We describe a sequential process for light-mediated formation and subsequent biochemical patterning of synthetic hydrogel matrices for three-dimensional cell culture applications. The construction and modification of hydrogels with cytocompatible photoclick chemistry is demonstrated. Additionally, facile techniques to quantify and observe patterns and determine cell viability within these hydrogels are presented.
Click chemistries have been investigated for use in numerous biomaterials applications, including drug delivery, tissue engineering, and cell culture. In particular, light-mediated click reactions, such as photoinitiated thiol−ene and thiol−yne reactions, afford spatiotemporal control over material properties and allow the design of systems with a high degree of user-directed property control. Fabrication and modification of hydrogel-based biomaterials using the precision afforded by light and the versatility offered by these thiol−X photoclick chemistries are of growing interest, particularly for the culture of cells within well-defined, biomimetic microenvironments. Here, we describe methods for the photoencapsulation of cells and subsequent photopatterning of biochemical cues within hydrogel matrices using versatile and modular building blocks polymerized by a thiol−ene photoclick reaction. Specifically, an approach is presented for constructing hydrogels from allyloxycarbonyl (Alloc)-functionalized peptide crosslinks and pendant peptide moieties and thiol-functionalized poly(ethylene glycol) (PEG) that rapidly polymerize in the presence of lithium acylphosphinate photoinitiator and cytocompatible doses of long wavelength ultraviolet (UV) light. Facile techniques to visualize photopatterning and quantify the concentration of peptides added are described. Additionally, methods are established for encapsulating cells, specifically human mesenchymal stem cells, and determining their viability and activity. While the formation and initial patterning of thiol-alloc hydrogels are shown here, these techniques broadly may be applied to a number of other light and radical-initiated material systems (e.g., thiol-norbornene, thiol-acrylate) to generate patterned substrates.
Click chemistries are increasingly used in the design of materials for numerous biomedical applications, including drug delivery, tissue engineering, and controlled cell culture, owing to their selective, efficient, and often cytocompatible reactivities.1-3 Photoclick chemistries that utilize light to trigger or initiate reactions (e.g., azide-alkyne,4 thiol−ene,5 and tetrazole-alkene6) are of particular interest for the formation or modification of biomaterials. Rapid rates under mild conditions and control of when and where they take place with light make these reactions well-suited for user-directed control of biomaterial properties in the presence of cells.7,8 In particular, thiol−ene photoclick chemistries have been used to generate hydrogel-based biomaterials with robust mechanical properties5,9 and for the encapsulation of a wide variety of cell types, including, but not limited to, human mesenchymal stem cells (hMSCs), fibroblasts, chondrocytes, and pancreatic cells, with promise for cell culture and delivery.10,11 Further, these chemistries have been used for the spatial patterning of biochemical cues to mimic key aspects of native cell microenvironments and facilitate appropriate cell-matrix interactions, including adhesion, differentiation, and invasion.3,12
For the construction of thiol−ene hydrogels with light, peptides containing cysteines (thiol) commonly are reacted with polymers functionalized with acrylates or norbornenes ('ene') for rapid, photoinitiated polymerization under cytocompatible conditions.13 Expanding this toolbox, we sought to establish methods for hydrogel formation with new versatile and accessible building blocks that required minimal synthetic processing or were commercially available toward their broad use as synthetic extracellular matrices.14 Specifically, peptides were modified with allyloxycarbonyl (Alloc)-protected lysines: one for pendant, integrin-binding groups to promote cell adhesion [K(alloc)GWGRGDS = Pep1Alloc] or two for non-degradable or cell-degradable crosslinks [K(alloc)RGKGRKGK(alloc)G or KK(alloc)GGPQGIWGQGK(alloc)K = Pep2Alloc, respectively]. With these sequences, conditions were established for rapid reaction (1-5 min) with four-arm thiol-modified poly(ethylene glycol) (PEG4SH) using cytocompatible doses of long wavelength UV light (10 mW/cm2 at 365 nm) and the photoinitiator lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP). The resulting hydrogels were stable under cell culture conditions for weeks. To enable cell-driven degradation and remodeling, an enzymatically-cleavable peptide was incorporated within the gel crosslinks (i.e., GPQGIWGQ),15 and a model primary cell, human mesenchymal stem cells (hMSCs), remained highly viable after encapsulation and during culture within these matrices. Further, peptides have been spatially patterned within these materials, and hMSCs remain viable and metabolically active under photopatterning conditions. Alternate pendant peptide sequences, not shown here (e.g., IKVAV, YIGSR, GFOGER, etc.), also may be incorporated within matrices to probe additional cell interactions with the surrounding microenvironment. These results are promising for the application of these hydrogel-based materials for three-dimensional (3D) cell culture and delivery to study and direct cell-matrix interactions for a variety of cell types.
