Methods for shaping laser sintered carbohydrate materials and compositions and uses thereof
Selective laser sintering of carbohydrate powders forms stable, complex vascular networks within biomaterials, addressing the limitations of existing methods by creating intricate, cell-compatible structures that replicate natural vasculature and enhance nutrient transport.
Patent Information
- Application Number
- JP2022571206
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2020-05-22
- Filing Date
- 2021-05-21
- Publication Date
- 2024-12-10
- Estimated Expiration
- 2041-05-21
AI Technical Summary
Existing methods for creating tissue scaffolds with vascular networks lack control over architecture, are harsh on cells, and fail to replicate the complex, three-dimensional structures of natural vasculature, limiting nutrient transport and integration with host vasculature.
A method using selective laser sintering (SLS) of carbohydrate powders, including isomalt and dextran, to form stable, continuous filaments that create intricate vascular networks within biomaterials, which are then used as templates for matrix materials to form interconnected vascular networks.
The method allows for the creation of complex, three-dimensional vascular networks that mimic natural vasculature, supporting cell viability and nutrient transport, and can be integrated with host tissues without cytotoxic conditions.
Smart Images

Figure 0007597292000004 
Figure 0007597292000005 
Figure 0007597292000006
Abstract
Description
[Technical field]
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to U.S. Patent Application No. 63 / 029,241, filed in the United States on May 22, 2020.
[0002] Federally Sponsored Research and Development Statement This invention was made with Government support from a National Research Service Award (NRSA) F31 Postdoctoral Research Fellowship under Grant No. HL140905 awarded by the National Institutes of Health (NIH). The Government may have certain rights in this invention. [Background technology]
[0003] The ultimate goal of the field of tissue engineering is the creation of patient-specific artificial organs that restore function to diseased or injured patients. Over the past two decades, significant advances in biomaterials and cell / tissue biology have led to some important clinical successes in the field. Tissue engineered thin avascular tissues (e.g., skin, cornea, and bladder) have been transplanted into patients with excellent functional recovery. However, solid organs of large dimensions present a much greater challenge to tissue engineer due to the elaborate and complex nature of their internal fluidic networks (i.e., vasculature). Convective transport of oxygen and nutrients through the vasculature is essential for the survival and function of cells that occupy the interior of large tissue engineered tissues. In the absence of nutrient transport through the vasculature, tissues thicker than a few hundred microns rapidly develop into a necrotic core. Thus, many researchers have devoted considerable efforts to tissue engineering artificial tissue constructs that contain open internal voids or channels to facilitate convective transport.
[0004] Several paradigms have emerged for the creation of tissue scaffolds with enhanced convective transport potential. In a widely used strategy, cells of interest are seeded into hydrogels composed of polymeric biomaterials along with fugitive particles or dissolved gases (e.g., salts) that can be selectively removed from the hydrogel. Upon removal of the fugitive material (e.g., by salt leaching or gas bubbling), an interconnected network of pores is obtained. Such interconnected pore networks have also been formed by critical point drying or electrospinning of polymers. These porous networks allow for convective transport of oxygen and nutrients to the tissue and provide a large surface area for them to diffuse into the hydrogel where the cells reside. However, these fabrication strategies offer little control over the scaffold architecture and often utilize harsh cytotoxic conditions or reagents. Furthermore, disorganized porous networks will not accommodate direct connections to the host vasculature when implanted in animal models.
[0005] Separately, needle molding techniques have been introduced to create linear channels in defined patterns in hydrogels. A polymer solution is dispensed around a series of needles and then cross-linked to form a solid gel. The needles are withdrawn after the gel has solidified, leaving open channels. These needle-molded hydrogels have become useful tools for studying transport in vascular channels, but they are unable to fully reproduce the complex, branched architecture of physiological vasculature. Soft lithography, another approach to patterning vasculature, utilizes photolithographic techniques that offer control of channel architecture with micron-scale resolution. However, the process requires expensive equipment and generally creates microfluidic conduits only as linear x, y, or z vectors that cannot mirror realistic vascular networks. Overall, neither soft lithography nor needle molding can create complex three-dimensional microfluidic networks that capture the architectural features of in vivo vasculature.
[0006] Additive manufacturing (AM, also known as three-dimensional printing or 3D printing) has been used to create fluidic networks within biomaterials. To that end, sacrificial templated is one promising modeling method. Using this technique, a defined pattern of fluidic channels can be created within a bulk material by encasing a removable template within the bulk material. FIGS. 1A-1D are a step-by-step schematic sequence illustrating sacrificial templated formation of a simple vascular network. FIG. 1A shows a template 101 that has been modeled via an existing AM method. In FIG. 1B, the template 101 is encased within a bulk material 103. After the template 101 has been selectively removed from within the bulk material 103, FIG. 1C shows a simple vascular network 105 that reflects the architecture of the template 101 defined by the bulk material 103. After modeling, arrows 107 in FIG. 1D represent fluid flow through the simple vascular network 105 within the bulk material 103.
[0007] In the first method of forming a removable template, as in Miller and Bellan, water-soluble carbohydrate glass is first extruded into a free-standing lattice and encapsulated in a biomaterial hydrogel. See: JS Miller et al., Rapid Casting of Patterned Vascular Networks for Perfusable Engineered Three-Dimensional Tissues, 11(9), Nat. Mater. 768-774 (2012) and LMBellan et al., Fabrication of an Artificial 3-Dimensional Vascular Network Using Sacrificial Sugar Structures, Soft Matter. 5(7), 1354 (2009). When the 3D printed carbohydrate glass lattice dissolved from within the bulk hydrogel, the geometry of the lattice was retained as open channels in the hydrogel matrix. Although this method can produce a variety of two-dimensional conduit networks, the resulting extruded sugar structures are fragile and the method cannot be fully scaled to produce a variety of three-dimensional architectures. Furthermore, the reproducibility of the final network geometry and the resolution of the printed filaments are both limited by the use of extrusion to form the carbohydrate glass network.
[0008] A second method of forming a removable mold utilizes a temporary fugitive ink that is extruded with one or more polymer scaffold materials, for example in Bertassoni and Kolesky. See: L.E. Bertassoni et al., Hydrogel Bioprinted Microchannel Networks for Vascularization of Tissue Engineering Constructs, 14(13), Lab Chip, pp. 2202-2211 (2014). DB. Kolesky, 3D Bioprinting of Vascularized, Heterogeneous Cell-Laden Tissue Constructs, 26(19), Adv. Mater. pp. 3124-3130 (2014). DB. Kolesky et al., Three-Dimensional Bioprinting of Thick Vascularized Tissues, Proc. Natl. Acad. Sci. USA, 201521342 (2016). Subsequent removal of the fugitive ink creates a corresponding void space in the bulk material. Similar to carbohydrate glass printing, a wide variety of 2D networks can be created, but it is not possible to print arbitrary complex three-dimensional architectures using fugitive ink extrusion techniques. Summary of the Invention [Means for solving the problem]
[0009] Some embodiments may be directed to compositions useful for forming structures in the form of a substantially interconnected vascular network. In some embodiments, the composition may include a powder comprising a carbohydrate powder and an anti-caking agent, where the powder has a granular form and has a specific energy of less than 6 millijoules per milliliter (mJ / mL).
[0010] In some embodiments, the carbohydrate powder may include at least one of dextran or isomalt.
[0011] In some embodiments, the anti-caking agent comprises at least one of corn starch, silicon dioxide, or xanthan gum.
[0012] In some embodiments, the powder may be configured to be a free-flowing powder, and the powder may have a maximum particle size of 250 micrometers (μm) or less.
[0013] Some embodiments may be directed to a three-dimensional structure comprising a powder system that includes a carbohydrate powder and an anti-caking agent.
[0014] In some embodiments, the three-dimensional structure may further comprise a surface coating comprising a hydrophobic polymer.
[0015] In some embodiments, the powder system may be fused together into a solid continuous filament using a laser.
[0016] In some embodiments, the powder system may be fused together into a solid filament network using a laser.
[0017] In some embodiments, the structure may undergo surface smoothing using a smoothing solution.
[0018] In some embodiments, the smoothened structure may then be coated in a hydrophobic polymer.
[0019] Some embodiments may be directed to a structure comprising a substantially interconnected vascular network, including a matrix material in which the structural material is disposed. In some embodiments, the structural material may be capable of dissolving or decomposing in water. In some embodiments, the structural material may be formed from a powder that fuses into a solid when irradiated with an energy beam. In some embodiments, the powder system may have a granular morphology and have a specific energy of less than 6 mJ / mL.
[0020] In some embodiments, the powders may be fused together into a solid continuous filament using a laser.
[0021] In some embodiments, the powders may be fused together into a solid continuous three-dimensional filament network using a laser.
[0022] In some embodiments, the surface of the structure may have a smooth topography.
[0023] In some embodiments, the powder may further include an anti-caking agent.
[0024] In some embodiments, the structural material may include at least one carbohydrate.
[0025] In some embodiments, the structural material may include at least one of dextran or isomalt.
[0026] In some embodiments, the structural material may include at least one of photoresist, agarose, gelatin, carbohydrates, sucrose, glucose, fructose, lactose, isomalt, dextran, cellulose, methylcellulose, poly(lactic acid), or poly(ethylene glycol).
[0027] In some embodiments, the anti-caking agent may include one or more materials that improve at least one of the flow, frictional properties, or particle packing of the construction material.
[0028] In some embodiments, the anti-caking agent may include at least one of corn starch, silicon dioxide, or xanthan gum.
[0029] In some embodiments, the particle size of the powder may be less than or equal to 250 μm in diameter.
[0030] In some embodiments, the surface of the structure may have a hydrophobic coating.
[0031] In some embodiments, the hydrophobic coating may include at least one of polycaprolactone, poly(L-lactide), polylactic acid, poly(lactic-co-glycolic acid), collagen, gelatin, zein, shellac, starch, wax, or petrolatum.
[0032] In some embodiments, the matrix material may comprise at least one of polyamide, poly(2-hydroxyethyl methacrylate), poly(vinyl alcohol), polyacrylamide, poly(ethylene glycol), polyurethane, collagen, agarose, albumin, alginate, chitosan, starch, hyaluronic acid, gelatin, fibrin, matrigel, glycerin, glycol, mannitol, inositol, xylitol, adonitol, glycine, arginine, a biological polymer molecule, or a peptide amphiphile, or a monomer, dimer, or oligomer thereof.
[0033] In some embodiments, the powder may be configured to be a free-flowing powder.
[0034] Some embodiments may be directed to a method of forming a substantially interconnected vascular network. In some embodiments, the method may include solidifying a powder by fusing with an energy beam to form a filament network; backfilling void spaces of the filament network with a matrix material; and removing the filaments to form a substantially interconnected vascular network including fluid channels. In some embodiments, the filament network may include a plurality of filaments, defining void spaces, and the plurality of filaments may be capable of dissolving or dissolving in water.
[0035] In some embodiments, the matrix material may be an aqueous solution containing the biomaterial.
[0036] Some embodiments of the method may further include crosslinking the biomaterial to form a hydrogel in the void space.
[0037] In some embodiments, the powder may include a carbohydrate powder and an anti-caking agent. In some embodiments, the powder may have a granular form and a specific energy of less than 6 mJ / mL. In some embodiments, the energy beam may be a laser.
[0038] Some embodiments of the method may further include surface smoothing the plurality of filaments with a smoothing solution, where the surface smoothing does not otherwise alter the architecture of the filaments.
[0039] In some embodiments, the smoothing solution may include at least one of isomalt, dextran, sucrose, glucose, lactose, trehalose, maple syrup, or cane syrup.
[0040] Some embodiments of the method may further include surface coating the plurality of filaments with a surface coating material, where the surface coating material does not backfill void spaces, and where the matrix material and the surface coating material may be distinct materials.
[0041] In some embodiments, the surface coating material may include at least one of polycaprolactone, poly(L-lactide), polylactic acid, poly(lactic-co-glycolic acid), collagen, gelatin, zein, shellac, starch, wax, or petrolatum.
[0042] In some embodiments, the filament network may be arranged in a three-dimensional branched pattern.
[0043] In some embodiments, the filament network may be arranged in an interpenetrating geometry.
[0044] In some embodiments, the filaments may be configured into an unsupported geometry during shaping.
[0045] In some embodiments, removal of the filament may be accomplished by dissolution or decomposition.
[0046] In some embodiments, the aqueous solution may further comprise a suspension of living cells, where removing the filaments does not damage the living cells.
[0047] In some embodiments, the matrix material may be a crosslinked biomaterial, the biomaterial comprising at least one of polyamide, poly(2-hydroxyethyl methacrylate), poly(vinyl alcohol), polyacrylamide, poly(ethylene glycol), polyurethane, collagen, agarose, albumin, alginate, chitosan, starch, hyaluronic acid, gelatin, fibrin, matrigel, glycerin, glycol, mannitol, inositol, xylitol, adonitol, glycine, arginine, a biological polymer molecule, or a peptide amphiphile, or a monomer, dimer, or oligomer thereof.
[0048] In some embodiments, the powder may be configured to be a free-flowing powder, and the powder may have a maximum particle size of 250 μm or less.
