Genetically Encoded Red and Far-red Fluorescent Indicators for Redox Signaling

Reactive oxygen species (ROS) are typical examples of “the dose makes the poison”. When they are generated under physiological concentrations, they play critical roles in various biological functions, such as cell differentiation, tissue regeneration and inflammation response. On the other hand, when ROS were overproduced, they can cause damage to biomolecules, including DNA, protein and lipids and lead to many diseases’ initiation and progression. Therefore, monitoring the ROS signaling in biological system are essential for studying redox biology to further tackling the mechanism of different diseases.
Genetically encoded redox indicators (GERIs) are emerging as powerful tools for redox signaling detection in living systems. GERIs can measure real-time signals in the natural cellular context with minimal toxicity. Besides, by attaching specific localization sequences to the N- or C-terminus, GERIs can be readily localized to different subcellular compartments with super spatial and temporal resolution. In addition, GERIs present as plasmids or viral vectors, which can be easily adapted to various biological study systems and shared across different labs.
Red and far-red fluorescent proteins (FPs) have several advantages. First, compared to shorter wavelength emitters (blue FP/green FP), they are less cytotoxicity to the cells. Second, the red and far-red spectrum are within the “optical window” for maximal cell and tissue penetration, leading to reduced sample scattering and autofluorescence. Third, they provide additional channel in combination with blue/green FP for multicolor and multiparameter imaging.
In this thesis, I focus on developing red and far-red GERIs for detection of different redox molecules. Based on our previous work—TrxRFP1, I performed directed evolution and enzymatic screening to develop a second generation thioredoxin biosensor, TrxRFP2. I also used similar strategy to fuse Trx2 enzyme with rxRFP1.1 to build a biosensor, MtrxRFP2, for detecting thioredoxin signaling in mitochondria compartment. I tested both TrxRFP2 and MtrxRFP2 in vitro and in different subcellular domain of cultured mammalian cell. This work expanded the collection of GERIs for thioredoxin signaling detection.
Another strategy was employed to develop a red GERI for peroxynitrite (ONOO-). Using a genetic code expansion technology, p-boronophenylalanine (pBoF) was introduced to a site close to the chromophore of a circularly permuted red FP scaffold (cpRFP). I performed thorough in vitro characterizations of this indicator. I also assessed its performance in mammalian cells in response to chemically and physiologically-related stimulations. X-ray crystallography and NMR methods were applied to investigate the response mechanism of this indicator.
Hydrogen peroxide (H2O2) play critical roles in many cellular processes and imaging the H2O2 signaling is always important for understanding the redox metabolism in biological system. I created a red GERIs for H2O2 detection based on the brightest RFP mScarlet-I. I firstly generated a circularly permutated FP—cpmScarlet. Next, I fused two H2O2 sensory domain OxyR at the N- and C- terminus of cpmScarlet to build up the H2O2 indicator. I utilized protein engineering and directed evolution to improve the indicator brightness and responsiveness. I finally identified a candidate showing good response to H2O2 with minimal photoactivation. This candidate was termed SHRIMP (scarlet-based hydrogen peroxide redox indicator with minimal photoconversion). I tested SHRIMP performance in HEK 293T cells and macrophages in response to chemically and physiologically-related stimulation. SHRIMP was also applied to isolated mouse islets to detect H2O2 generation with chemical stimulation. Last, SHRIMP was combined with green calcium indicator for multiparameter imaging in mammalian cells.
I also further expanded the GERIs spectrum to far-red region by creating a redox-active far-red fluorescent protein -- rxcpmMaroon1. I first fused zinc hook domain which contains active cysteines with a circularly permutated far-red fluorescent protein (cpmMaroon185). Next, I performed further site-specific and random mutagenesis to improve the protein expression and redox responsiveness of the variant. Last, I characterized rxcpmMaroon1 in vitro as well as in cultured mammalian cells to test its performance in response to redox stimulation.