Bottom-up tissue engineering is a promising approach for designing modularbiomimeticstructures that aim to recapitulate the intricate hierarchy and biofunctionality of native human tissues. In recent years, this field has seen exciting progress driven by an increasing knowledge of biological systems and their rational deconstruction into key core components. Relevant advances in the bottom-up assembly of unitary living blocks toward the creation of higher orderbioarchitecturesbased on multicellular-rich structures or multicomponent cell–biomaterial synergies are described. An up-to-date critical overview of long-term existing and rapidly emerging technologies for integrative bottom-up tissue engineering is provided, including discussion of their practical challenges and required advances. It is envisioned that a combination of cell–biomaterial constructs with bioadaptable features and biospecific 3D designs will contribute to the development of more robust and functional humanized tissues for therapies and disease models, as well as tools for fundamental biological studies.
Tissue engineering and regenerative medicine (TERM) strategies have been for long regarded as the next-generation medical treatments owing to their potential to repair, improve, or replace tissues/organs that exhibit defective functions resulting from trauma, chronic diseases or ageing.[1,2] Over the past decades, this field has witnessed a tremendous evolution motivated by an accumulating body of knowledge on human tissues development, homeostasis regulation, and inflammation/regeneration processes.[3–5] Adding to this, such fundaments have also been key for rapidly advancing the fabrication of complex microphysiological systems using well-designed bioinstructive materialsthat aim to recapitulate human disease hallmarks in vitro.
This deepened understanding of tissue-specific microenvironments and recognition of their fundamental modular nature has revealed that different cell populations and their supportive extracellular matrix (ECM) represent the core effectors in human biological systems and are essential for life.[7,8]During the processes of organogenesis and morphogenesis, such elements self-orchestrate tissue development from a nano- to macrostructural organization in a dynamic mode involving bothcell–cell crosstalk (e.g., via soluble mediators, vesicles, etc.) and dynamic cell–matrix biochemical and biophysical interactions. This highly bioregulated interplay remains active throughout our lifetime with constant ECM synthesis, biochemical modification, and remodeling in response to biological and environmental factors, as well as to tissue injury.[9] Loss-of-function at both cellular and ECM level dysregulates tissue homeostasis and prompts the onset of numerous life- threatening pathologies.
Recreating these building blocks and their intricately controlled crosstalk is essential when designing tissue engineered platforms that aim to repair/substitute the complex 3D architecture of tissue-specific microenvironments and their biological functions.[10] Despite significant advances, the inclusion of key biocomponents in ex vivo engineered constructs and their arrangement into structurally ordered and physiologically functional microtissues remains a demanding challenge.[11
Aiming to recapitulate such components, conventional tissue engineering strategies have explored “top-down ” approaches, involving cell seeding in supporting porous 3D scaffolds aiming to mimic the native ECM physiochemical and biomechanical cues.[12,13] In top-down tissue engineering, cells are expected to attach/proliferate, and ultimately completely populate a prefabricated 3D biodegradable scaffold, while simultaneously depositing de novo formed ECM along time. Up-to-date, numerous scaffold types have been engineered to better recapitulate tissues microarchitecture, physiology, and ECM soft/stiff spatiotemporal biomechanical rearrangements.[14] The latter often encompasses the inclusion of mechanical stimulation and/or biomolecules (e.g., growth factors, cytokines, etc.) for tailoring biomaterials to better recapitulate in vivo tissue microenvironments biomolecular signaling and mechanobiology.[15,16] However, cells immobilization in preformed ECM-mimetic supporting scaffolds is highly challenging, often resulting in low cell seeding density and heterogeneous spatial distribution.[17] Despite recent progress in biocompatible scaffolds fabrication techniques (e.g., 3D printing, two-photon polymerization, lithography, etc.) or on the use of bioderiveddecellularizedmatrix (dECM) tissuespecific templates, top-down scaffold-based approaches generally fail to mimic the unit-repetitive modular design found in native human tissues (i.e., nephrons, lobules, islets, etc.).[18]
挑战：1.预制的模拟ECM的支持支架中固定细胞，low cell seeding density heterogeneous spatial distribution（细胞种植密度低和空间分布不均）
2.😱自上而下支架的方法通常无法模拟在天然人体组织（即肾单位、小叶、胰岛等）中发现的单元重复模块化设计
These limitations and the accumulating knowledge of human developmental biology has supported bioinspired bottom-up approaches in the form of self-assembling multicellular modules and/or cell–ECM mimetic biomaterial constructs.
Scaffold-free cell-rich structures and scaffold-based cell– biomaterial bottom-up tissue engineering strategies are more suitable for replicating the natural intricacies and modularity of human tissues/organs.[19] These building blocks can then be combined into multiscale microtissues in a hierarchic and programmed assembly mode with a biospecific design. This is critical since organ-like microarchitectures are needed not only to replicate living structures functionality, but also to identify key parameters and their roles in determining engineered tissues function. Advancing biomimetic designs in this direction is essential for accelerating regenerative medicine approaches, but also in the context of developing preclinical drug screening models. Also, it is now recognized that hierarchical tissue organization naturally limits the accumulation of somatic mutations.[2021] Thus, bottom-up assembly of hierarchical constructs might also contribute to guide the differentiation of naive stem cells upon implantation and reduce the possible risks of cell-based therapies. From a bioengineering perspective, it is more valuable and intuitive to construct a repertoire of selected biofunctional cell-rich modules rather than immediately attempting to mimic the full complexity of natural tissues. When contemplating clinical translation, optimizing biofunctionality while minimizing complexity is essential.[22
Bottom-up tissue engineering provides a unique design flexibility, allowing to freely combine each building block to carry out distinct tasks in multiple layouts with tissue-biomimetic features, and only then assessing the final biofunctionality of multiscale-assembled microtissues. The cellular rich building blocks can be engineered via different approaches including self- or guided/programmed cell assembly (Figure 1)that enable user-controlled spatial distribution of different cell populations.
Ultimately, by establishing a library of biofunctional modular units that is representative of: i) a key biological function from a specific tissue or organ, ii) the underlying matrix that supports the cell modules and provides essential cues for their bioactivity, researchers will be able to advance tissue engineered constructs toward more realistic clinical applications. In this progress report, we provide an up-to-date outlook on current strategies for assembling bottom-up tissue engineered constructs and include a critical perspective on the key role that modular cell–biomaterial assemblies will play in the upcoming years for building biofunctional microtissues with a truly pro-regenerative potential.
The current understanding that cells naturally self-organize into highly ordered multicellular structures which precede tissue and organ formation has laid the foundation for the development of advanced methodologies that aim to recapitulate the high cellular density of human tissues
Cell sheet technologies take advantage of close cell–cell interactions to autonomously engineer microtissues without the use of biodegradable cell-supportive scaffolds. As mentioned earlier, their presence can hinder proper cell–cell communication, spatial arrangement and its degradation byproducts can influence cell physiology.[23,24] Their degradation can also originate areas rich in ECM deposition that can hinder communication between neighboring cell clusters.[25] Native tissues and organs are densely populated by numerous cell types that are enclosed in a vast framework of tissue-specific matrix that allows efficient cellular intercommunication, directing fate and function at the microscale. This cell-rich 3D environment with well-orchestrated ECM presentation leads to higher order physiological function seen in living tissues.[26] Pursuing this design philosophy is essential for building up different types of tissues such as cardiac, renal, and hepatic that generally operate at high cell densities.[27–29
细胞片技术利用紧密的细胞间相互作用来自主设计微组织，而无需使用可生物降解的细胞支持支架。如前所述，它们的存在会阻碍正常的细胞间交流，其空间排列及其降解副产物会影响细胞生理。[23，24]它们的降解也可能源于富含ECM沉积的区域，这可能会阻碍相邻细胞群之间的交流。[25]本地组织和器官密集地分布着大量的细胞类型，这些细胞类型被包裹在组织特定基质的巨大框架中，允许有效的细胞相互通信，在微观尺度上指导命运和功能。这种细胞丰富的 3D 环境具有精心策划的 ECM 呈现，导致在活组织中看到更高阶的生理功能。[26]追求这种设计理念对于建立通常在高细胞密度下运行的不同类型的组织，如心脏、肾脏和肝脏是必不可少的。
In this context, cell sheets arise asscaffold-freehigh cell density microstructures that aim to recapitulate the contiguous assembly of cells seen in living tissues, thus attempting to retain its structural and functional cues.[30] Here, cells are cultured and proliferate in adhesive substrates until confluent layers are generated. These cell-rich sheets are then harvested with methodologies that should maintain cell sheet integrity and allow it to be easily transferred. Thus, cell sheet harvesting with proteolytic enzymes (e.g., trypsin, dispase, etc.) has been discontinued because these can affect cell integrity, as well as disrupt essential intercellular junctions and cell surface proteins.[31,32
The emergence of smart surface engineering refined the control on cell detachment and enabled cell sheet harvesting in mild conditions. In this sense, Okano ’s group has popularized cell sheet applications by designing a temperature-dependent harvesting substrate, which is commercially available under the brand name UpCell.[27,33] The poly(N-isopropylacrylamide) coating undergoes a sharp change in wettability from 37 to 32 °C, which spontaneously peels the cell sheet from the surface at room temperatures.[23,25] Still, this one-time use system is expensive, detachment time may vary significantly, and it can be difficult to handle due to the low range of temperature that triggers the peeling process.[34,35] Thermally expandable hydrogel sheets or multilayered coatings are being explored as an alternative to this technology that facilitate transfer and stamping of cell sheets with spatially controlled cell adhesion.[36–38] Other sophisticated harvesting systems have been developed over the years exploring different cell-friendly stimuli, including electroactive substrates, pH-responsive coatings, and photoactivatable surfaces, that change their wettability on-demand.[39,40] In this context, hematoporphyrincontaining films for light-induced cell sheet harvesting of human-bone-marrow-derived mesenchymal/stromal stem cells (MSCs) were successfully developed.[41] Light-responsive titanium dioxide (TiO2) nanodots films also offer spatial control over cell detachment and can be used for aligning cell sheets in predetermined directions.[42] Alternatively, magnetic forces can hold cells in place until harvesting is intended.[6] Magnetite cationic liposomes modified with RGD (arginine-glycine-aspartic acid) cell-adhesive peptide were used as smart coatings for developing fibroblast cell sheets, which were stabilized under a magnetic field and could be moved as a contiguous microtissue on-demand.[43] Also, cells internalizing magnetic nanoparticles can be forced to assemble over different surfaces to create cell sheet-like structures.[44] More recently, Park and co-workers embedded magnetic nanoparticles in thin hydrogel sheets for efficient harvesting of endothelial progenitor cell sheets.[45] Using a bioinspired approach, researchers have also developed a cellulose-dopamine coating which enabled cellulase-assisted enzymatic harvesting with minimal cell damage as recently reported.[35
智能表面工程的出现改进了对细胞分离（脱离）的控制，使细胞片能够在温和的条件下收获。从这个意义上说，Okano的团队通过设计一种依赖温度的收割基板来推广细胞薄片的应用，其品牌是UpCell。[27，33]聚（N-异丙基丙烯酰胺）（ PIPAAm）涂层的润湿性在 37 至 32 °C 时发生急剧变化，在室温下会自发地将细胞片从表面剥离。 [23,25]尽管如此，这种一次性使用的系统还是很昂贵，脱离时间可能会有很大差异，而且由于引发剥离过程的温度范围较低，可能难以处理。[34，35]热膨胀水凝胶片或多层涂层正在被探索作为该技术的替代技术，其促进具有空间控制的细胞粘附性细胞片的转移和冲压。[36-38]多年来，已经开发出其他复杂的采集系统，探索不同的对细胞友好的刺激物，包括电活性基质、pH 响应涂层和光活化表面，它们可以按需改变它们的润湿性。[39，40]在本文中，成功地开发了用于光诱导人骨髓间充质/基质干细胞(MSCs)的细胞片采集的含血卟啉的薄膜。[41]光响应性二氧化钛(TiO2)纳米点膜还提供了对细胞分离的空间控制，并可用于在预定方向上排列细胞片。[42]或者，磁力可以使细胞保持在原位，直到打算采集为止。[6]用RGD(精氨酸-甘氨酸-天冬氨酸)细胞粘附肽修饰的磁性阳离子脂质体被用作开发成纤维细胞片的智能涂层，它们在磁场下稳定下来，可以根据需要作为连续的微组织移动。[43]此外，内含磁性纳米颗粒的细胞可以被迫在不同的表面上组装，形成细胞片状结构。[44]最近，Park和他的同事将磁性纳米颗粒嵌入薄的水凝胶片中，以有效地获取内皮祖细胞片。[45]利用生物启发的方法，研究人员还开发了一种纤维素-多巴胺涂层，如最近报道的那样，它能够在纤维素酶辅助下进行酶促收获，并将细胞损伤降至最低。
Monolayer cell sheets are built with millions of cells, but their sheet-like fragile structure is hard to manipulate and does not offer enough microtissue depth in comparison to thick native tissues.[46] However, researchers can assemble thicker constructs just by stacking cell sheets, taking advantage of the celldense vast ECM network that naturallyintertwines the different cell sheets into a contiguous and integral multilayered microtissue.[47] Their versatility extends beyond stacking single cell-type sheets, but also enables rich combinations of multiple cell types, thus more accurately mimicking heterogenous native tissues. The crosstalk signaling present in tissues plays a key role in influencing cell fate and potentiating biofunctionality.[48] Cells from closely related osseous tissues (i.e., periodontal ligament and jaw bone) assembled into a co-cultured cell sheet exhibited enhanced osteogenic potency and were more structurally similar to the native periodontal tissue in vivo.[49] Still, although some improvements can be achieved with randomized distribution of cell populations, reconstituting tissue-specific biological function relies on recapitulating its hierarchic cellular organization patterns. For instance, the liver’s ability to perform more than 500 different functions is associated with its repetitive functional units (liver lobules) that are organized in a hierarchic multiscale manner.[50] Precisely engineered culturing and harvesting substrates can direct the spatial arrangement of different cell populations in cell sheets design, contributing toward more complex and in vivo-like assemblies. In this context, straightforward approaches such as microcontact printing of ECM proteins using rationally designed stamps have been explored to enable spatial control over cellular adhesion layouts during cell sheet manufacturing.[51,52] By controlling nonadhesive and cell adhesive areas, researchers engineered hepatocyte modules surrounded by endothelial cells, achieving a co-patterned liver-like microtissue that maintained the ability to synthesize albumin and urea (Figure 2A,B).[53,54] In addition, other advanced methodologies (e.g., microfluidics chips and dielectrophoretic[55]/magnetic patterning[56]) that permit cell spatial organization were also explored for designing liver-like cell-rich lobules with biomimetic distribution[57] or for fabricating microvessel-like cell sheets that recapitulate key features of perfusable vascularized networks.[58] Apart from controlling cell distribution and harvesting in in vitro cultured substrates, cell sheets have also been recently processed as scaffolds-free 3D bioinks. Using an elegant approach, researchers used extrusion bioprinting to fabricate sheet-based constructs that showed an increased structural integrity in comparison to standard cellaggregates owing to their in vitro produced ECM.[59]
Still, engineering a functional microtissue requires more than just controlling the spatial location of different cells and cell types, but also cellular alignment. In fact, many tissues in the human body (e.g., skin, skeletal muscle, myocardium, brain, and cartilage) show regional and directional 3D anisotropy that is essential for their mechanical and biological functions.[64] For instance, the myocardium is assembled from multiple layers of cardiomyocytes aligned along several directions throughout the whole tissue in a 3D anisotropic organization that is required for efficient electrical propagation and synchronized contractility.[65] Researchers have demonstrated that layered cardiomyocyte sheets rapidly establish electrical communications via functional gap junctions, achieving 3D-like myocardial tissues with elongated cardiomyocytes and synchronized macroscopic pulsations.[66,67] These bioengineered microtissues possessed elongated cardiomyocytes resembling the native cardiac muscle, which maintained spontaneous pulsations for more than 1 year postimplantation.[31,68] Moreover, the cardiac microtissues show intrinsic angiogenic potential, which is essential for fostering favorable host tissue integration and preventing ischemia.[69,70] Alternatively, researchers have engineered a 3D anisotropic skeletal muscle tissue by stacking myoblast sheets with a defined alignment as seen in Figure 2C,D.[60] Interestingly, anisotropy from top layers was transferable to layers underneath due to the self-organization capacity of myoblasts. Also, cell sheets with different orientations were designed by stacking perpendicularly aligned differentiated myotube layers that do not self-organize, thus generating multioriented constructs.
