A bird glides across the sky, dipping and soaring through the clouds. In this diaphanous landscape devoid of stable landmarks, the bird relies on Earth’s magnetic field to navigate its surroundings. During its flight, the bird passes over a lake. There, beneath the rippling water, a group of bacteria collectively flit toward the lake’s surface—then in a sudden reversal, they flit down again. Looking at this coordinated movement, you’d think that, like the bird, something magnetic is at play. And you’d be right. 

Migrating via Magnet

Many organisms navigate by sensing the protective magnetic field enshrouding the planet. Birds, fish, ants—they all have a magnetic sense. But bacteria have been orienting themselves to Earth’s magnetic field long before these other life forms emerged. Hailing from diverse branches on the microbial Tree of Life, so-called magnetotactic bacteria (MTB) have, over eons, made homes in freshwater and marine environments the world over—in some places making up over 30% of the microbiome.

The throughline of this MTB menagerie is also the basis of the microbes’ magnet-inspired moniker. MTB contain chains of tiny membrane-encapsulated crystals made of ferrous compounds. These organelles, known as magnetosomes, help MTB passively align to and move along Earth’s magnetic field.

Whereas other forms of microbial navigation, like chemotaxis, are multi-directional, the movement dictated by magnetosomes (magnetotaxis) is mostly about up-and-down motion. “It might seem sort of nonsensical and [like it] doesn't have any purpose—we don't fully know the reason for it,” said Arash Komeili, Ph.D., a professor in the Department of Plant and Microbial Biology at UC Berkely who studies magnetosomes. “But I think the best guess that we have is that, in most places, the Earth's magnetic field would actually provide a useful guide for the bacteria as they navigate the environment."

For example, magnetotaxis helps microaerophilic MTB hover in areas with goldilocks concentrations of oxygen: not too much and not too little. In watery environments, this means hanging out in the oxic-anoxic transition zone, where oxygenated water at the surface bumps up against deeper oxygen-deficient water. Because these aqueous oxygen gradients are vertical, and Earth’s magnetic field lines run through them, aligning to the magnetic field saves MTB energy by simplifying the search for optimal oxygen concentrations. The bacteria simply swim up and down the field lines like tiny self-propelled elevators.

Indeed, when MTB find themselves in areas with too much oxygen, they rotate their flagella counterclockwise to propel downward. MTB in low oxygen environments do the opposite—their flagella spin clockwise and migrate upward. MTB also swim faster in the presence of magnetic fields, allowing for speedier navigation of oxygen gradients.

There may be other reasons why magnetosomes are useful, such as protecting against environmental stressors, detoxifying harmful molecules or helping MTB out-compete other bacteria by hoarding iron, though these functions require deeper investigation.

Mobilizing Magnetotactic Bacteria in the Clinic and Beyond

Uncovering the mysteries of MTB physiology is important for reasons beyond knowledge generation. MTB and their magnetic particles show immense potential for a burgeoning list of biomedical and biotechnological applications.  

Part of the appeal of MTB is their built-in compass. The microbes can be engineered to have specific properties (e.g., carry drugs), then administered and magnetically directed where to go in vivo. This feature has garnered particularly robust attention in the world of cancer imaging and therapeutics.  

Experiments with mice show that, under an applied magnetic field, MTB injected into the tail vein migrate to and accumulate in deep-seated liver tumors. Under an alternating magnetic field (i.e., a field that changes direction and strength repeatedly), the magnetic particles in the bacteria generate heat to kill tumor cells (magnetic hyperthermia). MTB can also be loaded with cancer-fighting drugs to further ablate tumors.  

“When we take any medicine, they either diffuse or go to circulation,” explained Jinxing Li, Ph.D., Red Cedar Distinguished Assistant Professor of Biomedical Engineering at Michigan State University’s Institute for Quantitative Health Science & Engineering. “So, I think 1 thing [MTB] can help overcome [is] the physical barriers of our circulation to focus the drug on a certain region.” Researchers have also devised strategies to kill pathogens/treat infections or remove pollutants from the environment with MTB.

Inspired by Nature: Making Microrobots

The potential goes beyond leveraging MTB in their natural form. For Li, MTB are a biological inspiration for synthetic innovations. “I think human beings always aspire to make something [that] replicates us, or something [in] nature,” Li mused. “So, we make sculptures. Now, we make robots. But we also think about how we can replicate something in the micro- [or] nano-scale world, especially in recent decades.” 

