Crystal jellyfish have an eerie beauty: thanks to a natural protein, they emit a faint green glow. For decades, researchers have used that green fluorescent protein and similar molecules to light up the field of biology, tracking what’s happening inside cells.
Now these ubiquitous tools are getting a glow-up: their quantum properties are being harnessed to make them similar to the fundamental bits of quantum computing. “These fluorescent proteins that everybody uses as a fluorescent label can actually be turned into a qubit,” says Peter Maurer, a quantum engineer at the University of Chicago in Illinois. The idea “sounds very science fiction,” says Maurer. But the physics isn’t new, and the approach has already been shown to work in principle.
Fluorescent-protein labels are currently one of the most important tools in biology laboratories around the world. They can monitor the location and activity of proteins, sense conditions inside a cell, check whether drug candidates are targeting the right spots and carry out a range of other tasks. But adding a quantum twist offers up fresh and exciting possibilities, say researchers.
On supporting science journalism
If you’re enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Quantum sensors can detect magnetic fields and are exquisitely sensitive, so protein versions might be able to pick up the tiny signals made by firing neurons or flows of ions, or spot minuscule quantities of free radicals that hint at cellular stress or serve as early signs of cancer. And researchers can turn these protein-based quantum sensors on and off remotely, making them useful tools for new imaging technologies and therapies.
Protein labels keep surprising researchers with more capabilities, says Jin Zhang, who develops biosensors at the University of California, San Diego (UCSD). “We often struggle with the sensitivity of fluorescent labels,” she says, so she is intrigued by what as-yet-unimagined science the quantum variants might unleash. “I’m still trying to envision the new applications these might bring.”
The effort is part of a larger field of quantum sensing for biological applications, which observers say is hot and progressing fast. Although the development of protein quantum sensors is at an early stage, the researchers doing this work say that there’s not much standing in its way: some of the proteins that could be used in this way are off-the-shelf, and the equipment for manipulating them is standard fare.
“In the past, it might have seemed like, ‘ah, that’s likely never going to work’,” says Ania Jayich, a physicist at the University of California, Santa Barbara, who works on other types of quantum sensor. “That’s not true any more.”
Diamonds forever?
Quantum physics is currently going through a second revolution. During the first, in the early 1900s, physicists started to unravel the bizarre properties of the quantum world, such as superposition, whereby something exists in several states simultaneously, and entanglement, in which quantum states become mysteriously linked. Now, in the second revolution, researchers are intentionally manipulating individual quantum properties to open the door to information-dense, high-precision applications in computing, communications and sensing.
Quantum computing needs qubits — basic units of quantum information — that aren’t disturbed by the world around them. Quantum sensing, by contrast, relies on qubits that are influenced by external factors, in specific ways that can be measured. Magnetic resonance imaging (MRI), for example, creates an image by manipulating and measuring a quantum property known as spin in the body’s hydrogen nuclei. Superconducting quantum interference devices (SQUIDs) are used to detect magnetic fields in the brain during magnetoencephalography scans in hospitals.
One of the most widely used quantum sensors today is the ‘NV diamond centre’ — a defect in a diamond crystal in which one carbon atom has been replaced by a nitrogen (N) and a neighbouring carbon is absent, forming a vacancy (V). The spin states of electrons in this centre can be manipulated using microwaves and lasers, such that magnetic fields, temperature and other environmental factors affect the light that the electrons emit in precise and well-understood ways. These sensors are extremely sensitive, versatile and stable even at room temperature — unlike many qubit systems, which require extreme cold. Today, sheets of NV diamonds or nanoscale crystals are used in the lab and in some commercial products, mainly in the physical sciences — for example, to map the performance of semiconductors.
By comparison, bioscience applications have proved harder to develop, because living systems are “warm and messy”, says Jayich, whose lab focuses on NV diamonds.
But that field is picking up. It is one of a handful of focus areas at the Chicago Quantum Institute at the University of Chicago, for example, and was given a funding boost by the US National Science Foundation in 2023. And it is the sole focus of the UK Quantum Biomedical Sensing Research Hub, launched in December 2024. “We’re at a really exciting time with quantum technologies, where a lot of the lab demonstrations are reaching a point where they’re ready for applications,” says physicist John Morton at University College London, who is co-director of the research hub.