Nevertheless, not all morphological aspects can be recapitulated by stacking cell sheets in a stratified manner. In fact, some architectural features of native tissues include tubular structures (e.g., trachea, blood/lymph vessels, or intestines) with specific 3D conformations and different cells at specific locations (i.e., wall versus lumen).[62] Inspired by organisms development, stacked cell sheet constructs can be maneuvered toward such configurations by twisting, rolling, or wrapping into desired tubular forms.[71] Multilayered cardiomyocyte sheets wrapped around a resected rat aorta formed a functional myocardial tube that integrated with the host tissue and exhibited well-defined sarcomeres with contractile behavior.[72] Their flexible nature can also be exploited for replacing the damaged epithelial lining in tracheas.[73] Vascular media and adventitia tubules have been assembled by rolling fibroblast and smooth muscle cell (SMCs) sheets, achieving 0.3 mm wall-thick constructs able to sustain supraphysiological mechanical stresses.[74] Other researchers have designed 3D anisotropic hMSCs tubules by rolling cell sheets cultured on aligned ECM substrates.[61] The tubular vascular grafts were then matured in static or dynamic (i.e., bioreactor) conditions. Ultrastructure analysis revealed distinct parallel grooves for bioreactor-matured tubules that fused in thicker walls (0.25 vs 0.17 mm in static samples) and exhibited mechanical and vasodilation features resembling the native arterial wall (Figure 2E,F). However, these studies do not include endothelial cells in the composition of the tubules, a critical aspect since these structures play a key role in their native counterparts.[74] T o address this, researchers have developed 3D macroscopic tubular tissues by a stress-induced rolling of cell sheets containing arrangements of endothelial cells, SMCs and fibroblasts in a layered fashion (Figure 2G,H).[62] Moreover, by controlling cell orientation in the 2D surface template, they could fabricate tubules with circumferentially and longitudinally oriented SMCs, thus mimicking the anisotropy seen in native tunica media and adventitia. Aiming to achieve more complex architectures, self-folding co-cultured cell sheets were obtained by culturing cells in origamiinspired micromolded alginate substrates that release cell assemblies upon enzymatic degradation with alginate lyase.[75] The resulting dodecahedron microstructures give rise to 3D co-culture cell-rich assemblies via a self-folding process mediated by cell–cell traction force (Figure 2I,J)
It is important to note that thick microtissues (i.e., >50–100 µm) often become necrotic before sufficient neovascularization develops, which can be overcome by incorporating endothelial cells, either in co-culture or as individual layers, capable of forming microvascular networks.[76] Moreover, laser-assisted bioprinting can be used for guiding human endothelial cells to form tubule-like structures on top of cell sheets.[77] Micropatterning of electrochemical-responsive substrates has also allowed the harvesting of human umbilical vein endothelial cells (HUVECs) in the form of capillary-like luminal structures.[78] Previous studies also demonstrated that functional vascularization improves integration with host tissues.[79] The characteristic intercapillary network of specific tissues is a key parameter to take into account in the bottomup design process.[80] For instance, the heart has a narrower intercapillary distance (<25 µm) than other tissues. Hence, when assembling cardiac tissues, it is important to optimize the ratio of noncardiomyocyte cells as they can affect the contractility of the engineered 3D construct and possibly cause arrhythmia.[76,81
Overall, multilayered cell sheet constructs can be stacked on top of one another to engineer specific stratifications seen in hierarchic organs. Cell sheets can also be deformed into tubular shapes that are prominent in certain tissues, as well as the vast vascular network. When in contact with vascular beds or host organs, cell-rich sheet constructs naturally integrate onto tissue interfaces and develop functional intervascularization. Over the last decade, cell sheets have been bioengineered into different tissues, ranging from small vessels, cardiac microtissues, hepatic-like lobules, skeletal muscle, cornea, and others.[24] Functional prevascularized and perfusable cardiac microtissues have been successfully developed, but the stacking process is still manual and operator dependent.[82,83] Recently, an advanced automated cell sheet stacking technology was developed by Okano ’s group, and is envisioned to shift current cell sheet designs toward large scale manufacturing and translation into clinical applications.[47] Adding to engineered cell sheets, other scaffold-free assemblies such as 3D multicellular aggregates (e.g., spheroids,[84] fibers,[85] dense membranes, etc.)[86] have been receiving an increasing focus for bottom-up tissue engineering and for establishing organotypic preclinical disease models
Cell-rich 3D aggregates are valuable building blocks for fabricating organotypic microtissues owing to their closer correlation to living organs gene expression patterns,[87] multidimensional cell–cell interplay, and pH/nutrient/oxygen diffusion gradients.[88] Engineered multicellular 3D constructs better recapitulate these intrinsic functions since cells are immediately driven toward 3D-like microaggregates in in vitro culture platforms, in contrast to the initial 2.5D monolayer setting of cell sheets.[89–91] The resulting structures display a more realistic physiological response at early timepoints and intrinsically include microenvironment specific (bio)chemical/physical cues that support their biological performance.[91
富含细胞的3D聚集体与活体器官基因表达模式、多维细胞-细胞相互作用和pH/营养/氧气扩散梯度更密切相关，因此是构建器官型微组织的有价值的构建块。[88]设计的多细胞3D结构更好地概括了这些内在功能，因为在体外培养平台中，细胞被立即驱动向类似3D的微聚集体，而不是最初的2.5D单层设置。[89-91]所得到的结构在早期时间点表现出更真实的生理反应，并且本质上包括支持其生物学性能的微环境特定(生物)化学/物理线索。
💡
意思是，在体外培养平台 ，除非特殊表面粘附处理，一般会直接成球或其它3D聚集体。
Cell-rich 3D clusters fabrication takes advantage of intercellular adhesion mechanisms (e.g., cadherin and integrin-mediated)[92,93] to create self-assembled structures comprising single, or multiple cell types, unlocking the potential to fabricate heterotypic cell constructs more similar to the cellular heterogeneity of human tissues.[94,95] Upon in vitro maturation, these seamless 3D microaggregates secrete de novo ECM frameworks in which cells reside and exchange interactions, adding to their biofunctionality and biomedical applicability. To date, numerous technologies have been developed for the rapid manufacture of scaffold-free 3D cellular aggregates via culture and proliferation in nonadhesive setups, including forced floating and hanging drop platforms, multiarrayed micromolds, and microfluidic chips.[96,97] These well-established methods have been employed by researchers for high-throughput generation of multicellular microaggregates exhibiting rod-, toroidal-, or honeycomb-like architectures, as well as spherical morphologies (i.e., 3D spheroids, Figure 1).[97–99
富含细胞的3D簇制造利用细胞间黏附机制(例如，钙粘附素和整合素介导的)[92，93]来创建包含单个或多个细胞类型的自组装结构，从而释放了构建更类似于人类组织细胞异质性的异型细胞结构的可能性。[94，95]在体外成熟后，这些无缝的3D微聚集体分泌新的ECM框架，细胞在其中驻留和交换相互作用，增加了它们的生物功能和生物医学适用性。迄今为止，已经开发了许多技术,通过在非粘附设置中培养和增殖来制造无支架的 3D 细胞聚集体，包括强制浮动和悬滴平台、多阵列微模具和微流控芯片。 [96,97] 这些成熟的方法已被研究人员用于高通量多细胞微聚集体的生成表现出杆状、环形或蜂窝状结构，以及球形形态（即 3D 球体，图 1）。 [97-99
In particular, self-assembled 3D spheroids have rapidly arisen as attractive cell-rich unitary building blocks for recapitulating in vivo organs functional units since 3D multicellular aggregates with spherical shapes are also observed during tissue morphogenesis.[100] 3D spheroids exhibit tissuespecific features, shape reproducibility, size versatility, ease of handling, bioprocessing, and potential for upscaled production.[97,101] The latter poses a critical aspect when researchers envision the clinical application of cell-dense multiscale constructs that require millions of 3D spheroidal functional units.[88,102,103
特别是，由于在组织形态发生过程中也观察到具有球形形状的 3D 多细胞聚集体，自组装 3D 球体已迅速成为具有吸引力的富含细胞的单一构件，用于概括体内器官功能单元。 [100] 3D 球体表现出组织特异性特征、形状再现性、尺寸多功能性、易于处理、生物加工和大规模生产的潜力。 [97,101] 当研究人员设想需要数百万个 3D 的细胞密集多尺度结构的临床应用时，后者是一个关键方面球形功能单元。[88,102,103
💡
3D球体优点：组织特异性特征、形状再现性、尺寸多功能性、易于处理、生物加工和大规模生产
To date, advanced monotypic (single culture) and heterotypic (co-culture) scaffold-free 3D multicellular spheroids have been developed as in vitro microphysiological systems for modeling pathophysiology of human diseases or as unitary building blocks for tissue engineering of cardiac, hepatic, vascular, or neuronal tissues, among others.[89,90,104] In the context of tissue engineering and regeneration, 3D spheroids have been employed as angiogenic stimulating units (i.e., via secretion of trophic factors—VEGF, PDGF)[93] or as functional blocks for the assembly of prevascularized microtissues.[90] Progresses in this field indicate that 3D spheroids comprising heterogeneous cell populations better recapitulate the complexity of human tissues and exhibit a more pro-regenerative capacity. In a recent study, 3D co-culture spherical constructs comprising HUVECS and human bone marrow MSCSs (hBM-MSCs) were fabricated for improving bone regeneration (Figure 3A). Upon implantation into chick femur bone defects endothelial-skeletal 3D clusters improved collagen II and angiogenic proteins expression in the osteogenic niche. More importantly, an increased mineralization and bone volume deposition was obtained for 3D HUVECHBMSCs co-culture spheroids when compared to the sham defect group (Figure 3A).[105] The establishment of functional tubular-vessel-like networks with positive blood perfusion have also been reported in co-cultured 3D spheroids comprising human osteoblasts (hOB) and human dermal microvascular endothelial cells (HDMECs) upon implantation into mouse dorsal skin models.[106] This evidences the tissue integrative properties of prevascularized 3D spheroid building blocks and the importance of taking this into consideration when engineering implantable cell-rich 3D assemblies. By using a similar strategy, 3D MSCs–OECs (outgrowth endothelial cells) co-culture spheroids were established in nonadherent agarose micromolds to function as elementary units for prevascularized microtissues formation. The established 3D spheroids demonstrated a significant pro-angiogenic potential in in vivo chick chorioallantoic membrane (CAM). assays, accompanied by close tissue integration, and their culture under xeno-free conditions potentiates their applicability in bottom-up tissue engineering (Figure 3B).[107] It is worth to reference that in vitro culture of cell-rich assemblies under xeno-free conditions is an important parameter for cell-based therapies application in a more realistic clinical setting.
Despite their recognized validity for regenerative medicine applications, these examples mainly employ nonlinked, individual spheroids. Also, the spherical morphology restricts the geometrical complexity of fabricated 3D multiscale microtissue assemblies.[89] For bioengineering larger and more intricately patterned tissues, 3D cell-rich spherical aggregates must be processed through advanced methodologies.[93] In this sense, researchers are advancing 3D spheroids postproduction bioprocessing by using microplatforms that promote spheroid-spheroid fusion and by taking advantage of automated biofabrication technologies (e.g., 3D bioprinting) that build-up spheroid-based living architectures with user-defined geometries.[102
尽管它们对再生医学应用的有效性得到认可，但这些示例主要使用非链接的单个球体。此外，球形形态限制了制造的 3D 多尺度微组织组件的几何复杂性。 [89]对于更大和更复杂图案组织的生物工程，必须通过先进的方法处理富含 3D 细胞的球形聚集体。 [93]从这个意义上说，研究人员正在通过使用促进球体-球体融合的微平台并利用自动生物制造技术（例如 3D 生物打印）来推进 3D 球体后期生产生物处理，该技术可以构建具有用户定义几何形状的基于球体的生活架构。
Spheroids fusion occurs when spatially adjacent spheroids establish physical contact and coalesce into a more cohesive single tissue, a spontaneous process that occurs during biological development and that is fundamental in myocardial and skeletal tissues formation.[102,108] This process is then followed by cell self-organization into distinct layers within the resulting microtissues, similarly to what occurs during human organs development.[108] Also, taking advantage of nonadhesive poly(dimethylsiloxane) (PDMS) microchamber arrays, the fabrication of robust neurospheroid networks was established through fused neuronal processes protruding from spatially adjacent 3D neurospheroids of rat cortical cells (Figure 3C). During optimization, the authors discovered that PDMS microchambers with 100 µm diameter and depth were optimal for promoting oxygen diffusion to 3D neurospheroids and for assuring constructs viability.[109] The interconnected/fused 3D spheroid neuronal network was then implanted in cortical tissue and exhibited excellent integration into cortical tissues as observed by the establishment of functional synaptic connections with host neurons. Random 3D microscale spheroids fusion has also been employed for creating 3D millimeter-scaled tissues with complex geometries (e.g., star, square, triangle, etc.) in nona dherent micromolded platforms (Figure 3D).[110,111] The microfabricated tissues comprising co-cultured HUVECs and MSCs 3D spheroids as unitary building blocks showed autonomous deformation, contractility and formed self-assembled vascular structures in high deformation regions.[111] This resulted in spatially modulated secretion of pro-angiogenic growth factors and specific vascular patterns.[111
To achieve more precise spatial positioning and user definedpatterns, 3D spheroids can be processed though 3D bioprinting/robotic assembly or magnetic-based manipulation.[2] In this sense, Mironov’s group bioprinted multicellular 3D spheroids into customized millimeter-sized vascular tubular constructs with well-defined topology and cellular composition as a result of spheroid-spheroid fusion (Figure 3E).[112] This groundbreaking approach, alongside with Kenzan ’s method[113] for bioprinting 3D spheroids (up to 500 µm) into needle arrays which are removed after spheroids fusion, are envisioned to contribute for the fabrication of evermore complex cell-rich biospecific constructs with controlled architectural features. Yet, to date, such strategies still require specialized equipment, extensive troubleshooting, and the maturation of bioprinted microtissues along time. Alternatively, 3D spheroids magnetic manipulation using preprimed cells with magnetoferritin nanoparticles has also been explored to generate structurally defined microtissue rings for tissue engineering, but the biofunctionality and long-term safety of such constructs remains to be elucidated.[114,115] Adding to the opportunities that arise from the establishment of dense 3D spheroid assemblies the development of 3D multicellular organoids in vitro is also becoming increasingly relevant in numerous fields of research including tissue engineering and regenerative medicine, disease modeling, and drug discovery.
Human 3D organoids are highly relevant as building blocks for bottom-up tissue engineering since their assembly is reminiscent of tissues organogenesis and morphogenesis. Generally, 3D organoids are composed of stem cells (i.e., human adult/ embryonic stem cells, or human pluripotent stem cells (hPSCs)) that undergo in vitro proliferation, directed differentiation, cell-sorting, lineage commitment, and self-assembly into higher order 3D architectures.[118] Interestingly, upon in vitro maturation, 3D organoids generate highly organized cell-rich structures that somewhat recapitulate the complex structural features and physiological responses of the tissues their cells were derived from.[119,120] Despite their organotypic features it is important to emphasize that current organoidgenesis in vitro is only possible under a highly controlled microenvironment often including a complex cell culture medium that contains multiple factors (i.e., small molecules and growth factors) for precisely guiding cellular differentiation and also an ECMmimetic hydrogel to further provide bioinstructive cues (i.e., Matrigel).[118] T o date, these multicellular systems have been increasingly employed in advanced bottom-up engineering of numerous tissue types such as kidney, liver, or pancreas. In fact, researchers have been recently able to develop pancreatic islet organoids from human embryonic stem cell derived pancreatic progenitor cells.[121] The cells spontaneously aggregated in vitro under controlled culture conditions and formed robust multicellular spheroid-shaped organoids with controlled size and cellular heterogeneity. Upon maturation the resulting 3D organoids exhibited functional insulin secretion after a glucose challenge. It is important to emphasize that such organoids were assembled using a unique hydrogel platform (i.e., Amikagel) and that the development of biofunctional organoids in fully scaffold-free conditions remains to be demonstrated to the best of our knowledge.[121] There is no doubt that these are highly promising microphysiological constructs, however most in vitro generated organoids still generally lack key cellular constituents of the tissues they aim to recapitulate. In this context, researchers are actively developing new strategies for including stromal, immunological, and vascular components in these 3D multicellular assemblies.[120,122] It is envisioned that advances in these important aspects will have a significant impact in the development of more physiomimetic building blocks for bottom-up tissue engineering.
Fiber-shaped constructs are highly valuable for bottom-up tissue engineering applications due to their ease of processability into higher order living architectures and similarity to multidirectional native anatomic structures, including blood vessels, neurons, lymph vessels, ligaments, and tendons.[123] In fact, the exacerbation of the length of one single dimension in fiber-shaped materials enables the spatially unconfined deposition of continuous structures. This unique potential has been leveraged by the rapidly emerging field of 3D printing/ bioprinting which explores the potential of long fiber-shaped materials to create personalized implants and tissue engineered cell-laden constructs with highly specific architectures and functionality.[124–126] Moreover, know-how from the textile industry, namely, knitting, weaving, and reeling has been transposed for the bottom-up tissue engineering of fibrous structures due to its potential for generating biofunctional micro/macroscale constructs with improved bioactivity and mechanical properties.[127] Interesting advances in this field include, but are not limited to, the fabrication of cell-laden hydrogel yarns for tendon bioengineering through mechanical stimulation[128] or the fabrication of vessel-like networks with cell-laden collagen fibers.[129
Despite the versatility of fiber-shaped constructs, few reports describe the fabrication and exploitation of fully scaffold-free cell-rich fibers. In this context, T akeuchi ’s group pioneered the production of meter-long cellular microfibers.[130] By using a double-coaxial microfluidics device, researchers were able to fabricate core–shell fibers comprising alginate crosslinked with calcium ions (diameters ranging from 100 to 200 µm) (Figure 4A). Such hollow structures were laden with mixtures of cell suspensions and ECM proteins, the latter capable of forming gelled structures at in vitro cell culture temperature conditions or in the presence of thrombin and maintained high cell viability. Cells from different sources including fibroblasts, cardiomyocytes, endothelial, and pancreatic cells could be processed in the shape of long fibers in the presence of specifically selected and optimized ECM proteins. In this approach, the alginate shell was easily removed from fiber constructs formed by fibroblasts or MIN6m9 cells (a mouse pancreatic beta cell line) using alginate lyase. Importantly, the generated cellular microfibers maintained shape and structural integrity after this process. Although this study represented a breakthrough in the use of elongated cellular fiber-shaped materials, the use of relatively stiff gel-forming ECM proteins as a determinant factor for the success of fibers’ formation did not warrant a fully scaffold-free character to the achieved microfibers. Moreover, the assembly of microfibers as higher order constructs was restricted to alginate-coated units, while the handling of cell/ ECM-only structures may pose additional challenges to the precise assembling of macrometric constructs.
In an effort to surpass the possible toxicity of enzymemediated methods to degrade alginate reported in primordial studies,[130] a method based on mechanical action to retrieve fibers composed of bone marrow MSCs and collagen was suggested.[134] While this study avoided the use of possibly toxic enzymes, the rapid formation of the cellular fibers (24 h) was still dependent on the addition of ECM proteins. Also based on the co-axial microfluidic-based extrusion of cell suspensions inside hollow alginate tubes, other researchers reported the formation of scaffold-free centimeter-long microfibers built solely from primary human chondrocytes (Figure 4B).[131] After four days of incubation and cellular aggregation, alginate tubes were removed by a postprocessing technique that involved their dissolution in sodium citrate. Cellular strands matured in vitro for 3 weeks in chondrogenic medium could be bioprinted into macrometric constructs with clinically relevant sizes. Interfiber fusion was observed as early as 12 h of contact, and complete merging was obtained after 7 days of in vitro maturation.
Alternative approaches to microfluidic-based fabrication of cellular fibers were suggested based on the patterning of growth channels prepared by laser micromachining[135] and more recently, by micromolding.[85] These strategies have been exclusively directed for the preparation of tendon single fibers, and are highly dependent on specific equipment and selective patterning of adhesive proteins into nonadhesive agarose molds.
The scaffold-free generation of fiber structures via 3D bioprinting as also been recently materialized through the extrusion of hMSCs cell-rich bioink into a medium containing alginate microgel particles (Figure 4C).[132] T o generate the supporting baths, microgels with different average sizes (7 and 409 µm) were assembled via dual crosslinking of methacrylated and oxidized alginate (OMA), and subsequently crosslinked with calcium. This disruptive approach allowed high-resolution bioprinting of hMSC-rich 3D filaments even with curved, corner, and X-shaped configurations due to the shear-thinning and self-healing properties of the engineered supporting bath, and regardless of microgels size (Figure 4C). The 3D bioprinted cell-rich filaments presented high cellular viability postprinting and the smaller microgels bath allowed to obtain maximum printing resolution of hMSCs filaments as exhibited by the 3D printed anatomic shaped constructs (Figure 4C). T o confer mechanical cues and stability during microtissues maturation postprinting the OMA bath was photocrosslinked. This allowed long-term culture and fusion of stem-cell-rich aggregates into denser 3D microtissues, as well as maintained their differentiation potential toward osteogenic and chondrogenic lineages.[132]
Nevertheless, most approaches targeting the preparation of microfibers are still focused on the achievement of multicellular bulk structures. A high degree of cellular compaction, however, may lead to poor oxygen and nutrients diffusion during maturation periods, culminating in loss of cellular viability and function in fibers’ core regions. T o tackle this challenge, microporous tissue strands were prepared by adding a porogen agent—alginate microbeads—to cellular suspensions before extrusion inside a hollow alginate tube (Figure 4D).[133] Fibers comprising adiposederived MSCs (ASCs) generated by this methodology showcased approximately 25% porosity and high pore interconnectivity. When compared to bulk nonporous strands, porous counterparts showed improved long-term in vitro cellular viability, as well as higher efficacy on the induction of both the osteogenic and chondrogenic differentiation on the aggregated cells (Figure 4D).
An overall analysis of current technologies available for the fabrication of scaffold-free cellular fibers enabled identifying opportunities in this rapidly growing, yet still poorly explored field. In this context, fiber-based textile engineering processes are particularly interesting for tailoring the mechanical properties (e.g., tensile stress/strength, Young’s modulus, and elongation at break) of multifiber assemblies to better recapitulate the biomechanical features of native tissues.[136] Relevant examples of textile techniques for such applications include knitting (e.g., allows control over interfiber porosity, higher through-plane strength, and stretchability), braiding (e.g., tailoring of load-bearing properties, tensile strength, and abrasion resistance), or the less mechanically demanding weaving process (e.g., tunable mechanical anisotropy and allows improved control over cell distribution), and winding of fibers onto 3D tubular constructs.[137,138] Furthermore, architectural manipulation of multiple fibers, each containing HUVECs, fibroblasts, or hepatocytes, allowed different cells to be interfaced in predefined patterns across braided assemblies.[138] The braiding process resulted in vastly different mechanical properties, with differences on the range of three orders of magnitude.[138] Future approaches based on the knitting and/or printing of co-cultured fibers or fibers comprising different cell types (or in distinct stages of differentiation/maturation) may pave the way for the fabrication of architecturally precise and hierarchic constructs that better recapitulate the features of human tissues with fibrillar structures.
From the former examples it becomes clear that the assembly of 3D cellular aggregates into higher order structures provides an exciting approach to generate cell-dense tissues from the bottom-up. However, from a critical perspective, cells selfassembly in both 2.5D sheets and 3D multicellular aggregates (e.g., spheroids, fibers, etc.) still occurs in a stochastic mode that does not fully recapitulate the highly controlled process of tissues morphogenesis at a unicellular level. Aiming to address this uncertainty, cutting-edge technologies for bottomup tissue engineering have focused on the synthetic engineering of cell membrane surface to imprint biospecific cues and preprogram cell-rich scaffold-free assemblies with precise cell–cell selectivity and spatiotemporal organization. In the following section, we highlight these emerging paradigms and present the most recent advances of these approaches.
In the context of bottom-up tissue engineering, the manipulation of cell surface to promote a programmed self-assembly into higher order 3D bioarchitectures, represents a powerful approach for assembling unitary cellular building blocks into highly controlled microtissues (Figure 1). Although cells can be coated with hard shells[139] we will mainly focus on soft surface modifications[140] and on precisely programmed surface engineering[141] allowing cells to be molded and organized into more complex structures.
Currently, cell surface modification is facilitated by the knowledge of various bioconjugation chemistries together with synthetic biology approaches (e.g., genetic engineering) and other methodologies that include the inclusion of multifunctional micro- and nanosized agents capable of connecting multiple cells. Adding to this, the vast array of available biorthogonal coupling chemistries, as well as their specific binding nature, allow for the construction of surface modified cellular modules leading to precise hierarchic architectures.[142] Because different cell types can be functionalized differently, researchers can control the cellular spatial arrangement in bottom-up assembled constructs.
In particular, cell surface glycoengineering arisen as an elegant metabolic cell engineering concept that exploits intrinsic sialic acid biosynthesis pathways in order to install reactive functional groups at cells surface. Cell functionalization takes place upon unnatural monosaccharides uptake and incorporation within the sialoglycan metabolism, which is responsible for continuous remodeling of natural sialic acid end-capping residues in live cells.[143] Bertozzi ’s group pioneered this biotechnological cellular hijacking tool by exposing at the cell surface biorthogonal ketone and azide groups available for oxime and click-chemistry conjugations, respectively.[144,145] Since then, advances in biorthogonal chemistry have presented this field with additional sialoglycan-compatible groups such as biotin, alkyne, alkene, and thiols. More recently, other innovative strategies incorporating arylazide photocrosslinkers, bifunctional sialoglycan analogues (e.g., azide-alkyne) and caging groups capable of neutralizing surface negative charge and inducing cell aggregation, have been reported.[143,146,147]
Among these strategies, azide-glycoengineering is by far the most explored alternative, owing to its high in vivo chemical stability, unique presence in the body, as well as the growing popularity of strain-promoted copper-free azide-alkyne cycloaddition (SPAAC) reactions with fast kinetics at physiological conditions under the absence of catalysts.[148] Still, a lack of cell selectivity in co-culture conditions is one of glycoengineering main limitations, but this can be overcome via delivery with cell-penetrating and targeted nanoassemblies (e.g., liposomes). In line with this, researchers were able to improve cell-selectivity and surface engineering efficacy by administering ligand-targeted liposomes loaded with azidosugars.[149] This allows for precise cell attachment of imaging/therapeutic agents, biomaterials, nanocarriers, or even other surface-modified cells.
