Development of a New Generation of Robust pH-Responsive Fluorescent Proteins for Use in Intracellular Imaging Studies

Cancer cells display abnormal alkaline intracellular pH (pH_i). Therapeutic approaches targeting intracellular alkalization therefore hold clinical promise, but these efforts have been hindered by tumor heterogeneity, highlighting an urgent need for tools to accurately map intratumoral pH_i. Furthermore, measuring the pH in subcellular compartments could improve the understanding of how pH is coupled to cancer cell metabolism and cellular processes. Genetically encoded tools, such as pH-responsive fluorescent proteins (FPs), have the potential to deliver these measurements. However, oxidizing intracellular conditions can cause FPs to misfold, aggregate, and become non-fluorescent. Here we describe the design, development, and testing of a new generation of genetically encoded, FP-based biosensors for the high-resolution visualization of intra- and subcellular pH.

We have engineered robust, pH-responsive FPs capable of withstanding oxidizing intracellular conditions. To do so, DNA constructs for GFP-derived proteins were modified to be cysteine free and contain point mutations to promote protein folding, monomericity, and pH sensitivity. Proteins were expressed using inducible E. coli systems and purified via immobilized metal affinity chromatography. Photophysical properties and pH responsiveness were measured by fluorimetry in buffering solutions of incremental pH. Candidates that were bright and responded well to pH changes were assessed for their stability in the oxidizing bacterial periplasm, where fluorescence was evaluated using confocal laser scanning microscopy. We report five new spectrally distinct protein variants, ranging from cyan to yellow in color, that exhibit bright fluorescence, low cytotoxicity, and 5 to 15-fold changes in fluorescence intensity over pH 6.5-8.5 while in the bacterial periplasm.

Moving forward, we plan to generate ratiometric biosensors with the ability to quantify pH_i independently of local sensor concentration by fusing a pH-stable FP to each pH-responsive FP for signal normalization. Ultimately, we hope that cytoplasmic expressed sensors may be used to measure intratumoral pH_i with a high degree of spatial localization, and that organelle targeted sensors may be used to examine cellular rates of glycolysis, oxidative phosphorylation, lipolysis, and other cancer-related metabolic pathways.