Science 346, 1255784 (2014)

Credit: SCIENCE

By measuring the specific temperature required to dissociate an interaction between a ligand and its target, the thermal shift assay has been used to detect and quantify drug-target engagement in cells. However, the application of this approach to provide unbiased identification of cellular targets of a drug remained a challenge. To address this, Savitski et al. combined the thermal shift assay with high-resolution quantitative MS to determine the effects of a drug on the thermal profile of the cellular proteome. The authors added either vehicle or drug to cells and heated lysates or intact cells across a range of ten different temperature points. Proteins at a particular melting temperature begin to denature and aggregate. These samples were labeled with a different TMT10 isotope tag for LC/MS/MS analysis to measure the fraction of soluble proteins at each temperature, producing a characteristic melting curve for each protein. Ligand binding to a protein creates a thermal shift in the curve, increasing the overall Tm for the protein. The authors tested whether their system could detect new off-targets of clinical kinase inhibitors such as vemurafenib and alectinib. Interestingly, both compounds targeted ferrochelatase (FECH), an enzyme in the heme biosynthesis pathway. FECH deficiency is associated with protoporphyria, a painful photosensitivity disorder. As vemurafenib and alectinib treatment are also known to cause phototoxicity, these findings suggest that the off-target inhibition of FECH may explain these side effects. Taken together, this technique offers an approach to provide proteome-wide identification of potential off-targets of drugs.