It started as a joke.

How do you put an elephant in a refrigerator? You open the door and put it in. So, how do you put in a giraffe? You open the door, take out the elephant, and put the giraffe in. The point is the absurdity of fitting something large into such a small space.

That’s why Maruša Mur, postdoctoral fellow at the Jožef Stefan Institute in Ljubljana, Slovenia, wanted to create a three-dimensional (3D) print of a tiny elephant inside a living cell.

And that’s exactly what she and her colleagues did.

The study, led by Mur and her supervisor, Matjaž Humar, PhD, represents something the scientific literature had not previously documented: 3D printing inside the cytoplasm of an individual living cell. Prior work had demonstrated 3D printing inside living organisms, including mouse tissue in 2020 and live insect and fish embryos in 2024, but always at a larger scale and in extracellular space, never inside the cytoplasm of an individual cell.

The achievement potentially opens the way to new tools for studying how foreign particles affect cells, delivering precisely engineered structures into any cell type, or, one day, building functional devices within the cytoplasm itself.

How to Put an Elephant in a Cell

The teams started with HeLa cells, the durable research cells from the “immortal” cell line of Henrietta Lacks. The technique Humar’s team developed grew out of earlier experiments with intracellular lasers. In 2015, while at Harvard Medical School and Massachusetts General Hospital in Boston, Humar had been part of a team that pioneered a method for inserting tiny optical resonators — microspheres small enough to act as lasers — into cells, allowing researchers to tag and track individual cells over time.

To get them inside, his team sometimes injected droplets using a micropipette. Then he began to wonder “whether these droplets could be used for some other purpose,” he said. “Biologists are injecting a lot of things into cells — usually solutions like dyes or DNA. We were the first to inject droplets. And then I thought, maybe we could inject a photoresist and use a laser to write into it.”

A photoresist is a material that polymerizes — or solidifies — when struck by light. The team used a femtosecond laser to harden the resin material inside a living cell, building a 3D structure layer by layer in seconds.

The technique Humar’s team used, called two-photon polymerization, delivers pulses of near infrared laser light so brief and so tightly focused that solidification occurs only at the precise point where light converges, a volume that can be smaller than a single bacterium. Structures can be built layer by layer with individual features as small as 100 nanometers. The technique has been used for years outside cells to make optical components, scaffolds for tissue engineering, and even tiny soft robots. But nobody had demonstrated it from the inside.

To get there, the team first had to find a material they could inject into a cell without killing it. Most photoresists are industrial products engineered for electronics manufacturing, not for interaction with living tissue. But a photoresist called IP-S worked, as it was tolerable to cells in liquid form and slightly water-soluble. After printing, unsolidified material would dissolve harmlessly throughout the cell, leaving only the finished structure behind.

“We absolutely didn’t know that any of these resists were biocompatible at all,” Humar said. “Fortunately, one of them worked perfectly. Maybe we were lucky.”

John Fourkas, PhD, Millard Alexander Professor in chemistry at the University of Maryland, College Park, Maryland, who was not part of the study, said the research team most likely lucked out, but for a specific reason: Most photoresists fail inside a cell because most are built from molecules called acrylates, which react quickly and indiscriminately with proteins. When placed in a cell, they latch onto cellular machinery at random. 

The IP-S resin Humar’s team stumbled upon is built from a related but slower-reacting class of molecules called methacrylates, which take orders of magnitude longer to react. 

The printing process itself, once the droplet is inside, is largely automated. A femtosecond laser scans the droplet in preprogrammed layers, solidifying the resin into whatever shape the researchers have designed. Printing a 10 µm structure takes 3-10 seconds. Finding each cell and positioning it under the laser is still done by hand.

The team printed an assortment of structures, including the elephant, barcodes made of stacked grids, and diffraction gratings that scatter laser light into distinctive patterns. In one time-lapse sequence, a cell carrying a printed elephant was observed to divide, passing the printed structure to one of its daughter cells.

“That was a surprise for us,” Humar said. “Even with a very large structure inside, the cell was so adaptable and flexible that it just divides without caring about what’s inside.” 

