The billion cell construct: will three-dimensional printing get us there?

Abstract

How structure relates to function--across spatial scales, from the single molecule to the whole organism--is a central theme in biology. Bioengineers, however, wrestle with the converse question: will function follow form? That is, we struggle to approximate the architecture of living tissues experimentally, hoping that the structure we create will lead to the function we desire. A new means to explore the relationship between form and function in living tissue has arrived with three-dimensional printing, but the technology is not without limitations.

Figure 1. Anatomical complexity remains unsolved.

(A) Leonardo da Vinci famously recognized the interpenetrating networks of lung vasculature and branched airways with his detailed drawings (c. 1500). Image courtesy of the European Union Leonardo Digitale. (B) Whole-lung vasculature can be reconstructed and visualized from computed tomography (CT) scans. Reprinted with permission from . (C) Air sac architecture of adult rat lung (electron micrograph of decellularized resin cast). Image courtesy of Laura Niklason, additional research available via , scale bar  =  1 mm. (D) Optical projection tomography image of an embryonic day 15 mouse lung undergoing branching morphogenesis. Epithelium (E-Cadherin, magenta), future conducting airways (SOX2, white). Image courtesy of Jichao Chen, additional research available via , scale bar  =  500 m.

Figure 2. Tissue engineering.

Investigations with engineered tissue constructs currently span at least eight orders of magnitude. Yet, the minimum therapeutic threshold for recapitulating solid organ function in humans is estimated at the level of 1–10 billion functioning parenchymal cells. We still have a ways to go.

Figure 3. Overview of 3D printing.

(A) A 3D model can be generated and visualized in a wide range of software packages. 3D model available under Creative Commons license via Thingiverse.com, courtesy of artists Barak Moshe and Faberdashery. (B) The surface topology is simplified to a mesh comprising a series of 3D coordinates (vertices) and the triangles (faces) that connect them. (C) The surface mesh is computationally sliced layer-by-layer to calculate machine instructions suitable for 3D printing. Machine instructions can be visualized en face or in cross-section (inset). (D) 3D printing via melt extrusion (inset) can easily achieve layer heights which surpass the resolution of human fingerprints. Scale bar  =  1 mm. (E) A selection of the diverse parameter space of 3D printing technologies. Many dozens of different combinations are in practice today.

Figure 4. Journey of a molecular nutrient through native tissues.

Cellular organization in vascularized tissues is commonly simplified into four regimes, which are rarely recapitulated together in engineered tissue constructs. Soluble blood components vary dramatically in size, concentration, and biochemistry, and each has distinct targets and mechanisms for negotiating tissue architecture. Artwork render and animation (Movie S1) performed with Blender.org open-source software.

Figure 5. Recapitulating whole organ vasculature.

It should be possible to create whole vascularized organoids by merging current anatomical mapping technologies with 3D printing. (A) A tissue or organ of interest is scanned via microcomputed tomography (micro-CT). Source 2D liver scans courtesy of Chris Chen and Sangeeta Bhatia, additional research available via . The resulting voxels (volumetric pixels) can be visualized and converted into a 3D surface topology. (B) Optionally, the 3D surface mesh can be fully parametrized in order to generate, de novo, similar vascular architectures as a new topology. (C) Native or synthetically generated vascular architectures are then computationally sliced and prepared for 3D printing directly (in sacrificial ink) or by boolean volumetric subtraction (in additive ink). After physical cleanup, 3D printing can yield cell-laden hydrogels containing living cells and perfusable vasculature. Shown here for clarity is an architecture with one inlet and zero outlets, but more complete or complex architectures with multiple inlets and outlets could be achieved with this same workflow.