Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Poly (ethylene glycol) hydrogel elasticity influences human mesenchymal stem cell behavior

Poly (ethylene glycol) hydrogel elasticity influences human mesenchymal stem cell behavior Coordinated investigations into the interactions between biologically mimicking (biomimetic) ma- terial constructs and stem cells advance the potential for the regeneration and possible direct replacement of diseased cells and tissues. Any clinically relevant therapies will require the develop- ment and optimization of methods that mass produce fully functional cells and tissues. Despite advances in the design and synthesis of biomaterial scaffolds, one of the biggest obstacles facing tissue engineering is understanding how specific extracellular cues produced by biomaterial scaf- folds influence the proliferation and differentiation of various cell sources. Matrix elasticity is one such tailorable property of synthetic scaffolds that is known to differ between tissues. Here, we in- vestigate the interactions between an elastically tailorable polyethylene glycol (PEG)-based hydro- gel platform and human bone marrow-derived mesenchymal stem cells (hMSCs). For these studies, two different hydrogel compositions with elastic moduli in the ranges of 50–60 kPa and 8– 10 kPa were implemented. Our findings demonstrate that the different elasticities in this platform can produce changes in hMSC morphology and proliferation, indicating that the platform can be implemented to produce changes in hMSC behavior and cell state for a broad range of tissue engineering and regenerative applications. Furthermore, we show that the platform’s different elasticities influence stem cell differentiation potential, particularly when promoting stem cell dif- ferentiation toward cell types from tissues with stiffer elasticity. These findings add to the evolving and expanding library of information on stem cell–biomaterial interactions and opens the door for continued exploration into PEG-based hydrogel scaffolds for tissue engineering and regenerative medicine applications. Keywords: biomaterial–cell interaction; scaffolds; stem cells Introduction of extracellular cues on cell proliferation and differentiation. The ex- tracellular matrix (ECM) is a highly defined and specialized micro- Parallel advances in biologically mimicking (biomimetic) material environment, which is essential for tissue development and function. constructs and in stem cell technologies enable the restoration and The ultimate decision of a cell to differentiate, proliferate, migrate, direct replacement of diseased cells and tissues. To achieve these apoptose or perform other functions is a coordinated response to the outcomes clinically, fully functional cells and tissues must be pro- physical and chemical interactions with these ECM effectors [2]. duced on a large scale [1]. Despite advances in the design and syn- Matrix elasticity is one mechanical property of the ECM that dif- thesis of biomaterial scaffolds, one of the biggest obstacles facing fers between tissues and can be manipulated in synthesized scaffolds tissue engineering is a lack of understanding regarding the influence V C The Author(s) 2018. Published by Oxford University Press. 167 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 168 Whitehead et al. to enhance tissue engineering success and applications [3]. Several having been used in approximately 700 clinical trials [24]. MSCs are studies have demonstrated that matrix elasticity can influence stem currently being investigated as potential cell sources to regenerate cell behavior and differentiation toward certain lineages, indicating bone tissue, cartilage, ligament tissue, muscle and adipose tissue the power of physical environment on cell state [4–10]. Notably, [25–29]. Engler et al. initially demonstrated that lineage specification in stem To analyze the interactions between bone marrow-derived cells can be directed by altering the elastic modulus of polyacryl- hMSCs and a hydrogel platform, we selected two hydrogel compo- amide (PA) gels, showing that elasticities of 0.1–1, 8–17 and sitions: 10% wt. PEG dimetharcylate (PEGDMA) MW 1000 and 25–40 kPa influence mesenchymal stem cell (MSC) differentiation 10% wt. PEGDMA MW 20 000 and the 3% wt. PEGDMA MW toward neurogenic, myogenic and osteogenic lineages, respectively 1000 and17% wt.PEGDMAMW20000,which yieldelastic [6]. Later, Wen et al. systematically modulated the porosity, ligand moduli in the ranges of 50–60 and 8–10 kPa, respectively. These density and stiffness of PA hydrogels, demonstrating that varying two hydrogel compositions were chosen because they are at the substrate porosity did not significantly change the osteogenic and upper and lower ends of the physiologically relevant elasticities. adipogenic differentiation of human adipose-derived stromal cells For conciseness, the hydrogels with an elastic modulus of 50– and marrow-derived mesenchymal stromal cells. These findings im- 60 kPa are referred to as ‘stiff hydrogels’ and the hydrogels with an ply that the stiffness of planar matrices regulates stem cell differenti- elastic modulus of 8–10 kPa are referred to as ‘soft hydrogels’. ation independently of protein tethering and porosity [11]. Despite Expanding on our previous work, here we demonstrate the utiliza- the studies mentioned above, there remains a lack of understanding tion of our PEG-based hydrogel blends to study the effect of elas- regarding the distinct roles of physical and chemical cues on specific ticity on the characteristics and differentiation potential of bone stem cell types. The influence of the dynamic extracellular cues of bi- marrow-derived MSCs [17]. We show that the hydrogels of differ- ologically relevant scaffolds on stem cell proliferation and differenti- ent elasticities produce changes in hMSC morphology and prolifer- ation remains unclear and further investigation is needed. ation, which provides support that the platform has the potential Understanding these interactions is paramount for the field of tissue to produce changes in hMSC behavior and cell state. Furthermore, engineering and regenerative medicine to more fully advance to clin- we find that the different elasticities can subtly influence stem cell ical applications. differentiation potential, primarily in cell types of stiffer elasticity. The importance of elasticity in influencing and directing cell be- Our findings enhance the fundamental understanding of stem cell– havior generates a need for tailorable biomaterial scaffolds. biomaterial interactions and open the door for the continued ex- Hydrogel-based biomaterials have rapidly become an attractive me- ploration of PEG-based hydrogel scaffold in tissue engineering and dium because their innate network closely resembles the structure of regenerative medicine. the ECM, their elasticity can be tailored, they allow for rapid diffu- sion of hydrophilic nutrients and they have a low content of dry mass, which reduces irritation and degradation [12]. These features Materials and methods allow the hydrogels to provide an environment that is like that of Materials and reagents the in vivo environment, as well as provide additional control of the Hydrogel synthesis and characterization physical and mechanical properties affecting cellular proliferation PEGDMA MW 1000 and MW 20 000 were purchased from and differentiation. To be effective, the hydrogel scaffold must be Polysciences and were used as received. The ultraviolet (UV) photoi- capable of promoting desirable cellular functions for specific appli- nitiator, 2-hydroxy-1-[4-(hydroxyethoxy) phenyl]-2-methyl-1 prop- cations without causing an inflammatory response. Different poly- anone (I2959) and fibronectin were purchased from Sigma Aldrich. mers used to engineer hydrogel scaffolds have different biological Methacrylate acid (MAA) was purchased from Fisher Scientific and properties, all with their own strengths and weaknesses. For exam- was passed through a basic alumina column prior to use to remove ple, polyethylene glycol (PEG) polymers are biocompatible and inhibitor. Heptane was purchased from Fisher Scientific. 1-Ethyl-3- bio-inert in nature. While PEG has been studied for multiple appli- (3-dimethylaminopropyl) carbodiimide (EDC), sulfo-N-hydroxysul- cations, the usefulness of PEG polymers for the formation of tailora- fosuccinimide (sulfo-NHS), 2-(N-morpholino)ethanesulfonic acid ble biomimetic scaffolds in tissue engineering and regenerative (MES) Buffer, and 1X phosphate buffer saline (PBS) were purchased medicine has not been fully investigated [13–15]. However, PEG from Thermo Fisher Scientific. acrylates are popular polymers utilized as hydrogel biomaterials for tissue engineering applications [16]. Previously, we demonstrated the generation of biomaterial scaffolds of varying elasticity by imple- Stem cell maintenance and characterization menting tailorable PEG hydrogels. Results showed that our hydrogel Human MSCs were provided by Dr. Bruce Bunnell from Tulane platform is compatible with multiple stem cell types, specifically University. Adipogenic differentiation media and osteogenic differ- mouse embryonic stem cells, human adipose stem cells and human entiation media were purchased from LaCell. MEM a, L-Glutamine, V R bone marrow-derived MSCs (hMSCs) [17]. Here, we further charac- Penicillin Streptomycin, ReadyProbes Cell Viability Imaging Kit terize the interactions of our hydrogel platform with hMSCs, pre- (Blue/Red), Alexa Fluor 555 Phalloidin and TRIzol reagent were senting an investigation into the specific interactions between purchased from ThermoFisher Scientific. Fetal bovine serum was hMSCs and our tailorable, affordable and reproducible PEG-based purchased from Atlanta Biologicals. Formalin was purchased from hydrogel platform. MSCs are adult, multipotent stem cells harvested Azer Scientific. Triton X-100 was purchased from Alfa Aesar. from bone marrow, adipose tissue, umbilical cords and muscle [18– Methanol was purchased from VWR. Bovine serum albumin was 31]. MSCs are known for their ability to differentiate into cell types purchased from Amresco. qScript cDNA SuperMix was purchased of the mesoderm lineage, with their differentiation into adipogenic, from Quanta Biosciences. Powerup SYBR green master mix was V R osteogenic and chondrogenic lineages being well described [22, 23]. purchased from Applied Biosystems. AlamarBlue reagent and 4’,6- These cells have the potential to be patient specific and, with several diamidino-2-phenylindole, dihydrochloride (DAPI) was purchased regenerative and immunosuppressive properties, clinically relevant, from ThermoFisher Scientific. Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Poly (ethylene glycol) hydrogel elasticity 169 Hydrogel preparation Maintenance of hMSCs Hydrogel solutions for the ‘stiff’ hydrogels (10% wt. PEGDMA Human MSCs were cultured on 10-cm polystyrene tissue culture MW 1000 and 10% wt. PEGDMA MW 20 000) and the ‘soft’ dishes in maintenance medium containing MEM a, L-glutamine, hydrogels (3% wt. PEGDMA MW 1000 and 17% wt. PEGDMA penicillin streptomycin and 16.5% FBS. The cells were incubated at MW 20 000) were prepared in deionized water (DH O) as reported 37 C with 5% CO . 