This past year has seen the publication of a number of new and improved voltage sensors, including ASAP1, MacQ, QuasAr1 and QuasAr2. The group of Viviana Gradinaru at the California Institute of Technology has now developed the most recent additions to the arsenal: Archer1 and Archer2.

Voltage sensors are the most direct sensors of neuronal activity. Ideally, they should possess high sensitivity, fast kinetics and high baseline fluorescence to allow monitoring of fast action potential trains and subthreshold events, which are synaptic events that do not trigger action potentials.

Archer1-EGFP expression in C. elegans AWC neurons. Credit: Figure from Flytzanis et al., Nature Publishing Group.

Archer1 and Archer2 possess these qualities. Both sensors are derived from archaerhodopsin-3 (Arch) and carry two or three mutations that were first identified in a mutagenesis screen of a related rhodopsin. In Archer1 and Archer2, these mutations lead to 3- to 5-times-higher baseline fluorescence, which allows imaging of these sensors at lower light intensities than that needed for wild-type Arch. They respond to artificial voltage changes with large fluorescence changes (85% and 60% ΔF/F per 100 mV for Archer1 and Archer2, respectively) that are in the range of the QuasArs' response and higher than the changes observed in ASAP1 and MacQ. Actual action potentials generate about half as much fluorescence change as the artificial situation.

Furthermore, Archer1 can track fast voltage changes (up to at least 150 Hz). This ability to track action potentials at high frequency places the Archer sensors in between ASAP1 and MacQ, whereas the QuasArs are best suited for neurons with lower action potential frequencies.

Finally, the improved Archer sensors display about 50- to 100-times-lower photocurrents than the original Arch when illuminated with red light. This property is useful because the activity of the monitored neurons is not influenced during the recording. On the other hand, Archer1 can serve as an optogenetic inhibitor when illuminated with green light. Under these conditions, Archer1 functions as an ion pump and can generate inhibitory currents. Thus, Archer1 can be used in a dual role: as a sensor and as an optogenetic inhibitor.

All of the mentioned voltage sensors have been shown to perform well in brain slices, but imaging neuronal activity in vivo is still difficult. Gradinaru and colleagues demonstrated the suitability of Archer1 for in vivo voltage monitoring in C. elegans. They expressed Archer1 in an olfactory neuron called AWC-ON and recorded the activity of this neuron in response to a suitable olfactory stimulus. Upon odor withdrawal, the researchers observed a slow increase in fluorescence that persisted for about 10 seconds. This response is consistent with previously reported measurements of calcium activity in these neurons and confirms Archer1 as a useful sensor in vivo, at least in neurons with slow electrical activity.

Researchers interested in visualizing voltage changes now have a variety of tools to choose from. Depending on the applications, some sensors may be better suited than others, but this will have to be tested empirically. In the case of the Archer sensors, it will be interesting to find out whether they can perform as well in the mammalian brain as in the C. elegans brain.