Herein, methods to photoencapsulate cells and subsequently photopattern biochemical cues within the proposed hydrogel system are presented (Figure 1). Techniques to observe and quantify these photopatterns also are demonstrated: notably, i) the quantitative and qualitative use of Ellman's assay to determine the modification of free thiols within patterned substrates and ii) the complementary qualitative use of fluorescent peptides (AF488Pep1Alloc) to observe these patterns in three dimensions. Further, assays to determine viability (live/dead viability/cytotoxicity staining) and metabolic activity (MTS; 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) are presented so that users may determine the cytocompatibility of photoencapsulation and photopatterning conditions for different cell lines within hydrogel matrices. While the protocol is demonstrated for a facile light-based photoclick hydrogel system, the techniques may be applied to numerous other radically-initiated hydrogel systems for photoencapsulation and photopatterning in the presence of cells.
1. Preparation of Materials for Hydrogel Formation
2. Hydrogel Formation and Photopatterning
3. Visualizing and Quantification of Photopatterning
4. Cell Encapsulation in Hydrogels and Photopatterning
5. Determining the Viability and Metabolic Activity of Encapsulated Cells
The setup and procedure to photoencapsulate cells and subsequently photopattern gels containing encapsulated cells is depicted in Figure 1, and an example for preparing stock solutions to form a 10 wt% gel is provided in Table 1. Using Table 1, the amount of monomers (PEG4SH, Pep2Alloc, ±Pep1Alloc) and photoinitiator (LAP) required to polymerize hydrogels is calculated. Based on these calculations, stock solutions of PEG4SH, Pep1Alloc, Pep2Alloc, and LAP are prepared and mixed with and without Pep1Alloc for forming and photopatterning gels, respectively (Figure 1A). Subsequently, cells are collected from plates, counted, and centrifuged in appropriate quantities for encapsulation (Figure 1B). The cell pellet is re-suspended in gel-forming solution (peptide/polymer/LAP in PBS), and cell/monomer mixtures are transferred to glass or syringe molds. Cells are encapsulated within the hydrogel upon application of light (1-5 minutes of 10 mW/cm2 at 365 nm) (Figure 1C). For photopatterning (Figure 1D), gels are soaked with Pep1Alloc and LAP for 30 min to 1 hr and 30 min, allowing diffusion of peptide and initiator into the polymerized matrix. These peptide-laden gels are covered with photomasks with desired patterns and exposed to a second dose of light (1 min) to conjugate the peptides to free thiols within the matrix. Pendant tethers are only covalently linked to the gel in regions exposed to light, facilitating appropriate cell-matrix interactions and mimicking key mechanical and biochemical properties of the native cell microenvironment toward probing cell function and fate in vitro.
Ellman's assay provides a facile and inexpensive method to quantify gel modification and peptide incorporation within photopatterned substrates. Patterned and non-patterned gels (i.e., gels with or without peptide modification) are soaked in Ellman's reaction buffer post-polymerization. Next, cysteine standards and the gels soaked in buffer are placed in 48-well plates and incubated with Ellman's reagent (Figure 2A, left). After 1 hour 30 minutes, aliquots of samples are placed in individual wells of a 96-well plate, and the absorbance is recorded at 405 nm. A calibration curve (Figure 2A, right) from the cysteine standards is plotted (absorbance vs concentration, linear fit), and the quantity of free thiols in gels may be determined based on their dilution factor. These free thiol concentrations for various polymerization and photopatterning conditions of a 10 wt% gel are shown in Figure 2B. Gels polymerized for 1 or 5 min without Pep1Alloc (blue bars, I = 1 min and II = 5 min) have statistically similar free thiol concentrations, indicating that a rapid reaction occurs and gelation is complete within 1 min (two-tailed t-test, p >0.05). Thus, additional exposure to light (2-5 minutes) does not result in further conversion of functional groups. Gels polymerized for 1 min without Pep1Alloc were soaked with Pep1Alloc (3 mg/ml) and LAP (2.2 mM) for 30 and 90 min (green bars, II = 30 min and IV = 90 min) and exposed to a second dose of light for 1 min. The decrease in free thiols (60-80% with respect to the 1 min condition) indicates efficient modification of gels with pendant cues under these conditions. If higher modification is desired, increased concentrations of Pep1Alloc solution can be used as accessibility of Pep1Alloc to free thiols in more dilute peptide solutions may limit conversion; for example, we have found that concentrations up to 20 mg/ml Pep1Alloc produce >90% conversion of free thiols.