Unsurprisingly, glycoengineering represents an enabling technology for various biomedical applications, namely, both live cell,[150,151] and extracellular vesicle labeling/tracking,[152] drug-based cancer theranostics,[153] as well as cell-based chemo- and immunotherapies.[154,155] Despite an impressive body of literature on these recent advances, tissue engineering concepts exploiting this biomachinery phenomenon are still at their youth. Glycoengineering of typically nonadherent human Jurkat cell surfaces imparted them with extra free thiol groups, thus stimulating spontaneous self-aggregation into clusters.[156] Importantly, this can allow allocation of typically nonadherent cell lines (e.g., immune cells) in bottom-up assemblies of heterogenous cell-rich building blocks as it will be further discussed. Alternatively, coupling of immunostimulants to live cells has recently been reported via a glycoengineering strategy.[154] The concept of incorporating immune cells in engineered assemblies, i.e., immunoengineering, is on the forefront of developing clinically relevant immunomodulatory tissues.[157] T ranslation of this technology to other cell types in the future, could be exploited for fast-generating spheroid building blocks or allowing more sophisticated and hierarchic assemblies via thiol-reactive incorporation (e.g., alkyne, norbornene, maleimides, or photocrosslinkable alkenes). Cells with norbornenebearing surfaces have now been successfully generated, which could allow for light-enabled precise spatial control of cell–cell interactions.[158] However, although norbornene-sugar incorporation efficiency is still limited in present attempts, other photocrosslinkable groups have shown promising results (i.e., acrylamide).[159] Cells exhibiting triple-orthogonal surface engineering have also been reported, which could allow for more complex heterogenous configurations of cell–cell interactions.[159] Alternatively, cell surface-grafting strategies are starting to emerge, either as controlled radical polymerization or anchoring telechelic synthetic polymers, in order to re-engineer cell surface functionality and interactions.
However, surface modification can elicit possible phenotypic and biofunctional alterations that are yet to be fully understood.[156] The usage of smart biodegradable cell linkages with on-demand dynamic anchoring of cells may help overcoming some of those concerns. With this rationale, a biocompatible and chemically detachable cell glue system based on biorthogonal linkers connecting glycoengineered cells was successfully developed.[162] Azide-containing cells were functionalized with bioreducible tetrazine and trans-cyclooctene linkers containing internal disulfide bonds, achieving fast cell–cell gluing upon mixing the two different cell groups. Then, the established cell glue network could be disassembled after natural glutathione (5 × 10−3 m) administration, showcasing the reversible glue behavior (Figure 5A). Another elegant way of obtaining on-demand control over cell–cell interactions was recently achieved by using bioengineered azide mammalian cells surface to contain β-cyclodextrin, which forms an inclusion complex with trans-azobenzene via host–guest interactions.[163] The azobenzene trans-to-cis conversion can then be triggered by photoactivation, dissociating the inclusion complex in a reversible manner. Using this rationale, photoactive azobenzene functionalized with cell recognition moieties (azo-aptamer) were used to enable cell–cell interactions between cyclodextrin-modified cells and azobenzene-bound cells, a process that could be easily reversed by light exposure. The photocontrolled manipulation of cell–cell interactions could also be useful in developing user-defined hierarchic cellular-rich assemblies.
This advanced technology has been explored in recent studies to direct drugs and nanoassemblies toward tumors, while also monitoring its progression.[166] Moreover, by exploiting the tumor-homing properties of MSCs, researchers have used glycoengineered MSCs as nanoparticle beacons, irreversibly trapping nanoconstructs in tumor sites.[150] Due to the high reactivity, bioorthogonal character and biocompatibility of this azide-DBCO chemistry, such rationale could be translated to glycoengineered tissue constructs (GTCs), where exposed azides allow postimplantation follow-up and signal for subsequent GTC-targeted nanoassemblies containing bioinstructive cues. Recent research efforts have achieved encouraging results supporting this novel strategy. For instance, glycoengineered azidechondrocytes could be readily tracked via near-infrared (NIR) imaging 4 weeks after subcutaneous implantation in mice.[167] Moreover, DBCO-650-labeling not only provided improved contrast imaging, but also preserved chondrogenic potency and showed minimal adverse effects on cartilage formation over traditional cell tracking fluorescent probes (i.e., DiD). In a similar approach, the migration of glycoengineered human ASCs (hASCs) in ischemic hindlimb mouse model was also tracked.[151] Again, NIRF-labeling via DBCO-Cy5 probe showed notable biocompatibility, in particular over DiD-based labeling, and allowed efficient in vivo monitoring of intramuscularly administered hASCs for 2 weeks, in particular their co-localization in ischemic sites. In a follow-up study, this technology was used to allocate glycol chitosan nanoparticles containing different imaging probes in hASCs.[168] However, rapid uptake of these clickable nanoassemblies was observed within 1 h of incubation. When envisioning the generation of nanoenabled cell clusters for tissue engineering, avoiding intracellular uptake is a critical parameter. T o this end, photoswitchable nanoparticles were covalently bound to living cell membranes via a similar glycoengineering strategy.[169] Site-specific membrane localization of upconversion nanoparticles in human embryonic kidney 293 (HEK293) cells was evident in fluorescence microscopy studies, which inspires future studies where spatial bioactive presentation can be tuned by external triggers and achieve localized differentiation in hierarchic constructs.
Tracking the fate of transplanted cells is essential for pursuing optimal tissue engineering applications, and this is facilitated in GTCs designs.[170] As demonstrated by Bertozzi ’s group, nonpenetrating azide-binding probes can enable 3D spatiotemporal in vivo imaging of developing zebrafish.[171] Because these probes are not transferable across cells, the addition of different reporters at various time intervals allow for differential labeling of tissue layers and display time-sensitive imaging of GTCs development. Another unique feature of glycoengineering is the sensitivity to different intracellular metabolisms that can distinguish cell subtypes and thus monitor stages of human breast cancer.[172] Similarly, incorporation of azido-sugars was significantly increased during cardiac hypertrophy.[173] Dysregulation of sialoglycan biosynthesis is also found in neurological disorders and central nervous system injuries.[174] These recent findings inspire novel applications in cell-based therapies and tissue engineering with innate pathophysiological-responsive monitoring.
Holistically, it is important to discuss the limitations of the aforementioned technologies in bottom-up tissue engineering and potential improvements. The dynamic nature of the cell membrane and its refined biomachinery does allow researchers to introduce superficial reactive moieties and use cells as building blocks in chemically driven interactions among other cells, biopolymers, and nanoassemblies. However, dynamic surface glycan recycling and inevitable cell division processes ultimately restrict the timeframe for appreciable surface reactivity. This issue raises concerns about whether long-term feeding of cells with unnatural sugars can lead to unexpected effects on cell and microtissue development.[175] Moreover, the importance of spacers on the stability of such cell-networks requires further studies. Indeed, the risk of using complementary cells for strain-promoted alkyne-azide cycloaddition with insufficient spacer length could result in membrane fusion among adjacent cells.[176] Hence, alternative cell surface engineering approaches using nanocarriers as crosslinking points have also been extensively investigated.
Nanosized fusogenic liposomes can efficiently fuse with cell membranes upon uptake, thus incorporating their phospholipidic content in living cell membranes.[177] This phenomenon has inspired Yousaf’s group to reprogram cell surfaces to display specific functional groups in order to accelerate 3D tissue assemblies.[178] This approach takes advantage of a pioneering biorthogonal chemistry based on ketone and oxyamine catalystfree conjugation with fast kinetics at physiological conditions and in the presence of serum.[17
Using this approach, researchers have reprogrammed nonadherent Jurkat cells with photolabile-oxyamine and ketone groups, achieving rapid multicellular 3D spheroid assemblies via intercellular oxime linkages.[164] Upon UV illumination, the microtissue disassembled into individual cells. In the same study, fibroblasts transformed with photolabile oxyamines readily adhered to aldehyde-containing interfaces. The reprogramed cells could then be selectively detached from the material upon UV illumination. In addition, large multilayered microtissues containing MSCs and fibroblasts sheets were easily assembled with this technology (Figure 5B). Photodisassembly of tissue multilayers can then be locally triggered resorting to photomasks during UV exposure, which enables the fine remodeling of microtissue hierarchy.
Liposomal fusion instructs chemoselective cell clustering, locking cells in place until sufficient matrix is produced and elicits microtissue formation and cell spreading. This oxime cell-coupling tool allows easy access to the generation of robust 3D modules with various geometries (e.g., circular, bar, and square) and accelerated assembly times.[179] Also, its fast kinetics have recently enabled in-flow spheroid and tissue assembly via microfluidics.[180] Cell cluster morphology and assembly times can be tuned by controlling flow rate, channel distance and cell density, and the cell modules are stable, thus not requiring seeding in Matrigel or other supporting ECM-mimetic scaffolds. Recently, scaffold-free tissue-like models (e.g., hepatic and cardiac) have been developed using this innovative tool.[181,182] Heterogenous cell populations can be rapidly programmed to self-assembly into dense multilayered tissue modules for biomedical applications. This flexible biotechnological tool is also compatible with current 3D bioprinting technologies where cells could act both as ink and glue, while biosacrificial layers could be achieved by including photo responsiveness within the oxime linkage, as demonstrated in the study by Luo et al.[164] It is important to note that the fusogenic capacity varies significantly among cell types.[183] Although this could represent a limitation to liposome-driven cell assembly strategies, this phenomenon could be exploited for differential cell modification in co-culture conditions. However, to date further studies are still required to ascertain the potential of this approach.
Alternatively, oligonucleotide-based technologies (e.g., DNA aptamers) have also been employed for establishing programmed cell–cell connectivity into higher order cell-rich 3D microtissue constructs. This approach is inspired by the orthogonal hydrogen bonding of nucleic base pairs naturally observed in DNA, an interaction that demands a specific template for complete binding of two complementary nucleotide sequences.[184] Such unique cross-reactivity has led to significant advances in other fields (e.g., DNA origami and patterning, synthetic nanopores, and molecular motors).[185] For bottomup tissue engineering, this selectivity has been materialized by anchoring single-stranded DNA (ssDNA) into living cells membrane to modulate cell–cell interactions between complementing aptamer sequences. This chemical reprogramming can be performed under typical cell culture conditions and does not require cells genetic manipulation. Such strategy has been explored in a seminal work, where ssDNA aptamer sequences were used as binding agents for building selective connectivity among cells. The complementary oligonucleotide sequences were introduced in different cells surface via selective chemistry between modified ssDNA aptamers (i.e., phosphine or difluorinated cyclooctyne) and glycoengineered nonadherent Jurkat cells decorated with azide moieties (via glycocalyx engineering with N-azidoacetylmannosamine sugars).[165] By using specifically matched ssDNA the authors were able to establish large 3D cell aggregates of typically nonadherent cells with DNA clustering being evident at cell–cell interfaces (Figure 5C). Interestingly, a precise control over cellular ratios (1:50) resulted in the formation of rosette-like microtissue assemblies with controlled cell neighboring (Figure 5C), confirming the DNA-mediated cell programming. Overall, these authors further demonstrated that the kinetic parameters of 3D microtissues assembly via defined cellular connectivity depend on DNA sequence complexity, density, and cell concentration.
或者，基于寡核苷酸的技术（例如 DNA 适体）也已被用于将程序化的细胞-细胞连接性建立到更高级的富含细胞的 3D 微组织结构中。这种方法的灵感来自于 DNA 中自然观察到的核酸碱基对的正交氢键，这种相互作用需要一个特定的模板才能完全结合两个互补的核苷酸序列。 [184]这种独特的交叉反应在其他领域（例如 DNA 折纸和图案化、合成纳米孔和分子马达）取得了重大进展。 [185]对于自下而上的组织工程，这种选择性已通过将单链 DNA (ssDNA) 锚定到活细胞膜中以调节互补适体序列之间的细胞-细胞相互作用来实现。这种化学重编程可以在典型的细胞培养条件下进行，不需要细胞基因操作。这种策略已在一项开创性工作中进行了探索，其中 ssDNA 适体序列被用作结合剂，用于在细胞之间建立选择性连接。通过修饰的ssDNA诱导体（即磷化或二氟化环辛烷）和用叠氮基团装饰的糖工程非粘附Jurkat细胞（通过用N-叠氮乙酰氨基糖的糖工程）之间的选择性化学反应，将互补的寡核苷酸序列引入不同的细胞表面。 [165]通过使用特别匹配的 ssDNA，作者能够建立典型的非贴壁细胞的大型 3D 细胞聚集体，其中 DNA 聚集在细胞-细胞界面处很明显（图 5C）。有趣的是，对细胞比例（1:50）的精确控制导致形成具有受控细胞相邻的玫瑰花状微组织组件（图 5C），证实了 DNA 介导的细胞编程。总体而言，这些作者进一步证明，通过定义的细胞连通性组装 3D 微组织的动力学参数取决于 DNA 序列的复杂性、密度和细胞浓度。
An important advantage of duplex DNA technology is the possibility to reverse DNA-mediated cellular assemblies into their unicellular building blocks via controlled melting or degradation.[165] This linkage reversibility can allow for cells selective isolation/purification,[186] or for templated cells inclusion in higher order ECM-mimetic structures as clusters which then disassemble and migrate to specific areas alike in some pathologies. This technology could therefore be of interest also for investigating fundamental cell migration studies. Using DNAprogrammed assembly of cells, it is also possible to incorporate components of the mesenchyme, such as fibroblasts, allowing the precise engineering of stem cell niches that capture stromal contributions. It is clear that this technology allows a precise control over individual cell–cell interactions and may enable direct examination and manipulation of juxtacrine cell–cell and cell–ECM interactions during tissue maturation
双链 DNA 技术的一个重要优势是可以通过受控的熔解或降解将 DNA 介导的细胞组装体逆转为单细胞结构单元。 [165]这种连锁可逆性可以允许细胞选择性分离/纯化，[186] 或模板细胞包含在更高阶的 ECM 模拟结构中作为簇，然后在某些病理中分解并迁移到特定区域。因此，这项技术也可用于研究基本的细胞迁移研究。使用 DNA 程序化的细胞组装，还可以结合间充质的成分，例如成纤维细胞，从而可以精确地改造干细胞生态位以捕获基质贡献。很明显，这项技术可以精确控制单个细胞-细胞的相互作用，并且可以在组织成熟期间直接检查和操纵近分泌细胞-细胞和细胞-ECM 相互作用
💡
这种利用核苷酸序列特异性结合还具有可逆解链（双链解链叫退火，或者其DNA变性条件）
These methods have unprecedentedly expanded our ability to precisely manipulate cells behavior or adhesion properties. Nevertheless, from a developmental biology perspective, multicellular self-assembly into complex tissue structures is driven mainly by genetically programmed routines activated at specific time frames to induce biologically relevant and dynamically orchestrated physiological responses.[187–189] Recent, approaches have therefore started to use synthetic biology tools to explore self-regulated cell–cell adhesion mechanisms in an attempt to better emulate in vivo biosystems features in in vitro bioengineered 3D microtissue assemblies. In fact, various genetically encoded circuits have already been developed for establishing complex biologically driven patterns at the cell population level. These include the encoding of: i) adhesion-driven assemblies (i.e., phase separation), of ii) lateral inhibition where neighboring cells have different fates, a characteristic process in vertebrates neuronal development, of iii) mechanically driven assemblies that can originate tree-like or fractal patterns, or the use of iv) “reaction-diffusion programing” of biological spatial pattern formation, also termed the “T uring/Gierer–Meinhardt model.”[190] The latter is based on activator–repressor species that govern spatial patterns formation via different diffusion rates. Such interactions are considered to be key in embryonic development and respond to changes in tissue size.[190] This mechanism is also particularly valuable since simple synthetically programmed genetic networks can originate complex and recurrent patterns which have the ability to self-regenerate when disturbed.[190] These strategies therefore add an attractive layer of complexity and dynamism to the formerly discussed cell surface engineering approaches and better recapitulate tissues morphogenesis
Taking inspiration on these processes, a recent study reported an elegant generation of genetically programmed multilayered 3D microtissue assemblies evocative of dynamic 3D structures observed during embryonic development.[141] Specifically, the authors designed genetic circuits that combined adhesion-driven (i.e., cadherins) and lateral inhibition via synNotch receptors at the cell–cell interface to create complex self-organized 3D microtissues with cell-signaling induced morphologic spatial rearrangements (Figure 5D). In turn, these dynamics generated new cell–cell interactions and reorganizations as a result of cell type diversification and asymmetry.
从这些过程中汲取灵感，最近的一项研究报告了一代优雅的基因编程多层 3D 微组织组件，唤起了在胚胎发育过程中观察到的动态 3D 结构。 [141]具体来说，作者设计了结合粘附驱动（即钙粘蛋白）和通过细胞-细胞界面处的 synNotch 受体的侧向抑制的遗传回路，以创建具有细胞信号传导诱导的形态空间重排的复杂自组织 3D 微组织（图 5D）。反过来，由于细胞类型多样化和不对称，这些动态产生了新的细胞间相互作用和重组。
It is clear that cutting-edge cell surface modifying technologies enable researchers to sculpt cellular behavior and allow the fabrication of precisely programmed cell-rich microtissues in in vitro culture. Yet, other biomaterial-based approaches are also moving bottom-up tissue engineering forward by imparting biointegrative/bioinstructive cues on cellular building blocks and by unlocking the build-up of highly organized cell– biomaterial assemblies. The following chapters will focus on different cell–biomaterial assemblies and their importance toward the creation of more in vivo-like microtissues
3.1 Macromolecular Cell Surface Functionalization via Layer-by-Layer (LbL)(通过逐层 (LbL) 实现大分子细胞表面功能化)
In recent decades, the LbL assembly technology has emerged as a simple, robust, and highly versatile engineering methodology to modify diverse inorganic and organic surfaces, including eukaryotic cells.[191] LbL technology has rapidly evolved to become a well-established tool for functionalizing surfaces,[192] and fabricating elegant and stable electrostatic, and nonelectrostatic-driven multilayered cellular-rich architectures with multiple functionalities and advantages.[193] Its simplicity and cost-effectiveness (no specific or expensive equipment is required), as well as mild processing conditions turned it into a powerful technology for cell surface functionalization in the context of bottom-up tissue engineering strategies. In fact, the LbL assembly process can be performed under physiological conditions entirely in aqueous solutions. This means that there is no need for the use of organic and/or harmful solvents or extreme pH, ionic strength, and temperature conditions, thus turning it into a very appealing tool when dealing with biomolecules which have not only limited solubility in nonaqueous solutions, but are also highly prone to lose their biological activity. In addition, it is a highly versatile technology both in terms of the substrates, building blocks, and intermolecular interactions that can be used to fabricate simple, organized, as well as more intricate architectures.[194–198] Moreover, an unprecedented source of building blocks including biological materials such as nucleic acids, enzymes and other proteins, peptides, polymers, viruses, or even cells can be used to functionalize the substrate surface, provided that the individual constituents show complementary interactions. Such versatility enables the fabrication of a plethora of continuous, molecularly uniform, and scalable thin films, as well as multilayered surfaces and devices with precisely tailored physicochemical, mechanical and biological properties, including multilayered thin films, free-standing multilayered membranes, core–shell particles, hollow multilayered capsules, or even 3D tissue-like structures across multiple length scales.[192] This is a key advantage over monolayer-based systems, thus turning LbL surface engineering technique into a highly suitable and powerful technology for a wider range of biomedical applications, including tissue and cell surface engineering, enabling a high degree of control over cell–surface, cell–biomaterial, and cell–cell interactions.[199–201]
However, notwithstanding the tremendous progress, only in the last few years the LbL assembly technology has extended well-beyond its importance for the functionalization of hard and soft inanimate charged surfaces, proving to be a suitable bottom-up strategy for functionalizing animate and dynamic surfaces, including living cells.
Such multilayered assemblies are intended to engineer hierarchically ordered 3D cellular architectures to emulate the complex organized structure and function of natural tissues and organs. Akashi ’s group has pioneered the research in this field by proposing the build-up of biocompatible tissue constructs comprising L929 fibroblast-based cellular multilayers and native ECM components, including cell adhesive proteins.[199] The authors have reported the use of fibronectin (FN) and gelatin (G) to prepare nanometer-size-based multilayered films on the cell surface by simply repeating the alternate immersion of a cell monolayer-modified glass substrate into FN and G aqueous solutions, under physiological conditions, using the dip-assisted LbL methodology (Figure 6A). In-between each protein deposition step, washing steps were required to remove weakly adsorbed molecules and avoid the cross-contamination of protein aqueous solutions. The successful growth of unlabeled and fluorescently labeled FN/G multilayered coatings comprising different number of layers (i.e., film thickness) was confirmed on a solid surface by employing quartz crystal microbalance (QCM) and fluorescence intensity, respectively. A linear increase in the frequency shift and fluorescence intensity was seen upon increasing the number of layers, thus indicating an increase in the thickness of the film covering the cell surface. The influence of the (FN/G)n multilayered coating on the possible build-up of 3D L929 fibroblast multilayers was assessed by confocal laser scanning microscopy (CLSM) and compared with solely FN-coated and uncoated cells.[199] It was found that, the (FN/G)n multilayered coating was homogeneous and that there was the need for at least 7 bilayers (≈6 nm film thickness, Figure 6A) and having the FN as the outermost layer (i.e., (FN/G)n/FN) to enable the adhesion of more cell layers, leading to higher order 3D fibroblast-based cellular assemblies. Furthermore, the nanofilms showed high intercellular adhesion, being easily peeled off from the glass substrate. However, the same behavior did not happen either on the single FN-coated (≈2.3 nm thick layer) or on the uncoated cells. This indicates that, after seeding the first cell monolayer, no additional cell layers could be included, irrespective of having the first cell monolayer uncoated or coated with a single FN layer. The reason behind this result is explained in light of the motifs displayed by the FN chemical structure. Although FN displays RGD moieties in its structure, such motifs are intrinsically required for regulating the adhesion of FN to the first layer of cells. Therefore, the FN per se is unable to bind to a second layer of cells, thus inhibiting the build-up of cellular multilayers.