[0049] Other aspects and advantages of the present invention will become apparent from the following description and appended claims. [Brief description of the drawings]
[0050] [Figure 1A] FIG. 1 is a schematic diagram of a prior art method. [Figure 1B] FIG. 1 is a schematic diagram of a prior art method. [Figure 1C] FIG. 1 is a schematic diagram of a prior art method. [Figure 1D] FIG. 1 is a schematic diagram of a prior art method. [Figure 2A] 1 is a schematic diagram of a method according to some embodiments. [Figure 2B] 1 is a schematic diagram of a method according to some embodiments. [Figure 2C] 1 is a schematic diagram of a method according to some embodiments. [Figure 2D] 1 is a schematic diagram of a method according to some embodiments. [Figure 2E] 1 is a schematic diagram of a method according to some embodiments. [Figure 2F] 1 is a schematic diagram of a method according to some embodiments. [Diagram 3] 1 is a graph of the specific energy of powders according to embodiments of the present disclosure. [Figure 4A] 1 is a scanning electron micrograph of a powder according to an embodiment of the present disclosure. [Figure 4B] 1 is a scanning electron micrograph of a powder according to an embodiment of the present disclosure. [Figure 4C] 1 is a scanning electron micrograph of a powder according to an embodiment of the present disclosure. [Figure 5A] FIG. 1 is a schematic diagram of selective laser sintering according to an embodiment of the present disclosure. [Figure 5B] FIG. 1 is a schematic diagram of selective laser sintering according to an embodiment of the present disclosure. [Figure 5C] FIG. 1 is a schematic diagram of selective laser sintering according to an embodiment of the present disclosure. [Figure 5D] FIG. 1 is a schematic diagram of selective laser sintering according to an embodiment of the present disclosure. [Figure 6] 1 is a photograph of a filament network according to one or more embodiments. [Figure 7] 1 is a graph showing filament width versus laser power according to one or more embodiments. [Figure 8A] 1 is a photograph of a filament network according to one or more embodiments. [Figure 8B]1 is a photograph of a filament network according to one or more embodiments. [Figure 9] 1 is a graph of stress versus strain for a structural material according to one or more embodiments. [Figure 10A] 1 is a scanning electron micrograph of a filament network surface according to one or more embodiments. [Figure 10B] 1 is a scanning electron micrograph of a filament network surface according to one or more embodiments. [Figure 11A] 1 is a graph of the effect of surface smoothing in accordance with one or more embodiments. [Figure 11B] 1 is a graph of the effect of surface smoothing in accordance with one or more embodiments. [Figure 11C] 1 is a graph of the effect of surface smoothing in accordance with one or more embodiments. [Figure 11D] 1 is a graph of the effect of surface smoothing in accordance with one or more embodiments. [Figure 12A] 1 is a photograph of an interconnected vascular network according to one or more embodiments. [Figure 12B] 1 is a photograph of an interconnected vascular network according to one or more embodiments. [Figure 12C] 1 is a photograph of an interconnected vascular network according to one or more embodiments. [Figure 12D] 1 is a photograph of an interconnected vascular network according to one or more embodiments. [Figure 12E] 1 is a photograph of an interconnected vascular network according to one or more embodiments. [Figure 12F] 1 is a photograph of an interconnected vascular network according to one or more embodiments. [Figure 13A] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 13B] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 13C]1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 13D] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 13E] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 14A] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 14B] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 14C] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 14D] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 14E] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 15A] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 15B] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 15C] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 15D] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 15E] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 15F] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 16A] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 16B] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 16C] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 16D]1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 16E] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 16F] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 17A] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 17B] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 17C] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 17D] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 17E] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 17F] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 18A] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 18B] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 19A] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 19B] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 20A] 1 illustrates an embodiment of a mutual tree attraction method. [Figure 20B] 1 illustrates an embodiment of a mutual tree attraction method. [Figure 20C] 1 illustrates an embodiment of a mutual tree attraction method. [Figure 20D] 1 illustrates an embodiment of a mutual tree attraction method. [Figure 21A]1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 21B] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 21C] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 21D] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 21E] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 21F] 1 illustrates an interconnected vascular network according to one or more embodiments. [Figure 21G] 1 illustrates an interconnected vascular network according to one or more embodiments. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0051] Throughout the drawings, like numbers are generally used to refer to like components.
[0052] A significant limitation of existing AM methods is the inability to build complex networks in all three dimensions in various materials. Liquid-phase materials deposited by extrusion, such as those used in the AM methods described above, are prone to deformation or collapse under their own weight. Furthermore, the viscosity of the liquid makes precise dispensing of small amounts extremely difficult (as would be required to form omnidirectional structures layer by layer). These inherent rheological constraints prohibit extrusion-based AM techniques such as those described above from forming architectures that contain unsupported overhangs, underhangs, and / or any three-dimensional branching, all of which are characteristic of mammalian vasculature. Furthermore, the 90-degree channel junctions formed as a result of extruded linear networks have fluid dynamics that are strikingly different from structures found in nature. The altered fluid dynamics of networks formed using existing AM techniques have undesirable effects on blood flow dynamics, shear stress experienced by endothelial cells, and the overall fluid resistance of the network.
[0053] In contrast, the sugar alcohol isomalt has been found to be compatible with selective laser sintering (SLS), undergoing a stable melt transition to form continuous fused solid filaments. Thus, embodiments described herein may be directed to compositions of matter and methods that can be used to generate patterned SLS, including workflows for fully automated fabrication of three-dimensional structures from various carbohydrate powders.
[0054] Moreover, SLS of carbohydrates as described herein is a novel approach to additively manufacture this type of material. The methods disclosed in the embodiments herein offer significant improvements over previous methods of fabricating carbohydrate structures, including improved resolution, structural complexity, reproducibility, and throughput.
[0055] To that end, some embodiments may be directed to SLS patterned structures in the form of solid continuous filaments or solid continuous three-dimensional structures, such as three-dimensional dendritic carbohydrate lattices or solid continuous filament networks. In some embodiments, the patterned structures may be used to template matrix materials. In some embodiments, the patterned structures may be used to cast vascularized tissue-engineered tissues (e.g., the structures may be described as matrices with embedded perfusable vascular networks). Additionally, some embodiments of the present disclosure may be directed to compositions of matter and methods of modeling tissue-engineered vascular networks free from the constraints of existing extrusion printing techniques, such as those described above. Compositions of matter embodiments may be compatible with SLS and undergo a stable melt transition to form continuously fused solid filaments. Additionally, embodiments described herein may include a workflow for fully automated modeling of three-dimensional structures from various carbohydrate powders. Finally, one or more embodiments are directed to a method called Mutual Tree Attraction for computationally generating dendritic substantially interconnected vascular networks.
[0056] Broadly, Figures 2A-2F briefly illustrate steps for forming a substantially interconnected vascular network that may be used in embodiments of the present method. As in Figures 2A-2F, embodiments of the present method may include solidifying powder granules by sintering or melting with an energy beam to form a three-dimensional structure to be used as a mold; surface smoothing the mold with a smoothing solution; surface coating the mold with a surface coating material; backfilling the void spaces of the mold with a matrix material; cross-linking the matrix material; and removing the mold to form a patterned structure with channels shaped like the mold. This method is named Selectively Laser Sintered-Carbohydrate Sacrificial Templating (SLS-CaST). Additionally, each of the steps briefly illustrated in Figures 2A-2F will also be discussed in more detail below.
[0057] In some embodiments, SLS may be used to form three-dimensional structures by tracing 2D patterns with a focused energy beam onto a powder bed. Powder embodiments may include one or more carbohydrate powders and may be fused together to form a structural material. Figure 2A shows granules of powder 1 fused together using an SLS system 2 by sintering or melting with a laser 3 to form a filament network 5. The shape traced by the laser 3 may be controlled by the movement of the stage 6, the movement of the laser 3, or both.
[0058] In some embodiments, a three-dimensional structure formed from a structural material by SLS may be used as a mold. Initially after formation, the surface of the three-dimensional structure may be rough. FIG. 2B shows the filament network 5 after removal from the 3D printing system. Void spaces 7, including between the filaments 9, are visible around the filament network 5. In this manner, the filament network 5 defines the void spaces 7. The surface 11 of the filament network 5 is rough.
[0059] In some embodiments, the surface of the filament network may be smoothed using a smoothing solution. Figure 2C shows the filament network 5 after undergoing surface smoothing to reduce the roughness of the surface 11. The surface smoothing results in the filament network 5 having a smooth structure.
[0060] In some embodiments, the surface of the filament network may be surface coated with a surface coating material. Figure 2D shows the filament network 5 after the surface coating material has formed a coating 13 on the surface 9 of the filament network 5. However, the coating 13 does not backfill the void spaces 7.
[0061] In some embodiments, the void space around the filament network is backfilled with a matrix material. Figure 2E shows a matrix material 15 surrounding the filament network 5 and backfilling the void space 7. Some embodiments of the matrix material may require crosslinking or other polymer treatment to solidify within the void space after introduction around the filament network.
[0062] In some embodiments, the filament network may be removed to form a substantially interconnected vascular network. The morphology of the final vascular network may include fluid channels through the matrix material that reflect the geometry of the filament network. FIG. 2F shows a substantially interconnected vascular network 17 that includes fluid channels 19 defined by coating 13 and matrix material 15. Arrows 21 represent one possible fluid flow direction through vascular network 17.
[0063] Powders according to one or more embodiments
[0064] Powder embodiments may include one or more carbohydrate powders. Powder embodiments may further include a pigment, an anti-caking agent, or both. Some powder embodiments may have the ability to flow, as may be quantified using specific energy. Powder embodiments may be in granular form, with multiple powder granules having specific geometric characteristics (e.g., average diameter). Each of these factors of the powder is described in more detail below.
[0065] Powder embodiments may include one carbohydrate powder or a mixture of two or more carbohydrate powders. The carbohydrate powder may comprise or consist of, for example, photoresist, agarose, gelatin, carbohydrates, sucrose, glucose, fructose, lactose, isomalt, dextran, cellulose, methylcellulose, poly(lactic acid) and / or poly(ethylene glycol).
[0066] As discussed above, previous studies have used sugars formed by melt extrusion of sacrificial templates, but the results were brittle and limited in the final three-dimensional shape. In contrast, isomalt (a sugar alcohol often used as an artificial sugar substitute) and dextran have been found to be compatible with SLS and undergo a stable melt transition to form continuous fused solid filaments. Thus, isomalt and dextran may be used to form three-dimensional structures, including complex three-dimensional structures such as vascular architecture. Thus, in some embodiments, the powder may include one or both of isomalt and dextran powders.
[0067] In some embodiments, the powder may have the ability to flow for dispensing during SLS. Specific energy is one measure of powder flowability. To that end, some embodiments of the present disclosure utilize powders with specific energies of less than 6 millijoules per milliliter (mJ / mL) (e.g., less than 5 mJ / mL or less than 4 mJ / mL), with or without anti-caking agents. In other words, some embodiments of the present disclosure utilize powders with specific energies of between 6 mJ / mL and 0.1 mJ / mL (e.g., between 5 mJ / mL and 0.1 mJ / mL, between 4 mJ / mL and 0.1 mJ / mL, or between 2 mJ / mL and 0.1 mJ / mL), with or without anti-caking agents.
[0068] The addition of anti-caking agents can effectively increase powder flow while preserving sintering quality. To that end, in some embodiments, one or more anti-caking agents can improve the properties of the powder, such as flow, fluidity, frictional properties, and / or particle packing of the construction material, as compared to a similar powder lacking the anti-caking agent.
[0069] Thus, according to some embodiments, the SLS powder may be a mixture of one or more carbohydrate powders and one or more anti-caking agents. In some embodiments, the anti-caking agents may include, but are not limited to, one or more of corn starch, silicon dioxide, or xanthan gum, or a mixture of two or more anti-caking agents.
[0070] Although nylon is a standard material for SLS because it has favorable powder rheological properties, nylon may not be desirable for certain applications, for example, because it is poorly biocompatible and not easily soluble. Thus, FIG. 3 compares the specific energy of nylon with that of isomalt and a mixture of 30 weight percent (wt%) cornstarch in isomalt (hereinafter referred to as "isomalt+cornstarch"), according to one or more embodiments of the present disclosure. As shown, isomalt powder granules have a higher specific energy and are therefore much more cohesive than standard SLS materials such as nylon, which may manifest as poor powder distribution or uneven layers. FIG. 3 also shows that the isomalt+cornstarch mixture has a flowability comparable to nylon. Thus, isomalt+cornstarch can be free-flowing like nylon. The data for isomalt and isomalt+cornstarch have a probability value (p-value) of p<0.05 as a t-test with sample size (n), n=3.
[0071] 4A-4C are scanning electron micrographs of nylon (FIG. 4A), isomalt (FIG. 4B) and cornstarch (FIG. 4C) illustrating different powder morphologies and helping to explain the differences in powder flowability. Nylon is a very free-flowing powder used in conventional polymer SLS processes. As shown in FIG. 4A, nylon has a relatively small, smooth, and regular morphology. In contrast, isomalt powder is irregular, polydisperse, and jagged as shown in FIG. 4B. As a result, isomalt powder has higher powder cohesion and relatively poor flowability compared to nylon. As discussed above, the addition of cornstarch significantly improves the flowability of powdered isomalt. FIG. 4C shows powdered cornstarch. It is hypothesized that the small, smooth cornstarch particles fit between the larger, jagged isomalt grains, reducing friction and flow resistance.
[0072] The inclusion of a pigment may enable the powder to more efficiently absorb radiation from an energy beam (e.g., a laser) during SLS. Additionally, the inclusion of a pigment may also enable successful shaping using a visible spectrum energy beam. Thus, some embodiments of the powder may include one or more pigments. In some embodiments, the pigment may be a biocompatible pigment. In some embodiments, the biocompatible pigment may include one or more of tartrazine, anthocyanin, and Allura Red AC (Red 40). Similar to the anti-caking agents discussed above, the pigment may be incorporated into the final three-dimensional structure during SLS.
[0073] According to some embodiments, the particle size described herein may be defined as the mean particle size, median particle size, modal particle size, weighted average particle size, average Feret diameter, Sauter mean diameter, the maximum of the probability density function of the particle size (using any applicable mathematical model, e.g., log-normal distribution, Weibull distribution, Rosin-Rammler distribution, log-hyperbolic distribution, skew log-Laplace model distribution, and further models derived therefrom), or some other statistical metric that may be used to determine and / or define the diameter of a particle.
[0074] In one or more embodiments, the particle size of the powder (e.g., measured as the average Feret diameter, etc., as discussed above) may be less than 250 micrometers (μm) (e.g., less than 200 μm, less than 100 μm, less than 50 μm, less than 10 μm, etc.). In other words, some embodiments of the present disclosure utilize powders having a particle size between 250 μm and 0.01 μm (e.g., between 200 μm and 0.01 μm, between 100 μm and 0.01 μm, between 50 μm and 0.01 μm, or between 10 μm and 0.01 μm).
[0075] In some embodiments, grinding, milling, mechanical separation and / or sieving may be performed to ensure that the powder has the intended particle size and / or particle size distribution. Such particle size operations may be performed on one or more components of the powder separately (i.e., before mixing), after combining two or more components of the powder, or both. Additionally, in some embodiments, the powder may be a mixture of carbohydrate powder grains, with or without an anti-caking agent, that have been ground and / or milled to reduce particle size and / or particle size prior to sintering.
[0076] In some embodiments, the powder may also be separated mechanically or by other means to select powder grains below a threshold size. One such method of separating powders by size may be sieving, although other methods may also be used. If the powder contains two or more component powders (i.e., carbohydrate powders and / or anticaking agents), these particle processing methods may be performed before and / or after mixing the component powders.
[0077] In some embodiments, the powder particle size can affect the powder layer height of a single layer, which can affect the final resolution of the three-dimensional structure. If the powder layer height of each additional layer of unsintered powder is too thick (i.e., contains too much powder), the powders cannot fully fuse together. Powder that is not well fused cannot properly add to the three-dimensional structure as it grows, and thus the process cannot form a continuous three-dimensional structure. In contrast, if each newly added layer is too thin (i.e., does not contain enough powder), the laser can over-sinter the previous layer, undesirably altering the final geometry.
[0078] According to one or more embodiments, the powder may be free-flowing to form a layer having a powder bed height (if in granular form) that can be successfully formed into a structural material by sintering. In some embodiments, the powder bed height may be less than 250 μm (e.g., less than 225 μm, less than 210 μm, less than 200 μm, less than 175 μm, less than 150 μm, etc.). In some embodiments, the powder bed height may also be greater than 100 μm (e.g., greater than 110 μm, greater than 120 μm, greater than 125 μm, greater than 150 μm). In some embodiments, the powder bed height may be in the range of 100 μm to 250 μm (e.g., in the range of 100 μm to 225 μm; in the range of 125 μm to 225 μm; in the range of 125 μm to 200 μm; in the range of 100 μm to 200 μm; in the range of 100 μm to 175 μm, etc.).