Uniquely, patterns of peptides added to the hydrogel may be rapidly imaged with Ellman's reagent (Figure 3A, under 5 min). However, visualization of the pattern is lost over time (greater than 5 min) owing to diffusion of the yellow TNB2- dianion through the gel. To improve imaging resolution and observe patterns in three dimensions (x-, y-, and z-planes) and in the presence of cells, fluorescent peptide addition (AF488Pep1Alloc) may be utilized. In Figure 3B x-, y- and z-projections of image stacks taken with a confocal microscope are shown, demonstrating pattern resolution (µm scale).
Approximately (94 ± 2) % and (94 ± 1) % of hMSCs encapsulated within degradable gels (Pep2Alloc = GPQG↓WGQ) remain viable (green cell bodies) 1 and 6 days after encapsulation, respectively, with few dead cells (red nuclei) observed (Figure 4A). Further, hMSC spreading is observed at 6 days post-encapsulation (Figure 4A, inset), indicating that cells can remodel and interact with these MMP-degradable matrices modified with integrin-binding RGDS. Metabolic activity assays performed on cells encapsulated in non-degradable gels (Pep2Alloc = RGKGRK) 1 and 3 days post-encapsulation (Figure 4B) provide a second measure of cell viability and demonstrate that cells remain active for the various photoencapsulation and photopatterning conditions tested with Ellman's assay (Figure 2B). In particular, there is no significant difference in metabolic activity between 30 min and 90 min incubations with Pep1Alloc+LAP (conditions III and IV) and the initial encapsulation (conditions I and II), indicating that the procedure is appropriate for applications of photopatterning in the presence of encapsulated cells (two-tailed t-test, p > 0.05).
Figure 1: Setup for encapsulating cells within hydrogels and subsequently photopatterning with biochemical cue. (A) Stock solutions of macromers and photoinitiator are prepared and mixed (PEG4SH = backbone, Pep2Alloc = crosslink, Pep1Alloc = pendant adhesive moiety (RGDS), LAP = photoinitiator). (B) Cells are collected for encapsulation. (C) Cells are mixed with macromer solutions without or with integrin-binding peptides (PEG4SH/Pep2Alloc/LAP or PEG4SH/Pep2Alloc/Pep1Alloc/LAP, respectively) and are encapsulated upon exposure to light. (D) Gels containing excess thiol groups during gel formation (here, PEG4SH/Pep2Alloc/LAP, formed with 2 mM excess thiol) may be patterned with biochemical cues by the subsequent addition of peptides functionalized with a single alloc group (Pep1Alloc, e.g., RGDS, IKVAV, etc.) to promote cell adhesion within specific regions of the gel. Here, patterning of gels with fluorescently-labeled RGDS is shown. Please click here to view a larger version of this figure.
Figure 2: Quantitative Ellman's assay setup and results to evaluate modification of patterned hydrogels. (A) Gels and cysteine standards are incubated in 48-well plates with Ellman's reagent. A linear standard curve is plotted to determine cysteine concentration in gel samples. (B) Excess free thiols are incorporated within hydrogels during gel formation and consumed upon patterning with a pendant alloc-peptide (I = 1 min polymerization; II = 5 min polymerization; III = 30 min incubation with Pep1Alloc; IV = 90 min incubation with Pep1Alloc; both III and IV polymerized and patterned with light for 1 min). The data shown illustrate the mean (n = 3) with error bars showing the standard error. Please click here to view a larger version of this figure.