2 2 previously [30]. 0.1% wt. UV photoinitiator, 2-hydroxyl-1-[4-(hy- droxyl) phenyl]-2-methyl-1 propanone (I2959), which is below con- Osteogenic and adipogenic differentiation centrations previously determined to be cyto-compatible [31], and Human MSCs were seeded on tissue culture plates and soft hydro- 2% wt. MAA were added to the hydrogel solution. Solution was 3 2 gels at a density of 2.010 cells/cm and grown until 80% conflu- sonicated for 20 minutes and then pipetted in between two photo- ence. Due to decreased proliferation of hMSCs on stiff hydrogels, masks separated by 0.55-mm stripes of teflon and UV polymerized 3 2 hMSCs were seeded on these specific gels at 4.010 cells/cm and at a wavelength of 365 nm and an intensity of 34 mW/cm . Stiff attached at 80% confluence. The appropriate differentiation media and soft hydrogels were UV polymerized for 10 and 20 minutes, re- (adipogenic differentiation media or osteogenic differentiation me- spectively. The hydrogels were then rinsed for 10 days in DH O (pe- dia) was added to the cells in all cases when cells demonstrated 80% riodically changed) to remove any un-reacted polymer or monomer. confluence. Differentiation media was changed every 72 hours until Prior to cell culture, hydrogels were functionalized with fibronectin time point for analysis. via EDC/Sulfo-NHS chemistry as described previously [17]. Cell viability assay V R Characterization of hydrogel swelling Cell viability was determined using the ReadyProbes Cell Viability Hydrogel swelling studies were performed as reported previously Imaging Kit (Blue/Red) and imaged on the EVOS FL imaging sys- [29, 32]. After UV polymerization, hydrogel films were cut into tem. Assay was done following manufacturer’s protocol. 19.5-mm discs and were weighed in air as well as in heptane (a sol- vent the PEG hydrogels will not swell in) to obtain the volume of the F-actin staining hydrogels immediately after UV polymerization. The hydrogels were Human MSCs were fixed with formalin and permeabilized using then rinsed for 10 days in DH O (periodically changed) to remove 0.2% Triton 100X. Alexa Fluor 555 Phalloidin was dissolved in any un-reacted polymer. Hydrogel discs were then dried for 5 days methanol to create a stock solution with a final concentration of under vacuum and subsequently weighed to obtain dry (or polymer) 200 units/ml. The final staining solution contained a 1:40 ratio of mass. The dried hydrogels were then swollen for 48 hours in DH O methanolic stock to PBS, with 1% BSA. Cells were protected from to reach swollen equilibrium. The polymer volume fraction in the direct light and incubated in staining solution for 15 minutes. DAPI swollen state,  and relaxed state  was calculated from the 2,s 2,r was added to each well at a final concentration of 1:2000 and incu- measured hydrogel mass in air and in heptane: bated for an additional 5 minutes. Cells were washed with PBS three times and imaged. W  W a;d n;d v ¼ (1) 2;s W  W a;s n;s W  W a;d n;d v ¼ (2) Cell attachment studies 2;r W  W a;r n;r 3 2 Human MSCs were seeded at a density of 2.010 cells/cm per sample and allowed to attach for 18 hours. Cells were fixed with where W is the hydrogel weight in dry state in air, W is the hy- a,d n,d formalin and permeabilized with 0.2% Triton X-100. Cells were in- drogel weight in dry state in heptane, W is the hydrogel weight in a,s cubated in a 1:1000 solution of DAPI and blocking buffer (0.2% swollen state in air, W is the hydrogel weight in swollen state in n,s Triton X-100 and 1% wt. BSA in 1X PBS) for 10 minutes. Cells heptane, W is the hydrogel weight in the relaxed state in air and a,r were washed with PBS three times, and 500 ml of PBS was added to W is the hydrogel weight in the relaxed state in heptane. The equi- n,r each well for imaging. The fluorescence was visualized and imaged librium volume swelling ratio (Q) was calculated by comparing the using the EVOS FL cell imaging system. Three images were taken ratio of the equilibrium swollen volume with the polymer volume at per well (top, middle and bottom). ImageJ was used to count the nu- the dry state [32]. Pore sizes were determined using the equation: clei per image. The average of the three images was taken for each 1=2 2C M 1=3 n c sample. n ¼ v l (3) 2;s Quantitative RT-PCR where n is the pore size,  is the polymer volume fraction in the 2,s RNA was collected and extracted from each cell type using TRIzol re- swollen state, C is Flory characteristic ratio, M is the average mo- n c agent following the manufacturer’s protocol. The RNA was quanti- lecular weight between crosslinks, M is the molecular weight of the fied using a Take3 plate on a BioTek plate reader. RNA monomer, and l is the bond-length along the backbone chain. The concentrations used for cDNA synthesis are shown in Supplementary M is found by using the Merrill and Peppas equation: Table S1. Due to low RNA concentrations in undifferentiated MSCs, t 2 each sample for that experiment was a pool of three wells from a 24- ln 1  v þ v þ v v 2;s 2;s 1 2;s 1 2 V ¼   (4) 1  well plate. cDNA was synthesized following the protocol provided by M M v v c n 2;s 2;s 2;r Quanta Biosciences for their cDNA SuperMix kit. The expression lev- v 2v 2;r 2;r els for each marker were quantified by qRT-PCR according to the where M is the number average molecular weight of the uncros- manufacturer’s protocol on an Applied Biosystems StepOne Plus in- slinked polymer, t is the specific volume of the polymer, is the molar strument (Primer pairs shown in Supplementary Table S2). Each reac- volume of the water, V is the polymer volume fraction in the re- tion was performed in triplicate for every sample and the relative laxed state, and v is the polymer–solvent interaction parameter. expression levels were determined by normalizing to gapdh. Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 170 Whitehead et al. Table 1. Pore sizes of stiff and soft hydrogels Composition Elastic modulus   Mc (g/mol) Q f (A) 2,r 2,s (kPa) [23] Stiff hydrogels 10% PEGDMA Mw 20, 000 / 50–60 0.18 6 5.11E3 0.094 6 2.31E3 1379.11 6 50.98 10.58 6 0.26 52.58 6 1.34 10% PEGDMA Mw 1000 Soft hydrogels 17% PEGDMA Mw 20, 000 / 8–10 0.19 6 7.71E3 0.043 6 1.72E3 4691.12 6 130.88 23.45 6 0.91 127.82 6 3.21 3% PEGDMA Mw 1000 Patel et al. 20% PEGDMA Mw 1000 388–390 0.24 6 0.01 0.18 6 1.30E3 256.13 6 8.32 5.56 6 0.04 17.77 6 0.32 reference hydrogels AlamarBlue Therefore, F-actin filaments of cells cultured on all three elasticity 3 2 conditions were stained and visualized (Fig. 1B). Cells on the soft hMSCs were seeded at a density of 2.010 cells/cm on all three elas- hydrogels maintained similar morphology to the tissue culture plate ticity conditions and grown under standard conditions for 72 hours. controls. In contrast, hMSCs cultured on stiff hydrogels displayed a At 72 hours, alamarBlue reagent was added to culture media at 10% more spindle-like morphology than MSCs cultured on tissue culture of the sample volume. Blanks for each sample were prepared by add- plates or soft hydrogels. ing equivalent amounts of culture media and alamarBlue reagent to ImageJ software was used to analyze the images from the F-actin wells containing corresponding elasticity conditions, without hMSCs. staining experiment to further confirm differences in cell number ob- Samples were incubated at 37 C and protected from direct light. served between the three elasticity conditions. The number of DAPI- Readings were taken at 1, 2, 3, 4 and 24 hours post alamarBlue re- stained nuclei in each image was counted and the average of three agent introduction. Fluorescence was measured at excitation 560/ samples per condition type was determined. Importantly, all hMSCs emission 590 using a BioTek Cytation 5 plate reader. shown in Fig. 1B were seeded at the same density, cultured for 72 hours and analyzed at the same exposure. As mentioned above, Statistical analysis the differences in image brightness observed from the stiff hydrogels All data are expressed as mean with error bars representing is attributed to the decreased porosity, which further obstructs visu- Standard Error (SE) for all quantitative comparison experiments. alization when viewed through an inverted microscope. The differ- Statistical analysis was carried out via one-way analysis of variance ence in brightness does not affect the cell count, as ImageJ was still (ANOVA) tests, using SPSS software v 24. P < 0.05 was considered able to differentiate individual nuclei (Fig. 1C). The cell count analy- statistically significant. Significant results were further analyzed via sis revealed a significant difference in the number of nuclei on stiff Tukey Honest Significant Difference (HSD) post hoc test and a P hydrogels compared to the tissue culture plate control, but no signif- values < 0.05 was considered significant. icant difference between the soft hydrogel and that same control was observed (Fig. 1D). To determine if this difference in cell number was the result of a Results difference in initial cell attachment, the number of adherent cells was Characterization of hydrogel swelling behavior counted 18 hours after seeding. ImageJ analysis of DAPI-stained cells Swelling behavior of the synthesized stiff (50–60 kPa) and soft on each surface revealed a significant increase in the number of cells (8–10 kPa) hydrogels was measured to determine the average molec- attached to both soft and stiff hydrogels compared to the tissue culture ular weight between crosslinks, network pore size and swelling ratio plate control (Fig. 2A–C). This indicates that attachment is not re- using standard swelling protocols reported previously. The results sponsible for the decrease in the number of cells present on the hydro- are summarized in Table 1. While the total percent polymer was gels after 72 hours. Alternatively, differences in rate of proliferation held constant at 20% wt. the amount of MW 1000 Da and MW could explain a difference in cell number. An AlamarBlue assay was 20 000 Da was varied to create more elastic hydrogels. As expected, utilized as an indicator of cellular proliferation, and the results show the molecular weight between crosslinks and the pore sizes was significantly less metabolic activity in cells cultured on stiff hydrogels larger in the soft hydrogels compared to the stiff hydrogels. The compared to soft hydrogels and tissue culture plates at 3, 4 and equilibrium swelling ratio (Q) of the soft hydrogel formulations is 24 hours (Fig. 2D). At 24 hours, metabolic activity was significantly twice that of the stiff hydrogels. higher in cells cultured on tissue culture plates than in cells cultured on both stiff and soft hydrogels. Given that proliferation is slower on the stiff hydrogels, the expression