Research teams are investigating, for example, how to use NV diamonds to conduct nanoscale MRI or to improve tools used to track magnetic tracers during surgery. And, by tweaking the exterior of the diamond crystals so that they bind to specific molecules in blood-plasma samples, researchers have developed experimental HIV tests that are 100,000 times as sensitive as standard diagnostics.
Plenty of researchers are experimenting with putting diamond quantum sensors inside cells. Maurer says about half of his lab is investigating new uses for NV diamonds and will continue to do so.
But NV diamond sensors have limitations: they tend to be clunky, around ten times bigger than a protein, and are hard to place precisely where you want them. Fluorescent proteins, by contrast, are small and can be generated exactly where they are needed inside cells using genetic-engineering techniques, putting them right next to whatever researchers wish to investigate. “The gain you get from that is huge,” says Jayich.
Quantum glow up
Around a decade ago, David Awschalom, director of the Chicago Quantum Institute, and his colleagues started to wonder if they could find molecules that act as qubits. Such qubits, he hoped, could be produced reliably through chemistry instead of being carved out of diamond or semiconductors. In 2020, his team reported in Science that it could get a synthesized organometallic molecule to behave like a qubit, and his team soon did the same with other molecules.
That work led Awschalom to team up with Maurer, who had put his physics knowledge to work on biological imaging, in pursuit of biological molecules that might perform the same trick. “It was essentially the same type of idea, but now with a system that was comfortable going into cells,” says Awschalom.
They zeroed in on ‘enhanced yellow fluorescent protein’ (EYFP), an off-the-shelf product that had been enhanced by biologists for a bright yellow glow. From a physics perspective, this molecule has an electron energy structure that is similar to that of existing qubits, says Awschalom.
Fluorescent proteins glow when their electrons are excited by laser light and then fall back to their relaxed energy state. Biologists typically insert the genetic instructions for the fluorescent-protein label next to the code for a protein of interest. Then, if the target protein is expressed, the label is expressed, too: shine a laser on the sample and it lights up like a Christmas tree. Variants have been developed with different colours. And protein engineers continue to develop versions that are useful sensors: their light can be affected by pH or mechanical forces inside cells, for example, or by the presence of calcium ions, which are crucial to cell signalling, or kinase enzymes involved in phosphorylation, an important on–off switch for protein activity. Fluorescent proteins with no quantum upgrade can’t, however, detect magnetic fields.
A small fraction of the time, the excited electrons in these fluorescent proteins shift into a metastable, non-fluorescent state called a triplet state (so-named for having three possible spin configurations). This causes the light to dim or blink. “People have known that that happens, and they hated it, because it makes your fluorescent beacon less bright,” says Maurer. For his purposes, this was an advantage, not an annoyance, because the triplet state enables the creation of a coherent superposition of spins — and that makes for a potentially useful quantum sensor. NV diamond quantum sensors also rely on a triplet state.
Awschalom says that, after some false starts, it was a relatively simple task to put the EYFP into the desired quantum superposition state using laser light and microwaves. Once the team understood the energy levels of the quantum states involved, he says, “literally the next day, it was working”. As hoped, the fluoresced light was affected by magnetic fields, varying in intensity by about 30%. The team showed that the quantum sensor worked in living bacterial cells at room temperature.
There are still plenty of hurdles to overcome. One issue is that fluorescent proteins are generally fragile: they degrade over time as you shine light at them. Maurer says that might be fixable. His team is also trying to boost the proteins’ sensitivity. Biologists had previously developed fluorescent proteins that spend as little time as possible in the triplet state; Maurer says they’re now planning on doing the reverse — creating variants and selecting for those that spend more time in the triplet state. They will also work to see whether, like NV diamonds, these proteins can be used to detect changes in other conditions reliably, including pH and temperature.
The ability to detect electromagnetic fields directly is particularly exciting, says Nathan Shaner, a biological engineer at UCSD who develops fluorescent proteins. “Something that’s really difficult to make is a robust, sensitive indicator for the action potential you get when neurons fire,” he says, for example. “It’s a tiny change on a tiny scale.”
Reinventing MRI
Maurer’s group isn’t the only one intrigued by the quantum properties of proteins. Andrew York, a physicist at the Chan Zuckerberg Biohub in San Francisco, California, noticed that red and green fluorescent proteins become slightly sensitive to magnetic fields when they carry a little extension: an organic compound called a flavin. These proteins respond to magnetic fields even at room temperature and in living creatures including nematode worms and bacteria.
