Eq. 3 (second order) to obtain a slope of around
8 μM
À1 s
À1
, which represents k on for the binding reaction
(Fig. 2d). However, the extrapolated k off values from these fits
were much lower than any measured k obs value, were close to zero
and thus associated with a moderate error. Therefore, k off was also
measured independently using a displacement experiment. A complex formed by 1 μM MDM2 and 1 μM p53TAD was mixed with
an excess (20 μM) of dansylated p53TAD peptide. Fitting this
kinetic trace to a single exponential gave a k obs of 0.63 s
À1 , which
118
Elin Karlsson and Per Jemth
term may be added to the exponential equation used to analyze
the transient, to account for this effect.
It is common to observe artifacts resulting from (bad) mixing
in the early parts of the trace. This is another reason to be careful
when interpreting rate constants in the range 300–400 s
À1 or
higher, in particular if the rate constants do not increase linearly
with increasing protein concentration under pseudo-first-order
conditions. Thus, a hyperbolic dependence of k obs as expected
from a three-state binding model (Schemes 2 and 3) may be due
to an instrumental mixing limitation rather than any conformational change in the protein(s). While such artifact-based changes
do not appear perfectly hyperbolic, they can be hard to distinguish
from a genuine three-state binding. Decreasing the temperature or
adding a cosolvent might reduce k obs values into a range where data
is reliable. Alternatively, the hyperbolic kinetics can be confirmed by
another kinetic method such as temperature jump that does not use
mixing as the source of equilibrium perturbation.
We have also noted an unusual artifact on a long timescale with
a “rate constant” approximately 0.2 s
À1
, i.e., in a range that could
interfere with slow conformational changes in kinetic experiments.
This artifact is not related to the more common photobleaching
effect. Interestingly, the amplitude changes with temperature and
reverses sign around room temperature while k obs for the phase
remains unaltered. The basis for the artifact is possibly related to
small differences in temperature between flow lines and the
observation cell.
3.6 Example Studies
and What They Tell Us
About IDP Binding
3.6.1 Case Study 1:
p53TAD and MDM2,
an Apparent One-Step
Binding
p53 is a central transcription factor in cell cycle regulation. The
transactivation domain of p53 (p53TAD) contains a conserved Trp
residue, which is directly involved in binding to the TAD-binding
domain of MDM2, a negative regulator of p53. p53TAD is intrinsically disordered in the free state but forms an α-helix upon binding to MDM2. Stopped-flow kinetic traces were obtained using a
fixed concentration of MDM2 (1 μM) and a range of p53TAD
concentrations (2–10 μM) and were well described by a single
exponential fit (Fig. 2).
Observed rate constants were fitted using Eq. 2 (pseudo-first
order) and Eq. 3 (second order) to obtain a slope of around
8 μM
À1 s
À1
, which represents k on for the binding reaction
(Fig. 2d). However, the extrapolated k off values from these fits
were much lower than any measured k obs value, were close to zero
and thus associated with a moderate error. Therefore, k off was also
measured independently using a displacement experiment. A complex formed by 1 μM MDM2 and 1 μM p53TAD was mixed with
an excess (20 μM) of dansylated p53TAD peptide. Fitting this
kinetic trace to a single exponential gave a k obs of 0.63 s
À1 , which
118
Elin Karlsson and Per Jemth