多层设计目的：设计分层有序的 3D 细胞结构，以模拟自然组织和器官的复杂组织结构和功能。
图中一层一层堆叠，ECM蛋白纳米薄膜就像胶水一样，将细胞拉伸平铺成片层，而不是细胞自组装成球。
❓意思是FN中的RGD基序只够满足第一层细胞粘附，对于第二层就不够了是么，待看文献
Besides the development of 3D hierarchically stacked constructs encompassing multilayers of L929 fibroblast cells,[199,207] several studies in the literature recalling to the use of the FN/G biomolecular recognition for LbL assembly of different cell types have been reported.[208,209] Those artificial 3D tissuelike constructs include the fabrication of vascularized blood vessels.[210] However, the fabrication of artificial 3D tissue constructs by solely resorting to the dipping LbL methodology is quite challenging. First of all, it is a time-consuming process owing to the numerous depositions and rinsing steps. As such, it raises film stability concerns since the deposited layers might be endocytosed before the adsorption of subsequent layers. Furthermore, it requires large amounts of materials for each adsorption step, as well as solid surfaces. Moreover, the multilayered coating is very thin, showing a film thickness in the nanometer-size range. Hence, researchers have been looking for alternative deposition methodologies to process such multilayered coatings in a fast pace and, simultaneously, develop thick and more robust 3D tissue models emulating the structure and function of natural tissues. One simple methodology that has been used to significantly speedup the LbL assembly process and which does not require the use of solid substrates, is the centrifugation-assisted LbL approach in which individual cells, collected by centrifugation after trypsinization, are alternatively incubated in the native ECM proteins FN and G using centrifugation.[208] After each deposition step, the cells are rinsed with buffer solution, to remove unbound molecules, followed by centrifugation. This deposition methodology has been used to successfully develop a variety of 3D human tissue models, including liver,[211] skin,[212] and blood/ lymph-vascularized human stromal models.[213] However, the centrifugation-assisted LbL methodology requires multiple centrifugation steps to separate the cells from the adsorption protein solutions, which may damage cell membranes and reduce viability, during the build-up process. Akashi and coworkers have investigated the effect of the centrifugation cycles on cell viability and have demonstrated that after submitting the uncoated hepatocyte carcinoma (HepG2) cell line to several centrifugation steps more than 90% of the cells were nonviable. However, in the case of the (FN/G)9-coated cells, the same number of centrifugation cycles ended-up with a cell viability over than 85% and extensively reduced the leakage of cytosolic enzyme lactate dehydrogenase, showing that the LbL coating extensively protects cells from centrifugation-derived physical stress (Figure 6B).[204] Despite these findings, one should not transpose the results gathered with the HepG2 cells and FN/G multilayered coating to other cell types and LbL constituents as the cell viability might by influenced by the cell type, LbL film composition and thickness. Nevertheless, physical stress can be avoided by resorting to filtration-assisted LbL technology which assures high viability and efficiency to the FN/G coated cells. The filtration-assisted LbL methodology has been employed to engineer 3D human tissue constructs, including vascularized cardiac microtissues,[214] liver,[211] and blood vessels.[215] Using a different approach, researchers also demonstrated that the LbL assembly technology can be combined with automatic inkjet printing of single cells and ECM proteins to precisely develop 3D human microtissue chips in a rapid and automatic mode.[216] Such technology hold great potential for in vitro high-throughput preclinical drug screening, as well as to study cell–biomaterial interactions.
In spite of immense reports on the fabrication of thin filmcoated cell surfaces, 3D cellular multilayers, and the establishment of human micro/macrotissue constructs by resorting to biological specific interaction between FN and G multilayers, one can also combine fibronectin with other natural ECM proteins, such as collagen,[217] or cytocompatible and negatively charged naturally occurring glycosaminoglycans, including heparin, or hyaluronic acid to modulate cell functions.[218,219] Adding to this, one could move beyond ECM proteins for cell coating, since they entail limited stability, high costs and batch-to-batch variability. Recent approaches focused on developing LbL-based multilayered micro/macrotissue constructs by resorting to a library of modular ECM-mimetic synthetic peptides, including peptide amphiphiles and multidomain peptides, comprising a repertoire of short different cell-binding motifs derived from ECM proteins. Those peptides are advantageous owing to their easy and cost-effective synthesis, biocompatibility, biodegradability, self-assembling capability into fibrillar nanostructures in aqueous media, and customized bioactivity, which turns them into suitable molecular building blocks for engineering artificial ECM-mimetic constructs to direct cell fate.[220–222] Such peptide library would encompass the widely studied FN-derived RGD and laminin-derived IKVAV (isoleucine-lysine-valine-alanine-valine) biofunctional peptide sequences, known to modulate cellular functions at the tissue and organ levels, among many others, opening new avenues in the molecular design of innovative ECM-like biomaterials for addressing a number of different tissue engineering applications.[223,224] When envisioning the assembly of dense LbL-built microtissues important aspects regarding nutrients/ oxygen availability and neo-vascularization must be considered to assure cellular viability and biofunctionality in denser microtissue constructs.
Cell surface functionalization and LbL-based microtissues assembly, via surface engineering can also be alternatively performed by resorting to protein/polymer combinations. These mainly involve the use of collagen ECM mimetic component to functionalize cells surface. In a recent approach, researchers developed collagen type I/alginate (COL/AA)5 multilayer thin films for generating 2.5D cell sheet constructs after in vitro maturation for 4 days.[225] The underlying hypothesis for successful films assembly was based on natural collagen interaction with the cell membrane and also its electrostatic interaction with negatively charged alginate biopolymers. The intrinsic collagen cell-selective interactions and reactivity toward negatively charged biopolymers was also leveraged for individual MSCs surface functionalization with multilayered thin films. In this work the collagen type I/hyaluronan (COL/HA) multilayered film did not fully covered cells surface, this important aspect resulted in an improved cytoprotection in unfavorable suspension culture conditions, supported MSCs colony-forming ability and also their osteogenic differentiation.[226] This study is particularly relevant for bottom-up tissue engineering approaches since the assembled mesh-like film was composed of two major ECM components. Such could lay the foundation for using ECM mimetic surface functionalized MSCs as biofunctional building blocks in multiscale multicomponent cell–biomaterial assemblies in the future
From another approach, various polymer/polymer combinations including cationic polymers have also been used for cell surface functionalization due to their electrostatically driven interaction with negatively charged cell membrane surfaces.[227] However, their use generally elicits cytotoxic effects and cell lysis, including the formation of nanosized cell membrane pores, which may result in cell death.[228–231] In addition, most polycations are known to quickly transpose the cell membrane and accumulate intracellularly, thus turning them generally useless for functionalizing the cell surface. Nevertheless, one cannot extrapolate and take this as the global picture. It is crucial to bear in mind that the propensity of the polycations to damage the cell membrane and induce cytotoxic effects on cells is strongly dependent on several properties, including polymer functional groups, concentration, molecular weight, conformation, charge density, hydrophobicity, deposition temperature and exposure time, as well as on the type of cells used.[205] This means that, depending on the assembly conditions and cell type, the same polycation can be cytotoxic or noncytotoxic to the cells, thus proving that the cell functions are extremely sensitive to the surface chemistry nature. In an attempt to mitigate polycations’ cytotoxicity and decrease the propensity to disrupt the cell membrane, other cell surface engineering approaches have emerged as promising tools for re-engineering the molecular landscapes of cell surfaces. One possibility concerns the decrease of the polycation charge density through the synthesis of block copolymers, grafting neutral polymers to polycations,[232] and further functionalization of the cell surface via electrostatic interactions. Moreover, a multilayered coating can be attempted directly on the cell surface by employing an automated filtration process, as well as exploiting the electrostatic interactions between the cationic graft copolymer and oppositely charged materials.[232,233] For instance, it has been demonstrated that poly(l-lysine)-graft-poly(ethylene glycol) (PLL-g-PEG) could be adsorbed on the cell membrane surface with minimal accumulation in the intracellular compartments, whereas PLL showed high cytotoxicity, destroyed the cell membrane, accumulated intracellularly, and decreased the cell viability.[205] The influence of the percentage of backbone lysine groups grafted to PEG chains (grafting degree) on pancreatic islet viability has been assessed after cell exposure to the copolymers. It was found that the PLL toxicity decreased while decreasing the charge density (Figure 6C). However, keeping fixed the concentration of the graft copolymer and the grafting degree, and increasing the PEG chain length led to a decrease of the PLL cytotoxicity. The authors also studied the influence of the PLL molecular weight on its cytotoxicity to cells. As expected, polycations with higher molecular weight showed higher toxicity in comparison with their lower molecular weight counterparts. Another strategy to modify the cells surface in a LbL fashion while assuring cell viability, comprises the use of PLL-g-PEGbiotin (PPB) cationic copolymer and streptavidin multilayers by exploiting the highly specific, stable, and strong biological interaction between streptavidin and biotin motifs (Figure 6D). After the electrostatic deposition of the PPB cationic copolymer onto the negatively charged cell surface, the streptavidin entity adsorbs onto it through biotin–streptavidin biospecific interactions. Moreover, since streptavidin encloses four binding sites for biotin, a biotin fraction still remains available, thus enabling the adsorption of further PPB/streptavidin layers until reaching the desired film thickness (Figure 6D). It has been demonstrated that such strategy is a very efficient and stable methodology to build-up biocompatible, PEGylated multilayer thin films on cell surfaces,[192] being an attractive methodology to control the extracellular microenvironments and regulate cell behavior. In a similar application, islet surface modification through the alternating adsorption of biospecific biotin-bovine serum albumin/streptavidin multilayers onto a hydrophobicdriven biotin-PEG-lipid functionalized lipid bilayer of the cell membrane has also been explored.[
Besides protein/protein, polymer/polymer, and protein/ polymer hybrid nanofilms generated by biomolecular recognition and electrostatic driving forces, other intermolecular interactions can also be used to drive film growth and functionalized cell surfaces. Covalent bond-based chemical approaches have also been used to coat cell surfaces using polymers functionalized with orthogonally reactive moieties. For instance, recent studies focused on the fabrication of a covalently crosslinked multilayered thin film of poly(vinyl alcohol) (PVA) by assembling thiol-modified PVA (PVA-SH) and pyridyl disulfidefunctionalized PVA (PVA-PD) multilayers on pancreatic islets surface.[235] However, before the generation of the multilayered coating, the cell membrane surface was first functionalized with maleimide reaction groups using a maleimide-PEG-functionalized phospholipid (Mal-PEG-lipid). Then, thiol-modified PVA was covalently bound to the Mal-PEG-lipid modified cell surface through thiol-ene reaction. Stable PVA-based multilayers were then generated on cells surface by sequential deposition of PVA-SH and PVA-PD multilayers via thiol-disulfide exchange reaction. This methodology has great potential for improving the transplantation of pancreatic islets (Langerhans), possibly eliminating the need for self-monitoring and insulin injection in patients with insulin-dependent diabetes mellitus (type I).
Moreover, similar assemblies can be developed through the sequential adsorption of other polymers via covalent bonds.[236,237] More recently, the efficient assembly of biorthogonal, covalently stabilized polymer multilayers on pancreatic islet surfaces through the sequential deposition of PEG-azide-functionalized hyperbranched alginate (ALG) biopolymer (PEG-azido-ALG) and methyl-2-diphenylphosphinoterephthalate-functionalized poly(amido amine) (MDTPAMAM) dendrimers via Staudinger ligation reaction was reported (Figure 6E).[206] These researchers also compared the assembly process by purely resorting to covalent bonds with that co-mediated by covalent bonds and electrostatic interactions by manipulating the degree of positive charge of MDTPAMAM via shielding with different amounts of glutaric anhydride (GA). The assemblies completely generated by covalent bonds showed a higher degree of surface inhomogeneity in comparison with those obtained by combining electrostatic and covalent interactions. Such behavior was assigned to the inhomogeneity of the basement coating created after the adsorption of the PEG-azide layer via NHS coupling onto pancreatic islets when compared with the coating obtained via electrostatic adsorption of PAMAM dendrimers.
Another very interesting approach to overcome the cytotoxicity induced by most polycations and trigger cell surface engineering concerns the development of hydrogen bonded LbL films. Inspired by this approach yeast living cell surfaces were functionalized with nonionic and biocompatible hydrogen-bonded tannic acid (TA) and poly(N-vinylpyrrolidone) (PVPON) multilayers.[238] However, before the deposition of the TA/PVPON multilayers, yeast cells were modified with a poly(ethyleneimine) (PEI) precursor layer to allow adhesion. It was found that, notwithstanding the cytotoxicity induced by the basement PEI monolayer, the assembly of at least (TA/ PVPON)3 bilayers sustained a high cell viability for at least 6 days of culture. Such behavior was expected and is the result of the low exposure of cells to toxic polycations, as well as to the highly permeable LbL shell generated by hydrogen-bonded layers. Moreover, the fact that the TA, a natural polyphenol, entails antioxidant properties and is capable of scavenging free radicals, extensively contributes to protect cells against damage. In a similar report, researchers have developed a rapid, conformal, and stable coating of various types of living pancreatic islets with hydrogen-bonded TA/PVPON[239] and TA/poly(N-vinylcaprolactam) (PVCL) multilayers,[240] under physiological conditions, aiming at treating Type 1 diabetes. It was demonstrated that the hydrogen-bonded multilayers were more effective than electrostatic-driven LbL assemblies, keeping cell viability and functionality up to 7 days in culture. Moreover, hydrogen-bonded shells showed immunomodulatory cryoprotective properties, as demonstrated by the reduced pro-inflammatory cytokine production when incubated with macrophages and diabetogenic cells.
Adding to macromolecular cell surface modification, other strategies exploiting cell–materials interplay, namely, nano- and microparticle–cell interactions have been investigated for bottom-up tissue engineering as it will be showcased in the following sections.
Nanoparticles are defined as colloidal materials with subcellular/sub-micrometer sizes ranging from 1 to 1000 nm.[241,242] In recent decades, these systems have attracted significant interest due to their numerous biomedical applications ranging from controlled delivery of hydrophilic/hydrophobic therapeutics, to multimodal bioimaging, biosensing and diagnostics.[243,244] Upto-date, functional nanosystems have been formulated from a plethora of inorganic and organic materials including synthetic block-copolymers,[245] natural origin biopolymers, and peptides/ proteins that endow them with unique bioactivity, biocompatibility, biodegradability, and chemical versatility.[246] This chemical flexibility and their high surface-to-volume ratio has been widely acknowledged and explored via precise chemical functionalization to imprint multifunctional features including cell/tissue targeting, adhesivity, and response to different stimuli.[247,248] So far, numerous types of nanoparticles have been fabricated to precisely adapt or respond to magnetic fields, temperature, ultrasound, pH/redox/hypoxic microenvironments, enzymes, light, among others.[249,250] This responsiveness may trigger changes in nanoparticles color, shape, size or originate complete disruption and prompt cargo release.[251] Adding to this, nanoparticles physicochemical properties have received significant focus due to their influence in the overall biological performance of these systems. Among these, particle size is particularly important at the nano–bio interface since multiscale interactions can be established with cells surface or intracellular organelles. Particles surface chemistry (i.e., charge, functionality, etc.) and shape (i.e., spherical, square, rod, elliptical, circular disks, etc.)[252] are too recognized to play a critical role in cellular internalization kinetics and cytotoxicity.
Owing to their physicochemical properties, nanoparticles have also been recently explored in the context of bottomup tissue engineering for promoting the assembly of cellular building blocks into higher order 3D clusters. The most recent studies have explored nanoparticles as membrane-adhesive mediators of cell aggregation or as structural supports for preformed 3D agglomerates. Different researchers have pursued this strategy in the form of either hyperbranched polyglycerols or dendrimeric intercellular linkers for rapidly forming multicellular structures.[253,254] Recently, IKVAV-functionalized polyamidoamine dendrimers coated with hyaluronic acid were used as adhesive particulates for rapidly compacting ASCs into 3D spheroid structures (Figure 7A).[255] These membrane-binding particles effectively maintained spheroids compactness in presence of competitive adhesion from tissue culture plates, while enhancing their proliferation and paracrine secretion of angiogenic factors. The potential of this nanobridging phenomenon has also been illustrated in recent applications where researchers have successfully glued together biological tissues and accelerated wound closures.[256,257] Catechol-functionalized liposomes have also been recently used to promote tissue–tissue adhesion and could be an interesting technology for bridging cell-rich 3D modules.[248] Conversely, polymeric nanoparticles adsorbed on cell membranes of spheroids can be used for slowing down spheroid spreading processes on adhesive substrates (Figure 7B).[258] Controlling spheroid fusion and compartmentalization could be an interesting feature for bottom-up bioprinting-based approaches using 3D spheroids as unitary building blocks.
由于它们的物理化学特性，最近还在自下而上组织工程的背景下探索了纳米颗粒，以促进细胞构建块组装成更高阶的 3D 簇。最近的研究已经探索了纳米粒子作为细胞聚集的膜粘附介质或作为预先形成的 3D 聚集体的结构支持。不同的研究人员以超支化聚甘油或树枝状细胞间连接体的形式采用这种策略，以快速形成多细胞结构。 [253,254] 最近，涂有透明质酸的 IKVAV 功能化聚酰胺胺树枝状大分子被用作粘合剂颗粒，用于将 ASC 快速压实成 3D 球状结构（图 7A）。 [255]这些膜结合颗粒在存在来自组织培养板的竞争性粘附的情况下有效地保持球体的致密性，同时增强它们的增殖和血管生成因子的旁分泌。这种纳米桥接现象的潜力也在最近的应用中得到了说明，研究人员成功地将生物组织粘合在一起并加速伤口闭合。[256,257] 儿茶酚功能化脂质体最近也被用于促进组织 - 组织粘附，可能是一项有趣的技术用于桥接富含细胞的 3D 模块。[248]相反，吸附在球体细胞膜上的聚合物纳米粒子可用于减缓球体在粘合剂基材上的扩散过程（图 7B）。 [258]对于使用 3D 球体作为单一构建块的自下而上生物打印方法，控制球体融合和划分可能是一个有趣的特征。
Attempting to explore nanoparticles for cell surface modification, Irvine’s group has developed maleimide-functionalized liposomes that readily attach to thiol groups expressed at T-cell membrane proteins.[260,261] With this approach, researchers managed to immobilize nanoparticles loaded with immunomodulatory drugs at the surface of T-cells, allowing for specialized cargo delivery to the linked T-cells and thus modulating its behavior locally.[261] More recently, smart protein nanogels that adhere to the plasma membrane were produced. These systems serve as reduction-responsive cell backpacks that release their cargo in the presence of the characteristic increase in redox activity at T-cells surfaces.[262] In terms of augmenting cell aggregation, protein–protein interactions can also drive colloidal assembly, which could be a pathway for establishing nanoenabled cell-dense modules with nanoparticulated reservoirs of bioinstructive signals.[263] On the other hand, a recent pioneering study focused on the use 20 nm carboxylated polystyrene nanoparticles as “nanostickers” capable of effectively maintaining single cells as large cohesive aggregates.[264] In this study, a cadherin-depleted S180 murine cell line characterized for its extremely poor cell– cell adhesion was used to illustrate the potential of nanostickers to glue cells together into cell-rich structures
为了探索用于细胞表面修饰的纳米颗粒，Irvine 的研究小组开发了马来酰亚胺功能化脂质体，该脂质体很容易附着在 T 细胞膜蛋白上表达的硫醇基团上。[260,261] 通过这种方法，研究人员设法将载有免疫调节药物的纳米颗粒固定在表面T 细胞，允许将专门的货物运送到连接的 T 细胞，从而在本地调节其行为。 [261]最近，生产了粘附在质膜上的智能蛋白质纳米凝胶。这些系统充当还原响应细胞背包，在 T 细胞表面氧化还原活性特征性增加的情况下释放其货物。 [262]在增强细胞聚集方面，蛋白质-蛋白质相互作用也可以驱动胶体组装，这可能是建立具有纳米颗粒生物指示信号储存器的纳米细胞致密模块的途径。[263]另一方面，最近的一项开创性研究侧重于使用 20 nm 羧化聚苯乙烯纳米粒子作为“纳米贴纸”，能够有效地将单个细胞保持为大的粘性聚集体。 [264]在这项研究中，一种以细胞-细胞粘附极差为特征的去除钙粘蛋白的 S180 鼠细胞系被用来说明纳米贴纸将细胞粘合在一起形成富含细胞的结构的潜力
Looking forward, advanced nanoconstructs anchored to cell membranes can extend their activity while function as bioreservoirs of small molecules/morphogens for driving stem cells differentiation or committing terminally differentiated cells to specific phenotypes. Previously, this elegant perspective of combining bioactive cell patterning and modular self-assembly was limited by a lack of affordable and efficient precision chemistry strategies. However, throughout the last decade, the maturing fields of nanotechnology, bioconjugate chemistry, and drug delivery have provided researchers with a remarkable toolbox of biodegradable/smart biomaterials, sophisticated surface modification strategies and frameworks for bioactive controlled release. With the advent of biorthogonality, emerging cellular backpacks and cutting-edge membrane-engineering strategies (e.g., glycoengineering, liposomal fusion, boronic acid, or maleimidebased chemistries), could translate to the form of scaffold-free, clickable modular assemblies containing different ratios of bioactive nanoparticles for bottom-up tissue engineering.
