Magnetic Control of Proteins: More than a Dream

Lots of scientists work to invent novel techniques or to engineer improved tools for familiar applications. But how does one invent a tool for applications that don’t even exist yet? Andrew York, Maria Ingaramo, and their team at Calico Life Sciences recently set out to do just that — by developing a new fluorescent protein tool that responds strongly to magnetic fields (Frank Hayward et al., 2024). They deposited their plasmid with Addgene, hoping to inspire others to explore the phenomenon, improve the tool, and develop new applications for magnetoresponsive proteins.

Motivated by curiosity

Although a wide range of fluorescent proteins and optogenetic tools exist, humans (like many organisms) are opaque, which makes most biomedical applications of these tools difficult or impossible. But if you’ve ever had an MRI, you know that magnetic fields can pass right through our tissues unimpeded. Andrew and Maria dreamed about the possibilities of a magnetoresponsive fluorescent protein. Controlling an optogenetic or fluorescent tool with magnets would enable new research and applications that don’t yet exist, but it wasn’t clear that this was even possible.

There is a variety of research on whether and how organisms respond to magnetic fields (Karki et al., 2021; Kattnig et al., 2016), and some previous attempts at magnetic control of biological systems (Wheeler et al., 2016; Duret et al., 2019). But the field has sometimes seen controversy, and the interactions of organic molecules like proteins with magnetic fields generally appears to be weak — if they exist at all. Andrew and Maria were encouraged by previous reports that described magnetic responses of cryptochrome proteins or organic fluorophores (Karki et al., 2021; Lee et al., 2011; Matsuoka et al., 2023), but these effects were very small or were possible only in non-biological conditions. So, they decided to search for a magnetic response among other proteins that might serve as a starting point for developing useful tools.

A lucky start

As Andrew and Maria explained, “to our surprise and delight,” the very first protein they tested — E. coli expressing good ol’ EGFP — showed a slight but detectable change in fluorescence when a small magnet was passed over the sample (Frank Hayward et al., 2024). The fluorescence intensity decreased whenever the magnet approached the sample (by less than 1%) and recovered after the magnet was removed (Figure 1). At first, Andrew and Maria were skeptical. The team was careful to test the possibility that this might just be an artifact and eliminate sources of interference, but the result persisted. (Even better, multiple other labs have reproduced their EGFP findings since the initial report in 2023.)

Figure 1: Fluorescence intensity of EGFP-FlavinTag in E. coli changes in response to a ~25 mT magnet. Gray regions indicate times when the magnetic field is on. Image reproduced from Frank Hayward et al. 2024 under CC-BY license.

They did more experiments to see if any small molecules act as a cofactor (among others, flavins are very effective) and to see if any other fluorescent proteins display a similar effect (many do). Fusing EGFP to the self-labeling FlavinTag made a reliable one-component magnetoresponsive system that could be expressed and purified. Although a ~1% change in signal might not make a very powerful tool, it was promising enough to keep exploring and to try developing a variant with an enhanced response.

Evolution of a new variant

Maria and the team set up a directed evolution and screening experiment with three of the proteins that had shown magnetic responses: EGFP, mScarlet, and AsLOV2. After several rounds of semi-random mutagenesis and screening, the EGFP and mScarlet variants were showing no obvious signs of improved response. Then, just when Andrew was about to suggest she deprioritize the project, Maria’s screens of AsLOV2 variants showed some significant gains!

After a few more rounds of mutagenesis, they eventually found a version of AsLOV2 with five mutations (C450P, L496V, Q513K, G528K, D540M) that shows a fluorescence change of ~75% in response to the magnetic field (Figure 2); they named this variant MagLOV. Andrew was thrilled to see that MagLOV’s “astonishing” magnetoresponse was clearly visible whether expressed in E. coli, in mammalian cells, or in purified form, and didn’t need any added cofactors. 

Figure 2: Fluorescence intensity and ΔF/F of MagLOV, a variant of AsLOV2 with mutations C450P, L496V, Q513K, G528K, D540M, in response to a ~10 mT magnet. Gray regions indicate times when the magnetic field is on. Image reproduced from Frank Hayward et al. 2024 under CC-BY license.

Video 1: Fluorescence movie of E. coli expressing MagLOV while a magnet is waved under the sample. The imaged area measures approximately 82 mm across. Video courtesy of Andrew York and Maria Ingaramo.

