15. Cool the sample to the starting temperature at 1
C/min,
recording the CD signal at the selected wavelength at closely
spaced intervals.
16. Record a full spectrum of the sample.
17. Wait for 30 min and record a full spectrum of the sample again.
18. Switch off the lamp and the instrument, and purge with N 2
according to the manufacturer’s instruction (usually 15–20
min).
19. Switch off N 2 flux.
Compare the high-temperature spectrum to that expected for
random coil and to the low-temperature spectrum to confirm that
the protein did fully unfold. Compare the initial and final two
low-temperature spectra. Does the sample fully refold? If so, this
will allow a thermodynamic analysis of the unfolding data [19]. The
refolding rate may differ substantially for each protein. If the two
low-temperature spectra differ, and the latter is closer to the initial
spectrum, then this implies that refolding is slow and future experiments on the protein should use slower cooling.
Calculate the difference spectrum of the spectra corresponding
to the low- and high-temperature limits. What was the wavelength
with the greatest difference in the signal? If this maximum difference is much larger than at the selected wavelength for recording
the CD signal, data may be improved by repeating the experiment
and selecting the maximum difference wavelength.
Inspect the plot of the CD signal at the selected wavelength
against temperature. Typically, there will be a single sharp transition, the midpoint of which corresponds to (or at least approximates) the T m . Where there is a more complex structure to the plot
(e.g., multiple transitions that may correspond to different
structured regions within the protein), this implies that less structurally resolved methods (e.g., DSF) will give results that cannot be
automatically analyzed for T m . In general, CD can be used to
provide a check on the T m derived by other methods and to
compare T m values of different batches or different protein variants.
Over what temperature range does the unfolding transition
occur? Knowledge of this range can be used to optimize future
thermostability experiments on this protein (e.g., is it possible to
monitor unfolding over smaller range and thus carry out the experiment more quickly and reduce potential for aggregation?).
Information About
the Tertiary Structure
Near-UV CD spectra show signals arising from aromatic residues,
and disulfide bonds are extremely sensitive to their environment.
The presence of defined signals in the near-UV region indicates that
the protein is folded, whereas their absence suggests a nondefined
tertiary structure (e.g., the protein is unfolded, misfolded, or in a
molten globule state). The protocols for far-UV spectra can be
Protein Quality Control
35
C/min,
recording the CD signal at the selected wavelength at closely
spaced intervals.
16. Record a full spectrum of the sample.
17. Wait for 30 min and record a full spectrum of the sample again.
18. Switch off the lamp and the instrument, and purge with N 2
according to the manufacturer’s instruction (usually 15–20
min).
19. Switch off N 2 flux.
Compare the high-temperature spectrum to that expected for
random coil and to the low-temperature spectrum to confirm that
the protein did fully unfold. Compare the initial and final two
low-temperature spectra. Does the sample fully refold? If so, this
will allow a thermodynamic analysis of the unfolding data [19]. The
refolding rate may differ substantially for each protein. If the two
low-temperature spectra differ, and the latter is closer to the initial
spectrum, then this implies that refolding is slow and future experiments on the protein should use slower cooling.
Calculate the difference spectrum of the spectra corresponding
to the low- and high-temperature limits. What was the wavelength
with the greatest difference in the signal? If this maximum difference is much larger than at the selected wavelength for recording
the CD signal, data may be improved by repeating the experiment
and selecting the maximum difference wavelength.
Inspect the plot of the CD signal at the selected wavelength
against temperature. Typically, there will be a single sharp transition, the midpoint of which corresponds to (or at least approximates) the T m . Where there is a more complex structure to the plot
(e.g., multiple transitions that may correspond to different
structured regions within the protein), this implies that less structurally resolved methods (e.g., DSF) will give results that cannot be
automatically analyzed for T m . In general, CD can be used to
provide a check on the T m derived by other methods and to
compare T m values of different batches or different protein variants.
Over what temperature range does the unfolding transition
occur? Knowledge of this range can be used to optimize future
thermostability experiments on this protein (e.g., is it possible to
monitor unfolding over smaller range and thus carry out the experiment more quickly and reduce potential for aggregation?).
Information About
the Tertiary Structure
Near-UV CD spectra show signals arising from aromatic residues,
and disulfide bonds are extremely sensitive to their environment.
The presence of defined signals in the near-UV region indicates that
the protein is folded, whereas their absence suggests a nondefined
tertiary structure (e.g., the protein is unfolded, misfolded, or in a
molten globule state). The protocols for far-UV spectra can be
Protein Quality Control
35