A cell’s familiar job can hide other possibilities. In laboratory culture, adult human airway cells assembled into tiny structures that could move themselves. Researchers called them Anthrobots. Their appeal lies in a simple question: what can ordinary cells build when their surroundings and arrangement change? [3]
A 2023 discovery, followed up in 2025
Gizem Gumuskaya and colleagues first published the peer-reviewed Anthrobot study online in Advanced Science on 30 November 2023. It appeared in the journal’s 2024 volume. A separate peer-reviewed life-cycle study followed on 6 June 2025. These dates distinguish the original finding from its later characterization. [1] [2]
The source material was adult human airway epithelium, the tissue lining respiratory passages. “Somatic” in the original paper’s title means body cells. The investigators explored how these cells organize in culture, where researchers can change the physical environment and observe the resulting forms. The experiments concerned living cell assemblies grown outside the body. [4]
The cells already had useful biological machinery
Airways contain cells bearing cilia: tiny projections that beat to move material along their surface. In the respiratory tract, that activity helps clear mucus and trapped particles. When cilia face outward on a small cellular sphere, their collective beating can help propel the sphere itself. The same kind of biological machinery is doing work in a different arrangement. [4]
Orientation matters because epithelial cells have different surfaces. The apical surface normally faces a lumen or an exposed space. In a conventional airway organoid, it may face the hollow interior. An “apical-out” organoid exposes that surface externally. This changes what the cilia contact and which parts of the tissue researchers can readily access. [5]
That approach already had a research history. A 2022 study generated outward-facing airway organoids from human bronchial epithelial cells in suspension culture. Those models supported investigations of respiratory viruses and antiviral responses. Anthrobot research uses related cellular plasticity to investigate movement, form and interactions with other tissues. An exposed ciliated surface alone is therefore part of a broader organoid field. [5]
How the moving assemblies were grown
In the original Anthrobot work, cells grew into multicellular spheres in a supportive extracellular matrix for about two weeks. The researchers then released the spheres into liquid culture, where cilia became outward-facing. Assemblies showed several shapes and movement patterns, including straight or curved travel, circling and wiggling. Reported diameters ranged from 30 to 500 micrometres, or 0.03–0.5 millimetres. [3]
The production was variable: about half of 2,281 characterized spheroids showed no motility during the observation period, despite most having cilia. Thus, a striking video of a moving Anthrobot should not be read as a claim that every cell cluster becomes a reliable miniature vehicle. Shape and movement were correlated; the study did not establish complete control over either. [1]
What the neuron scratch experiment showed
The researchers also brought Anthrobot groups into contact with a scratched layer of cultured human neurons. Larger assemblies, termed superbots, bridged parts of the gap. Neuronal regrowth appeared underneath connected bridges. This was a simplified, two-dimensional tissue assay, offering evidence of a local interaction between the assemblies and the cultured cells. [3]
The experiment was small. Of ten superbot bridges, five stayed connected to both sides across the three-day observation; those five showed gap closure underneath. A passive agarose contact control did not reproduce the effect. These observations leave the mechanism unresolved and do not demonstrate restored nervous-system function, brain repair or treatment effectiveness in people. [1]
The 2025 follow-up examined what changed inside
The later study tracked development, outward reorientation of cilia, movement and degradation. RNA sequencing compared gene activity in Anthrobots with that of their source cells. The researchers found substantial changes in expression, including genes associated with developmental patterning. This makes gene regulation an important part of the story, even though the construction process introduced no genetic edits. [2]
Sequence and expression answer different questions. DNA sequence concerns the genetic instructions present; expression concerns which instructions cells are using and how strongly. A new arrangement can therefore coincide with extensive molecular change while drawing on existing genetic machinery. Development-associated expression does not by itself establish that the assembly has become an embryo or a complete organ.
How Anthrobots relate to Xenobots
Earlier Xenobot research used cells from the frog Xenopus laevis. A 2021 study described self-organized constructs that moved using surface cilia, healed after damage and displayed collective behaviours. It also used computational modelling to investigate group activity. These results provided a separate experimental example of cells forming useful active structures beyond their ordinary tissue setting. [6]
The distinction between the platforms matters. The frog work and the human-airway work use different species and starting tissues. Capabilities reported for one cannot automatically be assigned to the other. Anthrobots extended the investigation to adult human-derived cells; their performance must be established through their own assays rather than inherited from the reputation of earlier biobots. [6]
Why call a cellular assembly a robot?
Here, “biological robot” describes a proposed engineering platform made of active living material. Its movement comes from cells and their organization. The term is useful when asking whether such structures can perform reproducible tasks, but the label itself proves no particular level of control, decision-making or practical usefulness. Those properties require separate experiments. [4]
A helpful assessment begins with the task: what did the construct actually do, under which conditions, and how often? Moving across a dish, contacting a tissue edge and promoting local cell coverage are distinct outcomes. Tracking them separately makes it easier to identify progress and to avoid turning an interesting behaviour into a broader claim than the experiment supports.
What could this contribute to biotechnology?
A realistic use is as a research model for cellular self-organization. Investigators could compare how donor differences, culture conditions or candidate compounds affect movement and form. Those are proposed ways to use the platform for controlled experiments. More ambitious ideas, such as targeted delivery or tissue repair inside a body, need their own demonstrations and validation.
Important next questions include why only some assemblies move, which signals contribute to neuronal gap closure, and whether a useful behaviour persists in a more complex tissue environment. Starting with human cells does not settle those questions. Anthrobots make them experimentally accessible while revealing an intriguing possibility: a cell’s normal role may describe only part of what it can do.
Sources
- Gumuskaya G et al. Motile Living Biobots Self-Construct from Adult Human Somatic Progenitor Seed Cells. Advanced Science 11, 2303575 (2024 issue). First published online 30 November 2023. doi:10.1002/advs.202303575.
- Gumuskaya G et al. The Morphological, Behavioral, and Transcriptomic Life Cycle of Anthrobots. Advanced Science 12, 2409330 (2025). First published 6 June 2025. doi:10.1002/advs.202409330.
- Tufts Now, Mike Silver. Scientists Build Tiny Biological Robots from Human Cells. 30 November 2023. Original institutional research account. Read the Tufts report.
- Anthrobots research team. Read the Fact Sheet. Undated original project explanation. Read the project fact sheet.
- Stroulios G et al. Apical-out airway organoids as a platform for studying viral infections and screening for antiviral drugs. Scientific Reports 12, 7673 (2022). doi:10.1038/s41598-022-11700-z.
- Blackiston D et al. A cellular platform for the development of synthetic living machines. Science Robotics 6, eabf1571 (2021). doi:10.1126/scirobotics.abf1571.
Written for @biotechnologyguy. The original study was first published in 2023; the follow-up was published in 2025.
0 Comments