Digital Light Processing-Based 3D Printing of Cell-Seeding Hydrogel

Subscriber access provided by GUILFORD COLLEGE

Manufacturing, Technology, and Devices

Digital light processing-based 3D printing of cellseeding hydrogel scaffolds with regionally varied stiffness Dai Xue, Jiaxin Zhang, Yancheng Wang, and Deqing Mei ACS Biomater. Sci. Eng., Just Accepted Manuscript • DOI: 10.1021/ acsbiomaterials.9b00696 • Publication Date (Web): 09 Jul 2019 Downloaded from pubs.acs.org on July 18, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Biomaterials Science & Engineering

Digital light processing-based 3D printing of cell-seeding hydrogel scaffolds with regionally varied stiffness Dai Xue2, Jiaxin Zhang3, Yancheng Wang1,2*, Deqing Mei1,2 1State

Key Laboratory of Fluid Power and Mechatronic Systems, School of Mechanical Engineering, Zhejiang University, Hangzhou, 310027, China 2Key Laboratory of Advanced Manufacturing Technology of Zhejiang Province, School of Mechanical Engineering, Zhejiang University, Hangzhou, 310027, China 3Department of Toxicology, Fourth Military Medical University, Xi’an, 710032, China *Email: [email protected]

Abstract Cell-seeding heterogeneous scaffolds with regionally varied stiffness play an important role in tissue engineering, e.g., bone and cartilage regeneration, that require the recapitulation of geometric complexity through biocompatible material to mimic the natural cell microenvironment in vivo. Here, we report the digital light processing (DLP)-based 3D printing of cell-seeding hydrogel scaffold with regionally varied stiffness by tuning the exposure time without changing the geometric architecture. Mechanical tests on printed poly (ethylene glycol) diacrylate (PEGDA) hydrogels homogeneous scaffold revealed that a 60 % increase of the elastic modulus can be achieved by setting the optimal exposure time. Furthermore, regulating the stiffness by varying the exposure time was demonstrated in the printed three-sectional heterogeneous scaffolds. Uniaxial compression tests showed that no fracture was observed even when the compression strain reached up to 25%, indicating that by adjusting the exposure time the undesired influence of the scaffold on mechanical integrity could be avoided. Then, 3T3 fibroblasts were seeded onto the scaffold, and the biocompatibility together with the physical support of the scaffolds were confirmed by observation and cell population assessment.

Keywords: 3D printing; Digital light processing; Cell seeding; Hydrogel scaffold; Regionally varied stiffness; Exposure time

ACS Paragon Plus Environment

ACS Biomaterials Science & Engineering 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 22

1. Introduction Interest in the fabrication of three-dimensional (3D) cell-seeding hydrogel scaffolds for biomedical applications, with user-defined microstructures and mechanical properties, is increasing.1-3 Such scaffolds can provide spatial support in vitro to rebuild cell microenvironments and fulfill their biological functionalities, such as muscle, bone, and cartilage.4,5 In order to emulate the natural properties that are required by the target tissue and/or organs, these scaffolds generally should be biocompatible and have the proper mechanical stiffness. A basic requirement for engineered scaffolds to realize their biological functionality is the proper geometrical architecture, such as pore size and porosity, to facilitate cell attachment, substance exchange, and the interactions between cells and extracellular matrix (ECM).6,7 Besides, cell microenvironments are usually heterogeneous, with the pore size and mechanical properties varying spatially.8,9 Therefore, by carefully reconstructing cell-seeding hydrogel scaffolds in vitro, these could help to improve tissue engineering, such as repairing bone or regenerating cartilage. Several approaches have been proposed to construct complex cell-seeding scaffolds, including salt leaching, freeze drying, and electro-spinning.10,11 Despite their high porosity and interconnectivity, these fabricated scaffolds turned out to be spatially random. The lack of user-defined flexibility and usability of biomaterials limits the application of these methods for the construction of intricate cell-seeding scaffold. 3D printing enables the construction of user-defined microstructures using biomaterials in tissue engineering, which offers considerable flexibility and biocompatibility. For instance, extrusion-based printing has been applied to print multi-layer woodpile scaffolds, which have been commonly used in bone and vasculature engineering.12-14 However, the printing resolution of extrusion-based printing is about 100 μm, and so cannot meet the precision of 10 μm or even less required for cell scaffolds. Selective laser sintering (SLS) offers superior resolution for scaffold fabrication, and possesses the ability to print metal-based materials for high strength and non-biodegradable applications.15,16 In addition, it should be noted that in the above-mentioned printing process, the features are created drop by drop or line by line and therefore the time needed to print more complex constructs at large sizes will increase severely. Digital light processing (DLP)-based 3D printing allows printing layer by layer, making high efficiency possible in printing complex user-defined microstructures. During printing, the light is


