A flat sheet of living cells, no thicker than a few layers, sits in a dish. Within hours, it begins to curl, crease, and fold into a tube, guided not by any external force but by the mechanical pull of the cells themselves. That transformation, once seen only in developing embryos, can now be programmed in the lab.
Two independent research groups have demonstrated that living tissue can be engineered to self-fold from two-dimensional sheets into precise three-dimensional shapes, using nothing more than the contractile forces cells naturally generate. The work, spanning techniques from DNA-based cell patterning to custom bioinks for 3D printing, opens a path toward building functional tissue for drug testing and, potentially, organ repair.
Cells as architects: the foundational discovery
The conceptual breakthrough came from a team at the University of California, San Francisco, led by researcher Alex Hughes. In a 2018 study published in Developmental Cell, the group used a method called DNA-programmed assembly of cells (DPAC) to place mesenchymal cells at exact positions within a flat tissue sheet made from embryonic chick tissue. As those cells condensed and compacted the surrounding extracellular matrix, the sheet folded along designed paths, producing tubes and other curved structures.
The critical insight was that cell-generated mechanical forces alone, without any external trigger like heat, light, or magnetic fields, could drive predictable shape change. The team paired their experiments with finite element modeling to show that the geometry of cell placement determined the geometry of the fold. In essence, they had written a set of instructions into the tissue’s architecture, and the cells carried them out.
From concept to printable platform
Building on that principle, a group at the University of Illinois Chicago developed a way to manufacture self-folding tissues using a modified 3D printing process. Their team created a composite bioink blending oxidized methacrylated alginate, gelatin methacrylate, and gelatin microspheres. The formulation allowed them to print flat, free-standing constructs that transformed over time as embedded cells pulled on the scaffold, a process sometimes called 4D bioprinting because the printed shape changes after fabrication.
In a study published in Matter in 2025, the UIC team reported that internal cell-contractile forces drove bending, twisting, and curling under normal physiological conditions, producing shapes including tubes, U-forms, S-curves, spirals, and arcs. A companion preprint on bioRxiv, indexed at PubMed, provides additional experimental details such as parameter sweeps and intermediate results.
Researcher Aixiang Ding, in a UIC institutional report on the work, emphasized the distinction between this internal-stimulus approach and earlier 4D systems that relied on external triggers. Because the cells themselves supply the driving force, the process avoids exposing tissue to stimuli that could damage it, a practical advantage for any future medical application.
A parallel clue from physics
A separate line of evidence, described in a preprint posted to arXiv, adds a useful physical framework. Because the preprint does not carry a DOI or named authors in the sources available for this article, readers should note that its claims have not been independently verified here. That work reported that detached fibroblast cell sheets contract in direction-dependent ways, behaving similarly to nematic liquid crystal elastomers, a class of materials well studied in soft robotics. The proof-of-concept study quantified how two-dimensional patterning choices could dictate the final three-dimensional shape of a living cell layer, reinforcing the broader idea that cell mechanics can be harnessed as a design tool.
What stands between the lab and the clinic
For all the elegance of these demonstrations, significant gaps remain before programmable tissue folding reaches patients.
Neither the UCSF nor the UIC study reports long-term data on whether folded tissues remain stable, viable, or functional after reaching their target shape. The UCSF work established the mechanism in embryonic chick tissue. The UIC platform showed shape control in bioprinted hydrogel constructs. But the critical next question, whether these structures can survive for weeks, integrate with surrounding tissue, and perform a biological function, remains unanswered in the peer-reviewed literature as of May 2026.
The relationship between the two programs is also worth noting. The UCSF study appeared years before the UIC work, and citation trails connect them, but no published statement from either group explains precisely how one technique informed the other or where they diverge in capability. Whether DPAC-style patterning and composite-bioink printing could be combined into a single hybrid platform, as some in the field have speculated, lacks direct experimental support.
Clinical trial timelines and regulatory requirements for programmable living tissues are absent from the primary literature. Press releases from both universities mention potential applications in organ repair and drug screening, but those statements are aspirational. No specific regulatory filings or preclinical milestones have been publicly announced.
Why convergence across labs strengthens the case
What makes this body of work notable is not any single experiment but the convergence. Two independent groups, using fundamentally different manufacturing methods, have arrived at the same core result: living cells can be arranged so their own mechanical forces sculpt flat tissue into targeted three-dimensional forms. That kind of replication across labs and techniques is exactly what moves a finding from curiosity to credible principle.
The practical stakes are substantial. Current approaches to building replacement tissue, from stacking cell layers to seeding pre-shaped scaffolds, struggle to replicate the complex folds and curves found in real organs. A method that lets tissue shape itself the way embryonic tissue does could eventually produce structures closer to what the body actually builds. For now, the technology sits at the boundary between a validated laboratory principle and a future medical tool. The next round of studies, focused on longevity, function, and animal models, will determine how quickly it crosses that line.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
