Purification and Characterization of Clover‐Ruby2 as a FRET Substrate for Protease Assays

Abstract

The use of fluorescent proteins (FPs) in biomedical research has significantly expanded our ability to monitor biological processes in real‐time. These processes include protein expression, localization, cellular trafficking, ligand binding, and other biological events where development of imaging methodology is still underway. While the green fluorescent protein (GFP) was the first molecule in this class to be discovered, there is currently a wide range of FPs available that vary in their spectroscopic, physical, and chemical properties. Förster (fluorescence) resonance energy transfer (FRET) is one area of biological imaging that relies on a donor FP and acceptor FP to facilitate the transfer of energy. While the cyan and yellow FPs have been more commonly used as a FRET pair, green and red FPs offer some advantages for live cell FRET imaging, such as lower phototoxicity, lower autofluorescence, and greater spectral separation. We have investigated the Clover‐Ruby2 protein as a substrate FRET pair for monitoring protease activity in both in vitro and in vivo applications. This is aligned with the goal of developing an in vitro model system to mimic protease events occurring in live cells. The Clover (C) and Ruby2 (R2) proteins have been expressed and purified from E. coli, both as individual proteins (C and R2), and as a tandem FRET pair (CR2) linked through a protease recognition region. To obtain optimal samples for these studies, purification strategies have been optimized using immobilized‐metal‐affinity‐, ion‐exchange‐, hydrophobic‐, and gel filtration‐chromatography. Characterization of the CR2 fusion protein using UV‐Vis absorbance and fluorescence spectroscopy at various steps during the purification procedure have provided results demonstrating the quality of FRET signal being dependent upon sample purity, salt concentration, and other factors that are still being investigated. Maturation of R2 was found to be most critical for obtaining quality dynamic FRET measurements, and it appears the level of R2 maturation achieved is also influenced by its chemical environment. Additionally, R2 was found to be become much more hydrophobic following the maturation process, as observed during hydrophobic chromatography (HC) separation where the non‐mature R2 showed very low affinity for HC media, and matured R2 had much greater affinity. Finally, gel filtration experiments demonstrated a significant degree of interaction of CR2 with free‐C and/or free–R2. This raised some concerns about using these proteins for FRET assays where the dynamic properties of the FRET signal is used to determine kinetic rates. If C and R2 interactions show an increase in FRET simply due to hydrophobicity, and non specific interactions, the FRET signal will be artificially increased, and the C and R2 pair may not dissociate in real‐time following the protease reaction. To overcome these limitations, we have developed a number of calibration procedures, obtained under different chemical conditions, to monitor the assays. This allowed normalization of the assays in different reaction environments, which is critical when working with proteases that have optimal activity under varied chemical reaction conditions.Support or Funding InformationNIH R15 GM080691 02