achieves maximal inhibition but avoids off-target cellular toxicity following 1–2 h incubation with cells, e.g., 10 μM. If such
a tool compound is not available for a target, compounds can
still be profiled by CETSA to look for novel binders, but the
results should be treated with caution in the absence of a
control that demonstrates exactly what change in thermal stability is induced upon intracellular binding.
5. To expedite the addition of protein to the PCR plate, prepare
protein stock solutions in a deep 384-well polypropylene
microplate. This allows the use of a multichannel pipette to
add 10 μL of the protein solution to the PCR plate.
6. Between runs, the temperature of the block in the LightCycler
will be higher than 20
C, so precooling the instrument can
increase reproducibility for proteins that melt at temperatures
<35
C.
7. Choose the pair of emission/excitation filters that matches the
SYPRO
® Orange maxima best. For the LightCycler 480 II, this
would be 465/480 nm. It may be necessary to optimize the
integration time and/or acquisitions per
C to achieve the
desired ramp rate. The ramp rate need not be 1
C/min
(we have successfully used up to 2
C/min to reduce acquisition time), but it is essential that all experiments being compared use identical parameters.
8. The optimal protein/dye concentration pair is target dependent. If the protein has a low melting temperature (<35
C), it
may be helpful to keep the plate cool as much as possible during
its preparation. If the melt curve is poor under all tested concentrations of protein and dye, it may be necessary to consider
different buffers for the DSF experiment or investigate alternative protein constructs/sources.
9. The melting temperature is defined as the midpoint of the
protein melt curve. The simplest way to determine this is
using the T m Calling function in the LightCycler software,
which will calculate and plot the negative of the first derivative
of the melt curve. The minimum of this curve is the melting
temperature. For large numbers of compounds, we use the
TSA package in the Genedata
® Screener software, but there
are alternative solutions that are freely available [3, 12].
10. If required, it is possible to screen several cell lines at this stage,
especially where expression levels of the target protein are
unknown. It is also possible to perform this step using purified
recombinant protein at 2 nM–5 nM, though any antibody pairs
identified will later need to be confirmed against cell lysates.
11. In most cases, we do not find it necessary to add protease
inhibitors to the SureFire Lysis Buffer, even for extended incubations of cell lysates up to 48 h.
Quantitative Detection of Ligand Binding by DSF and CETSA HT
213
a tool compound is not available for a target, compounds can
still be profiled by CETSA to look for novel binders, but the
results should be treated with caution in the absence of a
control that demonstrates exactly what change in thermal stability is induced upon intracellular binding.
5. To expedite the addition of protein to the PCR plate, prepare
protein stock solutions in a deep 384-well polypropylene
microplate. This allows the use of a multichannel pipette to
add 10 μL of the protein solution to the PCR plate.
6. Between runs, the temperature of the block in the LightCycler
will be higher than 20
C, so precooling the instrument can
increase reproducibility for proteins that melt at temperatures
<35
C.
7. Choose the pair of emission/excitation filters that matches the
SYPRO
® Orange maxima best. For the LightCycler 480 II, this
would be 465/480 nm. It may be necessary to optimize the
integration time and/or acquisitions per
C to achieve the
desired ramp rate. The ramp rate need not be 1
C/min
(we have successfully used up to 2
C/min to reduce acquisition time), but it is essential that all experiments being compared use identical parameters.
8. The optimal protein/dye concentration pair is target dependent. If the protein has a low melting temperature (<35
C), it
may be helpful to keep the plate cool as much as possible during
its preparation. If the melt curve is poor under all tested concentrations of protein and dye, it may be necessary to consider
different buffers for the DSF experiment or investigate alternative protein constructs/sources.
9. The melting temperature is defined as the midpoint of the
protein melt curve. The simplest way to determine this is
using the T m Calling function in the LightCycler software,
which will calculate and plot the negative of the first derivative
of the melt curve. The minimum of this curve is the melting
temperature. For large numbers of compounds, we use the
TSA package in the Genedata
® Screener software, but there
are alternative solutions that are freely available [3, 12].
10. If required, it is possible to screen several cell lines at this stage,
especially where expression levels of the target protein are
unknown. It is also possible to perform this step using purified
recombinant protein at 2 nM–5 nM, though any antibody pairs
identified will later need to be confirmed against cell lysates.
11. In most cases, we do not find it necessary to add protease
inhibitors to the SureFire Lysis Buffer, even for extended incubations of cell lysates up to 48 h.
Quantitative Detection of Ligand Binding by DSF and CETSA HT
213