For the first hours of an experiment, a microtubule sheet sat almost still in a well roughly 3.3 mm across, behaving as a load-bearing gel with negligible internal movement. Then, after roughly two hours, the flat material began to deform as newly made protein motors gathered into clusters and slid neighbouring filaments past one another.
The sheet buckled into a maze of blister-like protrusions roughly 50 micrometres wide and spaced about 200 micrometres apart. About half an hour later, the blisters fractured near their apexes. Their fragments reorganised into independently flowing active bundles.
The work, published in the peer-reviewed journal Nature Materials on 8 October 2026, describes a material whose movement changes over time in response to DNA instructions.
Rochelle Silverman, a lead researcher and co-author at the Weizmann Institute of Science, worked with Alexandra M. Tayar, also a Weizmann co-author, and collaborators including Vincent Noireaux, who led a University of Minnesota group that contributed the protein-production system. Hillel Aharoni contributed theoretical modelling of how gene-circuit parameters shape the emergence and evolution of the material's active states.
The team made its experimental sheet from microtubules, protein filaments that form part of cells' internal skeleton. They embedded a cell-free gene-expression system within the network. This cell-free transcription-translation system, known as CFE, was rebuilt from E. coli lysate and can make proteins on demand outside living cells.
DNA plasmids in the mixture encoded kinesin motors. As the proteins were made, they clustered and attached to the microtubule scaffold. The system also made cross-linking proteins and regulatory components within the material.
Kinesin clusters are needed to slide adjacent filaments. Once enough had accumulated, the initially quiet sheet deformed, buckled and developed flowing patterns. The independently flowing bundles then sustained turbulent-like extensile flows for up to eight hours, at speeds of roughly 0.3 to 0.5 micrometres per second.
The researchers also used confined channels. In 3D-printed channels, walls restricted the deformations while protrusions directed them, allowing individual structures to extend several millimetres.
Changing the DNA instructions changed the timing and pattern of movement. When kinesin-encoding DNA ranged from 0.1 to 13 nanomolar, higher concentrations moved the onset of motion from about four hours to about one hour. The result showed that the amount of genetic instruction could set when the material began moving.
Different genetic circuits produced sustained activity, pulses, arrested states and oscillations. In a third circuit with delayed negative feedback, changes in repressor DNA concentration led the material to settle into steady turbulence, oscillate with a period of about 30 minutes, or produce one pulse before arresting. The material can therefore pulse on a schedule encoded entirely in its DNA.
Mechanical conditions helped determine the size of the first wrinkles, while the rate of gene expression affected the pace at which deformation developed into flow. Expression rates, degradation kinetics and the structure of the regulatory circuit governed the active-matter states, the study found.
The researchers describe the result as a control system in which a biochemical circuit continually influences the force-generating components. It links gene regulation to mechanical behaviour in an engineered material outside a living cell. The study establishes gene circuits as programmable control layers for active matter.
The material still depends on a working cell-free protein-production environment and an energy supply for its molecular motors. The study does not establish whether the approach can be scaled up, how long it can operate before chemical fuel is depleted, or whether it can be designed for more complex mechanical tasks.
Other authors include Erez Zerbib and David Garenne.
Comments (0)