Adding to cell-membrane bound strategies, biomaterial- in-cell strategies using nanoparticles that are readily internalized by cells have also been explored for on-demand cell self-aggregation into engineered 3D microtissues. Specifically, PLL magnetic nanoparticles (MNPs) that immediately respond to static magnetic fields have been employed to generate magnetically levitated multicellular 3T3-L1 preadipocyte microtissues within 24 h of culture.[265] Interestingly, prolonged culture under adipogenic conditions resulted in the formation of 3D adipospheres exhibiting specific lipidic contents, thus evidencing biofunctionality. Also, levitated 3D spheroids derived from white adipose tissue (WAT) cells were able to recapitulate key aspects of WAT organogenesis. Positive and negative magnetophoresis could also be leveraged for assembling 3D cell-dense architectures (e.g., ringshaped, three-pointed star, rectangular, etc.) under different magnet configurations as it will be further discussed.[266] The use of magnetically active cells to generate 3D assemblies must however be subjected to further fundamental studies such as those already reported,[267] since our understanding of the long-term effects in cells biofunctionality and the degradation of these nanoparticles in living dynamic microtissues is still in its infancy.
除了细胞膜结合策略外，还探索了使用易于被细胞内化的纳米颗粒的细胞内生物材料策略，用于按需细胞自聚集到工程化的 3D 微组织中。具体而言，立即响应静磁场的 PLL 磁性纳米粒子 (MNP) 已被用于在培养 24 小时内产生磁悬浮的多细胞 3T3-L1 前脂肪细胞微组织。 [265]有趣的是，在脂肪形成条件下的长时间培养导致形成具有特定脂质含量的 3D 脂肪球，从而证明了生物功能性。此外，源自白色脂肪组织 (WAT) 细胞的悬浮 3D 球体能够概括 WAT 器官发生的关键方面。正磁泳和负磁泳也可用于在不同磁体配置下组装 3D 细胞密集结构（例如，环形、三尖星形、矩形等），这将进一步讨论。 [266]然而，使用磁性活性细胞生成 3D 组件必须经过进一步的基础研究，例如已经报道的那些，[267] 因为我们对细胞生物功能的长期影响和这些纳米颗粒在活的动态微组织中的降解的理解是仍处于起步阶段。
Even considering possible long-term effects, this technology remains highly attractive for more complex bottom-up tissue engineering. Recently, magnetic assembly was explored for precise fabrication of endothelial 3D spheroids and for their inclusion into prefabricated magnetic vascular tree-like templates. This magnetic confinement and prolonged culture induced 3D endothelial multicellular spheroids maturation and fusion into a higher order construct that exhibited morphologic and phenotypic stability.[268] The manipulation of spheroids into user-defined shapes/patterns was explored by using a disruptive approach that brought forward a paradigm shift in magnetic-based spheroid manipulation.[259] This involved the generation of Janus spheroids where cells and particles are segregated into distinct domains in a ECM-like microenvironment, aiming to reduce MNPs internalization and thereby avoid possible adverse effects on cell viability (Figure 7D). The resulting Janus constructs present lower MNPs internalization in comparison to standard spheroids and demonstrated the ability to fuse together into more complex 3D vascular tissue tube constructs via magnetic guidance.[259] It was also discovered that the strength of the magnet largely affected 3D spheroids fusion into the desired tubular shape, with constructs established with a weak magnet (i.e., 10% of maximum magnetic field) exhibiting faster inner ring contraction than their counterparts (Figure 7Di,ii).
即使考虑到可能的长期影响，这项技术对于更复杂的自下而上的组织工程仍然极具吸引力。最近，磁性组件被探索用于精确制造内皮 3D 球体并将它们包含在预制的磁性血管树状模板中。这种磁限制和延长培养诱导 3D 内皮多细胞球体成熟并融合成更高阶的结构，表现出形态和表型稳定性。 [268]通过使用一种颠覆性方法探索了将球体操纵为用户定义的形状/图案，该方法提出了基于磁性的球体操纵的范式转变。 [259]这涉及生成 Janus 球体，其中细胞和颗粒在类似 ECM 的微环境中被分离到不同的域中，旨在减少 MNP 内化，从而避免对细胞活力可能产生的不利影响（图 7D）。与标准球体相比，由此产生的 Janus 构建体呈现出较低的 MNP 内化，并证明了通过磁导融合在一起形成更复杂的 3D 血管组织管结构的能力。 [259]还发现磁铁的强度在很大程度上影响了 3D 球体融合成所需的管状形状，用弱磁铁（即最大磁场的 10%）建立的构造比它们的对应物表现出更快的内环收缩（图 7Di， ii)。
3.2.2 Bioinstructive and 3D Bioprocessed Nanoparticle-Based Assemblies(生物指导和 3D 生物加工的基于纳米粒子的组件)
Beyond engineering cells surface and operating as cellular aggregators, nanoparticles are also attractive for imparting different bioinstructive cues (e.g., growth factors, cytokines, mechanical, and electric/magnetic stimuli), further enabling engineered microtissues 3D maturation and improving biofunctionality.[269,270] Particularly, the incorporation of nanoparticles with stimuliresponsive features in bottom-up bioarchitectures development is highly valuable owing to their potential for loading different biochemical cargos and for spatiotemporally controlling biophysical signals presentation to cellular building blocks.
除了工程细胞表面和作为细胞聚集体发挥作用外，纳米粒子还具有吸引力，可提供不同的生物指导线索（例如，生长因子、细胞因子、机械和电/磁刺激），进一步使工程化微组织 3D 成熟和改善生物功能。 [269,270] 特别是，在自下而上的生物架构开发中加入具有刺激响应特征的纳米粒子是非常有价值的，因为它们具有装载不同生化货物和时空控制生物物理信号呈现给细胞构件的潜力。
In this context, Laponite nanosilicates (Laponite is a trademark of the company BYK Additives Ltd.) have been recently employed to generate bioinstructive gradients of different growth factors (e.g., VEGF, FGF, and PDGF) for promoting endothelial sprouting, or for establishing osteochondral interfaces in cell–biomaterial assemblies (via rhBMP-2 and TGF-β3).[272,273] In another elegant approach, researchers have exploited the potential of magneto-responsive nanoparticles for patterning biochemical gradients (i.e., BMP-2) in human MSCs–agarose scaffolds and established robust osteochondral constructs in vitro.[274] Strikingly, this strategy resulted in spatially controlled osteogenic gene expression and tissue mineralization, as well as the formation of a bone–cartilage interface after 28 days of maturation in vitro. Apart from magnetism, temperature and light (e.g., visible, near-infrared, etc.) also represent promising alternatives for on-demand generation of gradients of bioinstructive molecules that ultimately guide cell differentiation and modulate 3D microtissues physiology in vitro.[275–277] Recently, mechanoresponsive nanoparticles have too been formulated to provide bioactive molecules presentation upon deformation by biomechanical forces, thus unlocking the possibility to engineer constructs with biomechano-interactive signals similar to those naturally present in living tissues.[278] These studies represent a stepping stone in highlighting the versatility of stimuliresponsive nanoparticles for providing biochemical gradients in tissue-scale constructs, evidencing possible future trends in this field
Adding to their transported bioactive cargo, nanomaterials’ physicochemical features can also be exploited as modulation mechanisms to develop and mature on-demand 3D microtissues with complex biofunctionality. This has been the case of load-bearing and complex electroactive tissues (e.g., cardiac, nerve, and skeletal muscle).[279] In this regard, several electrically conductive nanomaterials, such as graphene, carbon nanotubes, and silicon/gold nanowires, provided promising outcomes in cardiac tissue engineering.[280,281] By impregnating gold nanowires in alginate scaffolds researchers were able to enhance the electrical interconnectivity of engineered cardiac 3D microtissues and improve cardiomyocytes phenotype, expression of contractile function markers and ability to generate synchronous contractions.[282] More recently, incorporation of silicon nanowires in human induced pluripotent stem-cell-derived cardiomyocytes enabled the assembly of electrically conductive cardiac 3D spheroids with enhanced cell–cell junctions, contractile machinery maturation, and decreased spontaneous beat rate, which could be beneficial for reducing arrhythmia post-transplantation.[283] Also, nanoparticles comprising conducting polymers (e.g., polypyrrole) were employed to trigger biomolecules delivery upon stimulation with weak external electrical fields, that are known to impact cell processes in angiogenesis, cardiomyogenesis, neurogenesis, and osteogenesis and could thus enable on-demand microtissue maturation along time.[28
除了运输的生物活性货物外，纳米材料的物理化学特性还可以作为调节机制来开发和成熟具有复杂生物功能的按需 3D 微组织。承重和复杂的电活性组织（例如心脏、神经和骨骼肌）就是这种情况。 [279]在这方面，几种导电纳米材料，如石墨烯、碳纳米管和硅/金纳米线，在心脏组织工程中提供了有希望的成果。 [280,281] 通过在海藻酸盐支架中浸渍金纳米线，研究人员能够增强工程的电互连性。心脏 3D 微组织和改善心肌细胞表型、收缩功能标志物的表达和产生同步收缩的能力。[282]最近，在人类诱导的多能干细胞衍生的心肌细胞中加入硅纳米线能够组装具有增强的细胞-细胞连接、收缩机械成熟和降低自发搏动率的导电心脏 3D 球体，这可能有利于减少心律失常[283]此外，包含导电聚合物（例如，聚吡咯）的纳米颗粒被用于在用弱外部电场刺激时触发生物分子递送，已知其会影响血管生成、心肌生成、神经发生和成骨中的细胞过程，因此可以实现按需微组织成熟随着时间的推移。 [28
Notwithstanding their potential for bottom-up tissue engineering, most of these studies describe the use of overly simplistic nanoparticle containing scaffolds, being unable to fully recapitulate native ECM composition, nor tissues complex architecture or anatomical scale. T o overcome such drawbacks, researchers are actively developing nanoparticle functionalized bioinks and exploiting advanced additive manufacturing techniques to build-up 3D microtissue constructs with tailored microenvironments allied to biospecific designs.[
尽管它们具有自下而上组织工程的潜力，但这些研究中的大多数描述了使用过于简单的含有纳米颗粒的支架，无法完全概括天然 ECM 组成，也无法完全概括组织复杂结构或解剖规模。为了克服这些缺点，研究人员正在积极开发纳米颗粒功能化生物墨水，并利用先进的增材制造技术来构建具有与生物特异性设计相关的定制微环境的 3D 微组织结构。
Nanoparticles incorporation in bioinks is highly valuable as it can improve the mechanical performance of natural-based biomaterials (e.g., gelatin and fibrin)[286] and allows bioactive cues’ tailored presentation, as well as to introduce electrical properties in the final assemblies.[286,287] Interestingly, it is being increasingly recognized that nanoparticles inclusion in bioinks also improves shear-thinning, a fundamental aspect for enhancing printability and resolution (Figure 8A).[288,289] From the plethora of nanomaterials currently being pursued in 3D nano-bioprinting strategies, Laponite nanoparticles are widely explored due to their innate bioactivity (i.e., osteogenic/ angiogenic), mechanical resistance, and sustained delivery of bioinstructive signals (e.g., growth factors).[290,291] Recent evidences corroborate that Laponite nanosilicates incorporation in PEG-based photocrosslinkable bioinks enabled sustained VEGF delivery which directed rapid HUVECs migration toward 3D bioprinted constructs.[292] Such nanoengineered Laponite/PEGdiacrylate bioinks displayed improved shear-thinning properties, high mechanical stability, increased printing fidelity, and suitable cytocompatibility post-crosslinking.[293] An improvement to this design was achieved by incorporating nanosized Laponite in GelMA/к-carrageenan and exploiting nanoengineered ionic covalent entanglement (NICE) in a bioink with innate bioactivity (i.e., GelMA cell adhesive domains).[294] With this synergistic approach, it was possible to bioprint large vertical structures and different human-scale 3D anatomical structures with improved mechanical strength, toughness and elasticity (Figure 8B). Cells laden in such bioprinted constructs also maintained high cell viability over 120 days, indicating the potential of this approach for long-term bottom-up tissue engineering applications.
在生物墨水中掺入纳米粒子非常有价值，因为它可以提高天然生物材料（例如明胶和纤维蛋白）的机械性能[286]，并允许生物活性线索的定制呈现，以及在最终组件中引入电特性。 286,287] 有趣的是，人们越来越认识到，生物墨水中包含的纳米颗粒还可以提高剪切稀化性，这是提高可打印性和分辨率的一个基本方面（图 8A）。[288,289] 从目前在 3D 纳米生物打印策略中追求的大量纳米材料来看, Laponite 纳米颗粒因其先天的生物活性（即成骨/血管生成）、机械阻力和生物指示信号（例如生长因子）的持续传递而被广泛探索。 [290,291] 最近的证据证实 Laponite 纳米硅酸盐掺入基于 PEG 的光交联bioinks 实现了持续的 VEGF 递送，从而引导 HUVEC 快速迁移到 3D b [292] ioprinted 构造。这种纳米工程的 Laponite/PEG 二丙烯酸酯生物墨水显示出改进的剪切稀化性能、高机械稳定性、提高的印刷保真度和合适的细胞相容性后交联。 [293]通过在 GelMA/к-角叉菜胶中加入纳米尺寸的 Laponite 并在具有先天生物活性的生物墨水（即 GelMA 细胞粘附域）中利用纳米工程离子共价纠缠 (NICE)，实现了对这一设计的改进。 [294]通过这种协同方法，可以生物打印大型垂直结构和不同的人体尺度 3D 解剖结构，并提高机械强度、韧性和弹性（图 8B）。装载在这种生物打印结构中的细胞在 120 天内也保持了高细胞活力，表明这种方法在自下而上的长期组织工程应用中的潜力。
Other silica-based nanoparticles with cationic surface charge have been exploited to improve the printability and shape fidelity of anionic polysaccharide bioinks (i.e., alginate/ gellan gum) via the establishment of electrostatic interactions. This approach resulted in enhanced mechanical robustness of large-scale ear-shaped 3D constructs and high cell viability. Importantly, the observed mechanical improvement was highly dependent on polymers molecular weight, nanoparticles surface chemistry, concentration and size, being mainly observed for particles (<100 nm).[298] Such indicates that various parameters must be carefully investigated to achieve optimal mechanical properties of printed constructs always taking as reference those of the native tissues one aims to recapitulate.[299
其他具有阳离子表面电荷的二氧化硅基纳米颗粒已被开发用于通过建立静电相互作用来提高阴离子多糖生物墨水（即藻酸盐/结冷胶）的可印刷性和形状保真度。 这种方法提高了大型耳形 3D 结构的机械强度和高细胞活力。重要的是，观察到的机械改进高度依赖于聚合物分子量、纳米颗粒表面化学、浓度和尺寸，主要观察到颗粒（<100 nm）。 [298]这表明，必须仔细研究各种参数，以实现印刷结构的最佳机械性能，始终以旨在概括的天然组织为参考。 [299
Recently, the use of inorganic nanoparticles to endow cellladen bioinks with electric conductivity has been explored for materialized 3D bioprinted constructs for application in cardiac tissue engineering. In this context, researchers explored the development of an electroactive bioink comprising inorganic gold nanorods (34 × 25 nm) and GelMA hydrogel. Gold nanoparticles inclusion improved bioink shear-thinning properties and printability, while also enhanced cardiac cell adhesion/organization versus its pristine GelMA counterpart (Figure 8C).[295] In addition, such gold nanocomposite bioink could install electroconductive bridges connecting adjacent cardiac bundles and promoted electrical communication in 3D printed constructs.
This resulted in increased cell–cell interactions, cardiac phenotypic expression as well as synchronized contractile frequency of bioprinted cardiac constructs (Figure 8C). Similarly, an alginate-based nanostructured bioink with embedded silver nanoparticles was used for bioprinting chondrocytes into 3D bionic ears, which could successfully receive electromagnetic signals (Figure 8D).[296] These are highly elegant approaches to generate electroactive microtissue constructs, nevertheless their in vivo applicability and long-term biocompatibility in cardiac microenvironment is an important aspect to be further evaluated.
最近，已经探索了使用无机纳米粒子来赋予载有细胞的生物墨水导电性，以用于在心脏组织工程中应用的物化 3D 生物打印结构。在这种情况下，研究人员探索了一种由无机金纳米棒（34 × 25 nm）和 GelMA 水凝胶组成的电活性生物墨水的开发。与原始 GelMA 对应物相比，包含金纳米粒子改善了生物墨水的剪切稀化特性和可印刷性，同时还增强了心脏细胞的粘附/组织（图 8C）。 [295]此外，这种金纳米复合生物墨水可以安装连接相邻心脏束的导电桥，并促进 3D 打印结构中的电通信。这导致细胞间相互作用、心脏表型表达以及生物打印心脏结构的同步收缩频率增加（图 8C）。同样，嵌入银纳米粒子的基于海藻酸盐的纳米结构生物墨水用于将软骨细胞生物打印到 3D 仿生耳中，这可以成功接收电磁信号（图 8D）。 [296]这些是生成电活性微组织结构的非常优雅的方法，但它们的体内适用性和在心脏微环境中的长期生物相容性是有待进一步评估的重要方面。
Adding to this, magnetic iron nanoparticles were exploited to recapitulate the anisotropy of human cartilage in bottomup engineered 3D constructs.[300] Such nanomaterials have been used to direct collagen alignment during bioprinting, thus enabling the fabrication of multilayered chondrocyteladen constructs with intercalating layers of aligned/random collagen fibers. Interestingly, bioprinting of such constructs led to enhanced collagen I and II expression, highlighting the importance of architectural control when envisioning to recapitulate intrinsically anisotropic human tissues in bottom-up engineered cell–biomaterial assemblies.[300] This attention on faithfully recapitulating tissues complex architecture using nano particles in the bioprinting process was further demonstrated by the inclusion of melanin nanoparticles (≈500 nm) in a silk fibroin/PEG-acrylate nanoengineered bioink. In this elegant strategy melanin nanoparticles innate light absorption was hypothesized to improve light-induced 3D bioprinting resolution in the z-axis while assuring embedded NIH/3T3 fibroblasts viability.[297] By reducing light penetration depth, nanoparticles enabled 3D projection stereolithography bioprinting of otherwise impracticable empty/tilted structures such as tubes or stairs, and enabled the fabrication of fully perfusable vascular networks with high printing precision (Figure 8E).
除此之外，磁性铁纳米粒子被利用来概括人体软骨在自下而上的工程 3D 构造中的各向异性。 [300]这种纳米材料已被用于在生物打印过程中指导胶原蛋白排列，从而能够制造具有插入层的排列/随机胶原纤维的多层软骨细胞结构。有趣的是，这种结构的生物打印导致胶原蛋白 I 和 II 表达增强，在设想在自下而上的工程细胞-生物材料组件中概括本质各向异性的人体组织时，突出了结构控制的重要性。 [300]通过在丝素蛋白/PEG-丙烯酸酯纳米工程生物墨水中加入黑色素纳米颗粒（≈500 nm），进一步证明了这种在生物打印过程中使用纳米颗粒忠实再现组织复杂结构的关注。在这个优雅的策略中，假设黑色素纳米粒子先天的光吸收可以提高 z 轴上光诱导的 3D 生物打印分辨率，同时确保嵌入的 NIH/3T3 成纤维细胞的活力。 [297]通过减少光穿透深度，纳米粒子能够对原本不切实际的空/倾斜结构（如管或楼梯）进行 3D 投影立体光刻生物打印，并能够制造具有高打印精度的完全可灌注血管网络（图 8E）。
In practice, the described bottom-up engineering technologies that are based on, or include nanomaterials, are envisioned to assist researchers in modulating mechanical properties and the local biochemical microenvironment/mechanical cues, allowing one to better guide living microtissues morphogenesis and maturation during in vitro culture before in vivo application. Advances in cell-membrane-functionalized nanoparticles,[301] may further contribute for selective particle–cell docking and for a higher control over cell–cell spatial patterning in cocultured 3D aggregates in a foreseeable future. Adding to these opportunities, it has been recently demonstrated that nanoparticles could potentially function as nanosensors for monitoring the early immune response to implanted tissue engineered constructs,[302] or for mapping oxygen distribution/dynamics within 3D bioprinted constructs.[303] Generating microtissues with built-in biocompatible nanosensors could be essential for further understanding and noninvasively monitor the physiological changes upon their implantation in real-time.