[0079] Solidification of the powder by sintering or melting according to one or more embodiments
[0080] SLS is a versatile form of 3D printing that can be used to form structural materials by tracing two-dimensional patterns with a focused energy beam onto a powder layer that may be used in some embodiments. Some embodiments of the method may use SLS to solidify a powder, such as the powders discussed above. In some embodiments, the powder can be sintered or melted to form a solid structural material that may be in the shape of a three-dimensional structure. In some embodiments, the three-dimensional structure may be used as a mold. In some embodiments, the energy beam used for SLS may be a laser beam. The SLS process and the properties of the resulting structural materials are described in more detail below.
[0081] In some embodiments, the powder absorbs the electromagnetic radiation delivered by the energy beam and can be fused (e.g., thermally melted or sintered) during SLS to form a structural material. To that end, in some embodiments, the powder is irradiated with an energy beam to form a structural material. Thus, the path of the energy beam in two dimensions can fuse regions of the powder to form a structural material. By patterning each successive layer in the third dimension, the structural material can be fused together in all three dimensions to form a three-dimensional structure. Thus, a three-dimensional structure can be created by repeating this process layer by layer, where an unsintered layer of the powder system spreads over the previously fused layer and is patterned by exposure to the energy beam to add to the three-dimensional structure.
[0082] 5A-5D illustrate schematics of a SLS three-dimensional printing process according to one or more embodiments. Each of these steps is briefly discussed and further detailed below.
[0083] FIG. 5A shows a digital rendering 23 performed by a computer 25 of the three-dimensional structure to be printed.
[0084] 5B shows a digital rendering 23 in a computer 25 that is converted into three two-dimensional digital slices 27. These digital slices 27 correspond to locations where the laser may solidify the powder into a three-dimensional structure during SLS.
[0085] FIG. 5C shows powder dispensing as part of the layer-by-layer building of a filament network by SLS.
[0086] Here, the SLS system 2 has a stage 6, a powder reservoir 29, and a sieve 31. The SLS system 2 has partially formed a filament network 5. The powder reservoir 29 stores the powder 1. The sieve 31 dispenses a powder layer 33 of the powder 1 onto the stage 6 in preparation for fusing / melting with an energy beam. The stage 6 may be a z-stage, an xy-stage, or an xyz-stage (where the z-direction is defined as the flow direction of the powder reservoir 29 from the sieve 31).
[0087] Although not shown here, sieve 31, stage 6, or both may be controlled by a microelectronic controller. The microelectronic controller may control one or more aspects of SLS 2, such as the flow rate of powder 1 onto stage 6, or the location where powder 1 is dispensed onto stage 6. Such a microelectronic controller may control one or more aspects of powder layer 33, such as the thickness of powder layer 33. Further details of microcontrollers within SLS system 2 are discussed further below.
[0088] FIG. 5D shows laser patterning as part of layer-by-layer building of a filament network by SLS.
[0089] Here, the SLS system 2 also includes a laser 3. The laser 3 sinters / fuses the powder layer 33 into a filament network 5.
[0090] Although not shown here, the laser 3, the stage 6, or both may be controlled by a microelectronic controller. The microelectronic controller may control one or more aspects of the SLS 2, such as various characteristics of the laser 3, or the location where the laser 3 sinters the powder layer 33. Such a microelectronic controller may control one or more aspects of the powder layer 33, such as the thickness of the powder layer 33. Further details of the microcontroller within the SLS system 2 are discussed further below.
[0091] Prior to commencing SLS building, in some embodiments, a 3D model may be digitally rendered using computer-aided design (CAD) software. Such rendering may then be exported into a file, such as, for example, an ".STL" file, that is compatible with a computer-aided 3D printing platform termed an SLS system. This SLS file can assist in translating the 3D model from CAD into precise instructions required for the various tasks of the SLS system outlined above. One such format may be the stereolithography (.stl) format, according to some embodiments.
[0092] The SLS file may also be processed by additional software that can convert the 3D model, for example, by slicing into a series of 2D cross-sectional slices that can each be used to pattern one layer of powder. In some embodiments, during this slicing process, parameters that govern the behavior of the SLS system during sintering may be defined. In some embodiments, the speed and / or power of the energy beam and / or the height of each layer may be set and / or customized by one or more software programs. The series of 2D slices may be expressed as a series of instructions to the SLS system that are expressed in a final format, which in some embodiments may be G-code language. The series of instructions may, in some embodiments, move the energy beam to a specific location or along a specific path, lower the printing platform, and / or move the powder distributor, among other instructions. In some embodiments, the series of instructions to the SLS system may be further modified by a custom developed program that can optimize the instructions specifically for SLS of carbohydrate powders to ensure proper interpretation by the SLS system and / or can define energy beam power density / energy beam scan speed settings.
[0093] The fabrication of three-dimensional structures made of structural materials from powders, including carbohydrate powders, according to the disclosed embodiments may be performed using an SLS system. The SLS system may be an automated, computer-aided 3D printing platform that uses additive manufacturing to fabricate three-dimensional structures from powders layer by layer according to a set of instructions. The energy beam of the SLS system may be mounted on a gantry or other such mechanism to precisely control the spatial positioning of the energy beam. According to some embodiments, the SLS system may also include a printing platform that is free to move in the z-direction, for example, so that a layer of powder can proceed downward after it is sintered. In some embodiments, the next layer of powder may be formed on the printing platform by depositing a metered amount of unsintered powder from a powder hopper and then smoothing the metered amount of unsintered powder using a dispenser.
[0094] The SLS process may be coordinated by a microelectronic controller in the SLS system, which controls motors, valves, mechanisms, etc. responsible for various tasks. In some embodiments, those tasks may include, for example, positioning, aiming, and controlling the output of the energy beam; positioning and moving the print platform; positioning and controlling mechanisms of the powder hopper, positioning and engaging the dispenser; and any of the other tasks for fusing the powder into a solid.
[0095] Solidification using an SLS set may, in some embodiments, begin with a first layer of powder that is manually deposited on the print platform and leveled by a user. Once the first layer of powder is on the print platform, a series of instructions may be executed in sequence such that the first layer of powder is patterned by an energy beam (e.g., a laser), followed by a second layer of powder that may be deposited from a hopper, and so on. Alternatively, in some embodiments, the first layer may be mechanically metered and dispensed by the SLS system such that the dispensing of the first layer is the first instruction in the series of instructions.
[0096] In some embodiments, powder may be added as subsequent layers by shaking a sieve connected to a powder reservoir suspended above the stage. Some embodiments may use a shaking action (e.g., of a sieve) to dispense the powder. Such a shaking action may aerate the powder and / or reduce powder compaction. After dispensing the powder from the suspended reservoir, in some embodiments, the dispensed amount of powder may be spread to provide a powder layer with approximately equal layer thickness. The distribution of the dispensed amount of powder may be performed, for example, by counter-rotating rollers. Finally, in some embodiments, excess powder (meaning more than is needed to form the powder layer) may be removed (e.g., by a spade mechanism), collected, and redistributed. The excess powder may be redistributed many times (e.g., tens to hundreds of times) without a noticeable decrease in print quality.
[0097] The SLS system may alternate between patterning the last layer of powder and depositing the next layer of powder until the final geometry is sintered.
[0098] The appearance and quality of three-dimensional structures formed by SLS can be affected by various factors, including laser power density, laser scanning speed, and / or powder bed height. According to one or more embodiments, proper control of these settings allows for consistent sintering of the powder to produce a three-dimensional structure. Inappropriate values (or combinations of values) of such factors can result in a final geometry that differs from the intended three-dimensional structure by reducing feature resolution, adding unintended features, removing intended features, powder not fusing completely, creating balling defects, distorting the final three-dimensional structure, adding cavities, and / or other undesired changes to the intended final geometry.
[0099] Those skilled in the art will appreciate that laser power density and laser scanning speed can be interrelated. For example, a low power laser moving at a low scanning speed can still deliver excessive power to an area. Excessive power can cause over-sintering, distortion, voids and / or other undesired changes to the intended final geometry.
[0100] Over-sintering is when particles outside the intended pattern are fused together with particles within the intended pattern that are fused to form the intended final geometry. Over-sintering can occur in the z-axis (i.e., build axis) and / or the x / y-axis (i.e., planar axes). In some embodiments, over-sintering can cause excessive fusion between successive powder layers, which can reduce the resolution of the final geometry along the build axis and / or add unintended features in the build axis and / or planar axes.
[0101] If the laser power density is too high and / or the laser scanning speed is too low, undesired changes to the final geometry, such as over-sintering, may occur. Additionally, low laser scanning speeds may result in irregular melt pools while the powder is being sintered. Irregular melt pools may cause further undesired changes to the intended final geometry, such as distortions and / or cavities.
[0102] Alternatively, if the laser power density is too low and / or the laser scanning speed is too fast, the powder cannot form a continuous fused final three-dimensional structure. These conditions can, in some embodiments, cause balling defects sometimes seen in SLS, where some powder, if insufficiently melted, can roll into separate spheres instead of forming the final three-dimensional structure.
[0103] In some embodiments, for a given powder, the width of the filament may be a function of laser power, translation speed, or both. Thus, some embodiments may control the laser power, translation speed, or both to control the filament dimensions.
[0104] In some embodiments, there may be an approximately linear relationship with a negative correlation between translation speed and filament diameter. Figures 6 and 7 demonstrate an embodiment in which the relationship between translation speed and filament diameter is approximately linear with a negative correlation. The negative correlation is evident in Figures 6 and 7 in that as the laser speed increases, the filament width decreases. Figure 6 shows that at a laser speed of 45 watts per square millimeter (W / mm 2 ) of three-dimensional structures containing multiple filaments printed at a constant power density of 45 W / mm. Increasing the translation speed of each successive filament from 500 to 2,750 millimeters per minute (mm / min) reduced the filament diameter from 800 to 400 μm (scale bar = 1 mm). Figure 6 is a representative image of five images taken after printing each three-dimensional structure. Figure 7 numerically plots the filament width versus laser speed for the five three-dimensional structures (including Figure 6). The plot is based on a linear ... 2 The graph shows a nearly linear relationship with a negative correlation between translation speed and filament diameter in the region spanning 400-800 μm at a constant power density of 1000 μm. The graph shows the average filament width and standard deviation for n=5 printing runs.
[0105] In some embodiments, the laser power density is 20 to 100 W / mm, for example 40 to 60 W / mm 2 range (e.g., 45 W / mm 2 ~60W / mm 2 Range: 40W / mm 2 ~55W / mm 2 Range: 45W / mm 2 ~55W / mm 2 Range: 40W / mm 2 ~50W / mm 2 or 50W / mm 2 ~60W / mm 2 The range may be between 0 and 1.
[0106] In some embodiments, the laser scanning speed may be in the range of 200 to 4000 mm / min, e.g., in the range of 1000 mm / min to 2000 mm / min (e.g., in the range of 1250 mm / min to 2000 mm / min, in the range of 1250 mm / min to 1750 mm / min, in the range of 1000 mm / min to 1750 mm / min, in the range of 1250 mm / min to 2000 mm / min, in the range of 1500 mm / min to 2000 mm / min, or in the range of 1000 mm / min to 1500 mm / min).
[0107] According to one or more embodiments, the SLS system may be used to print three-dimensional structures that exhibit uneven three-dimensional branching, smooth curvatures, and unsupported geometries. For example, Figures 8A and 8B (scale bar = 10 mm) are photographs of three-dimensional structures formed from structural materials printed using SLS of a powder containing a mixture of isomalt and 30 weight percent (wt%) cornstarch materials, according to one or more embodiments. The structures in Figures 8A and 8B both exhibit smooth curvatures, hierarchical branching, and unsupported overhangs, which are architectural motifs of mammalian vasculature.
[0108] The resolution of the final geometry of a three-dimensional structure, such as an interconnected vascular network, produced according to one or more embodiments described herein may be limited by the resolution of the SLS system. The resolution of the SLS system may be limited by factors such as the optics of the energy beam (e.g., laser spot size), which may define the approximate minimum diameter of the sintered filaments formed according to one or more embodiments. Alternative optics may be utilized to reduce the laser spot size, which may produce final geometries including smaller filaments according to some embodiments. Furthermore, the resolution of the SLS system may be limited by the thickness of the powder layer, which may be limited by the diameter of the powder. Thus, a thinner powder layer, and thus a more finely ground powder, may produce final geometries including smaller filaments according to some embodiments.
[0109] In some embodiments, the filament may have a minimum diameter of 300 μm (e.g., 200 μm, 100 μm, 50 μm, 10 μm, or 1 μm). In some embodiments, the filament (formed by a single pass of the energy beam) may have a maximum diameter of 1 mm (e.g., 5 mm, 10 mm, 50 mm, 100 mm, or 500 mm).
[0110] As in the embodiment shown in Figures 8A and 8B, a three-dimensional structure formed from a powder including an anti-caking agent may include both the anti-caking agent and the carbohydrate powder in the final structure. In other words, in some embodiments, the energy beam irradiation, such as that performed during SLS of a powder including a carbohydrate powder and an anti-caking agent, can sinter and / or melt both the carbohydrate powder and the anti-caking agent upon solidification into the final three-dimensional structure.
[0111] According to one or more embodiments, sintered structural materials shaped with isomalt or isomalt + cornstarch as presented herein have Young's moduli on the same order of magnitude as extruded carbohydrate glasses, are stiff and brittle, and are robust enough to support their own weight, survive multiple post-processing steps, and / or be transported between laboratories.
[0112] As shown in FIG. 9 and summarized in the table below, the sintered carbohydrates are rigid and brittle under compression, making them self-supporting and amenable to extensive handling, multiple coating steps and casting in viscous prepolymer solutions.
[0113] FIG. 9 shows graphs of stress (in megapascals (MPa)) versus strain (in percent elongation) collected during uniaxial compression of solid and macroporous carbohydrate cylinders according to an embodiment of this method. FIG. 9 also shows a schematic diagram of each geometry as an inset located adjacent to the corresponding stress versus strain curve. Young's modulus (MPa) was measured as the slope of the stress-strain curve in the linear region, yield stress (MPa) was measured as the peak stress value before failure, and yield strain (%) was measured as the corresponding strain. The accompanying results in the table summarize n=11 solid cylinders and 9 cylindrical lattices printed over three independent printing runs.
[0114] [Table 1]
[0115] According to one or more embodiments, the solidified three-dimensional structure may be capable of supporting its own weight and may be brittle and rigid. In some embodiments, the three-dimensional structure shaped from the structural material may have a Young's modulus in the range of 200 MPa to 1000 MPa (e.g., in the range of 500 MPa to 1000 MPa, in the range of 500 MPa to 700 MPa), for example approximately 600 MPa.
