In 1975, at Woods Hole, Massachusetts, marine biologist Richard Blakemore placed a drop of pond mud water under a microscope and saw something impossible: bacteria all swimming in the same direction.
Not random walks, not phototaxis or chemotaxis—they were swimming north. Every single bacterium, in unison, like a flock of tiny migratory birds.
Blakemore put a magnet next to the slide. The bacteria instantly turned and swam toward it. Remove the magnet, and they returned to their original heading.
He thought he was seeing things. He repeated the experiment. Same result.
That year, he published a paper in *Science* and named the phenomenon magnetotaxis. The bacteria became known as magnetotactic bacteria.
But the story is far more complex than "bacteria swim toward magnets." Fifty years later, in 2025, scientists found that these bacteria don't just passively get pushed around by magnetic fields—they actively sense them. A single cell, with no nervous system and no brain, possesses a sensing mechanism we only recently came to understand.
A Tiny Compass Inside the Cell
First, the hardware.
Magnetotactic bacteria carry a chain inside. Not a DNA strand, not a protein chain—a chain of magnets.
Each cell contains 10–20 magnetosomes: nanoscale magnetic crystals wrapped in a lipid membrane. The crystals are typically magnetite (Fe₃O₄) or greigite (Fe₃S₄), sized between 35 and 120 nanometers, lined up along the cell's long axis like a string of miniature magnetic beads.
The total magnetic moment of the whole chain is enough to make the bacterium align itself with Earth's magnetic field, like a compass.
Note: *align*, not be dragged. Earth's magnetic field is extremely weak—roughly 0.25–0.65 gauss, a thousandth of a fridge magnet. It has no force to "pull" bacteria along. What it does is more subtle: it orients the bacterium's body in one direction, and the bacterium swims forward with its own flagellum.
Dead magnetotactic bacteria also align with the field—but they don't move. Like a compass with a dead battery: the needle still points north, but it won't walk there on its own.
Why a Compass?
What problem does a bacterium living in lake-bottom mud need to solve?
Finding oxygen.
Magnetotactic bacteria are microaerophiles—they need oxygen, but only a little. Too much won't do; too little won't either. In lake-bottom mud, oxygen concentration forms a gradient from surface to depth: oxygen-rich above, oxygen-poor below. The bacteria must find that "just right" thin layer.
Searching for oxygen by random swimming in three dimensions is extremely inefficient. But if you can restrict your movement to one dimension, the search space drops from 3D to 1D.
Earth's magnetic field provides that one-dimensional constraint. In the Northern Hemisphere, field lines tilt downward, so swimming north means swimming down—toward deeper, lower-oxygen zones. In the Southern Hemisphere, field lines tilt upward, so Southern Hemisphere magnetotactic bacteria swim south.
A simple physical constraint transforms the search problem from "blindly wandering in 3D" into "walking along a line."
This is what evolution does best: it doesn't solve hard problems—it makes them disappear.
How Is the Chain Built?
Magnets attract each other. Scatter a pile of small magnets on a table and they won't form a neat line—they'll snap into a clump.
The same goes for magnetosomes. Without constraints, they would clump together rather than form a chain. A clump of magnets produces far less effective magnetic moment than a chain—linear alignment sums all the individual moments into one maximal total moment.
In 2005, scientists at the Max Planck Institute found the answer: a protein called MamJ.
MamJ acts like scaffolding, fixing each magnetosome onto the cytoskeleton one by one and forcing them into a straight line. When scientists knocked out the MamJ gene, the mutant bacteria still produced magnetosome crystals—normal in shape, size, and number—but the crystals no longer formed chains; they clumped. The result? The mutants could barely align with the magnetic field and got lost in the mud.
It's like having a pile of compass needles, but they're bunched up instead of arranged in a row. Each needle points north, but the whole has no direction.
Research published in *Nature Communications* in 2025 further found that different lineages of magnetotactic bacteria assemble their magnetosome chains in entirely different ways. The deep-branching *Desulfovibrio magneticus* RS-1 uses coiled-coil proteins (Mad20/23/25/26) and an actin-like protein (MamK/Mad28) to organize its chain—proteins that don't exist at all in common Alphaproteobacteria magnetotactic bacteria.
This means: the ability to make magnets comes from a common ancestor, but the way of arranging them into chains was independently invented at least twice.
That's the coolest part of evolution—good solutions get discovered repeatedly. Eyes evolved independently at least 40 times in the animal kingdom. Magnetosome chains evolved independently at least twice in bacteria. If you find an engineering problem solved multiple times independently in biology, that solution is probably optimal.
The 2025 Upset: They're Not Just Passive Compasses
For half a century, textbooks said: magnetotactic bacteria passively align with magnetic fields. Field turns, bacterium turns. No sensing, no decision—pure physics.
In August 2025, a paper in *iScience* overturned that assumption.