of the multipotency marker sox2 hMSC attachment to hydrogels was analyzed to see if there were significant changes in multipotency. Bone marrow-derived hMSCs were seeded on the hydrogel scaffolds Cells were seeded at the same density on each surface and cultured for and after 72 hours a viability assay was performed to determine if 72 hours before collecting RNA. Results of qRT-PCR of sox2 the cells survived on each of the three surfaces: tissue culture plates, (Fig. 2E) indicates that there is no statistically significant difference in soft hydrogels and stiff hydrogels. Based on propidium iodide stain- expression levels between each surface, demonstrating that the elastic- ing (dead cells stained red), we observe few, if any, dead cells ity conditions do not immediately influence the levels of certain multi- (Fig. 1A). The difference in image brightness observed from the stiff potency transcription factors. hydrogels is attributed to the decreased porosity, which further obstructs visualization. The difference in brightness does not alter the number of live/dead cells. Effect of elasticity on hMSC osteogenic differentiation Cell morphology can be an indicator of cellular state, and To be useful in tissue engineering and regenerative medicine, bioma- changes to this morphology could indicate changes in cell behavior. terial scaffolds must be able to support and potentially direct stem Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Poly (ethylene glycol) hydrogel elasticity 171 Elasticity Tissue Culture Plate Stiff Hydrogel Soft Hydrogel Cell Count Analysis Tissue Culture Plate Stiff Hydrogels Soft Hydrogels Figure 1. hMSCs attach to and survive on the different hydrogel compositions. (A) Viability assay of hMSCs cultured on tissue culture plates, stiff hydrogels and soft hydrogels for 72 hours. Live cell nuclei are shown in blue, while dead cell nuclei are shown in red. (B) Morphology of hMSCs cultured on the three elasticity conditions for 72 hours. The cell nuclei are shown in blue, while the F-actin filaments are shown in red. (C) Visual depiction of ImageJ analysis highlighting nuclei for count. (D) Cell count results from ImageJ quantification of cells seeded for 72 hours. *Tukey HSD resulting P < 0.05. n¼ 3. Scale bars: 400 mm cell differentiation toward desired lineages. Elasticity can play a role expression, an early marker of osteogenesis, between soft hydrogels in directing stem cell state, thus the effects of the hydrogel elasticities and stiff hydrogels (P < 0.05), between soft hydrogels and tissue cul- on hMSC differentiation toward an osteogenic lineage were investi- ture plates (P < 0.05) and between stiff hydrogels and tissue culture gated. Osteogenic differentiation was chemically induced in hMSCs plates (P < 0.01). seeded on all three elasticity conditions, and morphology was ana- lyzed using phase contrast microscopy (Fig. 3A). Due to the limited visibility in phase contrast images with hydrogels, phalloidin stain- Effect of elasticity on hMSC adipogenic differentiation ing was also used to visualize F-actin filaments (Fig. 3B). There was Since hMSCs also have the potential to be used for adipogenic tissue noticeable differentiation and calcium deposition on all three elastic- regeneration, we further assessed adipogenesis of these cells on each ity conditions. qRT-PCR of osteogenic markers runx2 and alp of the selected surfaces. Adipogenic differentiation was chemically (Fig. 3C) was performed on samples collected at day 7 of differentia- induced in cells seeded on all three elasticity conditions, and mor- tion. Analysis indicated no significant differences in the early osteo- phology was analyzed using phase contrast microscopy (Fig. 4A). As genic differentiation marker runx2 expression in hMSCs cultured on in the previous set of experiments, phalloidin staining was used to each surface. However, there were significant differences in alp visualize F-actin filaments and provide higher resolution images of Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Average nuclei/image 172 Whitehead et al. Tissue Culture Plate Stiff Hydrogel Soft Hydrogel B C Image Image Image 3 0 Tissue Culture Plate Stiff Hydrogels Soft Hydrogels Tissue Culture Plate Stiff Gels Soft Gels sox2 Expression 0.8 ** ** 0.6 ** * ** 0.4 3000 ** ** 0.2 1 hour 2 hours 3 hours 4 hours 24 Hours Tissue Culture Plate Stiff Hydrogel Soft Hydrogel Figure 2. hMSCs attach more readily to hydrogels, but display decreased proliferation, despite equal expression of sox2.(A) Example images of DAPI stain hMSCs 18 hours post-seeding on the different elasticity conditions. (B) Schematic representation of imaging method used for attachment studies. Three images were taken (as shown in panel A) of each sample, with three samples per surface. (C) Results of ImageJ quantification of nuclei per image. (D) Results of AlamarBlue analysis of hMSCs cultured on each surface. AlamarBlue was added after cells were cultured for 72 hours, and timepoints shown in graph represent hours after AlamarBlue introduction. (E) Quantitative reverse-transcriptase PCR analysis of sox2 expression in hMSCs cultured on each elasticity condition for 72 hours. *Tukey HSD resulting P < 0.05. **Tukey HSD resulting P < 0.01. n¼ 3 for C, D and E. Scale bars: 1000 mm cell morphology (Fig. 4B). Noticeable differentiation had taken were lower in the stiff hydrogels than in the soft hydrogels. place on each surface, with round globules, some of which are indi- Variability in the swelling characteristics between the two samples cated by green arrows in Fig. 4A and B, indicating vacuoles and adi- was attributed to the higher percentage of PEGDMA MW 20 000 pogenic differentiation. Phalloidin staining of the cells shows that in within the soft hydrogels (Table 1). Next, we characterized hMSC areas where lipid vacuoles formed there is a decrease in F-actin fila- interactions when cultured on the stiff and soft hydrogels. As the ments. This trend is seen on tissue culture plates, soft hydrogels and largest pores are nanometers in size and hMSCs have an approxi- stiff hydrogels. qRT-PCR of early adipogenic markers ppar-y and mate diameter range of 17.9–30.4 lm, there is no penetration of srebp1c was performed on samples collected at day 7 of differentia- hMSCs into the hydrogel network. Thus, hMSCs are cultured two- tion. Analysis indicates no significant differences in expression of dimensionally on the surface of these hydrogels. these early adipogenic markers between each surface (Fig. 4C). When seeded on both soft and stiff hydrogels, hMSCs were shown to attach and remain viable (Fig. 1A), further confirming the potential of this platform for use in cell culture and tissue genera- Discussion tion. However, changes in morphology were observed in cells cul- Previously, we demonstrated that PEGDMA hydrogels with elastici- tured on the different elasticities. Human MSCs cultured on the soft ties within a physiologically relevant range (8–60 kPa) can be gener- hydrogels maintained similar morphology to the tissue culture plate ated by varying the molecular weight of the polymer [17]. Here, we controls. In contrast, hMSCs cultured on stiff hydrogels displayed a characterized hydrogels at the upper and lower ends of this range, more spindle-like morphology as compared to hMSCs cultured on specifically in terms of their swelling behavior and interactions with tissue culture plates or soft hydrogels (Fig. 1B). These differences in hMSCs. The swelling behavior of the hydrogels was used to deter- morphology could indicate a change in cell behavior, such as sponta- mine the molecular weight between crosslinks (M ), the pore sizes neous differentiation. Furthermore, there appeared to be consis- (n) and the equilibrium swelling ratio (Q). All three, as predicted, tently fewer hMSCs on the stiff hydrogels after 72 hours of culture. Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Average Number of Cells/Image Relative Expression Normalized to gapdh Poly (ethylene glycol) hydrogel elasticity 173 Tissue Culture Plate Stiff Hydrogel Soft Hydrogel Tissue Culture Plate Stiff Hydrogel Soft Hydrogel ** * 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 runx2 apl alp Figure 3. hMSCs retain the ability to differentiate toward osteogenic lineages on all three elasticity conditions. (A) Phase contrast images of hMSCs at day 7 of os- teogenic differentiation. (B) Morphology of hMSCs at day 7 of osteogenic differentiation, corresponding to the phase contrast images in panel A. The cell nuclei are shown in blue, while the F-actin filaments are shown in orange. (C) Quantitative reverse-transcriptase PCR analysis of the osteogenic differentiation markers runx2 and alp in hMSCs at day 7 of osteogenic differentiation. *Tukey HSD resulting P < 0.05. **Tukey HSD resulting P < 0.01. n¼ 3. Scale bars: 200 mm ImageJ quantification of the hMSCs shown in Fig. 1B revealed sig- marker sox2 revealed no significant differences in expression across nificantly fewer cells on the stiff hydrogels (Fig. 1D). The decrease in the three elasticity conditions (Fig. 2E). However, further analysis of hMSCs could be the result of decreased attachment to the stiff multipotency markers and markers of possible differentiation line- hydrogels. However, attachment analysis 18 hours after seeding ac- ages could reveal that a subtle amount of spontaneous differentia- tually revealed an increased number of hMSCs attached to both hy- tion has taken place or longer time course studies may demonstrate drogel elasticities compared to tissue culture plate indicating that more significant changes in multipotency. For the scope of this the hydrogels have an impact on cell proliferation (Fig. 2A–C). study, the differentiation potential of hMSCs on the three elasticity AlamarBlue assays demonstrated that proliferation is signifi- conditions was analyzed through chemically induced differentiation cantly decreased in the cells cultured on the soft and stiff hydrogels. toward osteogenic and adipogenic lineages, rather than exploring Human MSC proliferation on stiff hydrogels was shown to be signif- the long-term effects of maintenance on each of these surfaces. icantly decreased 3-hours after the introduction of AlamarBlue On all three elasticity conditions, hMSCs cultured in osteogenic (Fig. 2D). The differences in morphology and proliferation observed differentiation media differentiated toward the osteogenic lineage, in hMSCs cultured on stiff hydrogels could be an indication of as evidenced by calcium deposition and expression of bone specific spontaneous differentiation. Quantitative Reverse-Transcriptase – markers. There was no significant difference in runx2 expression, polymerase chain reaction (RT-PCR) analysis of the multipotency which is an essential transcription factor for osteoblastic Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Relative Expression Normalized to gapdh 174 Whitehead et al. Tissue Culture Plate Stiff Hydrogel Soft Hydrogel Tissue Culture Plate Stiff Hydrogel