Magnetogenetics: Remote Control of Mouse Behavior with Magnetic Fields

Optogenetics revolutionized neural circuit research by giving researchers a light switch for individual neurons, but light has a problem: it does not pass through tissue. Implanted fibers are required, and that limits long-term, freely-moving experiments. Magnetogenetics asked a different question. Could a similar genetic switch be triggered by a non-invasive stimulus that reaches the entire brain, with no implant?

The candidate stimulus was a magnetic field. The candidate transducer was ferritin, a hollow iron-storage protein expressed naturally in nearly every cell. By genetically fusing ferritin to the cytoplasmic side of TRPV1, a temperature-sensitive cation channel, several groups proposed that an externally applied alternating magnetic field could push the ferritin core enough to open the channel and depolarize the neuron.

Two papers in 2016 showed magnetogenetic control in awake mice.

Wheeler et al., Nature Neuroscience, 2016 (vol. 19, pp. 756–761) built a TRPV1–ferritin fusion they called Magneto2.0 and expressed it virally in striatal neurons. A radio-frequency magnetic field at 465 kHz produced action potentials in slices, increased c-Fos in the targeted region, and shifted reward-driven turning behavior in freely-moving mice. The construct was published with full sequences, which made it the practical starting point for most labs that followed.

Stanley et al., Nature, 2016 (vol. 531, pp. 647–650), from the Friedman lab, used a structurally similar construct to drive bidirectional behavior: a chronic alternating magnetic field switched calcium-permeable currents on in glucose-sensing neurons of the ventromedial hypothalamus, raising or lowering blood sugar and feeding depending on which channel variant was used. This was the first demonstration that a magnetic field could control a metabolic phenotype, not just a circuit-level firing pattern.

Both papers were highly cited (each over 240 citations as of 2026) and seeded a wave of skepticism, replication attempts, and biophysics work over the following five years.

The biophysics of how a small ferritin core could deliver enough force, heat, or torque to gate an ion channel was contested. Critics noted that the iron content of a single ferritin (around 4500 iron atoms in vivo) is far below what classical magnetomechanical or magnetothermal models predict for channel opening at the field strengths used. Several groups failed to reproduce the original effects.

Hernández-Morales et al., Journal of Neuroscience, 2024 (PMID 38777598) ran the most thorough validation to date. Using patch-clamp electrophysiology on neurons expressing the Wheeler 2016 construct, they confirmed that a radio-frequency magnetic field does increase channel open probability, but the effect requires both the ferritin fusion and the TRPV1 channel and is gated by intracellular calcium dynamics rather than simple thermal heating. The paper neither vindicates nor rejects the original demonstrations: it shows the system is real but operates through a chemiosmotic pathway different from what the 2016 papers proposed.

For a behaviorist running a mouse maze paradigm, the appeal of magnetogenetics is that it removes the implanted fiber. A mouse can run a Morris water maze, a radial-arm maze, or a social-interaction arena while the field coil sits around the apparatus and modulates activity in a targeted brain region. There is no tether, no skull cap, no surgery for the recording side of the experiment after viral injection.

The cost is that the kinetics are slow (seconds, not milliseconds), spatial resolution is determined by viral targeting rather than fiber placement, and the field uniformity across a behavioral arena needs careful coil design. For circuit-level questions on long timescales (state changes, reward learning, feeding behavior) those tradeoffs are favorable. For sub-second precision, optogenetics remains the only option.

• Wheeler MA, Smith CJ, Ottolini M, Barker BS, Purohit AM, et al. Genetically targeted magnetic control of the nervous system. Nature Neuroscience. 2016;19:756–761. DOI: 10.1038/nn.4265.
• Stanley SA, Kelly L, Latcha KN, Schmidt SF, Yu X, et al. Bidirectional electromagnetic control of the hypothalamus regulates feeding and metabolism. Nature. 2016;531:647–650. DOI: 10.1038/nature17183.
• Hernández-Morales M, Morales-Weil K, Han SM, et al. Electrophysiological Mechanisms and Validation of Ferritin-Based Magnetogenetics for Remote Control of Neurons. Journal of Neuroscience. 2024. PMID: 38777598.