Developing improved fluorescent protein voltage sensors

A protein activity sensor has the important advantage that it can be specifically expressed in an individual cell type in the brain. For the past 15 years or so we have been working to improve protein sensors of membrane potential.

The first human-made genetically encoded voltage indicator was a mosaic constructed by inserting a fluorescent protein (FP) into a voltage sensitive protein residing in the plasma membrane. The sensor, FlaSh, (Siegel and Isacoff 1997), was a voltage gated potassium channel with GFP inserted following the 6th transmembrane segment. Aside from its importance as a proof-of-principle, FlaSh had the useful feature of a steep, sigmoidal fluorescence vs voltage relationship that could be tuned to select for different ranges of membrane potential. However, FlaSh also had drawbacks; its signal was relatively slow (tau ~100 msec), and small (ΔF/F <5%). But, most importantly it worked in frog oocytes but not at all in mammalian cells. In mammalian cells, FlaSh’s expression, and that of two other first generation sensors that were also based on mammalian ion channels, was mainly intracellular (Baker, Lee et al. 2007). This obstacle was overcome by changing the membrane resident voltage sensor to the voltage sensitive domain of the sea squirt Ciona intestinalis voltage sensitive phosphatase (Dimitrov, He et al. 2007). The first member of this family, VFSP2.1, expressed well in the plasma membrane of mammalian cells and signaled changes in membrane potential. However, its signal was small (2% for a 100 mV depolarization) and slow (tau > 20 msec).

We have improved on the signal size (ArcLight in Jin et al 2012) and showed that ArcLight provides useful in vivo signals in the mammalian brain (Storace et al, 2015). In addition, we have examined a number of FRET constructs and found some with relatively large and fast responses. (Sung et al, 2015).

We are presently working to improve signal size and speed as well as developing sensors that will target specific subcellular membranes (dendrites, axon terminals and cell body).

Storace, D., Sepehri Rad, M., Kang, B.E., Cohen, L.B., Hughes, T., and Baker, B.J. (2016) Toward better genetically encoded sensors of membrane potential. TiNS, in press.