Temperature of cold-denatuartion

°C (Celsius)

Differential Scanning Calorimetry (DSC)

50 mM potassium phosphate


100 μM

The temperature at which the protein denatures upon cooling (the lower temperature at which ΔG=0).
See ΔG(T*) assay for how ΔG is calculated.

ΔG(T*) assay details:
ΔG(T*) is Gibbs free energy at maximum stability (T* is temperature at which maximal ΔG value is reached).

Calorimetric data was fit in two stages to obtain Tm, ΔHm, and ΔCp (see Tm assay for details). From this modeled data, other parameters were derived.
The modified Gibbs-Helmotz equation (Becktel and Schellman, 1987) was used to calculate the temperature profiles of ΔGunfolding for each variant:

ΔG(T) = ΔH(Tm) (1-T/Tm) + ΔCp [T - Tm - T In(T/Tm)]

Tm assay details:
DSC (Nano DSC; TA Instruments) was used to assess the variants’ thermodynamic folding parameters: transition midpoint temperature or Tm, transition ΔHm, and ΔCp. The DSC experiments were performed with 100 μM AdK in 50 mM potassium phosphate (pH 7.5). Prior to measurement, samples were incubated for 8–12 hr at 5°C under mild reducing conditions with 100 mM dithiothreitol (DTT) to inhibit disulfide bond formation and better maintain a two-state unfolding mechanism. Samples were then degassed under reduced pressure for 15–30 min. Heating scans were performed under 3 atm pressure from 20°C–90°C at 1°C/min following a 2,200 s equilibration period after the cells were loaded. A cooling scan back to 20°C confirmed the reversibility of thermal denaturation for each variant. DSC experiments were minimally performed in triplicate for statistical averaging.
† Modeling of the calorimetric data was performed with commercially available software (NanoAnalyze v2.3.2, TA Instruments). From each protein scan, appropriate buffer scans were subtracted from the protein data. Data fits were performed in two stages to allow accurate fitting of the thermal parameters. First, datasets were corrected with a sigmoidal baseline correction to produce a level pre- and post-transition. The observed transition was fit to a simple two-state scaled model. The two-state scaled fit is used to provide a estimate of Tm and ΔHm and correction to the molar heat capacity. The second fitting is performed with the sigmoidal baseline correction replaced by a combination baseline to remove any slope observed pre- and post-transition. Fitting is made using a general model, where Tm, ΔHm and ΔCp are discretely parameterized. Each of the triplicate data were fit independently. Reported values indicate the average value from the three replicate measurements, reported error indicate the standard deviation of the three datasets. From the modeled data, several parameters were derived that were key to elucidating thermodynamic trends among the CPD variants.