Soft Hydrogel 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 ppar-γ srebp-1c Figure 4. hMSCs retain the ability to differentiate toward adipogenic lineages on all three elasticity conditions. (A) Phase contrast images of hMSCs at day 7 of adipogenic differentiation. (B) Morphology of hMSCs at day 7 of adipogenic differentiation, corresponding to the phase contrast images in panel A. The cell nuclei are shown in blue, while the F-actin filaments are shown in orange. Cells containing lipid vesicles demonstrated a rearrangement of F-actin filaments, indicated by green arrows. (C) quantitative reverse-transcriptase PCR analysis of the adipogenic differentiation markers ppar-c and srebp1-c in hMSCs at day 7 of adipo- genic differentiation. Results considered insignificant with P > 0.05. n¼ 3. Scalebars: 200 mm differentiation (Fig. 3C). However, there was a significant decrease In summary, the data from this study give insight into the proper- in alp expression in hMSCs cultured on both hydrogel elasticities ties and stem cell interactions of our previously established hydrogel (Fig. 3C). While lower levels of alp expression could platform; an inexpensive, highly tailorable platform that can be indicate decreased osteogenesis, given the observation of calcium de- adapted to any number of cell-material interaction studies and applica- position and cell morphology, it is also possible that a decrease in tions. There is a need within the field to investigate the roles of both alp expression is an indication of more rapid maturation of the the physical and chemical properties of different biomaterials on a vari- resulting cells. ety of stem cells. This article presents a focused investigation into the hMSCs cultured in adipogenic differentiation media also interactions between hMSCs and our specific PEG-based hydrogel plat- retained the ability to differentiate toward adipogenic lineages on all form. Changes in hMSC morphology and proliferation were observed three elasticity conditions. hMSCs on all three elasticity conditions in cells cultured on hydrogels, primarily those cultured on stiff hydro- began forming lipid vesicles characteristic of adipogenic differentia- gels. These results demonstrate that the elastic tailorability of this hy- tion (Fig. 4A and B). There was no significant difference in the ex- drogel platform can produce changes in hMSC behavior and cell state, pression of two key transcription factors involved in adipogenesis, indicating a potential for these hydrogels to be used to generate a con- ppar-c and srebp1-c. Ultimately, no difference in hMSC differentia- trolled environment for cell culture and tissue regeneration applica- tion toward adipogenic lineages was observed. tions. Furthermore, based on the differentiation studies, the different Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Relative Expression Normalized to gapdh Poly (ethylene glycol) hydrogel elasticity 175 8. Holst J, Watson S, Lord MS et al. letters Substrate elasticity provides me- hydrogel elasticities have subtle effects on stem cell differentiation. chanical signals for the expansion of hemopoietic stem and progenitor This effect is observed primarily in osteogenic differentiation, which cells. Nat Biotechnol 2010;28:1123–8. could indicate that cell lineages of higher elasticity are more susceptible 9. Saha K, Keung AJ, Irwin EF et al. Substrate modulus directs neural stem to elasticity changes. The results of this study further suggest that the cell behavior. Biophys J 2008;95:4426–38. hydrogel platform’s elasticity does affect stem cell behavior, opening 10. Yang C, DelRio FW, Ma H et al. Spatially patterned matrix elasticity the door to future investigations of the platform’s potential for control- directs stem cell fate. Proc Natl Acad Sci 2016;113:E4439–45. ling stem cell fate. While the platform is limited in terms of its ability 11. Wen JH, Vincent LG, Fuhrmann A et al. Interplay of matrix stiffness and to be tailored for responsiveness and degradation, the limitations are protein tethering in stem cell differentiation. Nat Mater 2014;13:979–87. appropriate for current and future studies focused on elasticity and 12. Annabi N, Tamayol A, Uquillas JA et al. 25th anniversary article: rational other specific cues, while maintaining affordable and reproducible bio- design and applications of hydrogels in regenerative medicine. Advanced Materials 2014;26:85–124. doi: 10.1002/adma.201303233. material scaffolds. In addition, PEG is an US Food and Drug 13. Bryant SJ, Anseth KS. Controlling the spatial distribution of ECM compo- Administration (FDA)-approved polymer, which gives the platform the nents in degradable PEG hydrogels for tissue engineering cartilage. potential to be used in the clinic, should future studies open the door J Biomed Mater Res A 2003;64A:70–9. for therapeutic uses of these hydrogels. The focus of this study was to 14. Elisseeff J, Anseth K, Sims D et al. Transdermal photopolymerization for understand cell–biomaterial interactions, with the goal of adding to a minimally invasive implantation. Proc Natl Acad Sci USA 1999;96:3104–7. library of materials for various applications in regenerative medicine. 15. Karmaker AC, Dibenedetto A, Goldberg AJ. Extent of conversion and its ef- A better understanding of the biomaterial scaffolds utilized in tissue re- fect on the mechanical performance of Bis-GMA/PEGDMA-based resins and generation, such as the studies shown here, is essential in optimizing their composites with continuous glass fibres. J Mater Sci 1997;8:333–401. their translational and clinical potential. 16. Zhu J. Bioactive modification of poly(ethylene glycol) hydrogels for tissue engineering. Biomaterials 2010;31:4639–56. 17. Patel NR, Whitehead AK, Newman JJ et al. Poly(ethylene glycol) hydro- gels with tailorable surface and mechanical properties for tissue engineer- Acknowledgements ing applications. ACS Biomater Sci Eng 2016;3:1494. The authors are grateful to Dr. Bruce Bunnell from Tulane University, for guid- 18. Asakura A, Komaki M, Rudnicki M. Muscle satellite cells are multipoten- ance and provision of hMSCs and to Dr. Jeff Gimble from Tulane University tial stem cells that exhibit myogenic, osteogenic, and adipogenic differen- and LaCell, LLC for guidance and reagents. This work was funded by the tiation. Differentiation 2001;68:245–53. Louisiana Biomedical Research Network (LBRN) NIH INBRE P20GM103424 19. Bieback K, Kern S, Klu ¨ ter H et al. Critical parameters for the isolation of and the Center for Cardiovascular Diseases and Sciences at LSU Health mesenchymal stem cells from umbilical cord blood. Stem Cells 2004;22: Shreveport in the form of the Partners Across Campus grants. We also received 625–34. funding from LaSPACE and NASA through a LaSPACE Research Enhancement 20. Lee OK, Kuo TK, Chen WM et al. Isolation of multipotent mesenchymal Award, LaSPACE Undergraduate Research Assistantship and two Graduate stem cells from umbilical cord blood. Blood 2004;103:1669–75. Student Research Assistantships, NASA Space Grant Award NNX158H828. 21. Zuk PA, Zhu M, Ashjian P et al. Human adipose tissue is a source of mul- We also acknowledge Louisiana Tech University’s College of Applied and tipotent stem cells. Molecular Biology of the Cell 2002;13:4279–95. Natural Sciences for support in the form of student minigrants. We would like 22. Mackay AM, Beck SC, Murphy JM et al. Chondrogenic differentiation of to thank both Newman and Caldorera-Moore labs for help and support, with cultured human mesenchymal stem cells from marrow. Tissue Eng 1998; special thanks to India Pursell and Rachel Eddy for assistance with cell culture 4:415–28. maintenance and Nehal Patel for technical support. 23. Pittenger MF, Mackay AM, Beck SC et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284. http://science.science mag.org/content/284/5411/143. 24. NIH. 645 Studies found for: Mesenchymal stem cells. https://clinicaltrials. Supplementary data gov/ct2/results?cond=mesenchymal+stem+cells&term=&cntry=&state=& city=&dist= (accessed 1 February 2018). Supplementary data are available at REGBIO online. 25. Dezawa M, Ishikawa H, Itokazu Y et al. Bone marrow stromal cells gener- Conflict of interest statement. None declared. ate muscle cells and repair muscle degeneration. Science (New York, N.Y.) 2005;309:314–7. 26. Fan H, Liu H, Toh SL et al. Enhanced differentiation of mesenchymal References stem cells co-cultured with ligament fibroblasts on gelatin/silk fibroin hy- brid scaffold. Biomaterials 2008;29:1017–27. 1. Ba ´ rtolo PJ, Chua CK, Almeida HA et al. Biomanufacturing for tissue engi- 27. Patrick CW. Tissue engineering strategies for adipose tissue repair. Anat neering: present and future trends. Virtual Phys Prototyp 2009;4:203–16. Rec 2001;263:361–6. 2. Lutolf MP, Hubbell JA. Synthetic biomaterials as instructive extracellular 28. Ponticiello MS, Schinagl RM, Kadiyala S et al. Gelatin-based resorbable microenvironments for morphogenesis in tissue engineering. Nat sponge as a carrier matrix for human mesenchymal stem cells in cartilage Biotechnol 2005;23:47–55. regeneration therapy. J Biomed Mater Res 2000;52:246–55. 3. Discher DE. Tissue cells feel and respon to the stiffness of their substrate. 29. Yousefi AM, James PF, Akbarzadeh R et al. Prospect of stem cells in bone Science 2005;310:1139–43. tissue engineering: a review. Stem Cells Int 2016;2016:1. 4. Banerjee A, Arha M, Choudhary S et al. The influence of hydrogel modu- 30. Bryant SJ, Bender RJ, Durand KL et al. Encapsulating chondrocytes in lus on the proliferation and differentiation of encapsulated neural stem degrading PEG hydrogels with high modulus: engineering gel structural cells. Biomaterials 2009;30:4695–9. changes to facilitate cartilaginous tissue production. Biotechnol Bioeng 5. Engler AJ, Rehfeldt F, Sen S et al. Microtissue elasticity: measurements by 2004;86:747–55. atomic force microscopy and its influence on cell differentiation. Methods 31. Caldorera-Moore M, Kang MK, Moore Z et al. Swelling behavior of Cell Biol 2007;83:521–45. nanoscale, shape- and size-specific, hydrogel particles fabricated using im- 6. Engler AJ, Sen S, Sweeney HL et al. Matrix elasticity directs stem cell line- print lithography. Soft Matter 2011;7:2879. age specification. Cell 2006;126:677–89. 32. Bryant SJ, Nuttelman CR, Anseth KS. Cytocompatibility of UV and visible 7. Gilbert PM, Havenstrite KL, Magnusson KEG et al. Substrate elasticity light photoinitiating systems on cultured NIH/3T3 fibroblasts in vitro. regulates skeletal muscle stem cell self-renewal in culture. Science 2010; J Biomater Sci 2000;11:439–57. 329:1078–81. Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Regenerative Biomaterials Oxford University Press

Poly (ethylene glycol) hydrogel elasticity influences human mesenchymal stem cell behavior

Loading next page...
 
/lp/ou_press/poly-ethylene-glycol-hydrogel-elasticity-influences-human-mesenchymal-BTLH6QV7XQ

References (37)

Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press.
