(143a) Engineering Flavin-Based Photoreceptors As An Emerging Class of Fluorescent Reporter Proteins

We report on the development of flavin-binding photoreceptor proteins as an emerging class of fluorescent reporters with potential applications in synthetic biology and biological engineering. Flavin-based fluorescent proteins (FbFPs) were recently developed as genetically encoded reporters that are characterized by small size (≈ 120 amino acids vs. 240 amino acids in GFP) and oxygen-independent maturation of fluorescence, which are key advantages over the widely used green fluorescent protein (GFP) [1,2]. Furthermore, we previously demonstrated that FbFPs present additional enhanced properties compared to GFP-based probes: specifically, broad pH tolerance (4-11), thermal stability (up to 60˚C), and rapid maturation of fluorescence (≈ 3 min. vs. ≈ 30 min. in case of GFP) [3]. In this way, FbFPs open up exciting avenues for investigating a broad class of biological systems that are poorly suited to GFP-based imaging (for example, thermophiles, anaerobic bioprocess such as fermentation, and cellular processes characterized by fast time-scales). However, FbFPs are at a nascent stage of development and are limited by modest brightness and a small library of available probes comprising only three known reporter proteins. These attributes substantially limit versatility and flexibility when selecting FbFPs for a specific application. Based on the considerable promise of FbFPs as robust fluorescent reporters under a broad range of environmental conditions, we sought to address the aforementioned issues using 1) directed evolution to enhance probe brightness and 2) mining sequenced microbial genomes to identify potential new members of the FbFP imaging toolkit.

In a first set of experiments, we applied directed evolution  to develop bright mutants of FbFPs. Specifically, we mutated nine amino acids in the fluorophore (flavin) binding pocket using site saturation mutagenesis and isolated two bright FbFP mutants (F37S and F37T) [4]. The F37S and F37T mutants exhibited a nearly 2-fold increase in overall brightness of fluorescence emission.  In a second set of experiments, with a view towards expanding FbFP library size, we applied genome mining techniques to identify homologous photoreceptor domains in bacterial, algal, and archaeal genomes. Specifically, sequence alignments between existing FbFPs were employed to query genome databases for homologous flavin-binding photoreceptors. Identified proteins were cloned and expressed in wild type Escherichia coli and screened for fluorescence using whole cell spectrofluorometry as well as fluorescence measurements on purified protein preparations. Using this approach, we discovered two entirely new FbFPs identified in blue light photoreceptors from the fresh water algae: Chlamydomonas reinhradtii and Vaucheria frigida. In particular, mutation of a critical cysteine residue in the flavin-binding pocket of the wild type (nonfluorescent) photoreceptors was found to inactivate the fluorescence-quenching photochemistry of the native photoreceptor protein and render the mutants fluorescent (quantum yields of 0.24 and 0.31 respectively). Additional improvements in the brightness of newly discovered FbFPs were sought via random mutagenesis using error-prone PCR and site directed mutagenesis.

In summary, we developed a biochemical framework to identify and engineer new and improved FbFP-variants using site saturation mutagenesis and genome mining approaches. We anticipate that our work will encourage and enable the broad application of FbFPs and engineered bright mutants as fluorescent reporters to investigate biological systems that are poorly suited to GFP-based labeling.

References

1. T. Drepper, et al. Nature Biotechnology, 25: 443:445 (2007).

2. S. Chapman, et al. PNAS, 105(50):20038:20043 (2008)

3. A. Mukherjee, et al. PLOS ONE, accepted

4. A. Mukherjee, et al. Journal of Biological Engineering, 6(1):20 (2012)