[0116] In some embodiments, the three-dimensional structure formed from the structural material may take the form of a filament network that may be formed with multiple filaments, a three-dimensional branched pattern, an interpenetrating geometry, and / or an unsupported geometry.
[0117] According to one or more embodiments of the present disclosure, the materials and / or methods may be applied to shape three-dimensional structures and thus final geometries that may include various freeform structures and / or patterned fluidic networks.
[0118] The structural material may be capable of dissolving and / or disintegrating upon interaction with a liquid, according to one or more embodiments. In some embodiments, the liquid may include, but is not limited to, or may utilize one or more of water, saline, or phosphate buffered saline (PBS). After solidification, the three-dimensional structure may be ready to be used as a template for sacrificial templating, according to one or more embodiments discussed below.
[0119] The three-dimensional structures, in one or more embodiments, may be environmentally stable such that they can be stored for several weeks before being used for sacrificial templating.
[0120] Surface smoothing according to one or more embodiments
[0121] After SLS according to embodiments of the present disclosure, the surface of the three-dimensional structure may initially have a large surface roughness. This surface roughness may be due in part to the three-dimensional structure being decorated by loosely bound granules of the powder. To remove these loosely bound granules and / or reduce the surface roughness, the three-dimensional structure may undergo surface smoothing with a smoothing solution. The smoothing solution may include one or more carbohydrates. The smoothing solution may include a solvent. The contact of the smoothing solution with the three-dimensional surface may be controlled. After surface smoothing according to some embodiments, the surface roughness of the three-dimensional structure may be reduced because the surface topography is smoother. In some embodiments, the surface smoothing may change the size and / or mass of the three-dimensional structure. Some embodiments of surface smoothing may deposit an additional carbohydrate coating on the outside of the three-dimensional structure. Aspects of the smoothing solution are described in detail below.
[0122] In some embodiments, the smoothing solution may include a carbohydrate. In some embodiments, the carbohydrate used in the smoothing solution may include, but is not limited to, isomalt, dextran, sucrose, glucose, lactose, trehalose, maple syrup, and / or sugar cane syrup. Furthermore, some embodiments of the smoothing solution may include two or more carbohydrates. The carbohydrate used in the smoothing solution may or may not be present in the powder used to shape the three-dimensional structure to be processed.
[0123] The smoothing solution according to some embodiments may include one or more solvents, which in one or more embodiments may include, but are not limited to, water and / or organic solvents.
[0124] In some embodiments, the smoothing solution may have a sufficiently high concentration of one or more carbohydrates included therein such that the smoothing solution does not dissolve or undesirably alter the three-dimensional structure.
[0125] The three-dimensional structure may be contacted with the smoothing solution by washing, spraying or immersion in some embodiments. Contact of the three-dimensional structure with the smoothing solution may continue for a contact time. In some embodiments, the contact time and / or contact method may be adjusted so that the contact is sufficient to remove loose granules and smooth the surface of the three-dimensional structure. In some embodiments, the contact time and / or contact method may be adjusted so that the contact does not dissolve or undesirably alter the three-dimensional structure.
[0126] For example, FIGS. 10A and 10B are scanning electron micrographs of three-dimensional structures formed by SLS before (FIG. 10A) and after (FIG. 10B) surface smoothing according to embodiments herein.
[0127] FIG. 10A is a scanning electron micrograph of a three-dimensional structure having the shape of a filament formed by SLS of a powder according to one or more embodiments discussed above. Upon visual inspection, FIG. 10A shows significant surface roughness. This surface roughness may be due to partially fused powder granules decorating the surface. Scale bars are 200 μm (magnification) and 1 mm (inset).
[0128] FIG. 10B is a scanning electron micrograph of the same filament after surface smoothing with a smoothing solution. In particular, the filament shown in FIG. 10A was treated with a smoothing solution consisting of a concentrated isomalt solution (60 weight percent (wt%) isomalt in water) according to an embodiment of the present disclosure. After surface smoothing, the filament was re-imaged as FIG. 10B. Upon visual inspection, FIG. 10B shows low to moderate surface roughness. As in FIG. 10A, the scale bars in FIG. 10B are 200 μm (magnification) and 1 mm (inset).
[0129] In one or more embodiments, the surface smoothing can leave the three-dimensional structure largely intact. In one or more embodiments, the surface smoothing can be one that does not significantly change the overall architecture. In one or more embodiments, the surface smoothing can be one that does not significantly bridge across regions of the three-dimensional structure. In one or more embodiments, the surface smoothing can reduce surface roughness. In one or more embodiments, the surface smoothing can remove loose granules of powder from the three-dimensional structure. The above features are evidenced by a comparison of micrographs of filaments taken before (FIG. 10A) and after (FIG. 10B) surface smoothing according to some embodiments. In particular, a comparison of FIGS. 10A and 10B shows surface smoothing that reduces surface roughness without significantly changing the overall architecture or bridging across the filaments.
[0130] In addition to the visually apparent improvement in filament surface roughness seen in Figures 10A and 10B, some embodiments of surface smoothing may modify the filament in one or more of the following quantifiable ways: reducing surface roughness, reducing filament width, or increasing filament mass. Figures 11A-11D show the effect of surface roughness (R a The effect of surface smoothing on the filament size (Figure 11A), filament width (Figures 11B and 11C), and filament mass (Figure 11D) are shown numerically. Data shown in Figures 11A, 11B, and 11D are mean ± standard deviation for n=9 (Figure 11A), n=24 (Figure 11B), and n=7 (Figure 11D). For Figures 11A and 11B, p<0.001 as paired t-test. For Figure 11C, p<0.01 as paired t-test.
[0131] Some embodiments of surface smoothing can result in reduced average surface roughness. To that end, FIG. 11A shows the R as a filament after surface smoothing according to an embodiment of the present disclosure. a This represents a nearly two-fold decrease.
[0132] Some embodiments of surface smoothing can result in a reduced size of the three-dimensional structure. In some embodiments, such size change can reflect a measurement of the width of the filament before and after surface smoothing, as in FIG. 11B. To that end, FIG. 11B shows a reduction in filament width between 100 μm and 200 μm after surface smoothing according to an embodiment of the present disclosure. Some embodiments may develop alternative metrics to quantify size changes that may depend on the geometry of the three-dimensional structure. Additionally, FIG. 11C is a graph of the change in filament width (μm) due to surface smoothing as a function of laser speed (mm / min). FIG. 11C shows that surface smoothing reduces the filament diameter by a nearly constant amount, approximately 100 μm.
[0133] Some embodiments of surface smoothing can result in an increase in the mass of the three-dimensional structure. To that end, FIG. 11D shows an increase in filament mass of between 10% and 20% after surface smoothing according to an embodiment of the present disclosure.
[0134] In some embodiments, surface smoothing allows an additional carbohydrate coating to be deposited on the outside of the three-dimensional structure, which may have the same or a different composition compared to the three-dimensional structure prior to surface smoothing.
[0135] In some embodiments, the three-dimensional structure may undergo surface smoothing by exposure to a smoothing solution as discussed above before undergoing further steps. In other embodiments, the three-dimensional structure may not undergo surface smoothing before undergoing further steps. The three-dimensional structure may, but does not, further comprise a smoothed surface as described above, according to one or more embodiments.
[0136] Surface Coating According to One or More Embodiments
[0137] To prevent premature dissolution of the three-dimensional structure and / or to establish a barrier between the three-dimensional structure and cells that may be contained in the matrix material, one or more embodiments of the method include a step of surface coating the three-dimensional structure with a surface coating material. Thus, according to one or more embodiments, the three-dimensional structure may further include a surface coating material. Such a surface coating material may coat the surface without backfilling the spaces between the regions of the three-dimensional structure and / or without substantially changing the shape of the three-dimensional structure. In one or more embodiments, the surface coating material may not significantly backfill the void space. Aspects of the surface coating are described in detail below.
[0138] In one or more embodiments, the surface coating material may include a polymer solution. The polymer solution may include, but is not limited to, polycaprolactone, poly(L-lactide), polylactic acid, poly(lactic-co-glycolic acid), amphiphilic copolymers including members of the Pluronic family, amphiphilic copolymers including ester or ether derivatives of poly(ethylene glycol) molecules, collagen, gelatin, zein, ghee, shellac, starch, wax, or petrolatum. In one or more embodiments, the surface coating material may include one or more polymer solutions.
[0139] Some embodiments of the surface coating material may include one or more solvents. In one or more embodiments, the surface coating material may include dichloromethane and / or chloroform.
[0140] In some embodiments, the surface coating may be applied, for example, by immersion (immersion + solvent evaporation), among other coating techniques.
[0141] Some embodiments of the surface coating material may be a hydrophobic material. In some embodiments, the surface coating material may include a hydrophobic polymer as a surface coating. The hydrophobic polymer may include, but is not limited to, polycaprolactone, poly(L-lactide), polylactic acid, poly(lactic-co-glycolic acid), collagen, gelatin, zein, shellac, starch, wax, or petrolatum. The surface coating may include one or more hydrophobic polymers. The hydrophobic polymer may be dissolved in one or more solvents, such as dichloromethane and / or chloroform.
[0142] In other embodiments, the surface coating material may be an amphiphilic material that contains both hydrophilic and hydrophobic functional groups. The amphiphilic material may include, but is not limited to, the amphiphilic materials included above, such as Pluronics (block copolymers of poly(propylene oxide) and poly(ethylene oxide)), and hydrophobic molecules combined with poly(ethylene glycol) (PEG), such as PEG stearate, PEG oleate, PEG laurate, PEG castor oil, or PEG myristate. The surface coating may include one or more amphiphilic materials. The amphiphilic material may be dissolved in one or more solvents, such as dichloromethane (DCM) and / or chloroform. For example, such solvents may be used for dip coating.
[0143] In some embodiments, the three-dimensional structure may be surface coated by exposure to a surface coating material as discussed above before undergoing further steps. In other embodiments, the three-dimensional structure may not be surface coated before undergoing further steps. The three-dimensional structure, according to one or more embodiments, may or may not further comprise a surface coating as described above.
[0144] The solidified, (optionally) surface-smoothed, and (optionally) surface-coated three-dimensional structure according to one or more embodiments may be brittle and rigid, and therefore may be able to continue to support its own weight. The three-dimensional structure may, in one or more embodiments, be sufficiently environmentally stable to be stored for several weeks before use. After solidification, (optionally) surface-smoothed, and (optionally) surface-coated, the three-dimensional structure may be used as a template for sacrificial templating according to one or more embodiments discussed below.
[0145] Backfilling of Void Space According to Embodiments
[0146] The three-dimensional structure can then be encased in a matrix material by surrounding the three-dimensional structure with the matrix material and allowing the matrix material to solidify. Thus, the void spaces of the three-dimensional structure may be backfilled with the matrix material. In some embodiments, this three-dimensional structure may serve as a template for the SLS-CaST process. Aspects of backfilling are detailed below, and examples are further presented in Figures 12A-12F.
[0147] In some embodiments, the matrix material may be an orthotropic bulk material, hi some embodiments, the matrix material may be a biomaterial, an elastomer, a plastic, a hydrogel, a biomaterial, a silicone, or some other curable material, or a mixture of one or more of these materials.
[0148] Some embodiments of the matrix material may be, according to some embodiments, an elastomer and / or plastic that may include polydimethylsiloxane, polycaprolactone foam, epoxy-based matrix, or monomers, dimers, or oligomers of any of these materials. In one or more embodiments, the elastomer and / or plastic may include one or more of these elastomers and / or plastics.
[0149] Some embodiments of the matrix material may be biomaterials that may include polyamide, poly(2-hydroxyethyl methacrylate), poly(vinyl alcohol), polyacrylamide, poly(ethylene glycol), polyurethane, collagen, agarose, albumin, alginate, chitosan, starch, hyaluronic acid, gelatin, fibrin, matrigel, glycerin, glycol, mannitol, inositol, xylitol, adonitol, glycine, arginine, biological polymer molecules, albumin, peptide amphiphiles, or monomers, dimers, or oligomers thereof. In one or more embodiments, the biomaterial may include one or more of these biomaterials. Some embodiments of such biomaterials may be biomaterials that are crosslinked after solidification.
[0150] In one or more embodiments, the biomaterial may be in solution, and thus the matrix material may further comprise a solvent, hi some embodiments, the solvent may be water or saline.
[0151] In some embodiments, the matrix material may be an aqueous solution of at least one biomaterial, which may further comprise a suspension of living cells.
[0152] In one or more embodiments, the vascular network may be formed from two or more matrix materials, each occupying a distinct region of the void space. For example, layers of different materials may be cast on top of each other. Examples of such embodiments are described in further detail below.
[0153] In one or more embodiments, the vascular network may be formed from a matrix material that includes two or more living cell types, with each cell type occupying a distinct region of the void space. Examples of such embodiments are detailed further below.
[0154] In one or more embodiments, the matrix material and the surface coating material may be different materials. In alternative embodiments, the matrix material and the surface coating material may be similar materials or the same materials.
[0155] In some embodiments, the method can be compatible with the cell-containing material without adversely affecting the viability of the encapsulated cells. Thus, some embodiments of the method may be used in one or more embodiments without using cytotoxic reagents or conditions. Furthermore, in some embodiments, the matrix material, the structural material, the (optional) smoothing solution, and the (optional) surface coating material can be compatible with the cell-containing material without adversely affecting the viability of the encapsulated cells. In some embodiments, the matrix material and / or the surface coating material can include live cells. Experiments described below have shown that not only are cells metabolically active near the patterned conduits formed by embodiments of the present disclosure, but that cells proliferate over time in the areas near the conduits.
[0156] Solidification of matrix material according to embodiments
[0157] Some embodiments may employ matrix materials that simply solidify over time and may not require additional processing steps, while other embodiments may utilize matrix materials that solidify after additional solidification processes. The additional solidification processes may include, but are not limited to, one or more of catalyst addition, light curing, heat curing, and enzymatic curing. The additional solidification processes may crosslink, branch, network, or otherwise chemically or physically change the matrix material to solidify and / or improve its stiffness, strength, toughness, hardness, elasticity, and / or durability. In some embodiments, solidification of the matrix material (including any additional solidification processes) may be successfully completed when the matrix material has a mold formed with the structural material disposed or embedded therein. Aspects of matrix solidification are described in more detail below.
[0158] Some embodiments may use a matrix material that simply solidifies over time. Some such embodiments may not require an additional processing step to solidify the matrix material. The exact solidification mechanism of such embodiments may include, but is not limited to, one or more of a chemical reaction and solvent evaporation within the matrix material.
[0159] Some embodiments may use a catalyst addition for the solidification of the matrix material. Some such embodiments may require mixing a catalyst with the matrix material prior to curing.