MagLOV shows a wealth of potential, and while many applications are still yet to be demonstrated (or invented), its behavior has already been successfully reproduced by several other labs. Andrew and Maria expect that further-improved variants of MagLOV will be possible with other engineering approaches and that magnetosensitive tools can be developed from other fluorescent proteins. 

Currently, the team is working to identify how MagLOV’s structure changes under magnetic field. Such changes could be harnessed or enhanced for future MagLOV-based tools, just as light-induced conformational changes are the basis of many optogenetic tools. And, since MagLOV’s parent protein AsLOV2 has already been adapted into a variety of optogenetic tools, they’re optimistic that MagLOV can be similarly adaptable into future magnetogenetic tools. 

Embracing science in the open

The York Lab operates with a strong commitment to exploration and collaboration. Andrew is a physicist with a background in optics and a track record of developing new microscopy tools for biological applications across a wide range of topics. He embraces open science, self-publishing directly on his website (and often posting updates on social media) and freely sharing code, raw data, and more. The team shares results early and often, including works-in-progress still in need of further study and development like MagLOV.

The results with MagLOV raise broader questions like “How and why do fluorescent proteins respond to magnetic fields?” and “What are magnetoresponsive fluorescent proteins good for?”, and while Andrew and Maria have lots of ideas and are happy to speculate, they know they can’t answer these questions alone. They readily admit they don’t fully understand the physical mechanism, though they have collected a variety of data that might help formulate some hypotheses. They are excited to see other researchers from a range of fields, like quantum chemistry and protein biophysics, as well as inventors and biomedical engineers, join efforts to explore the mechanisms and develop potential applications. 

Andrew is a frequent collaborator and is happy to give advice and support for follow-up studies. “We don’t have a vision of passive impact,” he says, noting that his team regularly shares troubleshooting tips and works with other labs to help them adopt the tools his group develops. And though it took a lot of work (and some luck) to get here, he believes MagLOV is just a starting point for much more.

If you’re interested in working on these questions or collaborating, the York Lab is eager to hear from you.

Find the MagLOV plasmid here!

---

References and resources

References

• Duret, G., Polali, S., Anderson, E. D., Bell, A. M., Tzouanas, C. N., Avants, B. W., & Robinson, J. T. (2019). Magnetic Entropy as a Proposed Gating Mechanism for Magnetogenetic Ion Channels. Biophysical Journal, 116(3), 454–468. https://doi.org/10.1016/j.bpj.2019.01.003. 
• Frank Hayward, R., Rai, A., Lazzari-Dean, J. R., Lefebvre, A. E. Y. T., York, A. G., Ingaramo, M. (2024). Magnetic control of the brightness of fluorescent proteins. Zenodo. https://doi.org/10.5281/zenodo.11406498. andrewgyork.github.io/gfp_magnetofluorescence.
• Karki, N., Vergish, S., & Zoltowski, B. D. (2021). Cryptochromes: Photochemical and structural insight into magnetoreception. Protein Science, 30(8), 1521–1534. https://doi.org/10.1002/pro.4124. 
• Kattnig, D. R., Evans, E. W., Déjean, V., Dodson, C. A., Wallace, M. I., Mackenzie, S. R., Timmel, C. R., & Hore, P. J. (2016). Chemical amplification of magnetic field effects relevant to avian magnetoreception. Nature Chemistry, 8(4), 384–391. https://doi.org/10.1038/nchem.2447. 
• Lee, H., Yang, N., & Cohen, A. E. (2011). Mapping Nanomagnetic Fields Using a Radical Pair Reaction. Nano Letters, 11(12), 5367–5372. https://doi.org/10.1021/nl202950h. 
• Matsuoka, R., Kimura, S., Miura, T., Ikoma, T., & Kusamoto, T. (2023) Single-Molecule Magnetoluminescence from a Spatially Confined Persistent Diradical Emitter. Journal of the American Chemical Society, 145(25), 13615–13622. https://doi.org/10.1021/jacs.3c01076. 
• Wheeler, M. A., Smith, C. J., Ottolini, M., Barker, B. S., Purohit, A. M., Grippo, R. M., Gaykema, R. P., Spano, A. J., Beenhakker, M. P., Kucenas, S., Patel, M. K., Deppmann, C. D., & Güler, A. D. (2016). Genetically targeted magnetic control of the nervous system. Nature Neuroscience, 19(5), 756–761. https://doi.org/10.1038/nn.4265. 

Additional resources on addgene.org

Find Optogenetics Plasmids

Find Biosensor Plasmids

Visit the Fluorescent Protein Guide Pages