Harrison Steel, an engineer at the University of Oxford, UK, teamed up with York and his colleagues to develop this idea. They have shown that the quantum details in this system are slightly different — they involve a triplet state and an entanglement between two electrons in neighbouring parts of the molecule. But, broadly, it’s the same principle: magnetic fields affect electron spin states, and this alters the intensity of the fluoresced light, turning the protein into a useful quantum sensor. The team made large batches of 3,000 slightly different proteins and selected the ones that were most sensitive to magnetic fields, focusing on four winners for further study.
The group also showed that each of their proteins flash in distinct ways when the radio waves and magnetic fields are turned off and on. They plan to develop a range of protein labels, each with a unique blink. That would be useful, they say, for ‘multiplexing’: looking at a range of a dozen or more proteins or conditions in the same sample. Researchers also have ‘quantum dots’ in their arsenal for this task: tiny semiconductor crystals that glow in a rainbow of colours, a kind of artificial version of fluorescent-protein labels. But these dots, like NV diamonds, are hard to place inside cells precisely (and, says Jayich, as sensors they aren’t as flexible or sensitive as NV diamonds).
Magnetically sensitive fluorescent-protein labels can also be used to improve imaging, Steel says. Currently, fluorescent-protein microscopy can provide amazingly detailed views of protein activity in tissue, but you need to be able to see the light clearly, which usually means slicing up a research animal or peering just a millimetre into its flesh. Deeper than that, the light scatters and the signal gets blurry — like trying to ‘see’ inside your hand by shining a torch through it.
Steel’s idea is to apply radio waves and a magnetic-field gradient to make fluorescent labels flash only at certain locations and times. Then, even though the light scatters on its way out of the flesh, he can use the known information about the magnetic field to precisely locate where the light was coming from, improving the resolution of the image. The approach blends the ability of an MRI machine to peer deep into flesh using radio and magnetic signals, with the protein-specific information of fluorescent labels.
This idea isn’t well suited to deep imaging of people, however: our bodies are so big that emitted light might be absorbed entirely before reaching the surface. And the degree of specificity Steel is after means inserting fluorescent labels by genetically tweaking cells, a technique reserved for life-saving therapies. But it could work on a living mouse.
So far, Steel and his colleagues have used their set-up to image fluorescing proteins in bacteria that have been embedded in a mouse-sized lump of plastic, with a resolution of about half a millimetre. But they have imaged only a one-dimensional line, rather than creating a whole 3D picture. Steel says they made this slapdash set-up after just a month of engineering — and “it’s like a very bad MRI machine”, he laughs. They plan to pin down control of the fields and interpretation of the signals to make a more functional 3D instrument, hopefully with even higher resolution, during the next year. “The physics is working, the challenge is just making something practical,” he says. “We know basically what to do with every step.”
Because these proteins can be manipulated using magnetic fields, the discoveries also open the door to ‘magnetogenetics’ — the idea of using a directed magnetic field to ‘turn on’ a label remotely and, say, alter the binding of nearby proteins to initiate a therapeutic response deep inside the body. “That’s very exciting to think about,” says Shaner. Other applications for these quantum sensors might become apparent in the future, he adds. “There’s huge potential there, somewhere. I think it’s not totally clear what’s possible, yet.”
The idea of perfecting quantum sensors that can be used inside cells is early days but promising, says Morton. “People have been trying to work out: what’s the next NV? What’s the next-generation quantum sensor?” Fluorescent proteins might be a contender, he says, but they have a long way to go to prove that their benefits outweigh their disadvantages, such as fragility, when compared with NV diamonds.
Maurer disagrees, pointing to the genetic ‘targetability’ of fluorescent proteins as a massive advantage. “It’s not that we abandoned diamond,” he says, but he thinks fluorescent proteins will win the day for quantum sensing inside cells.
Jayich is also excited about the possibilities, especially as researchers enhance and perfect these proteins. “This is just the beginning,” she says. “Even where it’s at now, it’s going to be already better than the other quantum sensors for certain applications in biology. It’s not crazy. It’s really exciting.”
This article is reproduced with permission and was first published on March 3, 2026.