[0160] Some embodiments may use light curing (also called photopolymerization) for solidification of the matrix material. In some embodiments, the electromagnetic radiation used for light curing may include visible light or ultraviolet (UV) light. In accordance with one or more embodiments described above, despite the opacity of the mold formed of the structural material, light curing can be successfully used around the mold to solidify the matrix material by photopolymerization, such as polyethylene glycol diacrylate (PEGDA).
[0161] Some embodiments may use heat curing (also called thermal polymerization) for solidification of the matrix material. According to one or more embodiments, thermally crosslinked matrix materials such as agarose may be compatible with SLS-CaST.
[0162] Some embodiments may use enzymatic curing for solidification of the matrix material. Some such embodiments may require mixing of an enzyme with the matrix material prior to curing. Previous studies have shown that gels of enzymatically polymerized silk fibroin and fibrin are difficult to pattern using extrusion. However, matrix materials that require enzymatic polymerization, such as silk fibroin and fibrin, may be well-compatible with the SLS-CaST method according to one or more embodiments of the present disclosure.
[0163] Other additional solidification processes to solidify and / or crosslink the matrix material may also be compatible with the SLS-CaST process.
[0164] Removal of template according to embodiment
[0165] After solidification of the matrix material around the three-dimensional structure that may be formed with the structural material, in one or more embodiments, the three-dimensional structure may be removed from within the matrix material. This removal may create an open cavity within the matrix material in the shape of the three-dimensional structure to form a perfusable compartment within the matrix material. In such cases, the three-dimensional structure herein may function as a mold. The structural material may be formed by an embodiment of the SLS process as described above.
[0166] Removal of the template from within the matrix material may be performed, according to one or more embodiments, by dissolving and / or disintegrating the structural material with a liquid, which in some embodiments may include or utilize one or more of, but is not limited to, water, saline, or phosphate buffered saline (PBS).
[0167] In some embodiments, the step of removing the template may be non-damaging to the living cells within the matrix material.
[0168] Incorporation of cells within a sacrificially templated matrix material, according to embodiments
[0169] The sacrificial template paradigm decouples three-dimensional printing from cell manipulation. Thus, laser-sintered filament networks can be used to pattern perfusable vascular networks in a variety of matrix materials, including biocompatible hydrogels of natural and synthetic origin, that can support cells in both the luminal (i.e., along the inner surface of the conduit) and parenchymal (i.e., outside the conduit; containing the bulk of the tissue) spaces.
[0170] In some embodiments, parenchymal cells can be homogenously incorporated into an interconnected vascular network. Some such embodiments may be formed by mixing the cells into the matrix material prior to solidification, for example, by mixing the cells into a pre-hydrogel solution.
[0171] In some embodiments, the area of the vascular network proximate to the flow passage may be seeded with endothelial cells. Some such embodiments may involve seeding the interconnected vascular network with a high density cell suspension (e.g., 30×10 per ml). 6 In some embodiments, the interconnected vascular network may be formed by injecting a cell suspension (such as a cell suspension of 100 cells) into the interconnected vascular network. In some embodiments, the interconnected vascular network may be spun after the cell suspension is injected into the interconnected vascular network to coat all of the channels. In such embodiments, endothelial cells (e.g., human umbilical vein endothelial cells (HUVECs)) may adhere to the luminal surface of the vascular network and form a monolayer along each channel.
[0172] Both healthy and diseased tissues of the body are characterized by spatially defined zones of cells and extracellular matrix (ECM). For example, the musculoskeletal system includes the interface between bone, cartilage and tendons, injured tissues may have areas of fibrous scar tissue, and tumors within tissues have unique microenvironments of cells and ECM. Culturing parenchymal fibroblasts alongside luminal endothelial cells is a plausible model due to the critical role played by such stromal cells in stabilizing the self-organization of endothelial cells into higher order structures such as putative capillary networks. Thus, some embodiments may include interconnected vascular networks with a spatially heterogeneous organization of cells and matrix materials.
[0173] Some embodiments may include both parenchymal cells and endothelial cells. In some embodiments, the parenchymal cells may be homogenously incorporated by mixing the cells into the matrix material prior to solidification, followed by incorporation of endothelial cells by flowing a cell-laden fluid through the vascular network after formation. In one or more embodiments, the vascular network may be formed from two or more matrix materials, each occupying a distinct region of the void space. Each of the matrix materials may or may not be seeded with one or more types of parenchymal cells.
[0174] As an example, luminal cells (HUVEC (GFP)) and parenchymal cells (IMR-90 fibroblasts (Histone 2B (H2B)-mOrange2), 2.5 × 10 per milliliter ( / mL) 6 Independent seeding of cell populations of up to 100 μm (10 μm) was demonstrated for dendritic vascular networks patterned within a gelatin methacrylate (GelMA; 10 wt%) matrix material.
[0175] After 11 days of perfusion culture (with a flow rate gradient), the endothelial layer was stable and well maintained, showing the formation of putative endothelial sprouts, growing into the fibroblast-laden parenchymal compartment. Furthermore, uniform interstitial seeding of IMR-90 fibroblasts was observed. Sectional confocal fluorescence microscopy further showed seeding of HUVECs across a series of channel branches in a dendritic network. HUVECs were well distributed around the periphery of the individual channels across the central channel of the network. Thus, the endothelialized channels showed a uniform coverage of endothelial cells around the entire periphery of the channel, across the various branches of the network.
[0176] In addition to homogeneously seeding cells in the stroma zone, embodiments may include a combination of strategically sequential gel polymerizations, distinct cell populations, and distinct matrix materials to form a single, spatially patterned, interconnected vascular network. Such interconnected vascular networks with spatially heterogeneous configurations of cells and materials may be useful for applications such as interface tissue engineering or screening tests performed under perfusion culture.
[0177] For example, extravascular tissue transition zones were cast along shared vascular networks to form integrated, perfusable constructs. Control over spatial tissue patterning was achieved by injecting fibroblasts (IMR-90 (H2B-mOrange2); 5 × 10 6 The bottom zone contained cancer aggregates (344SQ(H2B-mVenus); 15,000 aggregates / ml) along with endothelial cells (HUVEC(GFP); 5 × 10 6 / ml) alongside fibroblasts (10 × 10 6 / ml), and the intermediate zone containing supporting fibroblasts (5 × 10 6 / ml) together with endothelial cells (15 × 10 6This was evident upon seeding of the upper zone with 100 μg / ml of erythrocytes. Images of the individual fluorescent channels showed well-defined interfaces between the zones, and multiple days of perfusion into the branching vascular network demonstrated channel patency and robust lamination between the layers.
[0178] Examples of matrix materials in SLS-CaST according to embodiments
[0179] The photographs in Figures 12A-12F show similarly shaped channels in multiple matrix materials of various compositions, stiffnesses, and / or solidification mechanisms formed using the SLS-CaST method according to one or more embodiments. Here, the SLS-CaST procedure used a single mold design in the shape of a filament network to form channels that form an interconnected vascular network through each of these matrix materials. Each of the matrix materials was backfilled and solidified around the mold according to the embodiments described above, and the mold was then dissolved. Figures 12A-12F highlight the interconnected vascular network through each matrix material by perfusing the interconnected vascular network with blue dye. As seen in Figures 12A-12F, perfusion into the patterned channel network can show channel patency and connectivity of the interconnected vascular network within each matrix material. Additionally, Figures 12A-12F show various matrix material solidifications with various solidification methods. Details of sample fabrication will be discussed further.
[0180] 12A and 12B show interconnected vascular networks within rigid elastomers or plastics, specifically polycaprolactone (PCL) foam in FIG. 12A and polydimethylsiloxane (PDMS) in FIG. 12B (scale bar=5 mm), according to one or more embodiments. Thus, rigid elastomers or plastics may be used as matrix materials for SLS-CaST methods, according to embodiments of the present disclosure. According to one or more embodiments, channels patterned within rigid elastomers or plastics may have applications for imaging phantoms, as microfluidic devices, or as scaffolds for bone tissue engineering.
[0181] 12C-12F show interconnected vascular networks within matrix materials including a series of natural and synthetically derived hydrogels (e.g., biomaterials) of various stiffness and solidification mechanisms, according to one or more embodiments. Thus, natural or synthetically derived hydrogels may be used as matrix materials in the SLS-CaST method, according to embodiments of the present disclosure. These embodiments highlight the versatility of SLS-CaST to shape interconnected vascular networks from matrix materials with a variety of crosslinking mechanisms, mechanical properties, and cell-matrix interactions. In particular, according to embodiments of the present disclosure, the hydrogel utilized in FIG. 12C is PEGDA, in FIG. 12D is agarose, in FIG. 12E is silk fibroin, and in FIG. 12F is fibrin.
[0182] An interconnected vascular network within a biomaterial may be used as a perfusable model tissue, according to some embodiments. Thus, each interconnected vascular network shown in Figures 12C-12F was formed within the biomaterial without the use of cytotoxic reagents or conditions. In one or more embodiments, the biomaterial in which the interconnected vascular network may be formed may further include a suspension of living cells.
[0183] Additionally, Figures 12A-12F show matrix materials with different additional solidification treatments according to embodiments herein. Figure 12A is an embodiment of the present disclosure where the matrix material (polycaprolactone (PCL)) was initially prepared with a solvent (chloroform) and solidified mostly by solvent evaporation upon overnight curing. In Figure 12B, the matrix material (PDMS) was initially prepared with a catalyst according to embodiments of the present disclosure and solidified upon 48 hours of curing. Figure 12C shows an embodiment of the present disclosure where photopolymerization was successfully performed to crosslink PEGDA by photopolymerization using incident light from various angles. Figure 12D shows an embodiment of the present disclosure where thermal curing was successfully used to crosslink agarose by exposure to high temperatures. Finally, Figures 12E and 12F show patterned channels formed in silk fibroin and fibrin according to embodiments of the present disclosure, thus illustrating successful enzymatic polymerization of matrix material backfilling around the template. The vascular network formed using PCL in Figure 12A was imaged using micro-computed tomography (μCT) because the matrix material is opaque. In contrast, vascular networks formed with various transparent matrix materials were perfused with a blue liquid and imaged (Figures 12B-12F).
[0184] Complex channel architectures fabricated by SLS-CaST, according to embodiments
[0185] One or more embodiments herein can also pattern more complex channel configurations that encompass some of the prominent architectural motifs of natural mammalian vascular networks. Referring now to Figures 13-16, hierarchically branched and interconnected vascular networks are shown, according to one or more embodiments. Each of Figures 13-16 was formed, according to one or more embodiments, using the SLS-CaST method discussed above from a matrix material of 2 wt% low melting point agarose. Furthermore, in some embodiments, the matrix material can define an interconnected vascular network with multiple fluidically independent flow paths. Figures 17A-17D show such an interconnected vascular network with two flow paths. Further details of the fabrication method for the substantially interconnected vascular networks of Figures 13-17 are discussed further.
[0186] Axial tapering of the channels, while important for printing using extrusion, can be easily incorporated into the sintered carbohydrate mold according to one or more embodiments and thus realized downstream in the interconnected vascular network formed in the matrix material. To that end, Figs. 13A-13E show an embodiment of an interconnected vascular network in a matrix material with tapered channels and smooth curvatures. Fig. 13A shows a perspective schematic of a substantially interconnected vascular network including two planes showing the location of two cross sections. The interconnected vascular network is 10 mm x 20 mm x 4 mm and has a volume of 800 microliters (μL). Cross-sectional planes A and B are also displayed. Fig. 13B is a perspective view photograph of a filament network shaped by SLS for use as a mold. Fig. 13C is a photograph of a top view of the resulting interconnected vascular network showing the scale bar = 2 mm. Fig. 13D is a photograph of a cross section obtained in plane A and Fig. 13E is a photograph of a cross section obtained in plane B. Cross sections of the matrix material in the planes shown in Figures 13C and 13D show channels with annular profiles that decrease in diameter as one approaches the center of the matrix material. This change can be seen by comparing a cross section occurring closer to the center of the matrix material (Figure 13D) with a cross section occurring further away from the center of the matrix material (Figure 13C), which decreases in diameter as one approaches the center of the matrix material.
[0187] In the case of existing extruded sacrificial molds, filament crossings require each filament to be deposited on top of each other (i.e., in separate planes, requiring redundant material dispensing). Thus, the inter-conduit junctions patterned with the extruded mold retain this stacked log cabin morphology, which is significantly different from native inter-conduit junctions. In contrast, according to one or more embodiments, the SLS mold can provide seamless branching transitions even for junctions between many adjacent channels. Thus, FIGS. 14A-14E show an embodiment of a three-dimensional interconnected vascular network that shows a smooth transition from a single inlet and a single outlet to four daughter branches. FIG. 14A shows a perspective schematic of such a substantially interconnected vascular network. The interconnected vascular network is 7 mm×13 mm×7 mm and has a volume of 650 μL. FIG. 14B is a top-view photograph of a filament network shaped by SLS for use as a mold. Figure 14C is a perspective view photograph of the resulting substantially interconnected vascular network showing scale bar = 5 mm. Figure 14D is a photograph showing a top view of the same vascular network. Figure 14E is a photograph of a cross section taken through four daughter branches near the center of the vascular network. Figure 14E clearly shows separate round channels for each of the four daughter branches.
[0188] In some embodiments, SLS-CaST may be used to create interconnected vascular networks with hierarchical channel branching in all three dimensions as shown in Figures 15A-F and 16A-F. In contrast, neither multiple branching repeats nor suspended overhang geometries can be easily created with existing extrusion printing processes. Such embodiments may be useful, for example, for replicating mammalian vasculature. For such embodiments, carbohydrate templates for creating these example architectures may require one or more of unsupported overhangs and / or underhangs, branches, or curved channels oriented at an angle to the x, y, and z axes of the template.
[0189] To that end, Figures 15A-15F show an interconnected vascular network in the form of a three-dimensional hierarchical network including multiple branching repeats and suspended overhang geometries, according to one or more embodiments. Thus, Figures 15A-15F show an embodiment of a three-dimensional interconnected vascular network showing eight daughter branching repeats with suspended overhang geometries. Figure 15A shows a perspective schematic of such a substantially interconnected vascular network. The interconnected vascular network is 16 mm x 25 mm x 8 mm and has a volume of 3.2 milliliters (mL). Figures 15B and 15C are photographs showing a perspective view and a top view, respectively, of a filament network shaped by SLS for use as a template. Figure 15D is a perspective view photograph of the resulting substantially interconnected vascular network showing the scale bar = 10 mm. Figures 15F and 15F are photographs showing a top view and a side view, respectively, of the same vascular network.
[0190] 16A-16F further show interconnected vascular networks having a shape of a twisted three-dimensional hierarchical network according to one or more embodiments. Moreover, the vascular networks of FIGS. 16A-16F have smooth, non-uniform branching tapering in all three dimensions and a higher degree of channel tortuosity compared to FIGS. 15A-15D. FIG. 16A shows a perspective schematic of such a substantially interconnected vascular network. The interconnected vascular network is 16 mm×25 mm×10 mm and has a volume of 4 mL. FIGS. 16B and 16C are photographs showing a perspective view and a top view, respectively, of a filament network shaped by SLS for use as a template. FIG. 16D is a perspective view photograph of the resulting substantially interconnected vascular network showing the scale bar=10 mm. FIGS. 16E and 16F are photographs showing a top view and a side view, respectively, of the same vascular network.
