Optogenetics began with methods to alter neuronal activity with light, using e.g. channelrhodopsins. In a broader sense, optogenetic approaches also include the use of genetically encoded biosensors to monitor the activity of neurons or other cell types by measuring fluorescence or bioluminescence. Genetically encoded calcium indicators (GECIs) are used frequently to monitor neuronal activity, but other cellular parameters such as membrane voltage or second messenger activity can also be recorded optically. The use of optogenetic sensors is not restricted to neuroscience, but plays increasingly important roles in immunology, cardiology and cancer research.
The first experiments to measure intracellular calcium levels via protein expression were based on aequorin, a bioluminescent protein from the jellyfish Aequorea. To produce light, however, this enzyme needs the 'fuel' compound coelenteracine, which has to be added to the preparation. This is not practical in intact animals, and in addition, the temporal resolution of bioluminescence imaging is relatively poor (seconds-minutes). The first genetically encoded fluorescent calcium indicator (GECI) to be used to image activity in an animal was cameleon, designed by Atsushi Miyawaki, Roger Tsien and coworkers in 1997.[1] Cameleon was first used successfully in an animal by Rex Kerr, William Schafer and coworkers to record from neurons and muscle cells of the nematode C. elegans.[2] Cameleon was subsequently used to record neural activity in flies[3] and zebrafish.[4] In mammals, the first GECI to be used in vivo was GCaMP,[5] first developed by Junichi Nakai and coworkers in 2001.[6] GCaMP has undergone numerous improvements, notably by a team of scientists at the Janelia Farm Research Campus (GENIE project, HHMI), and GCaMP6[7] in particular has become widely used in neuroscience. Very recently, G protein-coupled receptors have been harnessed to generate a series of highly specific indicators for various neurotransmitters.[8][9]
Genetically encoded sensors are fusion proteins, consisting of a ligand binding domain (sensor) and a fluorescent protein, attached by a short linker (flexible peptide). When the sensor domain binds the correct ligand, it changes conformation. This movement is transferred to the fluorescent protein and the resulting deformation leads to a change in fluorescence. The efficiency of this process depends critically on the length of the linker region, which has to be optimized in a labor-intensive process. The fluorescent protein is often circularly permuted, i.e. new C-terminal and N-terminal ends were created. Single-wavelength sensors are easy to use for qualitative measurements, but difficult to calibrate for quantitative measurements of ligand concentration.
A second class of sensors relies on Förster resonance energy transfer (FRET) between two fluorescent proteins (FP) of different color. The shorter wavelength FP (donor) is excited with blue light from a laser or LED. If the second FP (acceptor) is very close, the energy is transferred to the acceptor, resulting in yellow or red fluorescence. When the acceptor FP moves further away, the donor emits green fluorescence. The sensor domain is typically spliced between the two FPs, resulting in a hinge-type movement upon ligand binding that changes the distance between donor and acceptor. The imaging procedure is more complex for FRET sensors, but the fluorescence ratio can be calibrated to measure the absolute concentration of a ligand. Read-out via fluorescence lifetime imaging (FLIM) of donor fluorescence is also possible, as the FRET process speeds up the fluorescence decay.
Indicators have been designed to measure ion concentrations, membrane potential, neurotransmitters, and various intracellular signaling molecules. The following list provides only examples for each class; many more have been published.
A recent review of GPCR-based genetically encoded fluorescent indicators for neuromodulators [9]