The platform, which the researchers call human vascular organoid sheets, or hVOS, is built from two cell types derived from human pluripotent stem cells: endothelial cells, which line the interior of blood vessels, and smooth muscle cells, which form the muscular wall that gives vessels strength and control over blood flow. Conventional vascular organoids, while valuable for modeling human vascular biology, offer limited control over the ratio of cell types and their spatial arrangement. The hVOS approach overcomes this by using extrusion-based 3D bioprinting to deposit the two cell populations at a defined ratio within a bioink based on gelatin methacryloyl, a photocrosslinkable gelatin derivative widely used in tissue engineering. The printing is carried out under chemically defined culture conditions, which improves reproducibility and reduces the variability that has plagued earlier organoid systems.

Once printed, the cells do something remarkable: they self-assemble. Within the soft GelMA matrix, the endothelial cells and smooth muscle cells migrate, sort themselves, and organize into stable, interconnected vascular structures that progressively mature over time in culture. Functional assays and immunofluorescence imaging confirmed that the networks developed the hallmarks of genuine vasculature, including proper endothelial junctions and vessel-like architecture. The researchers then turned to single-cell RNA sequencing to interrogate the cellular composition at single-cell resolution. This revealed a diverse landscape of vascular and stromal populations and, importantly, identified transcriptional programs associated with vascular maturation, mechanotransduction, and hypoxic adaptation. In other words, the cells inside the printed sheets were not merely surviving; they were sensing their mechanical environment, responding to low oxygen, and actively remodeling toward a more mature vascular state.

The critical question, of course, is whether such engineered tissue can actually help a living organism. To find out, the team transplanted the hVOS into a murine model of hindlimb ischemia, a standard experimental setup in which blood flow to a mouse’s hindlimb is surgically restricted, mimicking the human condition of severe peripheral arterial disease. Using laser speckle perfusion imaging, a noninvasive optical technique that maps blood flow across tissue in real time, the researchers showed that animals receiving the bioprinted vascular sheets recovered significantly better blood perfusion than controls. The treatment also increased limb salvage, meaning fewer limbs deteriorated to the point of necrosis and loss, and histological analysis confirmed enhanced repair of the ischemic tissue.

Perhaps the most striking evidence came from intravital two-photon imaging, a technique that allows researchers to watch living tissue at cellular depth in a living animal. The team labeled the hVOS-derived cells with green fluorescent protein, and at both 14 and 28 days after transplantation they detected circulating dextran, a fluorescent tracer injected into the mouse’s bloodstream, flowing within the GFP-labeled vascular structures derived from the graft. This is the crucial demonstration: the human-derived vessels printed in the laboratory had become functionally connected to the host circulatory system, carrying the animal’s own blood. An engineered vascular graft that remains isolated from the host circulation is of limited use; one that is perfused by the host is a genuine piece of working plumbing.

But the study did not stop at showing that the graft worked. It asked what happened to the human cells inside the graft after transplantation, a question that has remained poorly understood in the field. By recovering graft-derived human cells from the mice and performing single-cell RNA sequencing on them, the researchers uncovered a process of substantial adaptive remodeling. The transplanted endothelial cells shifted toward venous-biased and inflammatory states, accompanied by activation of NF-κB signaling and stress-associated transcriptional programs. NF-κB is a central regulator of inflammatory responses, and its activation suggests that the graft endothelium was responding to the inflammatory cues of the wounded, ischemic environment. Rather than being a sign of failure, this remodeling appears to reflect the graft’s active participation in the repair process, with endothelial cells adjusting their identity and function to the demands of the host tissue.

The smooth muscle cells and fibroblasts within the graft underwent coordinated changes as well, showing transcriptional signatures associated with wound healing and extracellular matrix remodeling. The extracellular matrix, the fibrous scaffold that cells deposit around themselves, is central to tissue repair, providing both structural support and biochemical signals. The finding that graft-derived stromal cells were actively engaged in matrix remodeling suggests that the transplanted sheets did more than sprout vessels; they participated in the broader regenerative response of the injured tissue. Together, these single-cell results paint a picture of an engineered graft that is not a static implant but a dynamic, responsive tissue that negotiates with its new environment.

The implications for regenerative medicine are considerable. Because the hVOS are produced by bioprinting, the platform is design-flexible: the ratio of endothelial cells to smooth muscle cells, the geometry of the printed sheet, and the composition of the bioink can all be tuned. This reproducibility and controllability address two of the biggest obstacles that have held back vascular organoid technology from clinical translation, namely batch-to-batch variability and the inability to specify tissue architecture. A prevascularized construct that can be printed to order and implanted into ischemic tissue could, in principle, serve patients with peripheral arterial disease, diabetic wounds, myocardial infarction, or any condition in which inadequate blood supply limits healing.

At the same time, the study’s single-cell findings carry a cautionary and intellectually fascinating message for the field. The in vivo microenvironment reshaped the function of the graft’s endothelial cells, pushing them toward inflammatory and venous-like states. This means that the therapeutic properties of an engineered tissue cannot be fully predicted from its properties in the culture dish. The body actively reprograms transplanted cells, and understanding that reprogramming is essential both for optimizing graft design and for safety. If endothelial cells in a graft adopt strongly inflammatory states, for example, researchers will need to determine whether that inflammation is a productive part of vascular remodeling or a risk factor for graft dysfunction. The hVOS platform, by making graft-derived human cells recoverable and analyzable at single-cell resolution, provides exactly the tool needed to answer such questions systematically.

The work, published open access in BMC Medicine with contributions from Tsinghua University, Shanxi Medical University, and collaborating institutions, represents a convergence of stem cell biology, biofabrication, and computational genomics. Human pluripotent stem cells supply an unlimited and genetically defined source of vascular cells; extrusion bioprinting supplies the architectural control; and single-cell transcriptomics supplies a molecular accounting of what the engineered tissue does before and after it meets the body. As the field of vascular regenerative medicine moves toward clinical applications, studies like this one suggest that the future will belong not to simple cell injections but to engineered, multicellular, prevascularized tissues that are designed to integrate, adapt, and heal. The bioprinted vascular sheets described here are still at the preclinical stage, tested in mice, and substantial work remains before any human application. But the demonstration that printed human vessels can hook up to a living circulation and remodel themselves to serve the host is a milestone that makes the prospect of growing replacement vasculature feel considerably less like science fiction.

Subject of Research: 3D bioprinted human vascular organoid sheets for ischemic tissue repair and vascular regeneration

Article Title: 3D bioprinted human vascular organoid sheets promote functional ischemic repair and exhibit adaptive in vivo remodeling

Article References: Fu, X., Yan, L., Chen, Z., Dou, B., Zhou, D., Zhou, X., Qu, K., Gao, C., Wang, P., Zhang, F., Zou, Z., Wang, T., Li, G., Ouyang, L., & Na, J. (2026). 3D bioprinted human vascular organoid sheets promote functional ischemic repair and exhibit adaptive in vivo remodeling. BMC Medicine, 24(1), Article 524. https://doi.org/10.1186/s12916-026-05256-2

Image Credits: AI Generated

DOI: 10.1186/s12916-026-05256-2

Keywords: 3D bioprinting, vascular organoids, human pluripotent stem cells, endothelial cells, smooth muscle cells, hindlimb ischemia, regenerative medicine, tissue engineering, single-cell RNA sequencing, angiogenesis, GelMA bioink, in vivo remodeling