UC Berkeley

Bioluminescence is the spectacular natural phenomenon in which living organisms produce and emit light. Natural bioluminescent protein systems have been discovered and characterized in over 30 species, ranging from fireflies to fungi and bacteria. Significant effort has been put towards engineering and improving these natural systems so that they may be used as tools for interrogating biological processes. As opposed to fluorescence, because bioluminescence requires no excitation light to produce a signal, it has proven extremely useful for imaging in highly autofluorescent samples, such as within deep tissues of whole organisms. To date, however, most bioluminescent imaging systems have been developed solely for the study of eukaryotic systems, and few have focused on the study of bacteria. Bacteria naturally colonize highly diverse and complex environments, from gastrointestinal tracts to soil and plant surfaces, that have proven difficult to study with currently available fluorescent tools. To allow for the study of bacterial signaling within these complex environments, new bioluminescent sensors developed specifically for bacterial signaling are needed.

Here, we describe the development and application of bioluminescent sensors for the bacterial cyclic di-nucleotide (CDN) signaling molecule cyclic di-GMP. Cyclic di-GMP is nearly ubiquitous in bacteria and plays a key role in the controlling motility and biofilm formation. As a first-generation bioluminescent sensor system, we developed intensity-based bioluminescent sensors for cyclic di-GMP. These sensors proved useful as in vitro tools for studying cyclic di-GMP, however were not amenable to live cell imaging. To move beyond purely in vitro systems, we developed next-generation ratiometric bioluminescent sensors for cyclic di-GMP. This next-generation system led to significantly improved sensor properties and allowed for the imaging of small numbers of live bacterial cells in an animal tissue-like model system. Finally, to expand these sensor systems to CDNs other than cyclic di-GMP, we applied a novel directed evolution approach to find sensors that respond to cyclic GMP-AMP. The first round of directed evolution was not successful, but work is ongoing on this front. Collectively, the work presented here lays the groundwork for using bioluminescent sensor systems to interrogate signaling in bacteria in their natural environments, which was previously not possible.