Non-invasive monitoring of deep-tissue developmental, metabolic, and pathogenic processes will advance modern biology. Imaging of live mammals using fluorescent probes is more feasible within the near-infrared (NIR) transparency window (NIRW: 650-900 nm) where hemoglobin and melanin absorbance significantly decreases and water absorbance is still low. Chromophores in genetically-encoded probes can be formed either autocatalytically from amino acids, as in a case of green fluorescent protein (GFP)-like proteins, or be bound to apoproteins. The most red-shifted fluorescent proteins (FPs) of the GFP-like family have excitation and emission spectra completely or partially outside of the NIRW and suffer from low brightness and modest photostability. Natural bacterial phytochrome photoreceptors (BphPs) utilize low molecular weight biliverdin as a chromophore. BphPs binding biliverdin provide many advantages over other chromophore containing proteins. First, unlike the chromophores of non-bacterial phytochromes, biliverdin is ubiquitous in mammals. This makes BphP applications in mammalian cells, tissues, and whole mammals as easy as conventional GFP-like FPs, without supplying chromophore through an external solution. Second, BphPs exhibit NIR absorbance and fluorescence, which are red-shifted relative to that of any other phytochromes, and lie within the NIRW. This makes BphPs spectrally complementary to other existing biophotonic tools such as all GFP- like FPs and available optogenetic tools. Third, independent domain architecture and pronounced conformational changes upon biliverdin photoisomerization make BphPs attractive templates to design various photoactivatable NIRFPs. Based on our analysis of the photochemistry and structural changes of BphPs we plan to develop three new types of the BphP-based NIRFPs. These include three bright and spectrally resolvable NIRFPs, putatively called short-, medium-, and long-NIRFPs (Aim 1); photoactivatable with non- phototoxic NIR light PA-NIRFPs that are initially dark but become fluorescent either in short-, medium-, or long- NIR spectral regions, and photoswitchable either irreversibly (PS-NIRFPs) or repeatedly (RS-NIRFPs) between these NIR regions (Aim 2); and NIR reporters for protein interactions and phosphorylation based on a reversible bimolecular fluorescence complementation approach utilizing monomerized versions of NIRFPs (Aim 3). We will apply directed molecular evolution approaches based on rational structure-based design and random mutagenesis of candidate proteins, followed by flow cytometry bacterial cell sorting, screening colonies on Petri dishes, and multiwell plate protein characterization. These conventional techniques will allow screening for standard FP properties such as excitation and emission wavelengths, overall brightness, photostability, pH-stability, and folding at physiological temperatures. New high-throughput screening methods will be developed to specifically optimize BphP-based NIRFPs. Selection of NIRFPs with high quantum yield, a crucial parameter for BphP-derived FPs, will be performed using time-resolved fluorescence lifetime measurements of thousands of colonies simultaneously. To screen for a high affinity to biliverdin, which does not penetrate through the inner bacterial membrane, a pulse-chase production of biliverdin using heme oxygenase co-expression and targeting of BphP NIRFPs to bacterial periplasmic space accessible for exogenous biliverdin will be employed. Promising NIRFP candidates will be directly screened in mammalian cells using shuttle vectors to optimize protein folding and stability in mammalian cells, affinity to endogenous biliverdin and low cytotoxicity. Optimized NIR probes will be tested in mouse tumor models and applied to studies in living mammals. The resulting NIR probes will extend fluorescence imaging methods to deep-tissue in vivo macroscopy including multicolor cell and tissue labeling, cell photoactivation and tracking, detection of enzymatic activities and protein interactions in mammalian tissues and whole animals.
We will develop several novel types of fluorescent proteins and biosensors for noninvasive imaging and detection of the molecular and cellular interactions in cells, tissues, and whole animals. These probes will emit in near-infrared and utilize the biliverdin chromophore, which is abundant in mammalian tissues. The probes will avoid auto fluorescence in live cells and in vivo because of tissue transparency in near-infrared. These near-infrared fluorescent proteins and biosensors will extend the methods developed for cell microscopy into deep-tissue imaging, including multicolor labeling, photoactivation and tracking, and detection of protein interactions and enzymatic activities.
Showing the most recent 10 out of 19 publications