Figure 3: Ellman's Assay and fluorescent images to visualize photopatterned hydrogels. (A) Ellman's reagent may be used to rapidly detect patterns (lines) in the x- and y-planes (yellow = unpatterned region, Scale bar = 1 mm). (B) Fluorescent peptides may be used to observe patterns in the x-, y-, and z-planes (green = patterned region; 200 µm scale bar; 10X water-dipping objective; Ex/Em 488/525 nm). Please click here to view a larger version of this figure.
Figure 4: Viability and metabolic activity of cells within non-degradable photopatterned hydrogels. (A) Example confocal z-stack (z-projection; 10X water-dipping objective) of viable (green; Ex/Em 488/525 nm) and dead hMSCs (red; Ex/Em 543/580 nm) encapsulated within hydrogels 24 hr post-encapsulation (Scale bar = 200 µm). Cells spread within these hydrogels 6 days after encapsulation (inset image, Scale bar = 50 µm). (B) Cells are metabolically active 1 and 3 days (dark and light bars, respectively) post-encapsulation for the various polymerization (I = 1 min polymerization; II = 5 min polymerization) and patterning conditions (III = 30 min incubation with Pep1Alloc; IV = 90 min incubation with Pep1Alloc; both polymerized and patterned with light for 1 min). The data shown illustrate the mean (n = 3) with error bars showing the standard error. Please click here to view a larger version of this figure.
Table 1: Example setup to calculate volumes of stock solutions for making hydrogels. Cells within the spreadsheet that are highlighted blue indicate user-defined parameters; the other quantities are calculated based on these settings. The final volumes for each stock solution to make a gel are highlighted orange. The white cells contain formulas used to calculate the final volumes based on the user-defined parameters. Note that a concentration of 2.2 mM LAP is equivalent to 0.067 wt%.
The procedure presented here demonstrates techniques to photoencapsulate cells within hydrogels formed by thiol−ene click chemistry and subsequently photopattern the gels with biochemical cues. The use of light to initially form hydrogels allows homogeneous mixing and suspension of the cells within the polymer solution prior to polymerization. Rapid polymerization 'locks' the gel in the shape of the defined mold and encapsulates suspended cells within the hydrogel network. Gels also may be molded into numerous different shapes (e.g., glass slides or syringe tips) depending on the final application desired. For example, cells encapsulated for 3D culture in hydrogels attached to glass slides are particularly useful for imaging applications as light attenuation is limited within a thin sample. Syringe molds can be used for rapid encapsulation of cells, allowing a larger number of samples to be prepared in a short time (compared to glass slides) that can be used for experiments requiring large numbers of cells such as flow cytometry or qPCR. Subsequently, these gels then may be patterned with biochemical cues to elicit desired cellular responses such as differentiation or invasion.30,31
Assays for viability and metabolic activity indicate the survival of cells for the material system and patterning conditions presented. Note that metabolic activity was monitored until day 3 in non-degradable gels (RGKGRK peptide crosslink) to assess the initial effects of polymerization and patterning conditions on cell function. Additionally, a membrane integrity assay (live/dead viability/cytotoxicity staining) of hMSCs at 1 and 6 days post-encapsulation in gels formulated with a degradable peptide crosslink (GPQGIWGQ) supports that cells remain viable and spread at 1 week in culture. The viability of additional cell lines has been reported for photopatterning conditions32 similar to those used here and can be evaluated for the described hydrogel system using live/dead staining and metabolic activity assays. While we have not observed issues with cell viability using this materials system and related procedures (4.2 and 4.3), some cell types may be sensitive to free-radical and/or light exposure. In this case, users may consider using non-photoinitiated materials systems, such as azide-alkyne,12 FXIII,31 or Diels Alder-based hydrogel formation chemistries.33
Facile techniques to detect and quantify patterning of biochemical cues within gels also are presented (3.1 and 3.2). Ellman's assay is of particular interest because the reagents are commercially available and no extra synthetic processing steps or more expensive reagents (e.g., fluorescently-labeled peptide) are required. Ellman's assay can be used to precisely determine the modification of free thiols with biochemical cues under different photopatterning conditions, as well as to rapidly visualize patterns. For quantifying peptide incorporation, the thiol functional group concentration before and after patterning, as a quantitative measure of peptide incorporation, is directly assessed with Ellman's assay. While this type of quantification can be done with fluorescently-labeled peptides,34 imaging-based quantification requires more time-consuming handling and analysis steps (e.g., synthesis of a fluorescently-labeled peptide and generation of a calibration curve to relate fluorescence to peptide concentration using image analysis). For imaging peptide incorporation, Ellman's reagent can be directly applied to samples and immediately visualized. While limited to the x- and y-planes for pattern visualization, the technique can be used as a simple, routine method to determine if matrices containing free thiol groups have been patterned. It is important to note that Ellman's assay is not considered cytocompatible, so while it may be used to observe and quantify photopatterns, it cannot be done in the presence of cells. For imaging patterning in three dimensions and in the presence of cells, the conjugation of fluorescent peptides within hydrogel matrices remains a powerful and widely-used approach. Resolution of patterns can be evaluated in the x-, y-, and z-planes using confocal microscopy, and this method is cytocompatible so that cells within patterned regions or non-patterned regions can be identified. Taken together, Ellman's assay and imaging-based techniques are complementary tools for researchers to assess both quantitatively and qualitatively the photopatterning of biochemical cues within the materials system.
