Supervillain Magneto, of X-Men fame, uses his mind to control magnets. Now, scientists have invented a protein namesake that allows them to use a magnetic field to control the firing of neurons. The Magneto ion channel quickly and reversibly activated neurons in mouse brain slices, in zebrafish, and even in freely moving, awake mice, as reported in the March 7 Nature Neuroscience online. “This is not for mind control,” senior author Ali Güler reassured this reporter. Nevertheless, Güler and colleagues at the University of Virginia in Charlottesville used Magneto to motivate mice. They expressed the paramagnetic protein in neurons studded with dopamine receptors and located deep in the brain. Magneto activated reward circuits, luring the mice to spend more time within a magnetic field than outside it. Essentially, the mice were made to “like” the pull of magnetism.
“Oh, this paper is cheating,” quipped John Cirrito of Washington University in St. Louis, Missouri, who was not involved in the work. “Anything with ‘Magneto’ in front of it automatically sounds cooler.” But all the same, Cirrito thought the technology had potential. “I see it as an incredible research tool,” he said. “I can see lots of applications for it.”
What Güler, first author Michael Wheeler, and colleagues offer is a new alternative to chemogenetics and optogenetics. Those techniques have given neuroscientists control over neurons expressing transgenes that respond to small molecules or light, respectively (see Nov 2012 news series). Each has its downsides. With chemogenetics, it takes minutes for an injected activator molecule to reach the right neurons, and scientists cannot control when the molecule dissipates. Optogenetics works quickly, but in order to reach neurons deep in the brain, neuroscientists must implant fiber optics to deliver the light, and that light only reaches neurons near the cable’s end.
To create a fast-acting, noninvasive neuron activator using magnets, Wheeler started with the TRPV4 (transient receptor potential vanilloid 1) ion channel, which opens to let sodium and calcium cations through in response to stretching of the plasma membrane (see image above). The authors reasoned that if they could tug on the channel with a magnetic force, they could mimic stretching and pop it open. To make TRPV4 sensitive to magnetic fields, they linked it to the iron-storing protein ferritin. They tried 21 different fusion constructs, placing ferritin in TRPV4’s intercellular loops or at its termini, and expressed them in human embryonic kidney (HEK) cells.
Most of the fusions were toxic, perhaps because they constantly leaked ions or gummed up the cell’s translational machinery, Güler said. With three channels that did not kill the cells outright, the authors imaged intracellular calcium to identify just one that opened up in response to a 50-milliTesla magnetic field. That is stronger than a refrigerator magnet, but much weaker than the one to three Tesla fields generated by most magnetic resonance imaging machines. This construct, with ferritin on TRPV4’s carboxyl terminus, is what they dubbed Magneto.
It let in less calcium than the researchers expected, though. They suspected some of the fusion channel might be stuck in the endoplasmic reticulum, and indeed, adding a bit of sequence to target Magneto to the plasma membrane improved its membrane localization and calcium trafficking. The authors called this particular mutant Magneto2.0.
Wheeler and colleagues tested Magneto2.0 in a variety of situations. First, they expressed the fusion gene in excitatory neurons of mice, and co-author Manoj Patel recorded action potentials from brain slices. In the 50-milliTesla field, he found the neurons spiked quickly, and rapidly stopped when he turned off the magnetic field.
Next, Wheeler expressed Magneto2.0 in the Rohon-Beard sensory neurons of zebrafish. When these neurons fire, fish larvae curl up. When Wheeler placed the larvae in a 500-milliTesla magnetic field, they coiled more than 10 times as often as they did without the magnetic stimulation.
Finally, the authors tested Magneto2.0 in freely moving mice, expressing it in medium spiny neurons of the striatum. These neurons express a dopamine receptor, and are suspected of controlling reward responses. To test this, Wheeler gave the mice a choice: They could hang out in a chamber with a 50- to 250-milliTesla electromagnetic gradient, or one with no magnetic field. The mice preferred the magnetized chamber. In other words, Wheeler created mice that feel good in a magnetic field.
“I like the technique a lot,” commented Yun Li of the National Institute on Drug Abuse in Baltimore. She pointed out that unlike light delivered by skinny fiber optics, the magnetic field can reach every Magneto-expressing neuron in the brain.
Cirrito praised the speed. “It is turned on crazy fast,” he said.
However, scientists also see plenty of room for improvement. Magneto is a prototype, and Güler is already making plans for Magneto3.0. He wants to eliminate the binding sites for TRPV4 ligands that could activate the channel. He also plans to try other positions for ferritin in hopes of boosting Magneto’s sensitivity. With Magneto2.0, the researchers had to place their mice in narrow channels with high-powered electromagnets near the animals’ heads. It would be hard to get any effects from the magnets in a larger organism, Güler said.
Both Cirrito and Li expressed concern about the timing of the calcium influx in Wheeler’s initial experiments with HEK cells. While a bit of calcium entered the cell as soon as the scientists activated the electromagnets, once turned off, the levels of cytoplasmic calcium continued to rise for as long as Wheeler watched the cells, which in this case was five minutes. This might be because cytoplasmic calcium prodded the endoplasmic reticulum, a major calcium storehouse, to let go of its own ions. Plus, some Magneto2.0 remains in the ER, so any magnetic field should let calcium out of that organelle into the cytosol as well. Scientists do not yet know exactly why the calcium rises so high, whether the same occurs in Magneto neurons, or what the long-term consequences would be. Güler said extended periods of high calcium might cause excitotoxicity. Li said she would prefer to see the calcium distribution normalize within minutes of turning off the magnet. She suggested that since some Magneto2.0 remains in the ER, improving its plasma membrane targeting could help by eliminating an intracellular source of calcium.
