*4.3. PU*

PU is a family of synthetic polymers that are composed of organic units and joined by carbamate (i.e., urethane) links. PUs can be classified into two groups: biodegradable or nonbiodegradable. The traditional PUs are thermoresponsive (or thermoplastic, i.e., melt when heated) nonbiodegradable polymers that do not biodegrade when implanted in vivo. Historically, nonbiodegradable (or unbiodegradable) PU has been widely used in some biomedical fields due to its excellent mechanical and bioinert properties. Two examples of these biomedical applications are intravenous perfusion tubes

and inanimate artificial hearts with strong mechanical strength [138,139]. However, these polymers cannot be printed using the existing 3D bioprinters.

In 2006, a brand new biodegradable elastomeric PU was developed by my group. This new PU is made of PEG and PCL monomers with excellent biocompatibility, biodegradability, bioprintability, and biostability for complex bioartificial organ manufacturing [136–138]. It can be 3D printed alone or with some other natural or synthetic polymers, such as gelatin, collagen, gelatin/alginate, etc. In one of our former studies, a hybrid hierarchical PU–cell/hydrogel construct was automatically created using an extrusion-based double-nozzle, low-temperature 3D printer (Figure 10) [28,49]. The PU mainly acted as a supportive template for cell, especially ASC, accommodation, growth, migration, proliferation, and di fferentiation. It is extremely useful for some antistress or antisuture applications, such as pulsatile culture of 3D printed hierarchical vascular and neural networks and in vivo anastomosis with the host vasculatures/neural networks [108–110]. The vascularized and innervated networks can be applied to 3D bioprinting of a variety of complex organs, such as the brain, heart, lung, and kidney.

**Figure 10.** A large-scale 3D printed complex organ with vascularized liver tissue constructed through the double-nozzle low-temperature 3D bioprinter created in Prof. Wang's laboratory at Tsinghua University: (**a**) the double-nozzle low-temperature 3D bioprinter; (**b**) working principle of an elliptical hybrid hierarchical polyurethane-cell/hydrogel construct built via the double-nozzle low-temperature 3D bioprinter; (**c**) a CAD model containing the branched vascular network; (**d**) a cross section of the CAD model containing five sub-branched channels; (**e**) working platform of the 3D bioprinter containing two nozzles and the 3D printed semielliptical hybrid hierarchical polyurethane-cell/hydrogel

construct; (**f**) an elliptical sample containing both a cell-laden natural hydrogel and a synthetic polyurethane (PU) overcoat; (**g**) several layers of the elliptical sample in the middle section containing a hepatocyte-laden gelatin-based hydrogel and a PU overcoat; (**h**) several layers of the elliptical sample in the middle section containing an ASC-laden gelatin-based hydrogel and a PU overcoat; (**i**) hepatocytes encapsulated in the gelatin-based hydrogel; (**j**) a magnified photo of (i), showing the alginate/fibrin fibers around the hepatocytes; (**k**) ASCs encapsulated in the gelatin-based hydrogel growing into the micropores of the PU layer; (**l**) ASCs on the inner surface of the branched channels; ( **m**) pulsatile culture of two elliptical samples; (**n**) two samples cultured in the bioreactor; (**o**) static culture of the ASCs encapsulated in the gelatin-based hydrogel; (**p**) pulsatile culture of the ASCs encapsulated in the gelatin-based hydrogel. Images reproduced with permission from [28,49].

Using similar techniques, a 3D printed PU/PEO sca ffold was reported later in 2014. The PEO component could be used to increase the viscosity of the PU solution. The optimal ratio of PU/PEO for an extrusion-based FDM technology was 76/24. A resulting PU/PEO sca ffold, containing an average pore size of 1–4 μm, was helpful to maintain cell morphology attached to it. Chondrocytes preferred to adhere onto the sca ffold due to its hydrophilic property and suitable pore size [139]. In another study, neural stem cells (NSCs) attached well on the PU-based sca ffold. The PU-based sca ffold could provide NSCs with a proper environment to adhere, proliferate, and migrate on it and repair the disordered or damaged central nerve directly [140].

The control of multinozzle 3D printer parameters can provide a strong connection between di fferent polymer systems, such as the PLGA solution–gelatin hydrogel, and PU solution–cell-laden gelatin-based hydrogel. Supportive synthetic polymers, such as PLGA and PU, and the ECM-like gelatin-based hydrogels can be adjusted to degrade at di fferent time points during the organ construction and maturation stages. With proper selection of natural and synthetic polymer components, the 3D printed bioartificial organs can avoid all the risks of vascular rupture, stress shrinkage, immune rejection, and other negative reactions during the in vivo implantation stages [108–110]. These are all long-awaited breakthroughs in bioartificial organ manufacturing, as well as other pertinent hot research areas, such as tissue engineering, biomaterials, drug screening, organ transplantation, and pathological analysis, with respect to in vitro complex organ automatic building processes and in vivo defective/failed organ repair/regeneration/replacement/restoration applications.

### **5. Challenges and Perspectives**

Despite remarkable progress achieved in 3D bioprinting over the last decade, there are still a number of challenges that remain in the manufacturing of o ff-the-shelf bioartificial organs for clinical applications. These challenges include (1) the extreme di fficulties in establishment of an axial anastomosed vasculature, which anatomically mimics those in a natural organ, providing the incorporated cells with water, gas, and nutrients and removing metabolites from the cells; (2) the limited sophisticated multinozzle 3D printers, which are capable of assembling as many homogeneous and heterogeneous cells along with di fferent polymers in a way similar to those in a natural organ; (3) the unmatched physical, biological, and physiological properties of 3D printed vascular, neural, and lymphatic networks for defective/failed organ restoration; (4) the uncertain roles of superfluous stem cells, growth factors, and other bioactive agents existing in the bioartificial organs for in vivo implantation [141–143].

In the future, the demand for 3D organ bioprinting, both for production in large amounts and small quantities as well as individual manufacture (i.e., customized), will definitely grow. More research needs to focus on mimicking the complex anatomical, material, and biological aspects of each human organ to be replaced, especially the hierarchical vascular, neural, and lymphatic networks for fluid and bioactive signal transportation. It is imperative to develop new sophisticated polymers with updating 3D bioprinters to recapitulate more complex micro-, meso-, and macro-3D milieus of natural organs with respect to geometrical features, material constituents, and environmental factors [144–146].

In the future, suitable choice of natural and synthetic polymers for 3D bioprinting of each bioartificial organ will still play a critical role. Much more integration of the existing natural and synthetic polymers for additional physiological functionality realization in a special natural organ is still on the way. High cell viability is a crucial prerequisite for complex 3D organ bioprinting with a large volume of homogeneous and heterogeneous cells. It is necessary to further clarify the special usages of various polymers for different tissue incorporation. Harsh 3D bioprinting conditions should be avoided or overcome through additional approaches. Patient-specific cells, especially stem cells and ECMs (or dECMs), are preferable to eliminate the immune reactions [147–152].

In the future, much more work needs to be done on stem cell extraction, culture, and induction. Because a human organ is composed of at least two types of cells, specific architectural design and multiple cell assembling are essential for each of the layer-by-layer 3D building processes. Sequential induction of stem cell differentiation in a predefined 3D construct needs to be normalized in a suitable laboratory [108–110]. In particular, ASCs have set an outstanding example for their vascularization and neuralization capabilities in 3D printed constructs. Further optimization of growth factors for different stem cell inductions needs to be carried out and standardized.