Li’s lab recently developed 3D printed microrobots with a trio of functions, including magnetic actuation, magnetic particle imaging and magnetic hyperthermia. The “TriMag” microrobots  consist of acrylic hydrogels embedded with iron-containing nanoparticles reminiscent of the magnetosomes in MTB. In mice, the biodegradable microrobots—with their sperm-like morphology of a head attached to a long, spiral tail—can permeate and localize in tissues and heat-kill tumor cells without harming normal tissue. Li envisions designing microrobots that help excise tissue at surgical sites or remove blood clots by navigating the microvascular environment of the body.

Compared to natural MTB-based systems, the precision and ability to integrate specific tools into microrobots is an advantage. But MTB naturally move extremely efficiently, harvest energy from their surroundings and readily sense and respond to their environment. In many ways, scientists working on building systems from scratch are striving to mimic the basic functions MTB evolved over millions of years. Evolution is, after all, the optimal optimizer. 

“Bacteria, they’re my dream machines,” Li said. Fleshing out the natural history and magnetic mechanisms of MTB may lead to synthetic or semi-synthetic systems that combine the best of both worlds. “If we understand how [the bacteria] sense, how they swim and how they take energy from the environment, we might be able to build synthetic machines with specific functions. I believe there are real possibilities in that direction."

The Origins of Magnetosomes

With that in mind, scientists are eager to uncover a central question in the world of MTB research: why and how did they evolve? Some researchers approach this question by studying magnetofossils (i.e., fossilized magnetosomes from ancient MTB). Analyzing these relics of MTB past reveals how long magnetosomes have been around. Other researchers seek insights through genetics.


“We have a lot of information about the genes that are involved in making magnetosomes, and they turn out to be pretty unique to magnetotactic bacteria,” Komeili said. “And so, you can look at all the diversity of magnetotactic bacteria in the world and try to understand how far back it goes based on the evolution of the magnetosome genes.”

Combining magnetofossil research with genetics has seeded postulations about the origins of magnetosomes. For example, it is possible that primitive magnetosomes from billions of years ago evolved to store iron. As Earth’s environment changed and grew more oxygen-rich (hello, Great Oxidation Event), the bacteria with magnetic abilities were better suited to navigate along those gradients. The result is the MTB of today.

Moving Beyond Magnetospirillum

And the MTB of today are a vast and heterogenous bunch. Research has focused primarily on members of the Magnetospirillum genus naturally found in shallow freshwater streams and sediments. These organisms generally make cuboctahedral magnetic crystals numbering 10-50 per cell. But other types of MTB can make hundreds of crystals, with shapes running the gamut from bullets to prisms.  

Komeili’s team has been working to understand the so-called deep-branching MTB—the ones far from Magnetospirillum on the phylogenetic tree. Studies in Magnetospirillum revealed a conserved set of genes (mam genes) that are essential for producing magnetosomes. Additional research suggests that no matter how deep in the phylogenetic tree MTB are, they have at least some of these genes. 

However, deep-branching MTB have other genes sprinkled next to or among their mam genes that are absent in Magnetospirillum spp. Could these genes be responsible for taking a “starter magnetosome” and giving it different properties? Using Desulfovibrio magneticus RS-1, a deep-branching MTB with tooth-shaped magnetosomes, Komeili’s lab discovered a set of crystal-associated proteins (Mad proteins), which they hypothesized controlled crystal shape. When they deleted the genes for the proteins, the crystal shape stayed the same, but how they were arranged changed entirely.

“Instead of being in a line, some of the mutations put them into little necklaces. Some of the mutations [cause them to] clump up into a ball. Some of them make a chain that looks normal, but it's not in the right place in the cell,” Komeili said. “It turned out that all of those proteins are actually part of an elaborate system to take individual crystals and put them into a chain inside of the cell.” He noted that chains are the most conserved magnetosome orientation because they help cells efficiently align to the magnetic field without “flopping around too much.”  

While Magnetospirillum spp. also have proteins that string magnetosomes into a chain, there is practically no overlap with those observed in D. magneticus RS-1. Komeili thinks the simplest explanation is that the genetic systems evolved independently from one another. He posits that early bacteria may have had a common way to make magnetic particles. Then, as navigating environmental gradients by way of the magnetic field became important, different lineages evolved different orientation strategies.  

As sequences from diverse MTB become available, researchers like Komeili can continue coloring in the outline of magnetosome mechanisms and evolution. And the more richly colored that picture becomes, the better we can leverage the ancient mechanics of MTB to face modern challenges.

---

 The ASM Mechanism Discovery Meeting at ASM Microbe 2026 is the place to be to learn from and connect with researchers across disciplines exploring the molecular mechanisms that drive microbial life and interactions. Want to join us? Registration is open!