Page 3 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


modulated using a liquid crystal display (LCD) or a digital micro-mirror device (DMD), and projected onto a photo-sensitive biocompatible pre-polymer. In-plane polymerization occurs in the projection area and coordinated by movement along the z-axis, complex constructs can be printed.17 The printing speed varies from 25 to 1000 mm/min, and depends mainly on the light power and material sensitivity.18 Superior resolution (like 1 μm or sub-micro levels) has been reported by focusing the projected light through an optical system.19,20 Thus, due to the excellent printing speed and resolution, DLP-based 3D printing process offers much flexibility and versatility for constructing complex cell scaffolds with the required properties in tissue engineering.21-23 In order to match the geometric and mechanical properties of natural tissues, the designed cell-seeding hydrogel scaffold needs to be heterogeneous and printed with regionally varied stiffness. The common method to modulate the stiffness in DLP-based 3D printing is pre-polymer modification, such as adding metal particle or cellulose nanocrystal (CNC).24,25 However, changes in the rheology and transparency of the pre-polymer will affect the mechanical integrity of the printed scaffold, especially at the interfaces of converting layers. Though hierarchical design, e.g. hierarchical pore size and porosity, is validated as a direct approach to achieve regionally varied stiffness in 3D printing, but the consequent loss of geometric fidelity may affect the functionality of the printed scaffold adversely.26,27 In addition, the ability to regulate stiffness has been demonstrated for printing through gray-scale images, but the method is not suitable for manufacturing complex scaffolds.28,29 Therefore, a method to print 3D cell-seeding hydrogel scaffolds with stiffness regionally varied stiffness is still urgently required and was the goal of this study. In this work, we propose an approach to print an octahedral type of cell-seeding hydrogel scaffold with regionally varied stiffness through a single pre-polymer material and identical geometry. This approach was inspired by studying the effect of exposure time on the printing size in DLP-based 3D printing, where a size plateau was found between under-curing and over-curing. During the size plateau, we hypothesized that longer exposure times would increase the cross-linking density of the pre-polymer and in turn would consequently change the stiffness of the printed construct. Therefore, the printing process was studied first to find the suitable range of exposure times at the size plateau, and three sets of exposure times were selected to print scaffolds. A pre-polymer

containing

poly(ethylene

glycol)

diacrylate

(PEGDA)

and

lithium

phenyl-2,4,6-trimethylbenzoyiposphinate (LAP) was adopted for printing, as it has been shown to



Page 4 of 22

have favorable biocompatibility.30,31 By combining two and three sets of exposure times in one printing, a heterogeneous scaffold with regionally varied stiffness could be made. Then, the mechanical properties of the printed homogeneous and heterogeneous scaffolds were evaluated by measuring elastic moduli and cyclic compression. 3T3 fibroblasts were seeded and cell proliferation and population were assessed to evaluate the potential application of this approach and of printed cell-seeding hydrogel scaffolds in tissue engineering.

2. Materials and Methods 2.1 Material Preparation PEGDA (number average molar mass Mn = 700) and poly-D-lysine were purchased from Sigma-Aldrich (USA). The matrigel was purchased from BD Biosciences (USA). Lithium phenyl-2,4,6-trimethylbenzoyiposphinate (LAP) was used as photoinitiator and synthesized following previous work.32 2,4,6-Trimethylbenzoyl chloride (Sigma) was added drop-wise into continuously stirred dimethyl phenylphosphonite (Sigma) at room temperature, and stirred for 18 hours. Then the solution was heated to 50 °C and an excess of lithium bromide dissolved in 2-butanone was added. The solution was cooled to room temperature when the solid precipitate was formed, and held overnight. Then, the solid precipitate, LAP, was filtered and washed three times before vacuum drying. Finally, the PEGDA and LAP were added into phosphate-buffered saline (PBS) at concentrations of 20% (v/v) and 1% (w/v), respectively. Yellow food-grade dye (0.75% (v/v), Wilton, USA) was added in the pro-polymer as light absorber to mitigate light dispersion and optimize printing resolution.