ISSN
2056-3418
eISSN
2056-3426
DOI
10.1093/rb/rby008
Publisher site
See Article on Publisher Site

Abstract

Coordinated investigations into the interactions between biologically mimicking (biomimetic) ma- terial constructs and stem cells advance the potential for the regeneration and possible direct replacement of diseased cells and tissues. Any clinically relevant therapies will require the develop- ment and optimization of methods that mass produce fully functional cells and tissues. Despite advances in the design and synthesis of biomaterial scaffolds, one of the biggest obstacles facing tissue engineering is understanding how specific extracellular cues produced by biomaterial scaf- folds influence the proliferation and differentiation of various cell sources. Matrix elasticity is one such tailorable property of synthetic scaffolds that is known to differ between tissues. Here, we in- vestigate the interactions between an elastically tailorable polyethylene glycol (PEG)-based hydro- gel platform and human bone marrow-derived mesenchymal stem cells (hMSCs). For these studies, two different hydrogel compositions with elastic moduli in the ranges of 50–60 kPa and 8– 10 kPa were implemented. Our findings demonstrate that the different elasticities in this platform can produce changes in hMSC morphology and proliferation, indicating that the platform can be implemented to produce changes in hMSC behavior and cell state for a broad range of tissue engineering and regenerative applications. Furthermore, we show that the platform’s different elasticities influence stem cell differentiation potential, particularly when promoting stem cell dif- ferentiation toward cell types from tissues with stiffer elasticity. These findings add to the evolving and expanding library of information on stem cell–biomaterial interactions and opens the door for continued exploration into PEG-based hydrogel scaffolds for tissue engineering and regenerative medicine applications. Keywords: biomaterial–cell interaction; scaffolds; stem cells Introduction of extracellular cues on cell proliferation and differentiation. The ex- tracellular matrix (ECM) is a highly defined and specialized micro- Parallel advances in biologically mimicking (biomimetic) material environment, which is essential for tissue development and function. constructs and in stem cell technologies enable the restoration and The ultimate decision of a cell to differentiate, proliferate, migrate, direct replacement of diseased cells and tissues. To achieve these apoptose or perform other functions is a coordinated response to the outcomes clinically, fully functional cells and tissues must be pro- physical and chemical interactions with these ECM effectors [2]. duced on a large scale [1]. Despite advances in the design and syn- Matrix elasticity is one mechanical property of the ECM that dif- thesis of biomaterial scaffolds, one of the biggest obstacles facing fers between tissues and can be manipulated in synthesized scaffolds tissue engineering is a lack of understanding regarding the influence V C The Author(s) 2018. Published by Oxford University Press. 167 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 168 Whitehead et al. to enhance tissue engineering success and applications [3]. Several having been used in approximately 700 clinical trials [24]. MSCs are studies have demonstrated that matrix elasticity can influence stem currently being investigated as potential cell sources to regenerate cell behavior and differentiation toward certain lineages, indicating bone tissue, cartilage, ligament tissue, muscle and adipose tissue the power of physical environment on cell state [4–10]. Notably, [25–29]. Engler et al. initially demonstrated that lineage specification in stem To analyze the interactions between bone marrow-derived cells can be directed by altering the elastic modulus of polyacryl- hMSCs and a hydrogel platform, we selected two hydrogel compo- amide (PA) gels, showing that elasticities of 0.1–1, 8–17 and sitions: 10% wt. PEG dimetharcylate (PEGDMA) MW 1000 and 25–40 kPa influence mesenchymal stem cell (MSC) differentiation 10% wt. PEGDMA MW 20 000 and the 3% wt. PEGDMA MW toward neurogenic, myogenic and osteogenic lineages, respectively 1000 and17% wt.PEGDMAMW20000,which yieldelastic [6]. Later, Wen et al. systematically modulated the porosity, ligand moduli in the ranges of 50–60 and 8–10 kPa, respectively. These density and stiffness of PA hydrogels, demonstrating that varying two hydrogel compositions were chosen because they are at the substrate porosity did not significantly change the osteogenic and upper and lower ends of the physiologically relevant elasticities. adipogenic differentiation of human adipose-derived stromal cells For conciseness, the hydrogels with an elastic modulus of 50– and marrow-derived mesenchymal stromal cells. These findings im- 60 kPa are referred to as ‘stiff hydrogels’ and the hydrogels with an ply that the stiffness of planar matrices regulates stem cell differenti- elastic modulus of 8–10 kPa are referred to as ‘soft hydrogels’. ation independently of protein tethering and porosity [11]. Despite Expanding on our previous work, here we demonstrate the utiliza- the studies mentioned above, there remains a lack of understanding tion of our PEG-based hydrogel blends to study the effect of elas- regarding the distinct roles of physical and chemical cues on specific ticity on the characteristics and differentiation potential of bone stem cell types. The influence of the dynamic extracellular cues of bi- marrow-derived MSCs [17]. We show that the hydrogels of differ- ologically relevant scaffolds on stem cell proliferation and differenti- ent elasticities produce changes in hMSC morphology and prolifer- ation remains unclear and further investigation is needed. ation, which provides support that the platform has the potential Understanding these interactions is paramount for the field of tissue to produce changes in hMSC behavior and cell state. Furthermore, engineering and regenerative medicine to more fully advance to clin- we find that the different elasticities can subtly influence stem cell ical applications. differentiation potential, primarily in cell types of stiffer elasticity. The importance of elasticity in influencing and directing cell be- Our findings enhance the fundamental understanding of stem cell– havior generates a need for tailorable biomaterial scaffolds. biomaterial interactions and open the door for the continued ex- Hydrogel-based biomaterials have rapidly become an attractive me- ploration of PEG-based hydrogel scaffold in tissue engineering and dium because their innate network closely resembles the structure of regenerative medicine. the ECM, their elasticity can be tailored, they allow for rapid diffu- sion of hydrophilic nutrients and they have a low content of dry mass, which reduces irritation and degradation [12]. These features Materials and methods allow the hydrogels to provide an environment that is like that of Materials and reagents the in vivo environment, as well as provide additional control of the Hydrogel synthesis and characterization physical and mechanical properties affecting cellular proliferation PEGDMA MW 1000 and MW 20 000 were purchased from and differentiation. To be effective, the hydrogel scaffold must be Polysciences and were used as received. The ultraviolet (UV) photoi- capable of promoting desirable cellular functions for specific appli- nitiator, 2-hydroxy-1-[4-(hydroxyethoxy) phenyl]-2-methyl-1 prop- cations without causing an inflammatory response. Different poly- anone (I2959) and fibronectin were purchased from Sigma Aldrich. mers used to engineer hydrogel scaffolds have different biological Methacrylate acid (MAA) was purchased from Fisher Scientific and properties, all with their own strengths and weaknesses. For exam- was passed through a basic alumina column prior to use to remove ple, polyethylene glycol (PEG) polymers are biocompatible and inhibitor. Heptane was purchased from Fisher Scientific. 1-Ethyl-3- bio-inert in nature. While PEG has been studied for multiple appli- (3-dimethylaminopropyl) carbodiimide (EDC), sulfo-N-hydroxysul- cations, the usefulness of PEG polymers for the formation of tailora- fosuccinimide (sulfo-NHS), 2-(N-morpholino)ethanesulfonic acid ble biomimetic scaffolds in tissue engineering and regenerative (MES) Buffer, and 1X phosphate buffer saline (PBS) were purchased medicine has not been fully investigated [13–15]. However, PEG from Thermo Fisher Scientific. acrylates are popular polymers utilized as hydrogel biomaterials for tissue engineering applications [16]. Previously, we demonstrated the generation of biomaterial scaffolds of varying elasticity by imple- Stem cell maintenance and characterization menting tailorable PEG hydrogels. Results showed that our hydrogel Human MSCs were provided by Dr. Bruce Bunnell from Tulane platform is compatible with multiple stem cell types, specifically University. Adipogenic differentiation media and osteogenic differ- mouse embryonic stem cells, human adipose stem cells and human entiation media were purchased from LaCell. MEM a, L-Glutamine, V R bone marrow-derived MSCs (hMSCs) [17]. Here, we further charac- Penicillin Streptomycin, ReadyProbes Cell Viability Imaging Kit terize the interactions of our hydrogel platform with hMSCs, pre- (Blue/Red), Alexa Fluor 555 Phalloidin and TRIzol reagent were senting an investigation into the specific interactions between purchased from ThermoFisher Scientific. Fetal bovine serum was hMSCs and our tailorable, affordable and reproducible PEG-based purchased from Atlanta Biologicals. Formalin was purchased from hydrogel platform. MSCs are adult, multipotent stem cells harvested Azer Scientific. Triton X-100 was purchased from Alfa Aesar. from bone marrow, adipose tissue, umbilical cords and muscle [18– Methanol was purchased from VWR. Bovine serum albumin was 31]. MSCs are known for their ability to differentiate into cell types purchased from Amresco. qScript cDNA SuperMix was purchased of the mesoderm lineage, with their differentiation into adipogenic, from Quanta Biosciences. Powerup SYBR green master mix was V R osteogenic and chondrogenic lineages being well described [22, 23]. purchased from Applied Biosystems. AlamarBlue reagent and 4’,6- These cells have the potential to be patient specific and, with several diamidino-2-phenylindole, dihydrochloride (DAPI) was purchased regenerative and immunosuppressive properties, clinically relevant, from ThermoFisher Scientific. Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Poly (ethylene glycol) hydrogel elasticity 169 Hydrogel preparation Maintenance of hMSCs Hydrogel solutions for the ‘stiff’ hydrogels (10% wt. PEGDMA Human MSCs were cultured on 10-cm polystyrene tissue culture MW 1000 and 10% wt. PEGDMA MW 20 000) and the ‘soft’ dishes in maintenance medium containing MEM a, L-glutamine, hydrogels (3% wt. PEGDMA MW 1000 and 17% wt. PEGDMA penicillin streptomycin and 16.5% FBS. The cells were incubated at MW 20 000) were prepared in deionized water (DH O) as reported 37 C with 5% CO . 