[0191] In some embodiments, the substantially interconnected vascular network may include a plurality of independent flow paths through the matrix material. In other words, in some embodiments, the matrix material may define a substantially interconnected vascular network having two or more fluidly separated fluid paths. Such a substantially interconnected vascular network may be formed using a filament network including a plurality of filaments that act as a template for the plurality of independent flow paths.
[0192] Some embodiments of such vascular networks can include a plurality of filaments that define a plurality of independent flow paths that are not intertwined. In other words, such vascular networks can include a plurality of independent flow paths, each flow path being defined by a filament network that can be separated without destroying any of the filament networks. Such embodiments can be formed by separately shaping a plurality of filament networks by SLS and placing the filament networks in close proximity immediately prior to solidification of the matrix material.
[0193] Some embodiments of such vascular networks may include a plurality of filament networks that define a plurality of independent flow paths that are "intertwined," meaning that the filament networks are geometrically connected and therefore cannot be separated without destroying at least one of the filament networks. In some embodiments, forming such a substantially interconnected vascular network may include shaping a plurality of entangled filament networks by SLS and casting the vascular network in a matrix material. In some embodiments, the entangled filament networks may be shaped to mirror their final configuration in the interconnected vascular network prior to the introduction of the matrix material.
[0194] FIG. 17A shows a perspective schematic of such a substantially interconnected vascular network with two filament networks. The interconnected vascular network is 18 mm×30 mm×8 mm and has a volume of 4.3 mL. FIG. 17B and 17C are photographs showing a perspective view and a top view, respectively, of a filament network shaped by SLS for use as a template. FIG. 17D is a perspective view photograph of the resulting substantially interconnected vascular network showing the scale bar=5 mm. FIG. 17E and 17F are photographs showing the top view and the side view, respectively, of the same vascular network with the scale bar=5 mm.
[0195] Upon closer inspection, it is apparent that the two entangled filament networks of Figures 17A-F could not be separately shaped by SLS and then entangled just prior to solidification of the matrix material. Instead, the two entangled filament networks of Figures 17A-F were co-shaped by SLS in a configuration that is approximately equivalent to their final configuration within the solid matrix material. In particular, Figure 17B shows two entangled filament networks shaped simultaneously by SLS in accordance with an embodiment of the present disclosure. The as-shaped filament network of Figure 17B mirrors the final configuration of the interconnected vascular network prior to introduction or solidification of the matrix material.
[0196] Model structure created by STS-CaST according to the embodiment
[0197] 18A-18B and 19A-19B show interconnected vascular networks fabricated by SLS-CaST from extracellular matrix (ECM) material containing living cells, according to one or more embodiments of the present disclosure. Such embodiments may be used to assess the metabolic activity of cells within sacrificially templated model tissues.
[0198] FIG. 18A shows the vascular network in a schematic manner, and FIG. 18B shows the vascular network after sectioning for staining and imaging. The interconnected vascular network is 6 mm×15 mm×6 mm. Prior to optical imaging, the vascular network is sectioned in the y-direction (parallel to the flow direction) into 300 μm thick slices, stained, and imaged by optical microscopy as shown in FIG. 19A and 19B. Cell staining was performed using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), and nuclear staining was performed using Nuclear Green™ LCS1.
[0199] 19A-19B are optical micrographs of two cell-laden vascular networks modeled using ECM matrix material having cell densities of 10e6 cells per milliliter (cells / mL) (FIG. 19A) and 60e6 cells / mL (FIG. 19B) in accordance with embodiments of the present disclosure.
[0200] Each sample was imaged after perfusion with cell media on days 0, 3, and 7. Thus, over time, Figures 19A-19B show the development of a characteristic annular region of highly active cells near the channel. The inset for day 0 shows the entire cross section with dashed lines indicating the location of the expanded region.
[0201] The purple signal shown in each figure reflects a formazan precipitate that forms when live cells actively metabolize the compound 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT).
[0202] In view of the above, one-step casting of the entire ECM by SLS-CaST as described in the embodiments of the present disclosure can eliminate the need for very time-consuming cell / ECM printing, allow non-experts in tissue modeling to carry out experiments with patterned vascular networks, and facilitate experiments with a wider range of cells, aggregates, and organoids. Indeed, cells that are not tolerant to extrusion and / or that benefit from culturing in aggregates can be successfully incorporated into vascularized model tissues using SLS-CaST according to one or more embodiments described herein. Furthermore, one or more embodiments of the present disclosure can decouple the modeling of vascular architecture from experimental design, providing users with unconstrained access to the remarkable full range of biomaterials that have been developed over the past decades.
[0203] Computer Design of Artificial Vascular Structures According to Embodiments
[0204] To design biomimetic vascular networks with tunable conduit topology, a mathematical model of leaf venation may be used to computationally grow branched vascular networks as disclosed herein. Networks modeled in this way may then be fabricated by the SLS-CaST method according to one or more embodiments. The prevailing hypothesis of leaf venation formation is the auxin canalization theory, which describes venation differentiation as a feedback loop between auxin hormone flow and venation expression. A space colonization algorithm inspired by auxin canalization has previously been developed for parametric tree growth. To better model natural vasculature, a modeling framework called Mutual Tree Attraction was developed to generate branched three-dimensional networks with a single inlet and a single outlet. Figures 20A-20D illustrate Mutual Tree Attraction for computationally growing arborescent vascular networks according to one or more embodiments. In particular, FIGS. 20A-20D show schematic diagrams of algorithmic growth of a dendritic vascular network having a 2D architecture via mutual tree attraction, in accordance with one or more embodiments of the present disclosure.
[0205] In one or more embodiments, the mutual tree attraction method may include the following, as shown generally in Figures 20A-20D: instantiating a two-dimensional or three-dimensional growth region in which a dendritic network is generated, and instantiating two or more seed nodes that serve as starting points for generating a series of connected nodes; calculating the positions of a series of nodes and edges starting from the seed nodes to be connected in a biomimetic branching tree-like form; forming a closed branching network (dendritic network) by connecting terminal nodes from each node tree; and generating cylindrical filaments that follow the node tree to form a three-dimensional model of a dendritic vascular network. In some embodiments, the method of computationally generating a dendritic vascular network discussed herein can generate a dendritic network architecture for an interconnected vascular network formed by the SLS-CaST method as discussed above.
[0206] In some embodiments, virtual hormone sources may be stochastically placed in the two-dimensional or three-dimensional growth region to affect the growth path and branching behavior of a series of nodes starting from the inlets and outlets. Thus, each seed node may be the root of a branching tree of nodes according to one or more embodiments. Through a series of discrete time steps, each seed point representing an inlet or outlet becomes the root of a branching tree of nodes filling the growth region, as described elsewhere. In some embodiments, the branching morphology may be tuned by adjusting the hormone source density and the attraction effect of the hormone source, which decays with distance. In FIG. 20A, two seed nodes 51 are located in an elliptical growth region 53. The first inset schematic shows the attraction field of the virtual hormone source 55 in the growth region 53. In the second inset schematic, the root 57 is extended along the average vector between the node 51 (here the seed node) and the virtual hormone source 55 to form a second node 59. The node 59 is a branching point formed by multiple nodes being differentially attracted to their respective nearby hormone sources 55.
[0207] To form a closed architecture, the actively growing branch tips of each tree may act as additional hormone sources to attract the branches of the opposing tree. Thus, the advancing tips of each node tree may act as additional hormone sources such that as two growing trees approach the center of the growing area, their terminal branches are attracted to each other. FIG. 20B shows this intermediate stage as each tree grows and approaches the other, according to some embodiments of the present disclosure. Each node tree includes a branch 67 terminating in an advancing tip 69. Each node serves as a hormone source for nodes from the opposing tree. Thus, the tips converge, resulting in continuous branching.
[0208] In some embodiments, a hormone source in the growth region may act on both advancing trees, while a hormone source at the tip may act on the opposing tree (i.e., self-attraction may be inhibited within a tree). To that end, Figure 20C shows the state of network 71 where two trees have come together to form a closed network.
[0209] Finally, the set of nodes and edges representing the vascular network may be assigned conduit diameters according to some embodiments, and inlets and outlets may be added. Thus, a three-dimensional model of the architecture may be formed that is compatible with a typical digital fabrication tool chain. FIG. 20D now shows an example of the final network 73 created by this method once conduit thicknesses have been assigned and inlets 75 and outlets 77 have been added.
[0210] Conduit thickness, according to one or more embodiments, may be assigned independently of the network growth step and may be derived from Murray's Law or other theoretical relationships. In one or more embodiments, filament dimensions may be determined through mathematical relationships, including but not limited to Murray's Law. The completed architecture may be referred to as a dendritic network given its tree-like morphology, which is reminiscent of natural flow systems ubiquitous in biology and beyond.
[0211] Fabrication of model mammalian vasculature using SLS-CaST according to embodiments
[0212] 21A-21G, the fabrication of an interconnected vascular network in the form of a filament network made with multiple filaments using a template formed by SLS to form channels in a matrix material by SLS-CaST, according to one or more embodiments, is illustrated, where FIGs. 21A-21F display the resulting dendritic architecture with extensive unsupported branching and filament tapering. In the resulting perforated or patterned matrix, the channels are all intact and continuous.
[0213] A perspective schematic of an interconnected vascular network can be seen in Figure 21 A. Figure 21 B is a top schematic of the same vascular network showing planes C and D, where plane C is closer to the inlet and plane D is closer to the center of the vascular network. The interconnected vascular network, according to one or more embodiments, was computationally modeled within a three-dimensional elliptical region.
[0214] 21C is a photograph of a mold for an interconnected vascular network in the shape of the schematic of a filament network or dendritic vascular network shown in FIGS. 21A and 21B (scale bar=10 mm). The interconnected vascular network modeled for FIG. 21A is shown in FIG. 21C as it was printed using SLS from powder, according to one or more embodiments. The mold for the interconnected vascular network features hierarchical branching between a single inlet and a single outlet to a total of 33 daughter conduits.
[0215] Figure 21D is a micro-computed tomography (μCT) cross section of the mold shown in Figure 21 C. Figure 21D shows the intactness of the 33 daughter ducts.
[0216] 21E-21G are photographs of the final interconnected vascular network formed by sacrificial templating of the dendritic carbohydrate template of FIGS. 21A-21D according to one or more embodiments. After being formed by SLS-CaST in agarose hydrogel (2 wt%) according to embodiments of the present disclosure, the interconnected vascular network photographed in FIG. 21E (scale bar=10 mm) was patent and well perfused with blue liquid. Cross sections of the completed vascular network shown in FIGS. 21E and 21F (schematically shown in FIG. 21B as planes C and D, respectively) illustrate well-perfused channels with approximately circular cross sections within the agarose hydrogel matrix material.
[0217] Theoretical and empirical evaluation of fluid convection through the final interconnected vascular network of Figures 21E and 21F revealed that these networks can effectively distribute flow across all channel sections over a range of flow rates (Q). Perfusion of a planar dendritic network sacrificially templated in agarose gel resulted in fluid flow through all of the branches. Branching fluid flows exhibited low velocities in the daughter branches, but upon merging, the combined flows increased in velocity. The parabolic velocity profile increased in direct proportion to the applied flow rate, as expected for laminar flow of Newtonian fluids. Furthermore, simulating convection flow through this vascular network using standard computational fluid dynamics (CFD) reproduced experimentally observed velocity profiles to a similar order of magnitude, indicating that such simulations can provide meaningful predictive insights into the velocity profiles and magnitudes occurring in dendritic architectures.
[0218] Examination of the results of three individual particle image velocimetry (PIV) experiments, independent gel replicates, revealed that the magnitude of the maximum reproducible velocity (v max ) and wall shear stress (WSS), which increased predictably in direct proportion to the applied flow rate.
[0219]
number
[0220] Larger and more complex dendritic networks were generated by mutual tree attraction, modeled as carbohydrate templates, and then incubated with cell-laden agarose gels (25e 6The gels were cast with 100 μl of 1000 cells / mL, 2% by weight. MTT staining of sections (500 μm thick) from the center of the gels shows patent channels and uniformly distributed metabolically active cells at day 0. The central vascular proximity of this dendritic architecture was quantified by segmenting MTT-stained tissue sections and generating a map of the distance from each point in the parenchymal space to the nearest vascular channel or outer surface. Representing the map as a histogram and corresponding cumulative distribution function showed that 65% of the central tissue volume of this architecture was within 1 mm of the conduits and 88% of the volume was within 1.5 mm, suggesting that this architecture can successfully support cells within a viable distance of approximately 1 mm, which is discussed elsewhere. Although smaller conduits and interductal distances can be readily modeled by SLS-CaST, this example highlights that the complexity of the vascular network scales with tissue volume and the need to introduce new metrics to account for vascular proximity.
[0221] Sacrificial Template for a Branched Vascular Network According to One or More Embodiments
[0222] In particular, according to one or more embodiments described above, SLS of carbohydrates allows for the creation of templates with previously unavailable architectures, including those with a wide range of three-dimensional branching and / or unsupported geometries. Thus, according to one or more embodiments, one application of SLS of carbohydrates may be to form sacrificial templates for creating interconnected vascular networks in a biocompatible matrix. Thus, the interconnected vascular networks formed by SLS-CaST in a biocompatible matrix may take the form of dendritic vascular architectures (e.g., those described above and shown in Figures 20A-20D and Figures 21A-21G) and may include features of these newly available architectures to more accurately mimic biological structures. In view of the above, SLS-CaST may be an excellent method for forming tissue engineered model tissues that can mimic natural physiology.
[0223] In one or more embodiments, a three-dimensional model of a dendritic vascular network may be computationally generated using the Mutual Tree Attraction algorithm described above. Thus, the conduit diameter may be provided by Murray's Law or other mathematical relationships, according to one or more embodiments. Furthermore, an interconnected vascular network having the geometry of the computationally generated dendritic vascular network may be formed in a biocompatible matrix material by SLS-CaST, according to one or more embodiments. The SLS-CaST method described herein may optionally include surface smoothing and / or surface coating.