Photoclick, or more broadly photoinitiated, chemistries for the formation and modification of hydrogels in the presence of cells are numerous. The molding, encapsulation, and patterning techniques presented here are not limited to the described material system and may be applied to alternative light-based chemistries, such as thiol-alkyne,35 azide-alkyne,4 and other thiol−ene chemistries (e.g., thiol-norbornene),10,13 as well as with different photoinitiators, such as Irgacure 2959, Eosin Y, and camphorquinone. Note, users may need to adjust procedure parameters (e.g., incubation times, polymerization times, cell density) to ensure that conditions remain cytocompatible for these other systems. Since the patterning process requires diffusion of the alloc-modified peptide(s) into the hydrogel (2.2.3-2.2.5), this process may prove most useful for the addition of integrin-binding moieties (e.g., peptides or extracellular matrix protein fragments) to the hydrogel, where attachment of the ligand to the network is required for the generation of traction forces by the cell and full integrin activation.36 Note, for biomolecules that may be similarly active in solution or upon immobilization (e.g., growth factors or cytokines), the incubation step for moiety diffusion (~1 hr) into the hydrogel could lead to signaling events that convolute patterning results. Other methods have been established for growth factor immobilization or local sequestration for their patterning.37-39Additionally, pattern resolution is dictated by the control over light exposure. Here, photomasks allow creation of patterns through the gel depth and in the x- and y-planes; however, greater spatial control over patterning of biochemical cues within gels can be achieved with alternative methods of irradiation such as the use of a two-photon confocal microscope to generate patterns in the x-, y-, and z- planes.34,40 Finally, it is important to note that while the material system utilized within this procedure is only initially modified with biochemical cues, orthogonal photoclick chemistries could be used to allow alterations in matrix properties over time.12 The procedure and techniques presented here add diversity to the current approaches for creating synthetic matrices with well-defined and spatiotemporally-controlled properties. In particular, the commercial availability of reagents and materials used within this procedure will be useful to a wide range of researchers interested in the use of hydrogel-based biomaterials for applications in controlled cell culture.
The authors have nothing to disclose.
This work was supported by the Delaware COBRE programs in Drug Discovery and in Advanced Biomaterials funded by Institutional Development Awards from the National Institute of Generals Medical Sciences at the National Institutes of Health (P20GM104316 and P30 GM110758-01, respectively), the Pew Charitable Trusts (00026178), a National Science Foundation Career Award (DMR-1253906), the Burroughs Wellcome Fund (1006787), and the National Science Foundation IGERT SBE2 program at the University of Delaware (fellowship to L. Sawicki). The authors thank the Delaware Biotechnology Institute BioImaging Center at the University of Delaware for training and access to confocal microscopy, Ms. Katherine Wiley for assistance during the video shoot, Mr. Matthew Rehmann for generously providing hMSCs isolated from bone marrow, Prof. Christopher J. Kloxin and Mr. Stephen Ma for generously providing photomasks, and Prof. Wilfred Chen for the use of the automated plate reader.