2.2 DLP-based 3D Printing System Setup Figure 1 shows a schematic view and the printing procedure of the lab-customized DLP-based 3D printing system used to construct the octahedral type of scaffold. The system consists of a DMD module (Texas Instruments, USA), a 405 nm light source (Luminus Devices, USA), optic lenses (Thorlabs, USA), and a customized printing platform mounted on a commercial motorized linear stage. The pre-polymer vessel is a petri dish with a polydimethylsiloxane (PDMS)-covered bottom.33 All the 3D digital models were created by SolidWorks; open-source software, Creation Workshop,


Page 5 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


was applied to control the printing process. The DMD contains millions of micro-mirrors, and each mirror can flip into two angles (ON and OFF state, respectively) according to the pixel color on the mask. Then the incident 405 nm light can be selectively reflected by the “ON”-state mirrors on the DMD, generating a light pattern that is geometrically similar to the working mask. To print a scaffold, the 3D model was first imported into the software, and sliced into a sequence of masks with a user-defined layer thickness; here, layer thickness was set to 50 μm to achieve optimal resolution and speed simultaneously. Then the exposure time for each layer was set. After this time, the pre-polymer will be solidified and the pattern of a single layer will be printed. Combined with lifting along the z-axis, the single-layer patterns would be stacked into a 3D construct.

Figure 1. Schematic view of the DLP-based 3D printing system, and the procedure for printing a user-defined construct.

2.3 Single-layer printing To investigate the effect of exposure time, printing thickness and size of each layer on the printing process, to begin, single-layer printing was implemented. Two conditions were considered for single-layer printing: (1) Single-layer printing without thickness constraint; here, the bottom of vessel was bonded with a 2 mm high PDMS fence, and a cover slide was attached inside the fence, and immersed in the pre-polymer. And the range of exposure time was set for 2.5 s to 3.7 s. (2) Single-layer printing at defined thickness; here, a drop of pre-polymer was placed in the vessel and between two PDMS spacers to set a layer thickness of 50 μm. Then a cover slide was placed on the



Page 6 of 22

pre-polymer drop and lowered till it touched both spacers. A wider range of exposure time that varies from 2.1s to 4.7 s was adopted for printing, and the exact range of exposure time will be concluded from the printing results where no under- or over-curing. A model consisting of a set of dots with designed diameter of 1 mm and with different heights was printed to investigate the relationship between exposure time and printing size under these two conditions. After printing, the cover slide together with the attached dots was removed from the vessel and gently flushed with PBS three times. Then, the dots were measured with a laser confocal microscope (Olympus, Japan); thus, the size, including height and diameter, could be determined. By comparing the measured and designed dot size, a reasonable range of exposure times can be estimated for the size plateau.

2.4 Samples and Scaffold Printing Based on the reasonable range determined from single-layer printing, three sets of exposure time (3.1, 3.5, and 3.9 s) were used to print the samples and octahedral type scaffolds. First, three groups of pillars with dimensions of 8 mm × 8 mm were printed. We hypothesized that the stiffness of the constructs can be adjusted by printing with different exposure times due to the variation in cross-linking density. Then, a 3D scaffold model, with a size of 7 mm × 7 mm × 12 mm, was designed and imported into the system for printing. By adopting appropriate exposure times, three groups of homogeneous scaffolds were printed. Two groups of heterogeneous scaffolds with regionally varied stiffness were printed by selecting two sets and three sets of exposure times for one printing, respectively. To assess cell population, three groups of hydrogel slabs (12 mm × 12 mm × 1 mm) used for cell adhesion were printed using three sets of exposure times. After printing, both the slab and the scaffold were flushed three times with sterile PBS and immersed in PBS (changed every 4 hours) until the dye was fully diffused out. Finally, the printed constructs were maintained in PBS for further use.

2.5 Mechanical Testing A measurement system consisting of a motion stage (Newport, USA) and a force transducer (Interface, USA) was set up to apply uniaxial compression to the printed PEGDA. The compression speed was set at 1 mm/min to obtain the stress–strain curves. The pillars were compressed until fractured and 40% compression strain was exerted on the printed scaffolds. The elastic moduli of the


Page 7 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


hydrogels were calculated from the slope between 2% and 12% strain in the pillar compression tests. Tests using five compression cycles were continuously conducted on both kinds of scaffolds to evaluate the structural integrity.

