Genetically encoded ion channels that respond to magnetic fields –‘Magnetogenetics’ — would enable wireless stimulation of specific neurons deep in the brain and thus provide a powerful tool for studying neural correlates of behavior in freely moving animals. A recently engineered magnetogenetic protein consisting of ferritin and TRPV4, dubbed Magneto2.0, was shown to elicit action potentials in neurons when exposed to a magnetic field. The iron-sequestering protein, ferritin serves as the magnetically sensitive domain, while TRPV4 is a cation selective channel that responds to temperature stimuli. However, the mechanism of how the protein senses magnetic field was not understood. Here, we propose a novel mechanism based on the magnetocaloric effect to explain the working of Magneto2.0: A magnetic field reduces the entropy of the ferritin nanoparticles when its magnetic spins align, resulting in an increase in temperature that in turn gates the heat-sensitive TRPV4 channel. This theory is supported by our calculations and experimental data showing that the observed responses are indeed thermally mediated.

In exploring this theory, I delve into aspects of nanoscale heat transfer, which deviate significantly from bulk thermal properties. Classical laws predict that there is no significant temperature gradient between a magnetically heated nanoparticle and the surrounding medium and that a single nanoparticle cannot generate enough heat to gate a channel. We measured the temperature and thermal conductance at the vicinity of heated nanoparticles using a novel thermosensor based on silicon microring resonator. A change in temperature shifts the resonant wavelength of the resonator. Temperature near the surface of heated nanoparticles attached directly to these resonators is measured based on the wavelength shift. We show that temperature near surface of the nanoparticles is much higher than that of the surrounding medium and that the thermal conductance at the nanoparticle-water interface is 13 orders of magnitude lower than expected from classical laws. This lowered conductance would enable a single ferritin to gate a nearby TRPV4 channel. In addition to reconciling biological observations with physical properties of magnetic nanoparticles, understanding this mechanism is essential for the design of future magnetogenetic tools with improved magnetic sensitivity.