2 2 previously [30]. 0.1% wt. UV photoinitiator, 2-hydroxyl-1-[4-(hy- droxyl) phenyl]-2-methyl-1 propanone (I2959), which is below con- Osteogenic and adipogenic differentiation centrations previously determined to be cyto-compatible [31], and Human MSCs were seeded on tissue culture plates and soft hydro- 2% wt. MAA were added to the hydrogel solution. Solution was 3 2 gels at a density of 2.010 cells/cm and grown until 80% conflu- sonicated for 20 minutes and then pipetted in between two photo- ence. Due to decreased proliferation of hMSCs on stiff hydrogels, masks separated by 0.55-mm stripes of teflon and UV polymerized 3 2 hMSCs were seeded on these specific gels at 4.010 cells/cm and at a wavelength of 365 nm and an intensity of 34 mW/cm . Stiff attached at 80% confluence. The appropriate differentiation media and soft hydrogels were UV polymerized for 10 and 20 minutes, re- (adipogenic differentiation media or osteogenic differentiation me- spectively. The hydrogels were then rinsed for 10 days in DH O (pe- dia) was added to the cells in all cases when cells demonstrated 80% riodically changed) to remove any un-reacted polymer or monomer. confluence. Differentiation media was changed every 72 hours until Prior to cell culture, hydrogels were functionalized with fibronectin time point for analysis. via EDC/Sulfo-NHS chemistry as described previously [17]. Cell viability assay V R Characterization of hydrogel swelling Cell viability was determined using the ReadyProbes Cell Viability Hydrogel swelling studies were performed as reported previously Imaging Kit (Blue/Red) and imaged on the EVOS FL imaging sys- [29, 32]. After UV polymerization, hydrogel films were cut into tem. Assay was done following manufacturer’s protocol. 19.5-mm discs and were weighed in air as well as in heptane (a sol- vent the PEG hydrogels will not swell in) to obtain the volume of the F-actin staining hydrogels immediately after UV polymerization. The hydrogels were Human MSCs were fixed with formalin and permeabilized using then rinsed for 10 days in DH O (periodically changed) to remove 0.2% Triton 100X. Alexa Fluor 555 Phalloidin was dissolved in any un-reacted polymer. Hydrogel discs were then dried for 5 days methanol to create a stock solution with a final concentration of under vacuum and subsequently weighed to obtain dry (or polymer) 200 units/ml. The final staining solution contained a 1:40 ratio of mass. The dried hydrogels were then swollen for 48 hours in DH O methanolic stock to PBS, with 1% BSA. Cells were protected from to reach swollen equilibrium. The polymer volume fraction in the direct light and incubated in staining solution for 15 minutes. DAPI swollen state,  and relaxed state  was calculated from the 2,s 2,r was added to each well at a final concentration of 1:2000 and incu- measured hydrogel mass in air and in heptane: bated for an additional 5 minutes. Cells were washed with PBS three times and imaged. W  W a;d n;d v ¼ (1) 2;s W  W a;s n;s W  W a;d n;d v ¼ (2) Cell attachment studies 2;r W  W a;r n;r 3 2 Human MSCs were seeded at a density of 2.010 cells/cm per sample and allowed to attach for 18 hours. Cells were fixed with where W is the hydrogel weight in dry state in air, W is the hy- a,d n,d formalin and permeabilized with 0.2% Triton X-100. Cells were in- drogel weight in dry state in heptane, W is the hydrogel weight in a,s cubated in a 1:1000 solution of DAPI and blocking buffer (0.2% swollen state in air, W is the hydrogel weight in swollen state in n,s Triton X-100 and 1% wt. BSA in 1X PBS) for 10 minutes. Cells heptane, W is the hydrogel weight in the relaxed state in air and a,r were washed with PBS three times, and 500 ml of PBS was added to W is the hydrogel weight in the relaxed state in heptane. The equi- n,r each well for imaging. The fluorescence was visualized and imaged librium volume swelling ratio (Q) was calculated by comparing the using the EVOS FL cell imaging system. Three images were taken ratio of the equilibrium swollen volume with the polymer volume at per well (top, middle and bottom). ImageJ was used to count the nu- the dry state [32]. Pore sizes were determined using the equation: clei per image. The average of the three images was taken for each 1=2 2C M 1=3 n c sample. n ¼ v l (3) 2;s Quantitative RT-PCR where n is the pore size,  is the polymer volume fraction in the 2,s RNA was collected and extracted from each cell type using TRIzol re- swollen state, C is Flory characteristic ratio, M is the average mo- n c agent following the manufacturer’s protocol. The RNA was quanti- lecular weight between crosslinks, M is the molecular weight of the fied using a Take3 plate on a BioTek plate reader. RNA monomer, and l is the bond-length along the backbone chain. The concentrations used for cDNA synthesis are shown in Supplementary M is found by using the Merrill and Peppas equation: Table S1. Due to low RNA concentrations in undifferentiated MSCs, t 2 each sample for that experiment was a pool of three wells from a 24- ln 1  v þ v þ v v 2;s 2;s 1 2;s 1 2 V ¼   (4) 1  well plate. cDNA was synthesized following the protocol provided by M M v v c n 2;s 2;s 2;r Quanta Biosciences for their cDNA SuperMix kit. The expression lev- v 2v 2;r 2;r els for each marker were quantified by qRT-PCR according to the where M is the number average molecular weight of the uncros- manufacturer’s protocol on an Applied Biosystems StepOne Plus in- slinked polymer, t is the specific volume of the polymer, is the molar strument (Primer pairs shown in Supplementary Table S2). Each reac- volume of the water, V is the polymer volume fraction in the re- tion was performed in triplicate for every sample and the relative laxed state, and v is the polymer–solvent interaction parameter. expression levels were determined by normalizing to gapdh. Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 170 Whitehead et al. Table 1. Pore sizes of stiff and soft hydrogels Composition Elastic modulus   Mc (g/mol) Q f (A) 2,r 2,s (kPa) [23] Stiff hydrogels 10% PEGDMA Mw 20, 000 / 50–60 0.18 6 5.11E3 0.094 6 2.31E3 1379.11 6 50.98 10.58 6 0.26 52.58 6 1.34 10% PEGDMA Mw 1000 Soft hydrogels 17% PEGDMA Mw 20, 000 / 8–10 0.19 6 7.71E3 0.043 6 1.72E3 4691.12 6 130.88 23.45 6 0.91 127.82 6 3.21 3% PEGDMA Mw 1000 Patel et al. 20% PEGDMA Mw 1000 388–390 0.24 6 0.01 0.18 6 1.30E3 256.13 6 8.32 5.56 6 0.04 17.77 6 0.32 reference hydrogels AlamarBlue Therefore, F-actin filaments of cells cultured on all three elasticity 3 2 conditions were stained and visualized (Fig. 1B). Cells on the soft hMSCs were seeded at a density of 2.010 cells/cm on all three elas- hydrogels maintained similar morphology to the tissue culture plate ticity conditions and grown under standard conditions for 72 hours. controls. In contrast, hMSCs cultured on stiff hydrogels displayed a At 72 hours, alamarBlue reagent was added to culture media at 10% more spindle-like morphology than MSCs cultured on tissue culture of the sample volume. Blanks for each sample were prepared by add- plates or soft hydrogels. ing equivalent amounts of culture media and alamarBlue reagent to ImageJ software was used to analyze the images from the F-actin wells containing corresponding elasticity conditions, without hMSCs. staining experiment to further confirm differences in cell number ob- Samples were incubated at 37 C and protected from direct light. served between the three elasticity conditions. The number of DAPI- Readings were taken at 1, 2, 3, 4 and 24 hours post alamarBlue re- stained nuclei in each image was counted and the average of three agent introduction. Fluorescence was measured at excitation 560/ samples per condition type was determined. Importantly, all hMSCs emission 590 using a BioTek Cytation 5 plate reader. shown in Fig. 1B were seeded at the same density, cultured for 72 hours and analyzed at the same exposure. As mentioned above, Statistical analysis the differences in image brightness observed from the stiff hydrogels All data are expressed as mean with error bars representing is attributed to the decreased porosity, which further obstructs visu- Standard Error (SE) for all quantitative comparison experiments. alization when viewed through an inverted microscope. The differ- Statistical analysis was carried out via one-way analysis of variance ence in brightness does not affect the cell count, as ImageJ was still (ANOVA) tests, using SPSS software v 24. P < 0.05 was considered able to differentiate individual nuclei (Fig. 1C). The cell count analy- statistically significant. Significant results were further analyzed via sis revealed a significant difference in the number of nuclei on stiff Tukey Honest Significant Difference (HSD) post hoc test and a P hydrogels compared to the tissue culture plate control, but no signif- values < 0.05 was considered significant. icant difference between the soft hydrogel and that same control was observed (Fig. 1D). To determine if this difference in cell number was the result of a Results difference in initial cell attachment, the number of adherent cells was Characterization of hydrogel swelling behavior counted 18 hours after seeding. ImageJ analysis of DAPI-stained cells Swelling behavior of the synthesized stiff (50–60 kPa) and soft on each surface revealed a significant increase in the number of cells (8–10 kPa) hydrogels was measured to determine the average molec- attached to both soft and stiff hydrogels compared to the tissue culture ular weight between crosslinks, network pore size and swelling ratio plate control (Fig. 2A–C). This indicates that attachment is not re- using standard swelling protocols reported previously. The results sponsible for the decrease in the number of cells present on the hydro- are summarized in Table 1. While the total percent polymer was gels after 72 hours. Alternatively, differences in rate of proliferation held constant at 20% wt. the amount of MW 1000 Da and MW could explain a difference in cell number. An AlamarBlue assay was 20 000 Da was varied to create more elastic hydrogels. As expected, utilized as an indicator of cellular proliferation, and the results show the molecular weight between crosslinks and the pore sizes was significantly less metabolic activity in cells cultured on stiff hydrogels larger in the soft hydrogels compared to the stiff hydrogels. The compared to soft hydrogels and tissue culture plates at 3, 4 and equilibrium swelling ratio (Q) of the soft hydrogel formulations is 24 hours (Fig. 2D). At 24 hours, metabolic activity was significantly twice that of the stiff hydrogels. higher in cells cultured on tissue culture plates than in cells cultured on both stiff and soft hydrogels. Given that proliferation is slower on the stiff hydrogels, the