Scientists have plenty to do to figure out exactly how the fusion channel works, commented Polina Anikeeva of the Massachusetts Institute of Technology. Anikeeva did not participate in the study but studies optogenetics and has used magnetic nanoparticles to disrupt amyloid-β aggregates (Loynachan et al., 2015). Scientists still do not fully understand how the TRPV4 receptor responds to a mechanical pull, she said. She was surprised to see such a strong effect from adding ferritin, because it does not hold a lot of iron and binds even less when scientists try to engineer it.
Güler plans to share his constructs with other researchers. Cirrito suggested that neuroscientists studying epilepsy might use it to mimic a seizure in mouse models by activating all of a certain neuron population at once. He said Magneto could be useful to study how neural activity affects Aβ deposition (see May 2015 news; Nov 2014 conference news). Cirrito also speculated that there might be a way to selectively deactivate neurons with Magneto, for example by expressing it in GABAergic neurons that shut down neural circuits.
Wheeler and Güler are not the only ones tinkering with magnetic gene control (Hughes et al., 2008; Huang et al., 2010; Stanley et al., 2015). Anikeeva and other researchers are developing magnetothermogenetics toward future wireless application of deep-brain stimulation to treat Parkinson’s and other diseases. This term refers to heating up magnetic nanoparticles with a magnetic field to activate temperature-sensitive ion channels (Chen et al., 2015). Anikeeva pointed out that the magnetic fields used by Wheeler, which are generated by direct current electromagnets, differ from the alternating current fields she uses. Scientists might be able to combine the techniques and control two different populations of neurons with different magnetic fields, she suggested.
Researchers doubt that Magneto might be used in the clinic any time soon. “In theory, an optimized version of Magneto could be used to treat ill-firing neurons,” Güler said. But in practice, that would require researchers to perfect safe gene therapy to deliver Magneto and find the right way to activate the channel in people.—Amber Dance
- In Optogenetic Mouse, Neural Activity Drives Up Amyloid Deposition
- Measuring Rapid Changes in Brain Aβ in Live Mice
- Loynachan CN, Romero G, Christiansen MG, Chen R, Ellison R, O'Malley TT, Froriep UP, Walsh DM, Anikeeva P. Targeted Magnetic Nanoparticles for Remote Magnetothermal Disruption of Amyloid-β Aggregates. Adv Healthc Mater. 2015 Aug 19; PubMed.
- Hughes S, McBain S, Dobson J, El Haj AJ. Selective activation of mechanosensitive ion channels using magnetic particles. J R Soc Interface. 2008 Aug 6;5(25):855-63. PubMed.
- Huang H, Delikanli S, Zeng H, Ferkey DM, Pralle A. Remote control of ion channels and neurons through magnetic-field heating of nanoparticles. Nat Nanotechnol. 2010 Aug;5(8):602-6. Epub 2010 Jun 27 PubMed.
- Stanley SA, Sauer J, Kane RS, Dordick JS, Friedman JM. Remote regulation of glucose homeostasis in mice using genetically encoded nanoparticles. Nat Med. 2015 Jan;21(1):92-8. Epub 2014 Dec 15 PubMed.
- Chen R, Romero G, Christiansen MG, Mohr A, Anikeeva P. Wireless magnetothermal deep brain stimulation. Science. 2015 Mar 27;347(6229):1477-80. Epub 2015 Mar 12 PubMed.
- Stanley SA, Gagner JE, Damanpour S, Yoshida M, Dordick JS, Friedman JM. Radio-wave heating of iron oxide nanoparticles can regulate plasma glucose in mice. Science. 2012 May 4;336(6081):604-8. PubMed.
- Rost BR, Schneider F, Grauel MK, Wozny C, Bentz CG, Blessing A, Rosenmund T, Jentsch TJ, Schmitz D, Hegemann P, Rosenmund C. Optogenetic acidification of synaptic vesicles and lysosomes. Nat Neurosci. 2015 Dec;18(12):1845-52. Epub 2015 Nov 9 PubMed.
- Lin SC, Brown RE, Hussain Shuler MG, Petersen CC, Kepecs A. Optogenetic Dissection of the Basal Forebrain Neuromodulatory Control of Cortical Activation, Plasticity, and Cognition. J Neurosci. 2015 Oct 14;35(41):13896-903. PubMed.
- Yoon HH, Park JH, Kim YH, Min J, Hwang E, Lee CJ, Francis Suh JK, Hwang O, Jeon SR. Optogenetic Inactivation of the Subthalamic Nucleus Improves Forelimb Akinesia in a Rat Model of Parkinson's Disease. Neurosurgery. 2014 Jan 23; PubMed.
- Packer AM, Roska B, Häusser M. Targeting neurons and photons for optogenetics. Nat Neurosci. 2013 Jul;16(7):805-15. Epub 2013 Jun 25 PubMed.
- Wheeler MA, Smith CJ, Ottolini M, Barker BS, Purohit AM, Grippo RM, Gaykema RP, Spano AJ, Beenhakker MP, Kucenas S, Patel MK, Deppmann CD, Güler AD. Genetically targeted magnetic control of the nervous system. Nat Neurosci. 2016 Mar 7; PubMed.