在实践中，所描述的基于或包括纳米材料的自下而上工程技术旨在帮助研究人员调节机械性能和局部生化微环境/机械线索，从而更好地指导活体微组织在体外的形态发生和成熟体内应用前的培养。在可预见的未来，细胞膜功能化纳米粒子[301] 的进展可能会进一步促进选择性粒子-细胞对接和对共培养 3D 聚集体中细胞-细胞空间模式的更高控制。除了这些机会之外，最近还证明，纳米颗粒可能会用作纳米传感器，用于监测对植入的组织工程结构的早期免疫反应，[302] 或用于绘制 3D 生物打印结构内的氧气分布/动力学。 [303]生成具有内置生物相容性纳米传感器的微组织对于进一步了解和无创实时监测其植入后的生理变化至关重要。
Microparticles, i.e., structures ranging from 1 to 1000 µm, have for long been the platform of choice for various biomedical applications,[304,305] with numerous reports exploring microcarrier formulations for focal delivery and controlled release of therapeutics, as injectable tissue-defect fillers, as biosensing tools, or as cell-expansion systems in static/dynamic in vitro cultures.
Along the last decades, the exponential evolution of precision particle fabrication techniques (i.e., electrodynamic jetting, molding/microarrays,[308] or microfluidics)[309,310] has contributed for improving the manufacture of microparticles with tailored surface chemistry, degradation and porosity.[311] Particulates with variable surface topography, stiffness, tunable size, and complex shapes have also been successfully engineered.[312,313] This technological progress, combined with particles physicochemical versatility, have provided the foundation for also exploring microparticles as injectable cell-encapsulating platforms and for in vitro disease modeling.[84,314] More recent endeavors have focused on exploring microparticles as building blocks for modular bottom-up tissue engineering by taking advantage of their role as orchestrators of cellular aggregation and of their bioinstructive cues for 3D microtissues maturation.[306,311,315] We will focus on cell–particle surface interactions at the material–bio interface rather than cell encapsulation which will be further discussed in hydrogel-based platforms.
在过去的几十年中，精密粒子制造技术（即电动喷射、模塑/微阵列、[308] 或微流体）[309,310] 的指数级发展有助于改进具有定制表面化学、降解和孔隙率的微粒的制造。第311章]具有可变表面形貌、刚度、可调尺寸和复杂形状的颗粒也已成功设计。[312,313] 这一技术进步与颗粒物理化学的多功能性相结合，为探索微粒作为可注射细胞封装平台和在体外疾病建模。[84,314] 最近的努力集中在探索微粒作为模块化自下而上组织工程的构建块，利用它们作为细胞聚集协调器的作用及其对 3D 微组织成熟的生物指导线索。[306,311,315] 我们将关注材料-生物界面处的细胞-颗粒表面相互作用，而不是细胞封装，这将在基于水凝胶的平台中进一步讨论。
In this sense, the establishment of cell–microparticle unitary blocks and their progression into modular 3D aggregates has been materialized by using cell-scaled solid microparticles as supporting/adhesive platforms (i.e., >2 µm to several hundred micrometers), as opposite to smaller particles which may be extensively internalized by cells when no cell-membrane anchoring technology is installed.[315] The control over microparticle–cell aggregates size and morphology is key when envisioning 3D microtissues build-up and is highly dependent on microparticle formulations monodispersity,[316] cell density,[84] and culture conditions (i.e., static/dynamic).[317] 3D cell–microparticle assemblies biointegration and overall biofunctionality is also vital in the context of bottom-up tissue engineering and recognized to be largely influenced by: i) cells and microparticles spatial distribution/density, ii) the existence of cell–particle anchoring ligands (i.e., antibodies, ECM mimetic proteins/ biopolymers, etc.),[305] and iii) the presence of bioinstructive morphogens (e.g., growth factors, cytokines, etc.).[318] These fundamental design considerations and processing know-how have enabled researchers to fabricate 3D constructs where particle–cells adhesion is random or directed.[319] Both approaches can warrant the application of bioengineered microtissues in bottom-up tissue engineering and regenerative medicine.
从这个意义上说，通过使用细胞尺度的固体微粒作为支撑/粘附平台（即 > 2 µm 到几百微米），已经实现了细胞-微粒整体块的建立及其向模块化 3D 聚集体的发展，这与较小的当没有安装细胞膜锚定技术时，这些颗粒可能会被细胞广泛内化。 [315]在设想 3D 微组织构建时，对微粒 - 细胞聚集体大小和形态的控制是关键，并且高度依赖于微粒制剂的单分散性、[316] 细胞密度、[84] 和培养条件（即静态/动态）。 [317 ] 3D 细胞-微粒组装体的生物整合和整体生物功能在自下而上的组织工程中也很重要，并且被认为在很大程度上受以下因素的影响：i) 细胞和微粒的空间分布/密度，ii) 细胞-粒子锚定配体的存在 (即抗体、ECM 模拟蛋白/生物聚合物等），[305] 和 iii) 存在生物指导性形态发生素（例如，生长因子、细胞因子等）。 [318]这些基本的设计考虑和加工技术使研究人员能够制造 3D 结构，其中粒子-细胞粘附是随机的或定向的。 [319]这两种方法都可以保证生物工程微组织在自下而上的组织工程和再生医学中的应用。
Random cellular adhesion to microfabricated spherical particles is by far the most explored alternative owing to its simplicity, low cost, and possibility to produce dense 3D microtissues which guide cell fate.[320] Using this concept, researchers have explored gelatin and heparin microparticles for loading BMP-4/noggin bioactive molecules and to support the establishment of pluripotent stem cell agglomerates in vitro. During culture, the controlled release of soluble mediators actively bioinstructed 3D cellular aggregates toward specific phenotypes, thus demonstrating the advantages of using microparticles as cell-adhesive and delivery platforms.[321] In a similar approach, the controlled release of TGF-β1 from gelatin microparticles impregnated in human-periosteum-derived cells (hPDCs) micromasses induced significant cellular differentiation toward chondrogenic lineage and the expression of chondrogenic biomarkers.[322] Microparticle-guided 3D microtissues random assembly was also recently explored through the fabrication of surface-decorated microparticles with human E-cadherin fusion protein cell–cell adhesion biomimetic ligands that improved cells proliferation and formation of dense MSCsrich 3D aggregates (Figure 9A).[323] However, when envisioning in vitro generated 3D microtissues implantation, the establishment of highly dense constructs is generally deleterious due to decreased oxygen/nutrients diffusion. Recent efforts have been made toward surpassing these issues through the development of VEGF165-functionalized microcarriers, or of oxygen releasing microparticles that increased the viability of 3D cellular masses cultured in microparticulated platforms.[324,325] The latter is a particularly elegant approach that has proven valuable for providing a timely supply of oxygen to microparticle adhered human-periosteal-derived cells which preserved their osteogenic differentiation even under hypoxia.[3
由于其简单、低成本和产生指导细胞命运的致密 3D 微组织的可能性，随机细胞粘附到微制造的球形颗粒是迄今为止探索最多的替代方法。 [320]利用这一概念，研究人员探索了用于装载 BMP-4/头蛋白生物活性分子的明胶和肝素微粒，并支持在体外建立多能干细胞团块。在培养过程中，可溶性介质的受控释放主动生物指导 3D 细胞聚集体朝向特定表型，从而展示了使用微粒作为细胞粘附和递送平台的优势。 [321]在类似的方法中，TGF-β1 从浸渍在人骨膜衍生细胞 (hPDC) 微团中的明胶微粒中的受控释放诱导显着的细胞分化为软骨形成谱系和软骨形成生物标志物的表达。 [322]微粒引导的 3D 微组织随机组装最近也通过制造表面修饰的微粒与人类 E-钙粘蛋白融合蛋白细胞-细胞粘附仿生配体进行了探索，该配体改善了细胞增殖和致密 MSCsrich 3D 聚集体的形成（图 9A）。 [323 ]然而，当设想体外生成的 3D 微组织植入时，由于氧/养分扩散减少，高密度结构的建立通常是有害的。最近通过开发 VEGF165 功能化微载体或释放氧气的微粒来解决这些问题，这些微载体增加了在微粒平台中培养的 3D 细胞团的活力。[324,325] 后者是一种特别优雅的方法，已被证明是有价值的。为微粒粘附的人骨膜衍生细胞提供及时的氧气供应，即使在缺氧条件下也能保持其成骨分化。
Adding to these technologies, open porous microparticle platforms have also received a significant focus for 3D microtissues assembly owing to their improved gases/nutrients and waste metabolites exchange with the surrounding microenvironment. Recently microporous particles technology was advanced with the fabrication of highly open porous polyhydroxyalkanoate (PHA) microspheres (OPMs; 300–360 µm in diameter, Figure 9B), which are able to harbor stem cells in their interconnected network while also assuring a higher cell viability and continuous proliferation in comparison to their less porous counterparts.[3
除了这些技术之外，开放式多孔微粒平台也因其改进的气体/营养物质和废物代谢物与周围微环境的交换而受到了 3D 微组织组装的关注。最近，微孔颗粒技术得到了进步，制造了高度开放的多孔聚羟基链烷酸酯 (PHA) 微球 (OPM；直径 300–360 µm，图 9B)，这些微球能够在相互连接的网络中容纳干细胞，同时确保更高的细胞活力
Aiming to further control microtissues shape, researchers used microarray platforms to promote the fusion of cell– particle microaggregates as an attempt to generate geometrically defined higher order assemblies.[320] This pooling and spatial confinement resulted in the bottom-up establishment of macrosized and shape defined tissue constructs (Figure 9C). Programmed particle–cell microagglomerates confinement and directed assembly into larger assemblies can also be established via magnetism, sound, mechanical forces, and light.
Despite these exciting advances, assuring a precise control over cells adhesion in specific microparticle regions remains remarkably challenging. While the dynamic nature of microparticles–cells interactions and aggregations allows researchers to explore ex vivo 3D microtissues assembly, due to its underlying concepts such interactions are inevitably random and with limited control over cell-specific adhesion. Recent studies have reported exciting results on the development of microparticles displaying anisotropic/heterogenic regions for programmed/guided cell adhesion. In particular, spatially designed cellular microenvironments in spherical particles volume have been generated by using segmented microcapillaries to eject microdroplets containing multiple types of collagen and sodium alginates.[328,329] This resulted in the fabrication of anisotropic microspheres with hemispheres containing two ECM microenvironments with different mechanical properties. On another approach, biomimetic anisotropic PCL particles (≈70 µm) exhibiting fuzzy and smooth surfaces in either side were used to selectively adhere fibroblasts and endothelial cells (Figure 9D). In vitro cell culture of endothelial cells added stepwise to Janus particles smooth side, induced different prostacyclin secretion by endothelial cells. These particles demonstrated higher affinity toward fibroblasts as opposite to hepatocytes, opening new avenues for cell isolation based solely on particles morphological features.[330] Selective cells isolation in antibody-functionalized microparticles has also been successfully demonstrated, further contributing toward the toolbox of available approaches for biospecific control over cell–particle adhesion.[314] In the long-run such strategies can be valuable for assembly of heterogeneous microtissues in which the build-up process is self-regulated by cells spatial distribution on particles surface. One could hypothesize that combining this approach with genetic cell surface engineering could yield living constructs with bioregulated architectures.
3.3.2. Bioinstructive and 3D Bioprocessed Microparticle-Based Assemblies(生物指导和 3D 生物加工微粒基组件)
As aforementioned, it is well established that the orchestrated presentation of different bioinstructive cues (e.g., growth factors, cytokines, mechanical, magnetic/electric, etc.), either via their spatiotemporally controlled presentation or in the form of biochemical gradients, is also essential for providing a close-to native microenvironment and further potentiating engineered microtissue maturation in vitro.[331] In this sense, besides functioning as building blocks for cellular aggregation and microtissues maturation, microparticles have also been employed as delivery systems for controlled release of bioactive molecules in bottom-up cell–biomaterial assemblies. This strategy has been successful in guiding the fate of microparticle-adhered cell unitary blocks, or through incorporation in cell-laden bioinks. The latter is particularly valuable, since the direct incorporation of growth factors in bioinks results in rapid diffusion, often reducing bioactive molecules overall concentration comparing to that achieved with sustained presentation via microparticle-mediated controlled delivery.[332] Researchers have explored these microparticles features for instance by incorporating gelatin microparticles loaded with either VEGF or BMP-2 into bioinks for achieving accelerated osteodifferentiation and regional angiogenesis on bioprinted 3D living constructs with well-defined architecture.[333,334] Adding to this, microparticles loaded with growth factor/cytokine rich cell-derived secretomes have also been employed as bioinstructive building blocks and are currently being envisioned for potentiating advanced 3D bioprinting strategies in the future.[3
如前所述，众所周知，不同生物指导线索（例如，生长因子、细胞因子、机械、磁/电等）的协调呈现，无论是通过它们的时空控制呈现还是以生化梯度的形式，也是必不可少的用于提供接近天然的微环境并进一步促进体外工程化微组织成熟。 [331]从这个意义上说，除了作为细胞聚集和微组织成熟的基石外，微粒还被用作自下而上的细胞-生物材料组件中生物活性分子的受控释放的递送系统。该策略已成功指导微粒粘附的细胞单元块的命运，或通过掺入载有细胞的生物墨水中。后者特别有价值，因为在生物墨水中直接掺入生长因子会导致快速扩散，与通过微粒介导的控制递送持续呈现所达到的浓度相比，通常会降低生物活性分子的总浓度。 [332]研究人员已经探索了这些微粒的特性，例如，通过将装载有 VEGF 或 BMP-2 的明胶微粒加入生物墨水中，以在具有明确结构的生物打印 3D 活体结构上实现加速骨分化和区域血管生成。 [333,334] 此外，装载有富含生长因子/细胞因子的细胞衍生分泌物也已被用作生物指导性构建块，目前正被设想用于在未来增强先进的 3D 生物打印策略。
In addition, to biochemical signals, the inclusion of mechanical cues in bottom-up engineered constructs via microparticles inclusion was also fruitfully demonstrated in recent works. An elegant approach, involved MSC-laden PLA microparticles embedding within GelMA-gellan gum bioinks, functioning as unitary building blocks for providing both cellular components and mechanical reinforcement. This approach increased up to twofold the compression modulus of bioprinted 3D constructs.[336] On a different setup, β-tricalcium phosphate microparticle incorporation were used for tuning 3D constructs stiffness,[337] as well as to induce osteodifferentiation and mimicking the mineral fraction present in calcified cartilage.
此外，对于生化信号，最近的工作也卓有成效地证明了通过微粒包含在自下而上的工程结构中包含机械线索。一种优雅的方法是将载有 MSC 的 PLA 微粒嵌入 GelMA-结冷胶生物墨水中，作为提供细胞成分和机械增强的单一构件。这种方法将生物打印 3D 结构的压缩模量提高了两倍。 [336]在不同的设置中，β-磷酸三钙微粒掺入用于调整 3D 构建体的刚度，[337] 以及诱导骨分化和模拟钙化软骨中存在的矿物质部分。
The inclusion of cell-laden microparticles in bioinks developed for advanced 3D biofabrication adds a layer of processability to these building blocks and allows the engineering of microparticle-based living constructs with biospecific tissue designs that would be otherwise difficult to obtain through other bottom-up processing technologies. However, it is important to emphasize that including dense polymeric microparticles in living tissue constructs is an unnatural approach and may generate nutrient/oxygen diffusion limitations, during long-term maturation of microtissues in vitro, an important parameter that must be considered at early design stages when modulating cell–microparticles density/ concentration.
为先进的 3D 生物制造开发的生物墨水中包含载有细胞的微粒，为这些构建模块增加了一层可加工性，并允许使用生物特异性组织设计来设计基于微粒的活体结构，否则这些结构很难通过其他自下而上的处理获得技术。然而，需要强调的是，在活组织结构中包含致密聚合物微粒是一种不自然的方法，并且可能会在体外微组织长期成熟期间产生营养/氧气扩散限制，这是在早期设计阶段必须考虑的重要参数当调节细胞微粒密度/浓度时。
Apart from particles for bottom-up tissue engineering approaches, also cell-laden hydrogels processed as nano/microgel particles or fibers provide a seamless supporting matrix to buildup and mature 3D microtissues as it will be discussed in the following sections.
除了用于自下而上的组织工程方法的颗粒外，作为纳米/微凝胶颗粒或纤维加工的载有细胞的水凝胶也为构建和成熟的 3D 微组织提供了无缝的支撑基质，这将在以下部分中进行讨论。
3.4. Hydrogel-Based Platforms for Biomimetic Bottom-Up Assembly(基于水凝胶的仿生自下而上组装平台)
Utilizing biomaterial-based ECM mimicking matrices for pursuing tissue-specific features and biomolecular gradients unravels the potential to achieve high-quality biological modules with flexible and augmented biofunctionality that can support cell adhesion, proliferation and ultimately de novo tissue morphogenesis. In essence, hydrogels are highly hydrated and self-supporting 3D networks with porous or fibrillar-like internal architectures similar to native ECM. These building blocks can be based on: i) natural sources, i.e., self-assembling peptides, engineered proteins, polysaccharides, and decellularized ECM; ii) synthetic polymeric materials and iii) hybrid hydrogels arising from the combination of different monomers/nanocomposites in the hydrogel network. The organization, composition, and structural features of the ECM vary significantly across tissues, hence, a universal building block may be unrealistic.[340] In a bottom-up perspective, hydrogels are advantageous due to their flexible design nature, derived from their polymeric framework and its amenability to be chemically tailored to better reproduce key ECM features (e.g., adhesion site density, biomolecule immobilization, matrix stiffness, stretchability, degradability, etc.). Moreover, their ease for obtaining cell-laden structures, sustaining proliferation, as well as providing 4D bioarchitectures that can be altered and matured through embedded cells activity or by specific stimuli are highly desirable in such bioengineering strategies
The specificity of ECM-like microenvironments materialized through the encapsulation of living healthy and fully functional cells poses several challenges to the development of spatially controlled cell-laden hydrogels: i) the effective threedimensionality of the intended patterns that ii) simultaneously enable the compatibility of the 3D-assembled hydrogels with cell encapsulation via cell adhesion, but that can also direct the fate and function of the engineered bioarchitecture. By now, the importance of scaling-up cell–matrix interactions to the third dimension is unquestionable because it affects the biological activity of the final construct.[341] Despite the unambiguous importance of exposing cells to 3D-controlled cues, the patterning of hydrogel matrices has been mostly achieved solely on a 2D perspective and frequently of a multistep nature or resorting to top-down reactions based on photolithographic methods. For example, photosensitive S-2-nitrobenzyl-cysteine moieties immobilized on a nonadhesive agarose matrix allow for UV-induced uncaging of sulfhydryl groups at specific regions, where maleimide-terminated peptides containing the fibronectin binding domain can be precisely tethered to the matrix framework via thiol-ene up to a 1.5 mm depth.[342] In this way, cell adhesion can be spatially controlled, unlike in tissue-derived materials (e.g., collagen and fibrin), which contain randomly dispersed cell adhesion domains. Currently, with emerging multiphoton-based lithography, researchers have now successfully reported true 3D patterning of multiple growth factors with subcellular precision and interconnectivity (Figure 10A).[342,343] Beyond that, orthogonal photodegradation has been exploited to engineer cell-laden hydrogels with intricate and perfusable cavities and microchannels with user-controlled shapes and location (Figure 10B–E).[344–346] Alternatively, in a truly biomimetic approach, Culver and co-workers have exploited two-photon lithography for manufacturing cell-laden PEG-based hydrogels with precise 3D cell immobilization based on native tissue sections as biological blueprints.[347] These hydrogels were rendered cell-degradable by imparting the PEG-diacrylate backbone with matrix metalloproteinase-sensitive peptide (GGPQGIWGQGK). Here, the researchers imaged cross-sections of different tissues (e.g., retina, cerebral cortex, heart, etc.) and were able to precisely match the hydrogel patterning to the original tissue vascular bed. This image-guided 3D photopatterning successfully recapitulated the neural stem cell niche seen in the subependymal zone of mice (Figure 10F). The versatility of this technology allowed them to pattern distinct bioactive peptides on regions of interest and accurately recreating vascular and neural progenitor staining of the biological tissue section. Moreover, the advent of photoinitiator-free strategies based on the [2 + 2] cycloaddition of maleimide groups at physiological conditions will foster the development of truly multiphoton click chemistry reactions for engineering cell microenvironment niches in hydrogel structures.
