A popular thermal-stability assay developed especially for the study of membrane proteins uses a thiol-specific probe, 7-diethylamino-3-(4-maleimidophenyl)-4-methylcoumarin (CPM). The fluorescence emission of CPM surges when it forms a covalent bond with the side chain of a free Cys, which becomes more readily accessible upon protein thermal denaturation. Interestingly, the melting temperatures of membrane proteins determined using the CPM assay in literature are closely clustered in the temperature range 45-55 °C. A thorough understanding of the mechanism behind the observed signal change is critical for the accurate interpretation of the protein unfolding. Here we used two α-helical membrane proteins, AqpZ and AcrB, as model systems to investigate the nature of the fluorescence surge in the CPM assay. We found that the transition temperatures measured using circular-dichroism (CD) spectroscopy and the CPM assay were significantly different. To eliminate potential artifact that might arise from the presence of detergent, we monitored the unfolding of two soluble proteins. We found that, contrary to current understanding, the presence of a sulfhydryl group was not a prerequisite for the CPM thermal-stability assay. The observed fluorescence increase is probably caused by binding of the fluorophore to hydrophobic patches exposed upon protein unfolding.