Thermofluor Data Analysis

This section provides an explanation about how to assess the raw data from a DSF experiment. First you will need to make a DSF test booking. This test requires (at least) 40 μL of protein at a concentration of 1 mg/mL or more.  We also require 1mL of the protein buffer (more correctly called the formulation – ideally the gel-filtration solution, or flow-through from a concentration step).

To access your DSF results:

  1. Download a copy of the BioRad CFX Software – a link to a .zip folder which contains the program is found at the bottom of every DSF booking.
  2. Once you’ve installed this, you will be able to open and view the .pcrd file attached to the completed booking.
  3. We recommend that you SAVE the data file to a safe place on your own PC.

In our standard DSF experiment (buffer screen 10 – conditions available on C6), we dilute your protein sample into a number of different buffers – testing the stability of the protein at different pH and with two levels of salt.  We do the melts in triplicate, as DSF is a low information technique, and the redundancy helps us to trust the results that we observe.  The primary number obtained from DSF is the temperature of the midpoint of the melt transition, Tm (correctly referred to as the temperature of hydrophobic exposure, Th).  However, the shape of the curve is telling as well.  The ideal curve is flat before the protein melts, then has a smooth, sharp sigmoidal melt transition.

Overview of a DSF experiment – left hand panel shows the melt curves, right hand panel shows the first derivative of the melt curves.  In this case we have coloured the lysozyme control curves blue, and the sample control curves are red.  There are some curves where the Tm is around 40°C, but most of the curves show a Tm of between 50 and 60°C

What should you look for:

A – Controls

  1. Process control (Lysozyme) in A1, A2, A3
  2. Protein only in A10, A11, A12
  3. Sample control (your protein in the original formulation) in H1, H2, H3
  4. Dye only in H10, H11, H12

B – Look to see if the melt curve of the protein in original buffer is well shaped.  If it is, look for the Tm – (which is the minimum of the first derivative curve)

C – Look to see if any of the other buffers/pH/Salt curves either have a ‘nicer’ curve shape (that means flatter at the start of the melting), or if they increase the Tm of the protein.

Controls in the DSF experiment – the blue curves are the lysozyme control – these should show a Tm of around 70-71°C.  The two flatter curves are the dye only and protein only controls.

Sample control in the DSF experiment – this is the protein in its original formulation.  This control is always found in wells H1, H2 and H3.  This curve shows a well shaped melt curve – the pre-transition is flat, there is a sharp, smooth melt then with the characteristic protein quench post-transition.   The Tm here is about 59°C.

Examples of ugly DSF curves – no melt transition to be seen.  We interpret this to mean that the protein is not well folded even at room temperature, so what you see is essentially just post-unfolding quenching.