[0224] Sacrificial Templating of Model Tissue Vasculature Using Sintered Carbohydrates According to One or More Embodiments
[0225] In one or more embodiments, the vascular network can be created in the tissue engineered tissue by forming the tissue engineered tissue around a laser sintered carbohydrate template and then selectively removing the template from within the tissue engineered tissue. In one or more embodiments, the template can be solidified by sintering into the shape of the desired interconnected vascular network (e.g., a dendritic network architecture, see: sacrificial template for dendritic vascular networks according to one or more embodiments outlined above), then encased in a matrix material and dissolved away so that the interconnected vascular network is retained as void space in the matrix material. The matrix material used can be any biocompatible hydrogel or other biomaterial of various compositions according to one or more embodiments. Immediately after sintering, the template can benefit from additional processing by being soluble in water. Rapid dissolution of the template in a matrix material with aqueous components such as aqueous prepolymers can be avoided according to one or more embodiments by coating the template with a hydrophobic polymer (e.g., polycaprolactone or poly(lactic acid)). According to one or more embodiments, prior to coating with a hydrophobic polymer, treatment with a smoothing solution may be performed to smooth the surface of the mold. After surface coating and / or surface treatment and backfilling of the mold with a matrix material, the mold may be dissolved out of the matrix material according to one or more embodiments. In some embodiments, the matrix material may be a hydrogel (with or without crosslinking) containing cells. The system formed according to one or more embodiments described herein may be a tissue engineered tissue construct or a model tissue.
[0226] Furthermore, embodiments of the present disclosure may be applied to create vascular networks in matrix materials, including any hydrogel, that may contain any cell type or combination of cell types. Some embodiments may solve the major problem of modeling vascular topologies in biocompatible materials. For example, embodiments of the present disclosure may create tissue engineered tissue constructs that replicate the internal architecture of the kidney, liver, pancreas, or lung, and / or contain suitable cell types.
[0227] Furthermore, the embodiments may be used to model model systems for scientific research, e.g., the study of vascular function and / or disease. For example, a simple vascular network formed using one or more embodiments may serve as a platform for scientific research experiments. Such experiments may examine such things as vessel formation and remodeling; angiogenesis; tumor intravasation and extravasation; and atherosclerosis.
[0228] Additionally, model tissues of various types generated by one or more embodiments of the present disclosure may aid in the evaluation of pharmaceutical agents, including novel pharmaceutical agents.
[0229] Outside the field of biomodeling, one or more embodiments of the present disclosure may be used to create microfluidic devices. In some embodiments, a silicone matrix may replace the hydrogel used in the biomimetic example discussed above. Formation of microfluidic devices using silicone matrices according to embodiments of the present disclosure to form channels in rigid long-lasting materials. One application of silicone-based microfluidic devices made according to some embodiments may be the formation of flow phantoms for imaging studies.
[0230] Development of open source SLS hardware according to the embodiment
[0231] An OpenSLS printer (described elsewhere) was modified for compatibility with carbohydrate materials. In summary, a commercial CO2 laser cutter was modified by replacing the cutting table with a custom powder handling module and overriding the on-board electronics with an open source 3D printer motherboard (RAMBo v.3.1, Ultimachine) running a custom version of the open source Marlin firmware (https: / / github.com / MarlinFirmware). A powder handling module with a build volume of 11 × 13 × 9 cm was designed and fabricated primarily using laser cut acrylic and 3D printed poly(lactic acid) (Ultimachine) filament. The detailed design of the powder handling module components is available in the OpenSLS repository (https: / / github.com / MillerLabFTW / OpenSLS) along with a bill of materials, wiring diagrams and custom firmware.
[0232] 3D MODEL DESIGN AND PROCESSING ACCORDING TO EMBODIMENTS
[0233] 3D models were designed using OpenSCAD (http: / / openscad.org), Blender (https: / / blender.org), or with custom generative algorithms. A custom Python add-on for Blender was developed to interactively and parametrically generate hierarchically branched architectures, inspired by ductal branching in the natural process of intussusception (https: / / github.com / MillerLabFTW / IntussusceptionAddon). The DNA helix model in Figure 8A was created by jharris (https: / / www.thingiverse.com / thing:412973), and the ash tree model in Figure 8B was created by dutchmogul (https: / / www.thingiverse.com / thing:1079821). Three-dimensional models were sliced into G-code using the open source Slic3r software (Slic3r.org). To assign different laser speeds and power settings to different regions of the model, each region was assigned to a different extruder in Slic3r. After slicing, the G-code was modified using a custom Python script to match the OpenSLS motor axis assignments and incorporate the desired speed and power parameters. The G-code was sent to the press via the open source Pronterface console (github.com / Kliment / Printrun).
[0234] Preparation and laser sintering of carbohydrates according to embodiments
[0235] Isomalt (Decomalt, Paris Gourmet) was ground in a blade coffee grinder (Krups® F203) and then strained through No. 35 and No. 60 mesh sieves (60 mesh = 250 μm mesh spacing).
[0236] The sieved isomalt powder was mixed with food grade corn starch (Argo®) in a 7:3 weight ratio and placed in a freeze-dryer (Labconco®) as a precaution against ambient moisture and stored at room temperature in an airtight container.
[0237] For building, a thin layer of the powdered carbohydrate mixture was manually spread across the build platform on top of a layer of painter's tape.
[0238] The appropriate G-code file was started in Pronterface and the layer-by-layer build began.
[0239] For each layer to be printed, a laser is used to pattern solid geometries by selectively melting carbohydrate powder.
[0240] After this patterning step, new powder is added to the subsequent layer by shaking a sieve reservoir suspended above the build volume. The shaking action displaces the powder and deposits it onto the build platform while aerating and preventing compaction. After dispensing the new powder from the suspended reservoir, the powder pile is spread into an even layer by counter-rotating rollers. Finally, the excess powder is removed by a plough mechanism and collected under the powder handling module. The powder can be recycled tens to hundreds of times without any noticeable degradation in print quality.
[0241] Typical printing occurs at laser translation speeds of 1,000–2,000 mm / min (16–32 millimeters per second (mm / s)) and 40–60 W / mm from a 40 watt (W) laser tube (LightObject). 2 A power density of 1000 Hz was used.
[0242] All carbohydrate structures were built using a bed height of 150 μm, except for those shown in FIGS. 8A and 8B, which were built using a 250 μm bed height.
[0243] During the build, nitrogen gas was gently flowed through the laser nosepiece to prevent the buildup of carbohydrate residue on the laser optics, and the laser cutter was constantly flushed with a constant stream of nitrogen.
[0244] After building, excess loose powder was removed with a brush or compressed air, and individual filaments were cleaned using a needle.
[0245] Characterization of Carbohydrate Powders and Printed Parts According to Embodiments
[0246] The flowability of powdered carbohydrates was investigated using a powder rheometer (FT4, Freeman Technology®) using a stability and variable flow test protocol. Pure isomalt powder was compared to isomalt + corn starch powder and a commercial nylon powder (PA650, Advanced Laser Materials). After preconditioning of the powder using the manufacturer's recommended protocol, flow resistance was measured over 7 cycles in an unconstrained geometry at an impeller tip speed of 100 mm / sec to obtain a measurement of specific energy. To compare flowability between materials, the specific energy (mJ / g) was multiplied by the adjusted bulk density (grams per milliliter (g / ml)) of each powder and reported as mJ / ml.
[0247] Scanning electron microscopy was performed to analyze the morphology of the powdered material and to characterize the surface properties of the sintered parts. Powder grains or flakes of the sintered parts were sputter coated with 10 nanometers (nm) of gold (Desk V Sputter Coater; Denton Vacuum®) and imaged with a Quanta400 Environmental Scanning Electron Microscope (ESEM) (FEI Company®).
[0248] To measure the surface roughness of the sintered carbohydrate, scanning electron microscope micrographs were thresholded and the profile of each edge was extracted using a custom MATLAB® function. The formula:
[0249]
number
[0250] Volumetric microtomography scans of the sintered carbohydrate and PCL foams were obtained on a SkyScan™ 1272 X-ray computed tomography (CT) scanner (Bruker®) using a 50 kilovolt (kV), 200 microamp (μA) X-ray source and an exposure time of 424 milliseconds (ms). 180 degree scans were performed with 0.3° increments. Volume reconstruction and visualization were performed using NRecon and CTVox (Bruker®) software, respectively.
[0251] Uniaxial compression tests were performed according to a modified version of ASTM standard D695-02a. The samples were subjected to compressive loading on a mechanical testing system (858 Mini Bionix, MTS) equipped with a 10 kilonewton (kN) load cell. After applying a preload of 25 Newtons (N), the samples were compressed along the long axis (i.e., axial direction) using a crosshead displacement rate of 0.5 millimeters per minute (mm / min).
[0252] Carbohydrate Post-Treatment and Sacrificial Templating According to Embodiments
[0253] The sintered carbohydrate was post-treated by treatment with a concentrated isomalt solution. The isomalt was dissolved in boiling deionized water (60 grams (g) of isomalt in 100 milliliters (ml) of water), cooled at room temperature, filtered (Steriflip®, 0.22 μm polyethersulfone), and stored at room temperature. The sintered carbohydrate was submerged in and removed from the solution for 10-20 seconds, and then excess liquid was removed using a stream of pressurized nitrogen gas or by wicking with a Kimwipe™.
[0254] In some embodiments, it may be desirable to add an additional hydrophobic coating to the carbohydrate template to preserve the carbohydrate architecture during solidification of the bulk matrix material (e.g., warm agarose solution may deform carbohydrate filaments before fully solidifying, especially in large gels). A hydrophobic coating was used for all exemplary embodiments using agarose gels (FIG. 12D, FIGS. 13-17, and FIG. 21) and for the demonstration gels (FIGS. 12A-12C and 12E-12F). Hydrophobic coating was achieved by immersing the template in a solution of PCL (CAPA™ 6500, Perstorp AB®) or poly(lactide-co-glycolide) (AP062, Akina, Inc.) dissolved in chloroform (Alfa Aesar™; 10-25 mg / ml). Excess coating was removed using pressurized nitrogen gas, and the coated template was dried for 5 minutes.
[0255] In some embodiments, it may be desirable to omit the hydrophobic coating on the carbohydrate template. Uncoated templates were used for experiments involving endothelial cell seeding and spatial patterning (FIGS. 18A-19B and 19A-19B).
[0256] The carbohydrate templates designed for sacrificial templating were attached to glass slides by dispensing additional PCL solution (100 mg / ml in chloroform) along the interface between the template and glass and allowing the solvent to evaporate.
[0257] In a typical sacrificial templating experiment, the prepolymer was dispensed by pipette around the carbohydrate template after post-treatment as described above.
[0258] In Figure 12A, PCL foams were prepared by forming a slurry of sodium chloride crystals in a PCL solution (40 wt% in chloroform), then dispensing the slurry around the carbohydrate template and allowing the solvent to evaporate overnight. In Figure 12B, PDMS (Sylgard™ 184, Dow Corning®) was prepared by dispensing a 10:1 base:catalyst mixture around the carbohydrate template and curing for 48 hours (h).
[0259] In Figure 12C, a mixture of PEGDA (6 kDa, 20 wt%; synthesized as reported elsewhere) or GelMA (10 wt%; synthesized as reported elsewhere) and photoinitiator (Irgacure® 2959, 0.05 wt%; Ciba®) was dispensed around the carbohydrate template and photopolymerized (100 megawatts per square centimeter (mW / cm)) for 30-60 s. 2 ), 320–500 nm; Omnicure® S2000), photopolymerized PEGDA and gelatin methacryloyl (GelMA) hydrogels were prepared.
[0260] In FIG. 12D, agarose hydrogels were prepared by dispensing a solution of low-melting point agarose (2 wt %; Gold Biotechnology®) heated to 60°C and cooled to 37-42°C around the carbohydrate template and then cooling to 4°C.
[0261] In FIG. 12E, silk fibroin hydrogels were prepared by mixing an aqueous silk solution (4-8 wt%; gift from Dr. David L. Kaplan, prepared as reported elsewhere) with 10 units per milliliter (U / ml) of horseradish peroxidase and 10 microliters per milliliter (μL / mL) of hydrogen peroxide (165 millimolar (mM); VWR Chemicals BDH®), dispensing around the carbohydrate template, and polymerizing at 37° C. for 10 minutes (min).
[0262] In FIG. 12F, fibrin hydrogels were prepared by dispensing a mixture of fibrinogen (20 milligrams per milliliter (mg / ml); Sigma-Aldrich®) and thrombin (50 U / ml; Sigma-Aldrich®) around the carbohydrate template and polymerizing at 37° C. for 10 minutes.
[0263] For all bulk materials, carbohydrate templates were sacrificed (dissolved away) by incubation in deionized water or PBS, assisted by gentle flushing through the channels with a syringe.
[0264] The PCL foam was then frozen and lyophilized.
[0265] The sacrificially templated channels were visualized by perfusion with either Microfil® silicone injection compound or colloidal ink suspended in gel. Microfil® (Flow Tech) was prepared by mixing color base, diluent and hardener (4, 5 and 0.45 ml, respectively), pre-curing for 30 min, and then perfusing through the channel using a syringe pump for 10 min until fully cured. An ink suspension was formed by mixing India ink (between 10 and 100 μL / mL) with PEGDA (3.4 kDa, 20 wt%) and photoinitiator (Irgacure 2959; 0.05 wt%) or with agarose (2 wt%). This suspension was perfused through the channel by syringe and photopolymerized (60-120 s, 100 mW / cm at various rotation angles) while in the channel. 2 ) or cooled. The carbohydrate-templated and sacrificially templated materials were photographed using a DSLR camera by submerging the hydrogels in water for photography.
[0266] Preparation of cell-laden gels according to embodiments
[0267] Agarose gel containing HepG2 cells according to embodiments
[0268] HepG2 hepatoblastoma cells (ATCC) were grown in Dulbecco's modified Eagle's medium (DMEM, 4.5 g / L glucose; Corning®) supplemented with 10% fetal bovine serum (FBS) (Atlanta Biologicals™), 1% penicillin / streptomycin (Life Technologies™) and 1x non-essential amino acids (Caisson Laboratories, Inc.).
[0269] Agarose gels containing cells were prepared by resuspending an appropriate number of cells in complete medium to half the desired gel volume.
[0270] This cell suspension was combined with an equal volume of sterile low melting point agarose (4% by weight in PBS, heated to 60 °C and then cooled to 37-42 °C before mixing with the cells).
[0271] The cell suspension and agarose were mixed vigorously using a positive displacement pipettor (Microman® E, Gilson™) for approximately 60 seconds, then dispensed around the carbohydrate template and cooled to 4° C. for 5-10 minutes.
[0272] The carbohydrate template was removed by incubation in medium for 10-15 min, assisted by gently flushing medium through the channels with a syringe.
[0273] Agarose gel comprising primary hepatocyte aggregates according to embodiments
[0274] Hepatocyte / normal human dermal fibroblast (NHDF) aggregates were harvested, encapsulated in agarose gel, and catheterized in a perfusion chamber as described above for HepG2 cells. All experiments used final cell densities of approximately 10e6 hepatocytes / mL and 5e6 NHDF / mL, and acellular agarose gels ("blanks") were cast for albumin measurements as a negative control.
[0275] For comparison of static and perfused gels, one set of gels was perfused with hepatocyte culture medium at 20 µL / min using a multi-channel peristaltic pump.
[0276] A second set of gels was catheterized identically and 1 mL syringes of hepatocyte medium were connected at the inlet and outlet to allow diffusional transport of nutrients into the gels.
