Assessment of membrane protein expression and stability using a split green fluorescent protein reporter

Membrane proteins are challenging targets for structural biologists. Finding optimal candidates for such studies requires extensive and laborious screening of protein expression and/or stability in detergent. The use of green fluorescent protein (GFP) as a reporter has enormously facilitated these studies; however, its 238 residues can potentially alter the intrinsic properties of the target (e.g., expression or stability). With the aim of minimizing undesired effects of full-length GFP, here we describe the utility of a split GFP reporter during precrystallization studies of membrane proteins. GFP fluorescence appeared by complementation of the first 15 residues of GFP (GFP11) (fused to the C terminus of a membrane protein target) with the remaining nonfluorescent GFP (GFP1–10). The signal obtained after sequential expression of SteT (l-serine/l-threonine exchanger of Bacillus subtilis) fused to GFP11 followed by GFP1–10 specifically measured the protein fraction inserted into the Escherichia coli cytoplasmic membrane, thereby discarding protein aggregates confined as inclusion bodies. Furthermore, in vitro complementation of purified SteT–GFP11 with purified GFP1–10 was exploited to rapidly assess the stability of wild-type and G294V mutant versions of SteT–GFP11 following detergent solubilization and purification. This method can be applied in a medium- to high-throughput manner with multiple samples.