2.6 Statistical Analysis All experimental tests were repeated three times and data are presented as means ± standard error of mean (SEM) of three independent experiments. Statistical significance was tested using one-way or two-way ANOVA. Statistical test was performed using Origin 9.0. P values of less than 0.05 were considered statistically significant. Since the relation of cell population versus time has been recognized as significant, so we performed statistical analysis on the relative cell population at each culture time.

2.7 Cell Seeding and Culturing. We used green fluorescent protein (GFP)-transfected 3T3 fibroblasts (Shanghai Institutes for Biological Sciences, China) to evaluate the biocompatibility of the scaffolds in supporting cell attachment and proliferation. The cells were firstly cultured in RPMI-1640 medium with a supplement of 10% fetal bovine serum (Hyclone, USA) and 100 units/mL penicillin–streptomycin, and then maintained at 37 °C under 5% CO2, and passaged according to protocol. Before seeding, the slabs and scaffolds were decontaminated by washing in 70% (v/v) ethanol for 12 hours and rinsed three times with sterile PBS. To achieve better cell adhesion, the scaffolds were coated with poly-D-lysine for 12 hours and matrigel for 1 hour in an incubator shaker.34 Then a cell suspension was prepared at a concentration of 1 × 105 cells/mL. Both slabs and scaffolds were removed from the well plate before cell seeding. The population assessment was carried out by Cell Counting Kit (CCK)-8 on the slab. To prepare the assessing samples, 50 μL of cell suspension were placed on the slab for all three groups and after 20 min, and the culture medium was perfused gently into the well along the wall until it fully covered the slab. For the scaffolds seeding, a three-step procedure was implemented to achieve a uniform and denser cell distribution. First, 150 μL of cell suspension were dropped from the top surface onto the scaffolds and held for 20 min. Then the scaffold was laid down and another 150 μL of suspension were dropped onto two adjacent lateral surfaces with a time



Page 8 of 22

interval of 20 min. After 20 min, culture medium was slowly perfused into the well until the scaffold was covered. Slabs and scaffolds were statically cultured in an incubator and the medium was changed every two days. The cell morphology and distribution on the scaffolds were acquired by fluorescent microscope (Olympus, Japan).

3. Results and Discussion 3.1 Characterization of the Printing Process When the pre-polymer is illuminated with 405 nm light, free radicals are generated from the initiator, and the PEGDA monomers are cross-linked into larger molecules by reacting with the free radicals. Thus a layer of liquid pre-polymer is converted into the solid state according to the pattern of light. The geometry of this printed layer is determined not only by the design mask, but is also affected by the generation rate and amount of free radical. For printing at a given layer thickness, pre-polymer and certain power density, almost the only parameter that matters is the exposure time. Insufficient exposure time will induce an incomplete single-layer structure and weak integrity of the multi-layer construct; this kind of defect can be attributed to under-curing. On the other hand, redundant exposure times will generate oversizing and merging of the printed construct, and even cause the layer to stick on the vessel bottom. Thus, over-curing will occur. Single-layer printing was conducted to study the effect of exposure time on printing thickness and diameter (bottom). We expected to find a reasonable range of exposure times for printing constructs without under- or over-curing. The demand of printing thickness should be firstly met to ensure the integrity between contiguous layers, and therefore single-layer printing without thickness constraint was employed. Figure 2a shows how printing thickness and diameter vary with exposure time. The result indicates that the printing thickness is not linearly proportional to exposure time, and that the slope decreases with longer exposure times, which is in good agreement with previous research.35,36 For a given layer thickness of 50 μm, an exposure time of 2.5 s is sufficient to acquire a thorough penetration for single-layer printing. However, the printing size in the xy plane is much smaller during the early stage, and the diameter is only 735 μm. Even though an exposure time of 3.7 s was applied, the dot still failed to reach the designed diameter (1 mm). As the printing thickness increases, the upper diameter narrows dramatically, which is caused by light attenuation deeper in


Page 9 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the pre-polymer. This result provides us with insight into the exposure time that must be used to print through the settled layer.