通过封装活的健康和功能齐全的细胞实现的类 ECM 微环境的特异性对空间控制的载细胞水凝胶的开发提出了几个挑战：i）预期模式的有效三维度，ii）同时实现兼容性通过细胞粘附进行细胞封装的 3D 组装水凝胶，但这也可以指导工程生物结构的命运和功能。到目前为止，将细胞-基质相互作用放大到三维的重要性是毋庸置疑的，因为它会影响最终构建体的生物活性。 [341]尽管将细胞暴露在 3D 控制的线索下具有明确的重要性，但水凝胶基质的图案化大多仅在 2D 视角下实现，并且通常具有多步性质或诉诸基于光刻方法的自上而下的反应。例如，固定在非粘附性琼脂糖基质上的光敏 S-2-硝基苄基-半胱氨酸部分允许 UV 诱导的巯基在特定区域的脱嵌，其中含有纤连蛋白结合结构域的马来酰亚胺末端肽可以通过以下方式精确地拴在基质框架上硫醇烯的深度可达 1.5 毫米。[342]通过这种方式，可以在空间上控制细胞粘附，这与包含随机分散的细胞粘附域的组织衍生材料（例如胶原蛋白和纤维蛋白）不同。目前，随着基于多光子的光刻技术的兴起，研究人员现已成功报道了具有亚细胞精度和互连性的多种生长因子的真正 3D 图案化（图 10A）。[342,343] 除此之外，正交光降解已被用于设计具有复杂结构的载细胞水凝胶。以及具有用户控制的形状和位置的可灌注腔和微通道（图 10B-E）。[344-346] 或者，在真正的仿生方法中，Culver 及其同事利用双光子光刻技术制造载有细胞的 PEG-基于天然组织切片作为生物蓝图的具有精确 3D 细胞固定的水凝胶。 [347]通过赋予 PEG-二丙烯酸酯骨架与基质金属蛋白酶敏感肽 (GGPQGIWGQGK)，这些水凝胶可被细胞降解。在这里，研究人员对不同组织（例如视网膜、大脑皮层、心脏等）的横截面进行了成像，并能够将水凝胶图案与原始组织血管床精确匹配。这种图像引导的 3D 光模式成功地概括了在小鼠室管膜下区看到的神经干细胞生态位（图 10F）。该技术的多功能性使他们能够在感兴趣的区域形成不同的生物活性肽，并准确地重建生物组织切片的血管和神经祖细胞染色。此外，基于在生理条件下马来酰亚胺基团的 [2 + 2] 环加成的无光引发剂策略的出现，将促进真正的多光子点击化学反应的发展，用于工程化水凝胶结构中的细胞微环境生态位。
Despite the relevance of top-down patterning techniques that enable the precise control of hydrogels chemistry on a 3D space and over time, the self-assembly of precisely designed hydrogels through spontaneous intermolecular interactions benefiting from easy injectability and tailoring of biophysical properties of ECM-mimetic structures could further potentiate these applications.[221,350] However, contrarily to the abovementioned examples of top-down patterning of 3D hydrogels, reports of exclusively bottom-up hydrogel assembly strategies targeting the precise positioning of biochemical patterns and gradients in 3D are still limited. Recent works have begun to utilize molecular self-assembly to generate hydrogel 3D bioarchitectures capable of continuously expanding their network similar to the growth of living tissues in nature.[351] Here, instead of growth factors, researchers manipulated oxygen concentration to engineer complex 3D shapes and also intricate out-of-plane buckling by inhibiting growth at specific regions. This is the first study for generating self-assembled hydrogel architectures inspired by the bottom-up morphogenesis process of living tissues, but cell-friendly conditions will still need to be attained for achieving relevant cell-laden hydrogel structures.
尽管自上而下的图案化技术能够在 3D 空间和时间上精确控制水凝胶化学，但精确设计的水凝胶通过自发的分子间相互作用进行自组装，这得益于易于注射和定制 ECM 模拟物的生物物理特性[221,350] 然而，与上述 3D 水凝胶自上而下图案化的示例相反，针对 3D 中生化模式和梯度的精确定位的专门自下而上水凝胶组装策略的报道仍然有限。最近的工作已经开始利用分子自组装来生成水凝胶 3D 生物结构，该结构能够不断扩展其网络，类似于自然界中活组织的生长。 [351]在这里，研究人员操纵氧气浓度而不是生长因子来设计复杂的 3D 形状，并通过抑制特定区域的生长来实现复杂的平面外屈曲。这是第一项受活组织自下而上形态发生过程启发而生成自组装水凝胶结构的研究，但仍需要达到对细胞友好的条件才能实现相关的载细胞水凝胶结构。
The ability of synthetic or recombinant short peptide sequences to function as hydrogelators has been used as the driving principle of one of the most commonly reported strategies to prepare bottom-up self-assembled hydrogels.[350,352–355] The use of amino acid sequences is an interesting approach due to their possible similarity with native ECM features capable of driving cell adhesion and other cell response phenomena, which include cell migration, apoptosis, mechanotransduction, vascularization, among others. Moreover, most of reported amino acid sequences used for the preparation of hydrogels are widely considered biocompatible, potentially biodegradable and nonimmunogenic.[356] The mechanism of hydrogelation of peptide blocks is generally regulated by noncovalent interactions which drive their assembly into fibrous structures that entangle in the form of 3D highly hydrated solid structures in the presence of stimuli such as ions, enzymes, temperature, pH changes or exposure to light.[356–359] Although the mechanisms driving the formation of supramolecular hydrogels are mostly based on noncovalent bonds, researchers’ recently reported the assembly of amphiphilic peptides through nonweak interactions convertible to covalent bonding.[360] Self-assembled networks based on nanofiber peptides include vast applications on neural, myocardial and wound regeneration.[361–363] The cell-friendly conditions under which the spontaneous assembly occurs, along with the ability to tailor the biochemical composition to mimic ECM cues, enabled the exploitation of bottom-up self-assembled peptide hydrogels as promising cell encapsulating matrices. The dispersion of hepatocyte-like spheroids in a RAD16-I peptide hydrogel showed that the generated 3D hydrogel significantly tailored cellular proliferation and presented matured differentiation profiles.[364] Indeed, it has been recently disclosed that the bioactivity of amphiphile mimetic peptides (e.g., brain-derived neurotrophic factor, BDNF) can be augmented in 3D hydrogels due to conformationally improved peptide–cell interactions in these systems, which not only encourages cell infiltration but increases functional maturation of the construct.[365] A co-assembly system of peptide amphiphiles designed to form nanofibers targeting cartilage repair was also proven promising for tissue regeneration based on a cell encapsulation approach.[366] Indeed, the decoration of peptide segments with the TGF-β1 binding epitope (i.e., HSNGLPL sequence), which is exposed in high density to encapsulated cells and damaged tissues due to the collapsing of the hydrophobic alkyl chains in the amphiphilic molecule, promoted the chondrogenic differentiation of mesenchymal stem cells encapsulated within the hydrogel and boosted the formation of hyaline cartilage in osteochondral defects in in vivo rabbit models. Hydrogels based on other peptide amphiphiles with bioactive domains as IKVAV , or with a tenascin-C-mimetic configuration, showed potential as cell encapsulation and regeneration matrices for inner ear and neural repair.
合成或重组短肽序列作为水凝胶剂的能力已被用作制备自下而上自组装水凝胶的最常见报告策略之一的驱动原理。 [350,352–355] 氨基酸序列的使用是一种有趣的方法，因为它们可能与能够驱动细胞粘附和其他细胞反应现象的天然 ECM 特征相似，包括细胞迁移、细胞凋亡、机械转导、血管化等。此外，大多数用于制备水凝胶的已报道氨基酸序列被广泛认为具有生物相容性、潜在可生物降解性和非免疫原性。 [356]肽块的水凝胶化机制通常由非共价相互作用调节，非共价相互作用驱动它们组装成纤维结构，在离子、酶、温度、pH 变化或暴露于光等刺激下以 3D 高度水合固体结构的形式纠缠在一起[356–359] 尽管驱动超分子水凝胶形成的机制主要基于非共价键，但研究人员最近报道了两亲肽通过可转换为共价键的非弱相互作用组装。 [360]基于纳米纤维肽的自组装网络包括在神经、心肌和伤口再生方面的广泛应用。[361-363] 自发组装发生的细胞友好条件，以及定制生化成分以模拟 ECM 线索的能力，使自下而上的自组装肽水凝胶作为有前途的细胞封装基质得以开发。肝细胞样球体在 RAD16-I 肽水凝胶中的分散表明，生成的 3D 水凝胶显着调整了细胞增殖并呈现出成熟的分化特征。 [364]事实上，最近发现两亲肽（例如脑源性神经营养因子，BDNF）的生物活性可以在 3D 水凝胶中增强，因为这些系统中的肽-细胞相互作用的构象得到改善，这不仅促进细胞浸润，而且增加结构的功能成熟度。 [365]设计用于形成靶向软骨修复的纳米纤维的肽两亲物的共组装系统也被证明有希望基于细胞封装方法进行组织再生。 [366]实际上，由于两亲性分子中疏水性烷基链的坍塌，以高密度暴露于封装细胞和受损组织中的 TGF-β1 结合表位（即 HSNGLPL 序列）修饰肽段，促进了软骨形成。包封在水凝胶中的间充质干细胞的分化，促进了体内兔模型骨软骨缺损中透明软骨的形成。基于其他肽两亲物的水凝胶具有 IKVAV 等生物活性结构域或具有生腱蛋白-C 模拟结构的水凝胶，显示出作为内耳和神经修复的细胞封装和再生基质的潜力。
Native tissues can display different degrees of anisotropy and spatially varying stiffness, that are key players in guiding cell migration, organization, and function.[340] Beyond their flexibility for designing self-assembled hydrogels with tunable biochemical and mechanical features, peptide amphiphiles can be also explored for developing biomimetic constructs. In an innovative study, researchers have exploited a C16G3RGDS peptide amphiphile for programing cell-driven contraction of compressed collagen hydrogels in defined regions.[371] Here, cells served as bioactuators capable of mechanically transforming hydrogels into curved structures, thus remodeling the dense collagen stroma toward a more native-like organization found in human cornea. In fact, these 4D self-curved tissues presented superior ECM organization and orthogonally aligned cells over planar substrates. Alternatively, an interesting study designed by DeForest group exploited fusion proteins as hydrogel crosslinkers to fabricate bioarchictectures that respond to userdefined stimuli such as calcium (e.g., calmodulin-based proteins) and light (e.g., photosensitive light, oxygen, and voltage sensing domain 2, LOV2).[372] In this study, the unique semisynthetic protein-PEG hydrogel matrices yielded tunable stiffness-patternable features which could be cycled on demand to investigate 3D cellular response to cyclic loading, a key aspect that native ECM is subjected to but was yet to be replicated in cell-laden hydrogel constructs (Figure 10G). In another sophisticated work, researchers have manufactured core–shell hyaluronic-based hydrogels with distinct biochemical and biophysical features.[373] Cell adhesion site density, enzymatic degradability, and mechanical stiffness could be hierarchically presented in a core–shell bioarchitecture by varying the macromer/crosslinker ratios and timing of introduction of the interfacial crosslinking agents that covalently functionalized hydrogels at the liquid-gel interface. The biorthogonal tetrazine-trans-cyclooctene ligation here used occurs in aqueous conditions with extremely fast kinetics and without using any specialized equipment, catalyst, or triggering stimuli. The responses of human mesenchymal stem cells encapsulated in these hydrogels were assessed according to the different properties of the core–shell bioarchitectures. For homogenous (i.e., core = shell) stiff hydrogels, cells remained spherical ≈7 days in those nondegradable and nonadhesive platforms, whereas MMP-degradable gels with few adhesion sites were sufficient to foster cells spreading. The researchers could then assemble heterogenous hydrogels with spatially resolved properties, such as stiff-to-soft transitions from shell to core, as well as increasing MMP degradability or adhesion site density toward the core. Recently, the same group has now devised multilayered hydrogel channels with three different cell populations spatially patterned across its interfaces, thus yielding functional bioarchitectures more closely resembling native arteries.[374] Another relevant strategy to obtain cellularly graded hydrogels exploits the self-healing phenomena that is characteristic of dynamic covalent hydrogel networks. Using this approach cell-laden hydrogels based on 2-acrylamidophenylboronic acid and poly(vinyl alcohol), which readily self-assembled in aqueous conditions were successfully fabricated.[373] Due to the dynamic nature of boronic-diol bonds, these hydrogels exhibited significant self-healing capacity, which was exploited for establishing a gradient of two different cell types (e.g., lung fibroblasts and breast cancer cells) initially cultured separately on different hydrogels. After cutting each cell-laden hydrogel block, two halves were merged together due to their natural self-healing features and simultaneous cell migration could be observed across the healed interface of the newly formed binary hydrogel. This current design could be further improved by including ECM-mimetic domains eliciting adhesion or matrix remodeling. In the future, the preparation of cell-laden hydrogels with distinct customizable microenvironments will undoubtedly be essential for manufacturing advanced bottom-up bioarchitectures.
Noncovalent interactions (e.g., host–guest inclusion complexes, avidin–biotin, and nucleotide base pairs) are attractive not only for self-assembling bottom-up hydrogels, but also because they can be engineered to endow the hydrogel networks with unique features owing to its reversible and adaptive nature. In host–guest interactions, the binding readiness between opposing pairs is convenient because it partially recapitulates the constant adaptiveness of native ECM and embedded cells, thus being promising for modulating the biochemical 3D environment.[375] For instance, the well-known host–guest β-cyclodextrin/adamantane pair can be used for tackling the dynamic mechano-regulation of cell–substrate interactions.[376] Adjusting ratios of host/guest monomers allowed the precise tailoring of the initial mechanical properties of the hydrogel, while the subsequent addition of free competing host molecules enabled softening of the material ondemand. In comparison to previously reported systems based on UV irradiation or temperature as sources of mechanomodulation, the authors hypothesize that this host–guest system could allow handling cells with high viability as compared to potentially damaging stimuli. Synthetic adamantane and cyclodextrin derivatives are commercially available and can be easily installed into a plethora of polymers, including polysaccharides such as hyaluronic acid that are frequently used for designing cell-bearing bottom-up hydrogels with shear-thinning properties.[377,378] The ability of such hydrogels to withstand high deformations has seldom been studied. T o this end, researchers have explored acrylated-β-cyclodextrin and adamantane-hyaluronic acid as macro-crosslinkers of ductile polymers prepared through the polymerization of N,N-dimethylacrylamide monomers.[379] This novel strategy yielded highly swollen cell-laden hydrogels with outstanding resistance to fatigue, i.e., up to 80% compressive deformation for over 1000 cycles. The high viability observed for encapsulated cells after deformation cycles makes these hydrogels promising materials for mechanostimulation and interesting candidates for reproducing tissues subjected to cyclic deformation such as cartilage.
Alternatively, natural host–guest pairs like streptavidin–biotin have also enabled the processing of cell-laden structures.[377,380] For example, living cells modified with avidin were used as hydrogel crosslinkers through direct interaction with biotinylated 1,4-benyl-dicarbonxamide supramolecular gelators capable of being easily modified and with the ability to self-assemble.[380] The stability achieved in this system, in opposition to the comparatively poor binding verified on polymeric molecules, drove selective and faster proliferation of different cell types. However, drawbacks associated with the avidin–biotin pair and cyclodextrin-based inclusion complexes include the difficult scale-up production of molecules modified with the streptavidin–biotin pair and the poor in vivo application of cyclodextrin-based pairs due to low binding affinity.[381] T o overcome such limitations, Kim ’s group suggested novel supramolecular hydrogel assemblies based on hyaluronic acid modified with pumpkin-shaped cucurbit[6]uril (CB[6]) or diaminohexane groups.[381] The highly selective and strong CB[6]–diaminohexane interaction allowed cell encapsulation on self-assembled hydrogels,[382] which could then be later modularly modified by treating the hydrogel with multifunctional tags-CB[6], which included RGD domains to promote cell adhesion, FITC probes for in vivo detection,[381] or TGF-β3 and dexamethasone to regulate mesenchymal stem cells’ chondrogenic differentiation.[383] Other interesting systems can comprise dock-and-lock (DnL) mechanisms based on engineered proteins and anchoring proteins attached to multiarm crosslinker polymers, which can instantly lock onto recombinant “docking” domains under physiological conditions.[384] Hydrogels prepared using this chemistry displayed a remarkable ability to recover from deformation cycles with selfhealing properties independently from mechanical disruption, while enabling cytocompatible encapsulation and injection of mesenchymal stem cells.
Owing to their several attractive features, including controllable sequences, precise recognition, and low toxicity, DNA building blocks have evolved considerably in the last decades for designing advanced hydrogel systems.[385] However, hydrogels based on spontaneous nucleotide pairing are still scarce, and the majority of reported technologies are dependent on enzymes,[386] pH changes and temperature variations to promote DNA fragmentation.[387] In fact, the decoration of macromolecules (e.g., polymeric chains and peptide sequences) with complementary nucleotide sequences has rendered the most effective and stimulus-free cell-compatible self-assembling of hydrogel micro- and macroscopic units.[388,389] In this context, progress in 3D bioprinting enabled by DNA fragments hybridization has progressed significantly over the last years. Researchers have developed a novel class of bioinks based on a polypeptide–DNA conjugate (bioink A) and complementary DNA linkers (bioink B).[389] Upon co-injection under physiological conditions both inks immediately formed a stable hydrogel that enabled high cell viability and the precise printing of multilayered structures with any intended shape. In addition, due to its polypeptide and nucleic backbone, the millimeter-sized bioarchitectures were naturally degradable through the action of nucleases or proteases. Recently, carboxymethylcellulose-based hydrogels crosslinked by self-complementary DNA interactions (duplex nucleic acids) and donor– acceptor (dopamine-bipyridinium) redox-mediated switchable bonds yielded stimuli-responsive networks with dynamic stiffness and simultaneous self-healing and shape memory properties.[390] Double network hydrogels that self-assemble entirely from noncovalent interactions, i.e., DNA hybridization and host–guest inclusion complexes between cucurbit[8] uril (CB) and phenylalanine functionalized carboxymethylcellulose have also been reported.[391] The resulting fully interpenetrating supramolecular network exhibited remarkable stretchability, ductility, shear-thinning and thixotropic properties. Moreover, the use of DNA motifs for programming constructs self-assembly is a unique strategy because such DNA motifs can potentially be recognized and processed by embedded cells or enzymes.[392] Hence, they can signal for the enzymatic machinery of cell lysates and instruct them to synthesize functional proteins, thus assembling biorelevant protein-producing hydrogels.[
由于它们的几个吸引人的特性，包括可控序列、精确识别和低毒性，DNA 构建模块在过去几十年中已经发生了相当大的发展，用于设计先进的水凝胶系统。 [385]然而，基于自发核苷酸配对的水凝胶仍然稀缺，大多数报道的技术都依赖于酶、[386] pH 变化和温度变化来促进 DNA 片段化。 [387]事实上，具有互补核苷酸序列的大分子（例如，聚合物链和肽序列）的装饰已经使水凝胶微观和宏观单元的最有效和无刺激的细胞相容自组装。 [388,389] 在这种情况下，在过去几年中，通过 DNA 片段杂交实现的 3D 生物打印取得了显着进展。研究人员开发了一种基于多肽-DNA 偶联物（bioink A）和互补 DNA 接头（bioink B）的新型生物墨水。 [389]在生理条件下共注射后，两种墨水立即形成稳定的水凝胶，可实现高细胞活力和任何预期形状的多层结构的精确印刷。此外，由于其多肽和核酸骨架，毫米大小的生物结构可以通过核酸酶或蛋白酶的作用自然降解。最近，基于羧甲基纤维素的水凝胶通过自互补 DNA 相互作用（双链体核酸）和供体-受体（多巴胺-联吡啶）氧化还原介导的可切换键交联，产生了具有动态刚度和同时自愈和形状记忆特性的刺激响应网络。 [390]还报道了完全由非共价相互作用自组装的双网络水凝胶，即 DNA 杂交和葫芦[8] 尿素 (CB) 和苯丙氨酸功能化羧甲基纤维素之间的主客体包合复合物。 [391]由此产生的完全互穿的超分子网络表现出显着的拉伸性、延展性、剪切稀化和触变性。此外，使用 DNA 基序对构建体进行编程是一种独特的策略，因为这些 DNA 基序可能被嵌入的细胞或酶识别和加工。 [392]因此，它们可以为细胞裂解物的酶促机制发出信号，并指导它们合成功能性蛋白质，从而组装出生物相关的蛋白质生产水凝胶。
Considering the versatility and interchangeable features of cell–biomaterial hydrogel assemblies, their combination into higher order 3D bioarchitectures with well-defined and complex geometries that recapitulate the anatomic features of different human tissues is envisioned to open new avenues toward their application in numerous clinical scenarios.This potential is starting to be materialized with the advent of biofabrication technologies that allow cells and biomaterials precise processing and positioning under conditions that assure maximum viability and biofunctionality
考虑到细胞-生物材料水凝胶组件的多功能性和可互换特性，将它们组合成具有明确定义和复杂几何形状的高阶 3D 生物架构，可以概括不同人体组织的解剖特征，从而为其在众多临床场景中的应用开辟新途径。随着生物制造技术的出现，这种潜力开始实现，该技术允许细胞和生物材料在确保最大生存能力和生物功能的条件下进行精确加工和定位
3.4.2. Advanced 3D Bioprocessed Cell-Laden Hydrogels(先进的 3D 生物处理载细胞水凝胶)
The latest advances obtained in cell-laden hydrogel bioprinting technologies have enabled the development of extremely sophisticated cell-hydrogel living architectures in native ECMmimetic microenvironments laden with multiple cell components and exhibiting evermore biospecific designs, as well as human anatomic-scale. Aiming to fabricate such constructs, researchers have produced pie-shaped alginate hydrogels with three cell types (i.e., amniotic-fluid-derived stem cells, smooth muscle cells, and aortic endothelial cells), containing spatially defined heterogeneous multiple cell distribution across the bioprinted construct by using an in-house modified thermal jet printer.[394] Inkjet bioprinting takes advantage from picoliter drop generation on-demand and high printing speeds for manufacturing 3D cell-laden hydrogel building blocks with high controlled spatial deposition of cells and biomaterials.[395] Beyond allowing one to spatially control cell distribution, this technology has also enabled the fabrication of fibroblastladen vascular-like tubes with appropriate horizontal/vertical bifurcations via an alginate-based bioink and piezoelectric jet printers.[396] In addition, inkjet-based bioprinting has been employed for producing layered structures with two different alginate-based bioinks, namely, 3D checkerboards, concentric/ halved circular patterns, and other complex layouts with high precision (≈100 µm).[397] These bioinks were also combined with MMP-sensitive PEG for producing 3D perfusable channels via a sacrificial layer-based method. Recently, inkjet-spray bioprinting that avoids the use of crosslinking baths has been developed for scalable manufacturing of layered hydrogel structures via an alginate/saponified GelMA-based bioink (Figure 11A).[395] By incorporating ECM-mimicking GelMA within the inkjet bioprinting process, this work improved the design over other attempts solely comprising alginate-based bioinks and could produce large-scale hydrogel structures with improved cell proliferation/spreading and collagen expression. This technology allowed faster bioprinting (≈3–23-fold) than the previously highlighted studies and is compatible with printing of heterogeneous hydrogel constructs. Moreover, GelMA blending improved hydrogels mechanical properties (i.e., compressive modulus, ≈10–25 kPa), an important aspect when foreseeing the biomedical applicability of such constructs. Despite this, the range of biomaterials compatible with inkjet processes is limited, and the fabrication of mechanically reinforced/thick 3D bioprinted constructs with tissue-like cell densities is highly challenging to materialize through this modality.[398]
在载有细胞的水凝胶生物打印技术方面取得的最新进展使我们能够在天然 ECM 模拟微环境中开发极其复杂的细胞水凝胶活体结构，该微环境中充满多种细胞成分，并展示出更多的生物特异性设计，以及人体解剖学规模。为了制造这种结构，研究人员生产了具有三种细胞类型（即羊水干细胞、平滑肌细胞和主动脉内皮细胞）的饼状藻酸盐水凝胶，在生物打印结构中包含空间定义的异质多细胞分布通过使用内部改进的热喷射打印机。 [394]喷墨生物打印利用按需生成皮升液滴和高打印速度来制造具有高度受控的细胞和生物材料空间沉积的载有 3D 细胞的水凝胶构件。 [395]除了允许人们在空间上控制细胞分布之外，该技术还能够通过基于藻酸盐的生物墨水和压电喷射打印机制造具有适当水平/垂直分叉的成纤维细胞血管样管。 [396]此外，基于喷墨的生物打印已被用于生产具有两种不同基于藻酸盐的生物墨水的分层结构，即 3D 棋盘、同心/半圆形图案和其他高精度 (≈100 µm) 的复杂布局。 [397]这些生物墨水还与 MMP 敏感 PEG 相结合，通过基于牺牲层的方法产生 3D 可灌注通道。最近，已开发出避免使用交联浴的喷墨喷射生物打印技术，用于通过基于藻酸盐/皂化 GelMA 的生物墨水大规模制造分层水凝胶结构（图 11A）。 [395]通过在喷墨生物打印过程中加入 ECM 模拟 GelMA，与仅包含基于藻酸盐的生物墨水的其他尝试相比，这项工作改进了设计，并且可以产生具有改善的细胞增殖/扩散和胶原蛋白表达的大规模水凝胶结构。与之前强调的研究相比，该技术允许更快的生物打印（≈3-23 倍），并且与异质水凝胶结构的打印兼容。此外，GelMA 混合改善了水凝胶的机械性能（即压缩模量，≈10-25 kPa），这是预测此类结构的生物医学适用性时的一个重要方面。尽管如此，与喷墨工艺兼容的生物材料的范围是有限的，并且通过这种方式实现具有组织样细胞密度的机械增强/厚 3D 生物打印结构的制造极具挑战性。 [398]
To bridge this gap, extrusion-based bioprinting is a valuable alternative as it allows bioink deposition at high cell densities and the manufacturing of large scale 3D constructs, being one of the most widely established bioprinting technologies due to its relative simplicity and versatility.[398] Using this approach cryopreservable cell-laden microgels have been employed as offthe-shelf bioinks for on-demand extrusion-based bioprinting of modular 3D constructs.[402] Alternatively, microfluidic-produced norbornene-functionalized hyaluronic acid/PEG-diacrylate microgels were employed as densely packed granular inks for 3D printing purposes, envisioning the design of particulated bioinks composed entirely of bundled microgels (Figure 11B). For obtaining the extrudable granular ink, cell-laden microgels containing NIH/3T3 fibroblasts were produced upon visible light photocrosslinking and subsequently compacted upon vacuum-driven filtration before printing, while maintaining high cell viability (≈70%) during both processes. Such shearthinning jammed microgel inks could be printed into diverse 3D constructs which could be stabilized with posterior interparticle photocrosslinking for producing mechanically robust architectures
However, from a critical perspective the packing density of these microgels must be carefully addressed to allow a compromise between 3D microtissues cellular density and the availability of nutrients/oxygen, as well as removal of metabolites during in vitro maturation. Aiming to address this important aspect, recent endeavors have focused on the potential of aqueous two-phase emulsion bioinks (i.e., immiscible GelMA/polyethylene oxide) for bioprinting hydrogel constructs with predesigned internal porous architectures.[403] The obtained pore-forming bioprinted 3D constructs displayed enhanced cell viability, spreading and proliferation over standard nonporous GelMA hydrogels across three different cell types (i.e., human hepatocellular carcinoma cells, human umbilical vein endothelial cells, and mouse embryonic fibroblasts). Although an elegant strategy for bioprinting pore-forming constructs with superior biological properties, these approaches still do not allow precise architectural control over vascular networks, a critical aspect in bottom-up tissue engineering. In this context, advanced multilayered coaxial extrusion systems have been employed for bioprinting core–shell hydrogels with prevascularized networks in a one-step process that allows researchers to obtain a more precise control over vascular frameworks, being superior to conventional sacrificial templating approaches.[404] In this approach, alginate comprising the innermost channel provided structural support during extrusion, while the final framework was permanently fixed by covalent photocrosslinking between GelMA/4-arm PEG-tetracrylate in the blend bioink. These researchers were able to assemble perfusable cell-laden hydrogel tubular 3D constructs that displayed highly organized cell spatial distribution, as well as tubular topology with gradually increasing, periodically varying or constant outer/inner diameters, thus achieving biologically relevant vascular networks with heterogeneous architectures.