expression of the multipotency marker sox2 hMSC attachment to hydrogels was analyzed to see if there were significant changes in multipotency. Bone marrow-derived hMSCs were seeded on the hydrogel scaffolds Cells were seeded at the same density on each surface and cultured for and after 72 hours a viability assay was performed to determine if 72 hours before collecting RNA. Results of qRT-PCR of sox2 the cells survived on each of the three surfaces: tissue culture plates, (Fig. 2E) indicates that there is no statistically significant difference in soft hydrogels and stiff hydrogels. Based on propidium iodide stain- expression levels between each surface, demonstrating that the elastic- ing (dead cells stained red), we observe few, if any, dead cells ity conditions do not immediately influence the levels of certain multi- (Fig. 1A). The difference in image brightness observed from the stiff potency transcription factors. hydrogels is attributed to the decreased porosity, which further obstructs visualization. The difference in brightness does not alter the number of live/dead cells. Effect of elasticity on hMSC osteogenic differentiation Cell morphology can be an indicator of cellular state, and To be useful in tissue engineering and regenerative medicine, bioma- changes to this morphology could indicate changes in cell behavior. terial scaffolds must be able to support and potentially direct stem Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Poly (ethylene glycol) hydrogel elasticity 171 Elasticity Tissue Culture Plate Stiff Hydrogel Soft Hydrogel Cell Count Analysis Tissue Culture Plate Stiff Hydrogels Soft Hydrogels Figure 1. hMSCs attach to and survive on the different hydrogel compositions. (A) Viability assay of hMSCs cultured on tissue culture plates, stiff hydrogels and soft hydrogels for 72 hours. Live cell nuclei are shown in blue, while dead cell nuclei are shown in red. (B) Morphology of hMSCs cultured on the three elasticity conditions for 72 hours. The cell nuclei are shown in blue, while the F-actin filaments are shown in red. (C) Visual depiction of ImageJ analysis highlighting nuclei for count. (D) Cell count results from ImageJ quantification of cells seeded for 72 hours. *Tukey HSD resulting P < 0.05. n¼ 3. Scale bars: 400 mm cell differentiation toward desired lineages. Elasticity can play a role expression, an early marker of osteogenesis, between soft hydrogels in directing stem cell state, thus the effects of the hydrogel elasticities and stiff hydrogels (P < 0.05), between soft hydrogels and tissue cul- on hMSC differentiation toward an osteogenic lineage were investi- ture plates (P < 0.05) and between stiff hydrogels and tissue culture gated. Osteogenic differentiation was chemically induced in hMSCs plates (P < 0.01). seeded on all three elasticity conditions, and morphology was ana- lyzed using phase contrast microscopy (Fig. 3A). Due to the limited visibility in phase contrast images with hydrogels, phalloidin stain- Effect of elasticity on hMSC adipogenic differentiation ing was also used to visualize F-actin filaments (Fig. 3B). There was Since hMSCs also have the potential to be used for adipogenic tissue noticeable differentiation and calcium deposition on all three elastic- regeneration, we further assessed adipogenesis of these cells on each ity conditions. qRT-PCR of osteogenic markers runx2 and alp of the selected surfaces. Adipogenic differentiation was chemically (Fig. 3C) was performed on samples collected at day 7 of differentia- induced in cells seeded on all three elasticity conditions, and mor- tion. Analysis indicated no significant differences in the early osteo- phology was analyzed using phase contrast microscopy (Fig. 4A). As genic differentiation marker runx2 expression in hMSCs cultured on in the previous set of experiments, phalloidin staining was used to each surface. However, there were significant differences in alp visualize F-actin filaments and provide higher resolution images of Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Average nuclei/image 172 Whitehead et al. Tissue Culture Plate Stiff Hydrogel Soft Hydrogel B C Image Image Image 3 0 Tissue Culture Plate Stiff Hydrogels Soft Hydrogels Tissue Culture Plate Stiff Gels Soft Gels sox2 Expression 0.8 ** ** 0.6 ** * ** 0.4 3000 ** ** 0.2 1 hour 2 hours 3 hours 4 hours 24 Hours Tissue Culture Plate Stiff Hydrogel Soft Hydrogel Figure 2. hMSCs attach more readily to hydrogels, but display decreased proliferation, despite equal expression of sox2.(A) Example images of DAPI stain hMSCs 18 hours post-seeding on the different elasticity conditions. (B) Schematic representation of imaging method used for attachment studies. Three images were taken (as shown in panel A) of each sample, with three samples per surface. (C) Results of ImageJ quantification of nuclei per image. (D) Results of AlamarBlue analysis of hMSCs cultured on each surface. AlamarBlue was added after cells were cultured for 72 hours, and timepoints shown in graph represent hours after AlamarBlue introduction. (E) Quantitative reverse-transcriptase PCR analysis of sox2 expression in hMSCs cultured on each elasticity condition for 72 hours. *Tukey HSD resulting P < 0.05. **Tukey HSD resulting P < 0.01. n¼ 3 for C, D and E. Scale bars: 1000 mm cell morphology (Fig. 4B). Noticeable differentiation had taken were lower in the stiff hydrogels than in the soft hydrogels. place on each surface, with round globules, some of which are indi- Variability in the swelling characteristics between the two samples cated by green arrows in Fig. 4A and B, indicating vacuoles and adi- was attributed to the higher percentage of PEGDMA MW 20 000 pogenic differentiation. Phalloidin staining of the cells shows that in within the soft hydrogels (Table 1). Next, we characterized hMSC areas where lipid vacuoles formed there is a decrease in F-actin fila- interactions when cultured on the stiff and soft hydrogels. As the ments. This trend is seen on tissue culture plates, soft hydrogels and largest pores are nanometers in size and hMSCs have an approxi- stiff hydrogels. qRT-PCR of early adipogenic markers ppar-y and mate diameter range of 17.9–30.4 lm, there is no penetration of srebp1c was performed on samples collected at day 7 of differentia- hMSCs into the hydrogel network. Thus, hMSCs are cultured two- tion. Analysis indicates no significant differences in expression of dimensionally on the surface of these hydrogels. these early adipogenic markers between each surface (Fig. 4C). When seeded on both soft and stiff hydrogels, hMSCs were shown to attach and remain viable (Fig. 1A), further confirming the potential of this platform for use in cell culture and tissue genera- Discussion tion. However, changes in morphology were observed in cells cul- Previously, we demonstrated that PEGDMA hydrogels with elastici- tured on the different elasticities. Human MSCs cultured on the soft ties within a physiologically relevant range (8–60 kPa) can be gener- hydrogels maintained similar morphology to the tissue culture plate ated by varying the molecular weight of the polymer [17]. Here, we controls. In contrast, hMSCs cultured on stiff hydrogels displayed a characterized hydrogels at the upper and lower ends of this range, more spindle-like morphology as compared to hMSCs cultured on specifically in terms of their swelling behavior and interactions with tissue culture plates or soft hydrogels (Fig. 1B). These differences in hMSCs. The swelling behavior of the hydrogels was used to deter- morphology could indicate a change in cell behavior, such as sponta- mine the molecular weight between crosslinks (M ), the pore sizes neous differentiation. Furthermore, there appeared to be consis- (n) and the equilibrium swelling ratio (Q). All three, as predicted, tently fewer hMSCs on the stiff hydrogels after 72 hours of culture. Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Average Number of Cells/Image Relative Expression Normalized to gapdh Poly (ethylene glycol) hydrogel elasticity 173 Tissue Culture Plate Stiff Hydrogel Soft Hydrogel Tissue Culture Plate Stiff Hydrogel Soft Hydrogel ** * 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 runx2 apl alp Figure 3. hMSCs retain the ability to differentiate toward osteogenic lineages on all three elasticity conditions. (A) Phase contrast images of hMSCs at day 7 of os- teogenic differentiation. (B) Morphology of hMSCs at day 7 of osteogenic differentiation, corresponding to the phase contrast images in panel A. The cell nuclei are shown in blue, while the F-actin filaments are shown in orange. (C) Quantitative reverse-transcriptase PCR analysis of the osteogenic differentiation markers runx2 and alp in hMSCs at day 7 of osteogenic differentiation. *Tukey HSD resulting P < 0.05. **Tukey HSD resulting P < 0.01. n¼ 3. Scale bars: 200 mm ImageJ quantification of the hMSCs shown in Fig. 1B revealed sig- marker sox2 revealed no significant differences in expression across nificantly fewer cells on the stiff hydrogels (Fig. 1D). The decrease in the three elasticity conditions (Fig. 2E). However, further analysis of hMSCs could be the result of decreased attachment to the stiff multipotency markers and markers of possible differentiation line- hydrogels. However, attachment analysis 18 hours after seeding ac- ages could reveal that a subtle amount of spontaneous differentia- tually revealed an increased number of hMSCs attached to both hy- tion has taken place or longer time course studies may demonstrate drogel elasticities compared to tissue culture plate indicating that more significant changes in multipotency. For the scope of this the hydrogels have an impact on cell proliferation (Fig. 2A–C). study, the differentiation potential of hMSCs on the three elasticity AlamarBlue assays demonstrated that proliferation is signifi- conditions was analyzed through chemically induced differentiation cantly decreased in the cells cultured on the soft and stiff hydrogels. toward osteogenic and adipogenic lineages, rather than exploring Human MSC proliferation on stiff hydrogels was shown to be signif- the long-term effects of maintenance on each of these surfaces. icantly decreased 3-hours after the introduction of AlamarBlue On all three elasticity conditions, hMSCs cultured in osteogenic (Fig. 2D). The differences in morphology and proliferation observed differentiation media differentiated toward the osteogenic lineage, in hMSCs cultured on stiff hydrogels could be an indication of as evidenced by calcium deposition and expression of bone specific spontaneous differentiation. Quantitative Reverse-Transcriptase – markers. There was no significant difference in runx2 expression, polymerase chain reaction (RT-PCR) analysis of the multipotency which is an essential transcription factor for osteoblastic Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Relative Expression Normalized to gapdh 174 Whitehead et al. Tissue Culture