Measuring Neuronal Activity In A Flash

Many of the methods used to map brain activity have the downside of low temporal resolution. Adam E. Cohen and coworkers at Harvard University address this by de­coupling the recording of the neuronal firing state from the readout in a method that they call flash memory (J. Am. Chem. Soc. 2014, DOI: 10.1021/ja411338t). The method uses an engineered microbial rhodopsin with three states—two nonfluorescent states, which exist in a voltage-dependent equilibrium, and a stable fluorescent state that can be reached via a photochemical conversion from only one of the non­fluorescent states. When the rhodopsin is illuminated with a flash of light at wavelength λ_write, part of the population is converted to the fluorescent state. The fluorescent population is frozen at the end of that pulse and can be read later by illuminating the rhodopsin with another pulse, λ_read, thereby sidestepping the temporal resolution problem. The fluorescence level during readout depends on the voltage during the write interval. The researchers added the sensor molecules to rat neurons and used them to record the cells’ electrical states. The team demonstrated that the observed fluorescence reflected whether or not the neuron fired during the recording period.