For most cell types, Humar noted, biologists have had no reliable way to introduce objects of this size directly into the cytoplasm. “I think others will come up with many other things that this method could be used for,” he said.

Happy Cells Don’t Run

Fourkas found the basic demonstration interesting but was skeptical when reading the study’s claims about cell health. Though the study reports that roughly 55% of cells survived 24 hours after the full printing process, the authors compare that percentage favorably to outcomes from standard laboratory techniques like electroporation.

But Fourkas argued that survival is not the same as health. In a 2015 Nature Photonics paper co-authored by Humar during his time at Harvard Medical School, HeLa cells that had engulfed polystyrene beads showed 98.4% viability after 24 hours — a figure that makes the current study’s 55% harder to dismiss.

He noted that after a cell containing a printed structure was divided, both resulting daughter cells moved unusually fast through the dish, including the daughter cell that received no printed structure. If the stress were purely mechanical and caused by the object itself, you would expect only the daughter cell carrying the structure to behave differently. The fact that both cells behaved differently suggested something subtler was wrong.

“There’s a big gap between ‘the cells survived’ and ‘the cells are fine,’” Fourkas said. HeLa cells, he explained, are a notoriously sturdy cell line that can tolerate conditions that would kill other cell types. They’re a reasonable place for a research design to start, but not to stop. 

Fourkas also saw potential in what the demonstration could eventually enable, particularly in studying how rigid foreign objects affect cells. Researchers have long wondered how microscopic particles — from industrial fibers to microplastics — alter cell behavior once inside cells, and existing methods offer little control over the size, shape, or material of objects being studied.

A printing technique that could place precisely shaped structures inside specific cells might offer “an interesting way to study how particles of different shapes affect them,” according to Fourkas. “They’re actually able to fabricate things inside of cells, and despite the fact there’s a lot of chemistry happening in the cell, you can still have living cells afterwards. That definitely opens up a new toolbox.”

What Comes Next

Humar wants to continue his work by automating the process, making use of software to find and inject cells without human intervention at each step, pushing throughput from roughly 100 cells per session toward thousands. He also envisions functional structures: drug-loaded resins, magnetically responsive materials, and conductive polymers for recording or stimulating electrical activity within the cytoplasm.

But what Humar realistically expects in the near-term is using the technique as a tool for basic research: “For studying cell mechanics and biology,” he said, “I see the most possibility there.”

Example: The mechanical properties of cancer cells are increasingly understood to drive how tumors invade, spread, and resist treatment. Shaking up cell interiors on demand could give researchers new ways to study that phenomenon.

Cole DeForest, PhD, Weyerhaeuser Endowed Professor of chemical engineering and bioengineering at the University of Washington, Seattle, who was not involved in the study, called the paper thought-provoking despite its limitations. For future studies, he’d like to see whether changing a printed structure’s geometry influences how cells migrate or divide. 

“They drew an elephant inside of a cell — okay, that’s great, but where do we go from there?” DeForest noted that his lab has developed radical-free, bio-orthogonal photochemistries, which could in principle be adapted for intracellular printing like Humar’s team demonstrated. The vision he finds most compelling is precursors that cells express themselves, eliminating the need to inject anything. “That removes one of those harsh insults,” he said, “where you’re not having to physically poke the cell.”

Humar said the work began the same way his intracellular laser research did a decade ago — not with a clinical target or a funding rationale, but with a simpler question: “‘Let’s see if this works.’ It was just pure curiosity,” he said. 

Intracellular lasers — a field that has grown steadily since the early 2010s — are now studied by research groups across the US, Europe, and Asia. Humar has plenty of next steps in mind. But what the elephant ultimately means for biology or medicine, he said, he is happy to find out alongside everyone else.

Fourkas and DeForest reported having no relevant financial relationships. The study was funded by the European Research Council under the European Union’s Horizon 2020 research and innovation program (grant agreement No 851143) and the Slovenian Research and Innovation Agency (grants N1-0362 and P1-0099). Disclosure information for study authors is available in the original study publication.