Plate Stiff Hydrogel Soft Hydrogel Tissue Culture Plate Stiff Hydrogel Soft Hydrogel 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 ppar-γ srebp-1c Figure 4. hMSCs retain the ability to differentiate toward adipogenic lineages on all three elasticity conditions. (A) Phase contrast images of hMSCs at day 7 of adipogenic differentiation. (B) Morphology of hMSCs at day 7 of adipogenic differentiation, corresponding to the phase contrast images in panel A. The cell nuclei are shown in blue, while the F-actin filaments are shown in orange. Cells containing lipid vesicles demonstrated a rearrangement of F-actin filaments, indicated by green arrows. (C) quantitative reverse-transcriptase PCR analysis of the adipogenic differentiation markers ppar-c and srebp1-c in hMSCs at day 7 of adipo- genic differentiation. Results considered insignificant with P > 0.05. n¼ 3. Scalebars: 200 mm differentiation (Fig. 3C). However, there was a significant decrease In summary, the data from this study give insight into the proper- in alp expression in hMSCs cultured on both hydrogel elasticities ties and stem cell interactions of our previously established hydrogel (Fig. 3C). While lower levels of alp expression could platform; an inexpensive, highly tailorable platform that can be indicate decreased osteogenesis, given the observation of calcium de- adapted to any number of cell-material interaction studies and applica- position and cell morphology, it is also possible that a decrease in tions. There is a need within the field to investigate the roles of both alp expression is an indication of more rapid maturation of the the physical and chemical properties of different biomaterials on a vari- resulting cells. ety of stem cells. This article presents a focused investigation into the hMSCs cultured in adipogenic differentiation media also interactions between hMSCs and our specific PEG-based hydrogel plat- retained the ability to differentiate toward adipogenic lineages on all form. Changes in hMSC morphology and proliferation were observed three elasticity conditions. hMSCs on all three elasticity conditions in cells cultured on hydrogels, primarily those cultured on stiff hydro- began forming lipid vesicles characteristic of adipogenic differentia- gels. These results demonstrate that the elastic tailorability of this hy- tion (Fig. 4A and B). There was no significant difference in the ex- drogel platform can produce changes in hMSC behavior and cell state, pression of two key transcription factors involved in adipogenesis, indicating a potential for these hydrogels to be used to generate a con- ppar-c and srebp1-c. Ultimately, no difference in hMSC differentia- trolled environment for cell culture and tissue regeneration applica- tion toward adipogenic lineages was observed. tions. Furthermore, based on the differentiation studies, the different Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Relative Expression Normalized to gapdh Poly (ethylene glycol) hydrogel elasticity 175 8. Holst J, Watson S, Lord MS et al. letters Substrate elasticity provides me- hydrogel elasticities have subtle effects on stem cell differentiation. chanical signals for the expansion of hemopoietic stem and progenitor This effect is observed primarily in osteogenic differentiation, which cells. Nat Biotechnol 2010;28:1123–8. could indicate that cell lineages of higher elasticity are more susceptible 9. Saha K, Keung AJ, Irwin EF et al. Substrate modulus directs neural stem to elasticity changes. The results of this study further suggest that the cell behavior. Biophys J 2008;95:4426–38. hydrogel platform’s elasticity does affect stem cell behavior, opening 10. Yang C, DelRio FW, Ma H et al. Spatially patterned matrix elasticity the door to future investigations of the platform’s potential for control- directs stem cell fate. Proc Natl Acad Sci 2016;113:E4439–45. ling stem cell fate. While the platform is limited in terms of its ability 11. Wen JH, Vincent LG, Fuhrmann A et al. Interplay of matrix stiffness and to be tailored for responsiveness and degradation, the limitations are protein tethering in stem cell differentiation. Nat Mater 2014;13:979–87. appropriate for current and future studies focused on elasticity and 12. Annabi N, Tamayol A, Uquillas JA et al. 25th anniversary article: rational other specific cues, while maintaining affordable and reproducible bio- design and applications of hydrogels in regenerative medicine. Advanced Materials 2014;26:85–124. doi: 10.1002/adma.201303233. material scaffolds. In addition, PEG is an US Food and Drug 13. Bryant SJ, Anseth KS. Controlling the spatial distribution of ECM compo- Administration (FDA)-approved polymer, which gives the platform the nents in degradable PEG hydrogels for tissue engineering cartilage. potential to be used in the clinic, should future studies open the door J Biomed Mater Res A 2003;64A:70–9. for therapeutic uses of these hydrogels. The focus of this study was to 14. Elisseeff J, Anseth K, Sims D et al. Transdermal photopolymerization for understand cell–biomaterial interactions, with the goal of adding to a minimally invasive implantation. Proc Natl Acad Sci USA 1999;96:3104–7. library of materials for various applications in regenerative medicine. 15. Karmaker AC, Dibenedetto A, Goldberg AJ. Extent of conversion and its ef- A better understanding of the biomaterial scaffolds utilized in tissue re- fect on the mechanical performance of Bis-GMA/PEGDMA-based resins and generation, such as the studies shown here, is essential in optimizing their composites with continuous glass fibres. J Mater Sci 1997;8:333–401. their translational and clinical potential. 16. Zhu J. Bioactive modification of poly(ethylene glycol) hydrogels for tissue engineering. Biomaterials 2010;31:4639–56. 17. Patel NR, Whitehead AK, Newman JJ et al. Poly(ethylene glycol) hydro- gels with tailorable surface and mechanical properties for tissue engineer- Acknowledgements ing applications. ACS Biomater Sci Eng 2016;3:1494. The authors are grateful to Dr. Bruce Bunnell from Tulane University, for guid- 18. Asakura A, Komaki M, Rudnicki M. Muscle satellite cells are multipoten- ance and provision of hMSCs and to Dr. Jeff Gimble from Tulane University tial stem cells that exhibit myogenic, osteogenic, and adipogenic differen- and LaCell, LLC for guidance and reagents. This work was funded by the tiation. Differentiation 2001;68:245–53. Louisiana Biomedical Research Network (LBRN) NIH INBRE P20GM103424 19. Bieback K, Kern S, Klu ¨ ter H et al. Critical parameters for the isolation of and the Center for Cardiovascular Diseases and Sciences at LSU Health mesenchymal stem cells from umbilical cord blood. Stem Cells 2004;22: Shreveport in the form of the Partners Across Campus grants. We also received 625–34. funding from LaSPACE and NASA through a LaSPACE Research Enhancement 20. Lee OK, Kuo TK, Chen WM et al. Isolation of multipotent mesenchymal Award, LaSPACE Undergraduate Research Assistantship and two Graduate stem cells from umbilical cord blood. Blood 2004;103:1669–75. Student Research Assistantships, NASA Space Grant Award NNX158H828. 21. Zuk PA, Zhu M, Ashjian P et al. Human adipose tissue is a source of mul- We also acknowledge Louisiana Tech University’s College of Applied and tipotent stem cells. Molecular Biology of the Cell 2002;13:4279–95. Natural Sciences for support in the form of student minigrants. We would like 22. Mackay AM, Beck SC, Murphy JM et al. Chondrogenic differentiation of to thank both Newman and Caldorera-Moore labs for help and support, with cultured human mesenchymal stem cells from marrow. Tissue Eng 1998; special thanks to India Pursell and Rachel Eddy for assistance with cell culture 4:415–28. maintenance and Nehal Patel for technical support. 23. Pittenger MF, Mackay AM, Beck SC et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284. http://science.science mag.org/content/284/5411/143. 24. NIH. 645 Studies found for: Mesenchymal stem cells. https://clinicaltrials. Supplementary data gov/ct2/results?cond=mesenchymal+stem+cells&term=&cntry=&state=& city=&dist= (accessed 1 February 2018). Supplementary data are available at REGBIO online. 25. Dezawa M, Ishikawa H, Itokazu Y et al. Bone marrow stromal cells gener- Conflict of interest statement. None declared. ate muscle cells and repair muscle degeneration. Science (New York, N.Y.) 2005;309:314–7. 26. Fan H, Liu H, Toh SL et al. Enhanced differentiation of mesenchymal References stem cells co-cultured with ligament fibroblasts on gelatin/silk fibroin hy- brid scaffold. Biomaterials 2008;29:1017–27. 1. Ba ´ rtolo PJ, Chua CK, Almeida HA et al. Biomanufacturing for tissue engi- 27. Patrick CW. Tissue engineering strategies for adipose tissue repair. Anat neering: present and future trends. Virtual Phys Prototyp 2009;4:203–16. Rec 2001;263:361–6. 2. Lutolf MP, Hubbell JA. Synthetic biomaterials as instructive extracellular 28. Ponticiello MS, Schinagl RM, Kadiyala S et al. Gelatin-based resorbable microenvironments for morphogenesis in tissue engineering. Nat sponge as a carrier matrix for human mesenchymal stem cells in cartilage Biotechnol 2005;23:47–55. regeneration therapy. J Biomed Mater Res 2000;52:246–55. 3. Discher DE. Tissue cells feel and respon to the stiffness of their substrate. 29. Yousefi AM, James PF, Akbarzadeh R et al. Prospect of stem cells in bone Science 2005;310:1139–43. tissue engineering: a review. Stem Cells Int 2016;2016:1. 4. Banerjee A, Arha M, Choudhary S et al. The influence of hydrogel modu- 30. Bryant SJ, Bender RJ, Durand KL et al. Encapsulating chondrocytes in lus on the proliferation and differentiation of encapsulated neural stem degrading PEG hydrogels with high modulus: engineering gel structural cells. Biomaterials 2009;30:4695–9. changes to facilitate cartilaginous tissue production. Biotechnol Bioeng 5. Engler AJ, Rehfeldt F, Sen S et al. Microtissue elasticity: measurements by 2004;86:747–55. atomic force microscopy and its influence on cell differentiation. Methods 31. Caldorera-Moore M, Kang MK, Moore Z et al. Swelling behavior of Cell Biol 2007;83:521–45. nanoscale, shape- and size-specific, hydrogel particles fabricated using im- 6. Engler AJ, Sen S, Sweeney HL et al. Matrix elasticity directs stem cell line- print lithography. Soft Matter 2011;7:2879. age specification. Cell 2006;126:677–89. 32. Bryant SJ, Nuttelman CR, Anseth KS. Cytocompatibility of UV and visible 7. Gilbert PM, Havenstrite KL, Magnusson KEG et al. Substrate elasticity light photoinitiating systems on cultured NIH/3T3 fibroblasts in vitro. regulates skeletal muscle stem cell self-renewal in culture. Science 2010; J Biomater Sci 2000;11:439–57. 329:1078–81. Downloaded from https://academic.oup.com/rb/article-abstract/5/3/167/4984511 by Ed 'DeepDyve' Gillespie user on 21 June 2018

Journal

Regenerative BiomaterialsOxford University Press

Published: Apr 24, 2018

There are no references for this article.