Figure 2. (a) Variation of printing thickness and diameter with exposure time in single-layer printing without thickness constraint; (b) three dots printed under increasing exposure time show under-curing, proper curing and over-curing; (c) detailed relationship between printed diameter and exposure time; (d) two conditions of single-layer printing: without thickness constraint (upper) and at a defined thickness (lower). All error bars represent SEM, and n=15 for all data points.

The process of photo-polymerization can be concluded to be the result of several reactions and diffusions, such as the reaction between pre-polymer and initiator, light propagation, free radical diffusion and chain elongation.37 For single-layer printing without thickness constraint, vertical diffusion will intensify in size extending along the z-axis for the printed construct. In contrast, the size extending along z-axis is physically limited for single-layer printing at a defined thickness, and mainly free radial diffusion together with chain elongation contribute to the increase in size in the projection plane. Based on the obtained basic exposure time, the effects of exposure time on diameter at a defined thickness were investigated. Figure 2b shows three groups of dots (boundary is marked by a dashed line) printed under exposure times of 2.5, 3.5 and 4.5 s, respectively. For the 2.5 s group, the printed diameter of the dot is 930 μm, and is 100 μm larger than the result without



Page 10 of 22

thickness constraint, shown in Figure 2a. For the 3.5 s set, the dot reaches the designed diameter, and the deviation is lower than 1%. If the exposure time keeps increasing, the diameter will exceed the designed value when the exposure time becomes 4.5 s. The detailed results for diameter versus exposure time are shown in Figure 2c. For the repeating test, five pillars were printed under each exposure time and totally 15 points of height and diameter were measured in every pillar. As expected, three periods can be observed in Figure 2c, and we defined those according to the deviation from the designed value as under-curing (smaller by 1%), proper curing, and over-curing (larger by 1%). Thus, the reasonable range of exposure times is from 3.1 s to 3.9 s. In the under-curing period, there is a greater increase in diameter with increasing exposure time. Probably, the reaction and diffusion attributable to photo-polymerization are occurring, and this trend is similar to the printing thickness results by exposure time. Then the diameter plateau is encountered, and this happens when the diameter approaches the designed value. Since the light beam is parallel, the free radicals are mainly generated in the illuminated area and then spread around. When dispersion occurs, the power density at the beam boundary is usually lower, as shown in Figure 2d (top). The distribution of light in the polymer has a Gaussian profile.37 Therefore, the under-curing can be explained as a result of insufficient energy when the whole dot is exposed for an identical time but the light is attenuated at the projection boundary. Under the condition of defined thickness, part of the dispersed light is reflected by the cover slide and the printed construct, and this leads to higher power density at the projection boundary (Figure 2d, bottom). This means that over-curing as a printing defect needs to be considered in multi-layer printing. During the properly cured periods, the diameter reaches a plateau but we assume that the in situ PEGDA crosslinking is still in progress and further increases the mechanical properties of the printed PEGDA hydrogel.

3.2 Mechanical Properties of the Printed PEGDA Hydrogel To validate our hypothesis that during the proper printing period, the hydrogel properties still vary with increases of the exposure time, mechanical tests were performed on the printed PEGDA pillars and the mechanical properties were evaluated by measuring the elastic modulus, ultimate strain, and stress. Firstly, we selected three sets of exposure times (3.1 s, 3.5 s and 3.9 s) to print three groups of pillars. Figure 3a shows the printed pillars with the same geometries. Then these pillars were compressed until fractured, and the strain–stress curves are shown in Figure 3b. As


Page 11 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


expected, the ultimate stress is slightly increased at longer exposure time, since a longer exposure time improves the cross-linking density in the printed PEGDA hydrogel. Besides, there is distinct increase in ultimate strain with decreasing exposure time. It is suggested that the denser cross-linked PEGDA printed under longer exposure times likely contributes to the higher stiffness of the printed construct. The calculated elastic moduli are 2950, 3140 and 3410 kPa (Figure 3c). Thus, the elastic modulus can be regulated by about 10% by changing the exposure time during the proper curing period. This result highlights the potential of using exposure times in a reasonable range to print scaffolds, as the stiffness can thus be regulated without altering the materials and geometry.

Figure 3. (a) Three printed groups of PEGDA pillars; (b) stress–strain curves; and (c) calculated elastic moduli under exposure times of 3.1, 3.5 and 3.9 s. *p

Digital Light Processing-Based 3D Printing of Cell-Seeding Hydrogel

Recommend Documents