A team at Aix-Marseille University in France studied a magnetotactic bacterium called SS-5. They did a simple experiment: measured swimming speed under Earth-strength magnetic field conditions, then measured again under shielded (zero-field) conditions.
Result: SS-5 swims faster in the geomagnetic field.
Passive alignment cannot explain this. Passive alignment changes direction only, not speed. If the bacterium were merely being physically oriented by the field, it should swim at the same speed in zero field and in field—just in different directions.
But SS-5 swims faster when a field is present. That means it senses the field's existence and adjusts its behavior accordingly.
The key evidence: when researchers genetically disrupted SS-5's magnetosome chain, the speed difference vanished. Same swimming speed with or without a field.
This shows the sensing signal comes from the magnetosome chain itself. The researchers proposed a mechanism of magnetomechanical signal transduction: the chain experiences a torque in the magnetic field, and this torque—through some mechanochemical coupling—is transmitted to the cell, altering the rotation speed of the flagellar motor.
In other words: the magnetosome chain is not just a compass; it's a sensor. It converts physical force into biochemical signal.
Think of the accelerometer in your phone—a MEMS device converting physical acceleration into electrical signals. Magnetotactic bacteria invented the same thing billions of years ago: converting magnetic torque into biochemical signal.
Single-Cell Magnetometry: 2026 Confirmation
In February 2026, a University of Basel team published in *Physical Review E* an experiment of almost unbelievable precision.
They fixed a single magnetotactic bacterium onto a microcantilever—like strapping a mosquito to a spring. Then they measured changes in the cantilever's vibration frequency. The smaller the frequency shift, the more stable the bacterium's magnetic moment.
The results precisely confirmed: the magnetic moment of a single magnetosome chain is exactly enough to align the bacterium in Earth's field. No more, no less—evolution wasted nothing.
The experiment also revealed something interesting: when the external field was reversed, individual magnetosomes in the chain flipped abruptly—like a few compasses in a string suddenly switching from north to south. In real environments, though, this doesn't happen: Earth's field isn't strong enough, and free-swimming bacteria simply rotate as a whole to realign—the flips never get the chance.
The researchers specifically noted that strong external fields can disrupt this alignment—important for potential microrobotic applications. If you want to remotely control magnetotactic bacteria with external fields, too-strong fields can actually make them "lose their way."
From Pond to Tumor: Applications of Bacterial Navigation
Magnetotactic bacteria's navigational ability isn't just a biological curiosity—it's becoming a medical tool.
The idea is simple: if you can guide bacteria to a specific location with magnetic fields, you can have them carry drugs there.
A team led by Jinxing Li at Michigan State University ran experiments: injecting magnetotactic bacteria into mouse tail veins, then guiding them with external fields. The bacteria traveled through the circulatory system and accumulated in deep liver tumors. Under alternating magnetic fields, their magnetic particles generate heat (magnetothermal therapy), killing tumor cells. The bacteria can also be loaded with anticancer drugs for a double strike.
Li puts it bluntly: "When we take any drug, it either diffuses or enters circulation. Magnetotactic bacteria can help overcome the physical barriers of circulation and concentrate drugs in specific regions."
His team is also developing 3D-printed microrobots—synthetic versions inspired by magnetotactic bacteria. Not living bacteria, but imitations of their navigation strategy: nanomagnets + propulsion + external magnetic field guidance.
Sensing Without a Brain
Back to the question that fascinates me most: how does a single cell—with no neurons, no synapses, no brain—"sense" a magnetic field?
Before 2025, the answer was "it doesn't sense; it just passively aligns." Now we know the answer is more complex: the magnetosome chain experiences a torque in the field, and that torque—through mechanisms we don't fully understand—is transmitted to the flagellar motor, changing swimming speed.
This is a mechanochemical signal transduction pathway. No electrical signals, no neurotransmitters—only force.
It reminds me of a deeper analogy. We're used to binding "sensing" to "nervous systems," as if a brain were required for sensation. But magnetotactic bacteria tell us: the essence of sensing is not neurons but signal transduction—converting one form of energy into another, and acting on it.
Your skin converts pressure into electrical signals. Your eyes convert photons into electrical signals. Magnetotactic bacteria convert magnetic torque into biochemical signals.
Different media, same logic.
From this perspective, the magnetosome chain is not just a MEMS sensor invented 200 million years ago—it's a prototype of sensing itself. Before nervous systems, before multicellular life, before almost everything we associate with "intelligence," single-celled organisms were already using physical forces to perceive the world.
We like to say "thinking with the body." Magnetotactic bacteria have no other option—they have only their bodies. But that body has been enough.
---
*References:*
- *Blakemore, R. (1975). Science 190: 377-379*
- *Scheffel et al. (2005). Nature 440: 110-114*
- *Gachon et al. (2025). iScience 28: 113377*
- *Zhang et al. (2025). Nature Communications 16: 4634*
- *Claus et al. (2026). Physical Review E*