Custom Peptides | Various Vendors | —- | Peptides may also be synthesized via standard SPPS techniques with materials from vendors including ChemImpex and ChemPep. |
4arm PEG Thiol, MW 20k | JenKem USA | 4ARM-SH | Listed under Multi-arm Homofunctional PEGs. PEG4SH may also be synthesized as previously referenced. |
Lithium Phenyl-2,4,6-trimethylbenzoylphosphinate | Colorado Photopolymer Solutions | Li-TPO | |
Dimethyl Phenylphosphonite | Sigma Aldrich | 149470 | Caution, causes severe skin burns and eye damage. Wear protective gloves, clothing, and eye protection. |
2.4.6-Trimethylbenzoyl Chloride | Sigma Aldrich | 682519 | Caution, causes severe skin burns and eye damage. Wear protective gloves, clothing, and eye protection. |
Lithium Bromide | Sigma Aldrich | 213225 | |
2-Butanone | Sigma Aldrich | 360473 | |
Flask, Round Bottom, 100 mL | Chemglass | CG-1506-05 | |
80 mL Filter Funnel, Buchner, Medium Frit | Chemglass | CG-1402-L-02 | Filter paper inside a regular glass funnel may be used if desired. |
Magnetic Stir Bars | Various Vendors | —- | |
Magnetic Stirring and Hot Plate | Various Vendors | —- | |
Vacuum Dessicator | Various Vendors | —- | |
Deuterium Oxide | Acros Organics | 16630 | |
Dulbecco's Phosphate Buffered Saline | ThermoFisher Scientific | 14190-250 | |
Penicillin Streptomycin | ThermoFisher Scientific | 15070-063 | |
Fungizone | ThermoFisher Scientific | 15290-018 | |
Trypsin-EDTA (0.5%), no phenol red | ThermoFisher Scientific | 15400-054 | Select appropriate medium and trypsin depending on cell type. |
Microcentrifuge tubes | Various Vendors | —- | 1.5 or 2 mL sizes, sterile. |
BD Syringe with Slip (Luer) Tips (Without Needle) | Fisher Scientific | 14-823-16H | Product number listed here is for a 1 mL syringe (16H), Various sizes are available (14-823-XX). |
Fisherfinest Premium Plain Glass Microscope Slides | Fisher Scientific | 12-544-1 | |
High-Purity Silicone Rubber, 0.010" Thick, 6" x 8" Sheet, 55A Durometer (Gasket) | McMaster Carr | 87315K62 | |
Ethanol, 200 Proof | Decon Labs | 2701 | |
RainX, Original | Amazon | 800002243 | May be purchased from other vendors. |
Sigmacote | Sigma Aldrich | SL2 | |
Photomasks | Advance Reproductions Corporation | —- | Photomask Division, different designs may be printed as desired. |
DTNB; Ellman's Reagent; 5,5-dithio-bis(2-nitrobenzoic acid) | ThermoFisher Scientific | PI-22582 | |
Sodium Phosphate Dibasic | Sigma Aldrich | S5136 | |
Ethylenediaminetetraacetic acid | Sigma Aldrich | E6758 | |
Sodium Hydroxide, Pellets/Certified ACS | Fisher Scientific | S318 | |
Orthophosphoric Acid | Alfa Aesar | 33266 | |
Cysteine Hydrochloride Monohydrate | Sigma Aldrich | C7880 | |
LIVE/DEAD Viability/Cytotoxicity Kit, for mammalian cells | ThermoFisher Scientific | L-3224 | |
CellTiter 96 Aqueous One Solution Assay | Promega | G3582 | |
Centrifuge | Various Vendors | —- | Capable of speeds at 90-110 x g. |
Multiwell Plate Reader | Various Vendors | —- | Capable of reading absorbance at 405 nm in a 96-well plate. |
Epifluorescent or Confocal Microscope | Various Vendors | —- | To visualize peptide patterns and cells within hydrogels. |
Omnicure Exfo Series 2000 | Excelitas Technologies | —- | Alternate light systems may be used to polymerize hydrogels. |
Zeiss Zen Lite Software | Zeiss | —- | Available at zeiss.com; compatible with images taken on Zeiss microscopes |
ImageJ | NIH | —- | Available at imagej.nih.gov; applicable for general image analysis |