High-Throughput Protein Interaction and Stability Screening

May 15, 2015 (Vol. 35, No. 10)

A New, Rapid Solution for 1,536-Well Thermal Shift Assays

From small molecule and fragment-library screening to antibody-antigen target specificity, ligand screening is an increasingly important part of the drug discovery process. Ligand screening allows researchers to determine protein binding affinities (and efficiencies) early in the compound screening process, enabling them to take only the most promising proteins through to further study.

In addition, protein stability screening (i.e., systematically determining conditions that promote the stability of the protein of interest) is imperative for profiling favorable crystallization conditions, developing storage buffer formulations, and protein quality control. 

Differential Scanning Fluorimetry

Historically, protein interaction and stability have been measured using a variety of techniques, including nuclear magnetic resonance (NMR), differential scanning calorimetry (DSC), and surface plasma resonance imaging (SPRI). This tutorial describes the use of differential scanning fluorimetry (DSF), also called thermal shift assay, for protein stability and ligand screening. It also highlights the application of DSF in high-throughput screening on the Roche LightCycler^® 1536 real-time PCR system.

DSF is a thermal-denaturation assay that measures the thermal stability of a target protein and the subsequent change in the protein’s melting temperature (T_m) upon the binding of a ligand to that protein. The thermal stability change is measured by generating a thermal-denaturation curve in the presence of a hydrophobic fluorescent dye, such as Sypro Orange. As the protein unfolds, the hydrophobic pockets of the protein are exposed, and the fluorescent dye binds nonspecifically to the exposed surfaces, resulting in an increase in fluorescence.

In addition to thermal-denaturation assays, the LightCycler 1536 system can be used to perform isothermal assays for the optimization of protein buffers and storage conditions. Of all available techniques, benefits of performing DSF on the LightCycler 1536 system for protein stability and interaction screening are twofold:

1. Speed: Assays can yield T_m data within 3 minutes.
2. Small Amount of Protein Required: Assays can be successfully performed using as little as 5 pmols of purified protein per well.

We believe this new application on the LightCycler 1536 system has relevance for biotech and pharmaceutical firms, contract research organizations (CRO), and large academic research groups currently screening proteins via other techniques.

Figure 1. Correlation plot shows the strength of consistency of T_m data generated using LightCycler 480 and LightCycler 1536 instruments to assess protein melting in a low temperature range. The plot shows 8 degrees of linearity and a correlation R² value of 0.99 in the T_m values.

Experimental Data

ased on publicly available information pertaining to differential scanning fluorimetry, we concluded that the DSF data are highly discordant and nonreproducible in low (between 50 and 60°C) and high temperature ranges as high as 75°C. In light of this information, we designed experiments that were run in parallel on LightCycler 480 and LightCycler 1536 real-time PCR instruments with the goal of demonstrating the ability of the two platforms to consistently capture protein melting changes within these—historically more difficult—temperature ranges.  A ligand concentration gradient was generated to enable the determination of slight variations in T_m for two proteins (unnamed here) known to have T_m values in low and high temperature ranges (Figures 1 and 2).

The adoption of high-throughput differential scanning fluorimetry (Thermal Shift Assays) with the LightCycler 1536 system enables significantly higher throughput when screening for ligand binding and protein stability, while reducing the amount of protein required. It may also significantly reduce overall cost to run the screens. Our experimental data show that the LightCycler 480 and LightCycler 1536 platforms exhibit an extremely high level of concordance, allowing researchers to scale assays from the LightCycler 480 instrument to the LightCycler 1536 instrument with confidence.

Figure 2. Correlation plot shows the strength of consistency of T_m data generated using LightCycler 480 and LightCycler 1536 instruments to assess a protein melting in a high temperature range. The plot shows 5 degrees of linearity and correlation R² value of 0.99 in T_m values.

Rama Badugu, Ph.D. ([email protected]), is application support consultant and David Lahm ([email protected]) is marketing manager, sequencing & genomics at Roche Life Science, Roche Diagnostics.

All Roche products mentioned are for life science research only. Not for use in diagnostic procedures. LIGHTCYCLER is a trademark of Roche. 581-61226-0515.