Adding to this discussion, it is important to emphasize that 3D bioprinting of native-like cannular tissues (e.g., gastrointestinal tract, trachea, urinary bladder, urethra, blood vessels, etc.) should attempt to replicate not only the multilayered layout of such biological architectures, but also its varying multicellular composition.[405] T o address this requirement, a digitally coded microfluidic-based multichannel coaxial extrusion system was developed to allow continuous bioprinting of perfusable tubular structures with on-demand control over the number of concentric layers (i.e., up to 3 multilayers) spanning across userdefined lengths, while using tissue relevant cell types in separate bioinks.[405] Using this elaborate system, vascular tissues were produced with human smooth muscle cells SMCs/HUVECs, while bioprinting tubular urothelial structures containing human urothelial cells (inner layer) and human bladder SMCs (outer layer). Bioprinting of geometrically defined multicellular tubular hydrogels represents an important step toward creating biomimetic cannular microtissue constructs. Still, replicating the typical multimaterial composition of native tissues represents one of the most challenging aspects in modular bottom-up tissue engineering. Aiming to tackle this, microfluidic-based extrusion platforms were engineered for extruding more than a single bioink simultaneously.[406] Multimaterial extrusion platforms can now successfully bioprint up to seven different cell-laden hydrogel bioinks with fabrication speeds (up to 15 times faster) unmatched by conventional devices.[399] With this technology, bioprinting of several complex structures, such as multilayered cuboids, blood vessel-like rings, and miniaturized organ-like constructs with multiple bioinks was successfully achieved (Figure 11C). In this work, researchers have demonstrated the outstanding capacity for rapid continuous bioprinting of constructs with various cell types and material compositions that can be spatially controlled over defined locations or gradients. However, printing speed in such system is still a limiting factor when considering large-scale constructs engineering.[286] Recently, the integration of a microfluidic device capable of rapidly switching between multiple bioinks in a stereolithography-based platform equipped with dynamic mirror microdevice (DMD) technology has been employed for enhancing multi-biomaterial bioprinting.[400] Such device enabled the rapid 3D bioprinting of constructs with various bioinks containing different cell types, namely, musculoskeletal interfaces (e.g., C2C12, fibroblasts, and HUVECs) and tendon–bone interfaces (e.g., human MSCs, fibroblasts, and osteoblasts) (Figure 11D). This work uniquely combines the multi-biomaterial processing in multichannel microfluidic systems with the fast fabrication times and high spatial resolution of stereolithography-based bioprinting. Alternatively, the DMD capacity for dynamically projecting complex layouts has been employed for rapid 3D bioprinting of prevascularized constructs with functional endothelial networks that can anastomose with host circulation.[407] With this technology, researchers bioprinted hepatic 3D constructs with well-defined hexagonal lobule units of hepatic cells derived from human-induced pluripotent stem cells.[408]
Notably, another important addition to the toolbox of advanced light-enabled bioprinting has been recently disclosed. Volumetric bioprinting represents a groundbreaking technology that can materialize cell-laden hydrogels at unprecedented speed and allows their manufacturing with innumerous complex geometrical features at high-resolution.[401] Inspired by the principles of computed tomography, the researchers irradiated a rotating cylinder containing photocrosslinkable bioink (i.e., GelMA), photoiniator and cells with a sequence of 2D light patterns that intersect and elicit spatially resolved 3D bioprinting at sites of multiple exposures. Due to this, construct dimensions do not dictate printing time, unlike conventional bioprinting approaches. In fact, human auricle models (4.14 cm3; Figure 11E) fabricated by the volumetric approach were printed at unparalleled rates versus extrusion-based printing (i.e., up to 250 times faster) and digital light processing photomanufacturing (i.e., up to 75 times faster), while also exhibited smoother surfaces unlike the other techniques which presented filament/voxel-paved surface artifacts. Interestingly, one of the unique features of volumetric bioprinting is to produce free-floating structures without using any sacrificial hydrogel templates. T o demonstrate the potential of this feature, a functional ball-and-cage cardiac valve that enabled unidirectional flow was fabricated. Such valve design cannot be reproduced by extrusion-based or DMD-based bioprinting technologies without resorting to sacrificial templates. Also, researchers were able to bioprint several complex constructs (e.g., trabecular bone and meniscus) with relevant cell types (e.g., MSCs and articular chondroprogenitor cells, respectively) and high cell viability. However, current limitations of this approach, such as limited spatial control over multicellular and multi-biomaterial distribution that are achieved in recent microfluidic-based bioprinting, should be carefully considered
via transient supporting bioinks (i.e., alginate).[409] Alternatively, in an innovative approach, researchers have demonstrated the potential of self-assembling peptides for bioprinting a variety of ECM proteins and biomolecules (i.e., fibronectin, collagen, keratin, elastin-like proteins, hyaluronic acid, etc.) with high cell viability, thus serving as a versatile tool box that can provide tunable bioink composition and structural control.[410] Because collagen self-assembles at neutral pH, collagen bioinks can be readily bioprinted via simple pH modulation. Interestingly, researchers have recently combined this principle with the second generation of freeform reversible embedding of suspended hydrogels (FRESH v2.0) and successfully demonstrated 3D bioprinting of organ-scale human heart with patient-specific anatomical architecture and synchronized contractions of cardiac ventricles containing human embryonic stem-cell-derived cardiomyocytes (Figure 11F). Alternatively, bioprinting tissue beads and organ-specific dECM hydrogels could represent other rapidly expanding parallel strategies attempting to harness the complex material composition of native tissues that could play a role in orchestrating their biospecific functions.[398,411] Due to their viscoelastic characteristics, such bioinks could also benefit from improved shear thinning properties, which are known to play a role in maintaining cell integrity during the bioprinting process.[412] Hydrogels with self-healing behavior (e.g., host–guest interactions) can also function as supporting baths for accommodating extruded bioinks.[413] Although hydrogel viscoelasticity has proven to be beneficial in influencing cell behavior (e.g., proliferation, spreading, differentiation, and bioactivity) due to mechanical similarities with ECM, they can be associated with lower mechanical stiffness which may difficult the bioprinting of large scale structures.[414] In this context, bioinks can be mechanically reinforced by intercalating deposition of melted robust biomaterials (i.e., PCL) facilitating bioprinting of voluminous tissues with enhanced structural integrity and without compromising cell viability.[415,416] Other approaches comprise biocompatible benzimidazole-based biomolecule that can accelerate exchange dynamics of hydrazone crosslinking in hyaluronic acid-based hydrogels, which can be exploited for designing self-gelling bioinks with improved stability.[417]
Although still in its infancy, 4D bioprinting attempts are also beginning to emerge in bottom-up tissue engineering approaches, motivated by establishment of shape-shifting polymers and advent of actuating constructs.[418] Recent reports focused on the development of shape-morphing bioinks comprising methacrylated alginate/hyaluronic acid biopolymers and mouse bone marrow stromal cells, which could self-fold under aqueous conditions from sheet-based configuration into hollow tubes.[419] Such swelling-driven capillary assemblies displayed small internal diameters (>20 µm) and are thus more similar to smaller blood vessels, which are currently unattained by typical bioprinting techniques. Recently, 4D bioprinting of alginate/GelMA-based bioinks patterned with alginate/polydopamine inks, were used to generate 3D cellladen constructs with programmed shape-morphing features via NIR light-triggered local deswelling of the hydrogel.[420] On the same context these approaches can be further improved since, stereolithography-based grayscale digital light processing is now emerging as an enabling tool for manufacturing functionally graded materials with location-specific properties. This could allow 4D bioprinting of constructs with programmable buckling/deformation sequences.[421] Multiphoton technology is also capable of eliciting topographical changes in protein-based hydrogels via inner contraction of the structural network, thus allowing manipulation of cell microenvironment in 4D, but is still constrained by complex operation conditions and slow patterning rate.[422,433] Recent studies are also starting to modulate the mechanical properties of bioprinted constructs and the impact of cell-generated forces in the maturation of multicellular structures.[423] With increased understanding of tissue developmental processes, cells can potentially serve as sophisticated bioactuators in bioprinting, where cell-driven microtissue contraction/compaction effects can be leveraged for increasing the effective spatial resolution of bioprinted construct features or implementing programmable structural design motifs with living cells as the driving force.[424] Alternatively, 4D biomolecule-driven microtissue maturation could be explored on the emerging research subfield of prokaryotic bioprinting. Microgels encapsulating bioengineered bacteria can serve as autonomous biofactories that can potentially be bioprinted and continuously provide 4D biochemical cue presentation (e.g., rhBMP-2) to neighboring human cells along time.[42
The combination of cell-rich/cell–biomaterial constructs into multiscale organotypic assemblies (i.e., from nano to macrolevel) arises as the most challenging, but conceivably, the most promising approach for better recapitulating native tissues structure, connective organization and physiology in 3D macroconstructs. Apart from improving biofunctionality and biomimicry, the bottom-up combination of multiblock 3D assemblies into higher order multiscale architectures with biospecific designs is envisioned to provide a seamless biointegration into host tissues upon constructs implantation (Figure 1). After this stage, obtaining a self-regulated response to local/systemic biomacromolecular cues is key for assuring 3D microtissues commitment toward tissue-specific functions. Unsurprisingly, mimicking human tissues bioarchitecture while assuring microenvironment-sensing and physiological response in implantable multiscale assemblies is remarkably complex.
An elegant approach that involves the deconstruction of nature into its fundamental building blocks (i.e., cells/materials/soluble factors) as means to manipulate each component and encode action/response biofeedback networks has been recognized as a valuable reductionistic alternative to fabricate multiscale macrotissues with better biofunctionality and higher translational potential
Gathering on these fundaments researchers used advanced 3D biofabrication technologies to fabricate large scale constructs with a multicomponent, multiscale bone bioarchitecture incorporating osteoinductive silica nanoplatelets within constructs. This was essential for promoting osteodifferentiation of hMSCs.[426] In this work, VEGF-functionalized GelMA cell-laden cylinders were stacked in a pyramidal construction (comprising 28 rods), ultimately obtaining a perfusable lumen with HUVECs and pericytes, surrounded by differentiated osteoblasts.
Also inspired by these concepts researchers generated compartmentalized multilayered micro/macrocapsules as hierarchic, self-regulated 3D living systems that support stem cells differentiation toward osteogenic or chondrogenic lineages depending on core/surrounding microenvironment cues and cell/biomaterial combinations.[427,428] This hierarchically structured and compartmentalized system is composed of PLLA microparticle–cell building blocks encapsulated in permselective liquefied microenvironments that assure gas/ nutrients/waste exchange and sustain enclosed 3D micro tissues viability during prolonged periods in vitro and in vivo.[427] In a similar line, the development of vascularized hierarchic microtissue assemblies based on cell confinement in ECM mimetic environments has also been recently described.[428] Such approach involved a highly controlled fabrication of cell laden alginate-collagen microcapsules in microfluidic chips, followed by their random impregnation into a hydrogel shell. The hierarchically assembled construct was then employed as a template for build-up and maturation of stromal cells and endothelial cells into primitive vascular networks.[429] Despite focusing on cancer cells and the establishment of a close-to-native tumor microenvironment, the underlying concept of this multicomponent platform can be translated for bottom-up tissue engineering and regenerative medicine applications. Nevertheless, the build-up of such 3D multicomponents was highly random and no control over vascular networks spatial distribution was obtained. Such is paramount when considering the effects of 3D engineered constructs vascularization in the overall outcome after implantation.
To overcome this inherent unpredictability, researchers are investigating precision chemistry approaches and biofabrication techniques that unlock the possibility to direct bottomup assembly of multicomponent, multiscale microtissues. In this sense, by using modular cell–biomaterial bioink combinations researchers have successfully bioprinted multiscale constructs with decoupled micro- and microenvironments and multiscale bioarchitectures.
Overall, combining distinct assembly mechanisms (i.e., self-assembly, guided assembly, direct assembly, etc.) and synchronizing them toward the development of spatially coded complex bioarchitectures, will be essential for manufacturing living multicomponent modules across several length scales.[10] The emerging tools for engineering heterogenous assemblies with high precision and biofunctionality at cellular and biomaterial scale are now more established than ever, hence their interface with pioneering assembly technologies that can manipulate such building blocks could collectively accelerate the design from micro- to macroscale living 3D constructs. In this process bioengineers will be challenged with design tradeoffs by having to assure the fabrication of biologically relevant modules while keeping production cost and scalability at reach
In a holistic view, human organs display a robust, yet highly regulated cellular framework that combines both structural and functional properties for maintaining a living organism. Interestingly, the human body does not require a pre-existing scaffold for generating full-sized organs and mature tissues. In fact, the natural assembly of tissues into hierarchic modules is intrinsically a bottom-up process through interaction among cellular building blocks and cell-fabricated matrix throughout development, growth, homeostasis, and aging. Nature does this balance effortlessly in the human body but achieving this complexity in de novo engineered architectures remains remarkably challenging.
The bioinspired philosophy in modular bottom-up engineering strategies provides immense design flexibility in the generation of unitary building blocks (i.e., cells and biomaterials) with tailored biochemical features, as well as specific biophysical cues which are potentiated by their interaction and combination. T o date, bottom-up engineered micro/macrotissue assemblies have proven their biomedical value and further developments are envisioned considering the constant improvement and discovery of new nano/micro assembly technologies, precision chemistry/ genetic engineering tools, 3D biofabrication approaches, as well as the increasing understanding of basic biological regeneration processes and tissues physiology.[430] We therefore anticipate an accelerated offer of bottom-up tools that allow the development of tissues or organs with higher complexity and functionalities. Such bioengineered structures may not necessarily exhibit the exact anatomical characteristics of native tissues but should possess the required physiological performance. In this sense, there is currently an analytical challenge regarding the specific biomarkers and pathways that should be characterized in engineered 3D micro/macrotissues. Fortunately, with an increasingly widespread access to the vast toolbox offered by Omics sciences (e.g., metabolomics, lipidomics, glycomics, transcriptomics, etc.) researchers will be able to better evaluate the quality of the fabricated 3D living assemblies and their maturation during in vitro culture and evolution postimplantation in vivo. The latter can also be complemented with the noninvasive follow up of implanted 3D constructs via sophisticated in vivo bioimaging probes. Still, regarding in vivo implantation, a significant effort has been put toward improving living 3D modular constructs biointegration into surrounding host tissues by controlling/stimulating angiogenesis with the inclusion of morphogens or mechanical cues.[431] This remains one of the most challenging aspects of bottom-up tissue engineering but with the advent of sophisticated volumetric light-based 3D biofabrication-based approaches for generating flow-functional multivascular networks,[401] we anticipate that the ex vivo development of functional prevascularized modular tissue constructs will improve in the upcoming years. The scalability and speed of 3D bioprocessing techniques have been recently improved with elegant approaches for rapid printing of anatomically sized living architectures.[432] Nonetheless, minimizing complexity while pursuing optimal microtissue biofunctionality and physiological responsiveness to its surrounding microenvironment will be critical for pushing more realistic bottom-up assembled constructs toward preclinical applications.
![B)通过脂质体融合技术使用化学表面编程技术构建由光氧胺MSCs和酮功能化成纤维细胞组成的多层微组织。这两个细胞群体通过羟胺连接组装成共培养的单层和面向用户的多层。经许可复制。[164]版权所有，施普林格自然。](https://cdn.popsy.co/images/https%3A%2F%2Fwww.notion.so%2Fimage%2Fhttps%253A%252F%252Fs3-us-west-2.amazonaws.com%252Fsecure.notion-static.com%252F52ea37ea-f153-4d30-a387-ad686effb8da%252FUntitled.png%3Ftable%3Dblock%26id%3D1c4eb3b7-7588-4af8-a8c8-4a6043e76c9e%26cache%3Dv2?width=1500&optimizer=image)