[0277] For long-term albumin measurements, effluent medium was collected from each perfused gel and static gels were flushed with fresh medium to capture accumulated albumin.
[0278] The dense dendritic gels were connected in a flow loop configuration containing 50 mL of hepatocyte culture medium that was continuously recirculated using a custom-designed peristaltic pump at 1 mL / min. Between 10 and 20 mL of medium was aseptically removed from the flow loop for albumin measurements each day and replaced with an equal volume of fresh medium.
[0279] Endothelialized GelMA Gel According to an Embodiment
[0280] Human umbilical vein endothelial cells (HUVECs) (Angio-Proteomie; passages 4–7) labeled with green fluorescent dye (GFP) were grown in complete Vasculife® medium (Lifeline® Cell Technologies) supplemented with 1% penicillin / streptomycin. IMR-90 lung fibroblasts (American Type Culture Collection) were grown in F-12K medium (American Type Culture Collection) supplemented with 10% FBS and 1% penicillin / streptomycin.
[0281] Sacrificially templated GelMA gels (10 wt%; planar dendritic architecture) were prepared with or without encapsulated IMR-90 fibroblasts (2.5e6 cells / mL) and catheterized in a custom perfusion chamber (as described in the Main Methods section) prior to HUVEC injection.
[0282] For seeding, HUVECs were resuspended to a density of 30e6 / mL and slowly injected into the channel by catheter. To promote uniform HUVEC adhesion, gels were seeded at 37°C for 6 h and rotated 90° every 15 min.
[0283] After the seeding period, gels were perfused with complete VascuLife® medium (HUVEC-only gels) or a 1:1 mixture of VascuLife®:F12-K medium (HUVEC gels with IMR-90) at a flow rate set at 5 μL / min for the first 24 h, 10 μL / min for the next 24 h, and 20 μL / min thereafter.
[0284] Spatially Patterned Gels According to Embodiments
[0285] GFP-HUVEC and IMR-90 fibroblasts were cultured as described above. 344SQ mouse metastatic lung adenocarcinoma cells were grown in RPMI1640 medium (Corning®) supplemented with 10% FBS and 1% penicillin / streptomycin.
[0286] Second generation lentiviruses were used under the oversight of the Rice University Institutional Biosafety Committee for protocol 662023 to stably transduce IMR-90 and 344SQ with plasmids encoding H2B-mOrange2 and H2B-mVenus, respectively.
[0287] 344SQ aggregates were formed as detailed elsewhere. Briefly, laser-ablated PDMS microwells were passivated using Pluronic F-127 solution (5 wt%) and seeded with 200 cells per well using gentle centrifugation (200 × g). After overnight aggregation at 37 °C, aggregates were harvested by gentle pipetting and loose cells were discarded by filtering through a 40 μm cell strainer.
[0288] For casting of spatially patterned gels, carbohydrate templates were prepared by adhering one open side to a glass slide and stretching Parafilm® across the opposing open side, leaving a small gap through which the prepolymer could be pipetted. Appropriate combinations of cells or beads (10 μm green, red, and blue fluorescent polystyrene beads; MagSpheres) were suspended in the appropriate hydrogel prepolymer and sequentially pipetted into the volume of the carbohydrate template.
[0289] Immediately after pipetting, the GelMA-containing gel region was partially photopolymerized (5 s light exposure), and photopolymerization was completed after casting of the entire construct (30 s exposure on each side).
[0290] After photopolymerization, the gels were incubated at 37° C. to completely crosslink the fibrin gel. The gels were catheterized in a perfusion chamber as described above and perfused with complete VascuLife® medium (5 μL / min).
[0291] Preparation and perfusion of cell-laden gels according to embodiments
[0292] Sacrificially templated gels as described above were prepared for multiday perfusion culture by catheterization in a custom-designed perfusion chamber cassette.
[0293] Generally, perfusion chamber volumes were formed from non-flowable PDMS (SE1700; Dow Corning®) extruded directly onto glass slides using a custom pneumatic extrusion printer (https: / / github.com / MillerLabFTW / ShopBot-PDMS-Printer) or from 3D printed poly(lactic acid) containing a molded PDMS gasket insert.
[0294] Each gel was placed in a perfusion chamber and catheterized using an intravenous cannula catheter (between 14 and 22 gauge depending on the vascular network architecture) or a flexible blunt luer tip between 15 and 20 gauge (Nordson Electron Fusion Devices).
[0295] In experiments using agarose gel, the gel was secured by dispensing additional cell-free agarose to fill the entire perfusion chamber to prevent the catheter from breaking or slipping out of the channel. After this additional agarose had solidified, the perfusion cassette was assembled by capping the chamber with a glass slide and tightening the screws to secure the assembly.
[0296] The endothelialized gel and gel containing a simple vascular network were connected in a linear flow path where complete medium was pumped from a syringe or medium bag through a sterile filter (0.22 μm polyethersulfone), through the gel, and into a waste container. Silicone microbore tubing (Tygon®, Cole-Parmer®) was used to connect the various components of the flow path.
[0297] Media was pumped at a flow rate of 25 L / min using either a syringe pump (NE-300; New Era Pump Systems™) or a multichannel peristaltic pump (Ismatec™, Cole-Parmer®).
[0298] The gels containing dendritic cells were connected in a flow loop configuration. For HepG2 flow loop perfusion (FIG. 6), 150 ml of complete medium was continuously circulated between the gel and an air-flushed PYREX® bottle using a 500 L / min peristaltic pump (Ismatec™).
[0299] For flow loop perfusion with long-term functional monitoring, a flow loop design was employed that allowed for sampling and replenishing the recirculating medium: an approximately constant volume of 50 mL was recirculated using a custom peristaltic pump, 20 mL was removed daily through a one-way sampling valve, and this volume was replenished from a connected IV bag.
[0300] Flow Loop Perfusion of Gels Patterned with Dendritic Vascular Networks, According to Embodiments
[0301] Gels containing dendritic networks were perfused in a flow loop using a peristaltic pump at a flow rate of approximately 500 μL / min. Vertical orientation of the gels improved perfusion of channels throughout the architecture compared to horizontally oriented gels. The gradient of phenol red color across the gel indicates the relative metabolite exchange available throughout the gel. Initially, gels were cast using medium without phenol red and perfused with medium containing phenol red. Immediately after perfusion began, perfusion medium is visible in the channels, but diffusion of nutrients and phenol red into the interstitial spaces is negligible. After 2 days, components of the perfusion medium can diffuse throughout the entire gel.
[0302] In gels containing metabolically active cells, a red color is observed in the vicinity of the perfusion channel where CO2 can be exchanged into the channel and carried away. In contrast, CO2 and metabolic waste products accumulate away from the channel, acidifying the gel and producing a yellow color. Gels containing no live cells confirmed that the perfusion medium diffused throughout the gel after 2 days, demonstrating that the phenol red gradient was due to differences in metabolite exchange rather than incomplete diffusion of phenol red.
[0303] For flow loop perfusion with long-term media sampling, a more advanced setup was designed with an IV bag containing media and a one-way sampling valve that could be added to the flow loop to replenish nutrients.
[0304] Immunostaining and Fluorescence Imaging According to an Embodiment
[0305] Wide-field fluorescent imaging of endothelialized spatially patterned gels was performed on a Ti-E inverted microscope (Nikon®) equipped with a Zyla™ 4.2 sCMOS camera (Andor® Technology Limited) using a motorized stage to obtain large-area scans.
[0306] Gel Sectioning and Imaging According to an Embodiment
[0307] Prior to staining, gels were removed from culture and live sectioned into slices with a vibrating microtome (VT1000S, Leica™). For sectioning, gels were attached to a metal stage using cyanoacrylate glue and submerged in room temperature DMEM (sectioning medium: serum-free, phenol red-free, 1% penicillin / streptomycin). Immediately after sectioning, gel slices were stained with MTT solution to measure metabolic production.
[0308] To prepare MTT stock solution, MTT (Bio Basic) was dissolved in PBS (5 mg / ml), sterile filtered, and stored at −20° C. MTT staining solution was prepared by mixing equal volumes of thawed stock solution and sectioning medium and added to the gel slices until they were completely submerged.
[0309] After 30 min (or 60 min for cold perfusion experiments), the MTT staining solution was aspirated and gel slices were fixed in paraformaldehyde (4%; Electron Microscopy Sciences, Inc.) for 30 min.
[0310] After fixation, slices were incubated with nuclear stain (3 μL / mL) (Nuclear Green™ LCS1; Abcam) and then washed in PBS (3×30 min) before imaging.
[0311] Stained gel slices were imaged under a stereomicroscope (SteREO Discovery.V8; Zeiss™) equipped with a DSLR camera (EOS 5DSR; Canon®).
[0312] While the present invention has been described with respect to a limited number of embodiments, those skilled in the art having the benefit of this disclosure will recognize that other embodiments may be devised that do not depart from the scope of the invention as disclosed herein. Accordingly, the scope of the present invention should be limited only by the appended claims.
Claims
1. A composition for forming a structure comprising an interconnected vascular network, comprising: a carbohydrate powder and a powder comprising an anti-caking agent; The powder has a granular form, A composition having a specific energy of less than 6 millijoules per milliliter (mJ / mL).
2. 10. The composition of claim 1, wherein the carbohydrate powder comprises at least one of dextran or isomalt.
3. 3. The composition of claim 1 or 2, wherein the anti-caking agent comprises at least one of corn starch, silicon dioxide, or xanthan gum.
4. The powder is configured to be a free-flowing powder; The composition according to any one of claims 1 to 3, wherein the powder has a maximum particle size of 250 micrometers (μm) or less.
5. A three-dimensional structure comprising the composition of any one of claims 1 to 3.
6. The three-dimensional structure of claim 5 further comprising a surface coating comprising a hydrophobic polymer.
7. The three-dimensional structure according to any one of claims 5 to 6, wherein the smoothened structure is subsequently coated in a hydrophobic polymer.
8. A structure comprising an interconnected vascular network, a matrix material in which the structural material is disposed; The structural material is capable of dissolving or decomposing in water; the structural material is formed from a powder that fuses into a solid when irradiated with an energy beam; The powder has a granular form, A structure having a specific energy of less than 6 mJ / mL.
9. The structure of claim 8 , wherein a surface of the structure has a smooth topography.
10. 10. The structure of claim 8 or 9, wherein the powder further comprises an anti-caking agent.
11. The structure of any one of claims 8 to 10, wherein the structural material comprises at least one carbohydrate.
12. 12. The structure of any one of claims 8 to 11, wherein the structural material comprises at least one of dextran or isomalt.
13. 13. The structure of any one of claims 8 to 12, wherein the structure material comprises at least one of photoresist, agarose, gelatin, carbohydrates, sucrose, glucose, fructose, lactose, isomalt, dextran, cellulose, methylcellulose, poly(lactic acid) or poly(ethylene glycol).
14. The structure of claim 10 , wherein the anti-caking agent comprises one or more materials that improve at least one of the flow, frictional properties, or particle packing of the structural material.
15. 15. The structure of claim 10 or 14, wherein the anti-caking agent comprises at least one of corn starch, silicon dioxide, or xanthan gum.
16. The structure of any one of claims 8 to 15, wherein the powder has a particle size of less than or equal to 250 μm in diameter.
17. The structure according to any one of claims 8 to 16, wherein a surface of the structure has a hydrophobic coating.
18. 18. The structure of claim 17, wherein the hydrophobic coating comprises at least one of polycaprolactone, poly(L-lactide), polylactic acid, poly(lactic-co-glycolic acid), collagen, gelatin, zein, shellac, starch, wax, or petrolatum.
19. 19. The structure of any one of claims 8 to 18, wherein the matrix material comprises at least one of polyamide, poly(2-hydroxyethyl methacrylate), poly(vinyl alcohol), polyacrylamide, poly(ethylene glycol), polyurethane, collagen, agarose, albumin, alginate, chitosan, starch, hyaluronic acid, gelatin, fibrin, matrigel, glycerin, glycol, mannitol, inositol, xylitol, adonitol, glycine, arginine, a biological polymer molecule, or a peptide amphiphile, or a monomer, dimer, or oligomer thereof.
20. The structure of any one of claims 8 to 19, wherein the powder is configured to be a free-flowing powder.
21. 1. A method for forming an interconnected vascular network, comprising: solidifying the powder by fusing with an energy beam to form a filament network, the filament network including a plurality of filaments that can dissolve or decompose in water and that define void spaces; backfilling the void spaces of the filament network with a matrix material; and removing the filaments to form the interconnected vascular network including fluid channels.
22. The method of claim 21 , wherein the matrix material is an aqueous solution containing a biomaterial.
23. 23. The method of claim 22, further comprising crosslinking the biomaterial to form a hydrogel in the void space.
24. the powder comprising a carbohydrate powder and an anti-caking agent; The powder has a granular form, has a specific energy of less than 6 mJ / mL; The method according to any one of claims 21 to 23, wherein the energy beam is a laser.
25. The method of any one of claims 21 to 24, further comprising surface smoothing the plurality of filaments with a smoothing solution.
26. 26. The method of claim 25, wherein the smoothing solution comprises at least one of isomalt, dextran, sucrose, glucose, lactose, trehalose, maple syrup, or cane syrup.
27. surface coating the plurality of filaments with a surface coating material; the surface coating material does not backfill the void space; The method of any one of claims 21 to 26, wherein the matrix material and the surface coating material are different materials.
28. 28. The method of claim 27, wherein the surface coating material comprises at least one of polycaprolactone, poly(L-lactide), polylactic acid, poly(lactic-co-glycolic acid), collagen, gelatin, zein, shellac, starch, wax, or petrolatum.
29. 29. The method of any one of claims 21 to 28, wherein the filament network is arranged in a three-dimensional branched pattern.
30. 30. The method of any one of claims 21 to 29, wherein the filament network is arranged in an interpenetrating geometry.
31. The method of any one of claims 21 to 30, wherein the filaments are configured into an unsupported geometry during shaping.
32. The method according to any one of claims 21 to 31, wherein the removal of the filaments is carried out by dissolution or decomposition.
33. 23. The method of claim 22, wherein the aqueous solution further comprises a suspension of living cells, and wherein removing the filament does not damage the living cells.
34. 23. The method of claim 22, wherein the matrix material is a crosslinked biomaterial, the biomaterial comprising at least one of polyamide, poly(2-hydroxyethyl methacrylate), poly(vinyl alcohol), polyacrylamide, poly(ethylene glycol), polyurethane, collagen, agarose, albumin, alginate, chitosan, starch, hyaluronic acid, gelatin, fibrin, matrigel, glycerin, glycol, mannitol, inositol, xylitol, adonitol, glycine, arginine, a biological polymer molecule, or a peptide amphiphile, or a monomer, dimer, or oligomer thereof.
35. The powder is configured to be a free-flowing powder; The method according to any one of claims 21 to 34, wherein the powder has a maximum particle size of 250 μm or less.
