Engineering and screening of genetically encoded FRET Calcium indicators

Fluorescent protein sensors have gained great importance in
research as they exhibit a number of advantages over synthetic
dyes. They can be targeted precisely, large populations of cells
can be imaged simultaneously, and they allow for chronic imaging
approaches. Many of them however still suffer from comparably low
signal changes. Improving fluorescent protein sensors can be
tedious and time-consuming. For this reason, great efforts have
been made not only to improve existing sensors, but also to develop
better strategies to improve them. In this work, a novel
large-scale bacterial based screening assay was established to
complement rational design. Sensor expression, stimulation, and
screening in bacteria, as well as the handling of large amounts of
data created by such a screening assay were optimized. While the
new assay can be adapted for other applications, it is especially
well suited for the screening of genetically encoded Ca2+
indicators of the basis of FRET (Förster Resonance Energy
Transfer). We used the assay to optimize such sensors, utilizing
the Ca2+ binding protein Troponin C fused between the fluorescent
proteins ECFP and cpCitrine. The resulting ‘Twitch’ sensor series
exhibited a large dynamic range of up to 1000% FRET ratio change,
great sensitivity and fast kinetics. In a second approach, we
attempted to develop a similar sensor deploying red-shifted
fluorescent proteins. To this end, further screening was conducted
to optimize the orange fluorescent protein mKOκ for FRET, and a
FRET sensor deploying mKOκ. The sensor we developed utilized
troponin C and the fluorescent protein Dreiklang (photoswitchable)
in addition to mKOκ. It was bright and exhibited a FRET ratio
change of approximately 170%. In summary, the screening procedures
presented in this thesis, will facilitate the development of a
range of genetically encoded biosensors, and were already employed
to develop a number of highly effective Ca2+ FRET indicators.