Hydrogels as artificial matrices for human embryonic stem cell self-renewal Ying J. Li,1 Eugene H. Chung,1 Ryan T. Rodriguez,2 Meri T. Firpo,3 Kevin E. Healy1,4 Department of Bioengineering, University of California Berkeley, Berkeley, California 2 Department of Obstetrics, Gynecology and Reproductive Sciences, Center for Reproductive Sciences, University of California San Francisco, San Francisco, California 3 Department of Medicine, Stem Cell Institute, University of Minnesota, Minneapolis, Minnesota 4 Department of Material Science and Engineering, University of California Berkeley, Berkeley, California
1
Received 14 December 2005; accepted 12 January 2006 Published online 1 June 2006 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.a.30732 Abstract: Human embryonic stem cells (hESCs) have the potential to differentiate into all cell types in the body and hold great promise for regenerative medicine; however, large-scale expansion of undifferentiated hESCs remains a major challenge. Self-renewal of hESCs requires culturing these cells on either mouse or human fibroblast cells (i.e., a feeder layer of cells), or on artificial extracellular matrices (ECMs) while supplementing the media with soluble growth factors. Here we report a completely synthetic ECM system composed of a semi-interpenetrating polymer network (sIPN), a polymer hydrogel, which was designed to allow the independent manipulation of cell adhesion ligand
presentation and matrix stiffness. In the short term, hESCs that were cultured on the sIPN adhered to the surface, remained viable, maintained the morphology, and expressed the markers of undifferentiated hESCs. This was the first demonstration that a completely synthetic ECM can support short-term self-renewal of hESCs. © 2006 Wiley Periodicals, Inc. J Biomed Mater Res 79A: 1–5, 2006
INTRODUCTION
by changes in morphology, loss of embryonic stem cell markers such as OCT-4 and stage-specific embryonic antigen (SSEA)-4, and loss of pluripotency.1,2 Compared with the cell-based feeder systems, artificial ECMs offer several advantages, including reduced risk of pathogen transmission and ease of scale-up. Efforts to understand the hESC cell-ECM interactions have included using animal-derived proteins such as Matrigel™ and laminin; however, these systems do not support self-renewal of some hESC lines3,5 and modifications of the biochemical and mechanical properties of these materials are limited.6 Here we report a completely synthetic ECM system composed of a semi-interpenetrating polymer network (sIPN), a polymer hydrogel, which was designed to allow the independent manipulation of cell adhesion ligand presentation and matrix stiffness.7,8 The sIPNs used in this study were hydrophilic, swelled in aqueous media, and mimicked native ECMs in several important ways. By varying the polymer components, sIPNs with a range of matrix stiffness and cell-adhesion ligand densities were created. Matrix stiffness and cell-adhesion ligand density were important design criteria for artificial ECMs, since
Human embryonic stem cells (hESCs) have the potential to differentiate into all cell types in the body and hold great promise for regenerative medicine if large-scale expansion of undifferentiated hESCs can be achieved.1,2 Self-renewal, i.e. undifferentiated growth, of hESCs requires culturing these cells with either mouse or human embryonic fibroblasts (MEFs),3 or on artificial extracellular matrices (ECMs) while supplementing the media with soluble growth factors.4 When hESCs are cultured in the absence of MEFs and directly on polystyrene tissue culture plates, the cells quickly undergo spontaneous differentiation. Spontaneous differentiation is characterized Correspondence to: K. E. Healy, 370 Hearst Memorial Mining Building, #1760, Berkeley, CA 94720; e-mail:
[email protected] Contract grant sponsor: National Institute of Health Grant; contract grant numbers: NIH AR47304, NIH RR017498 Contract grant sponsor: National Defense Science and Engineering Graduate Fellowship © 2006 Wiley Periodicals, Inc.
Key words: human embryonic stem cells; self-renewal; hydrogels; interpenetrating networks; artificial extracellular matrices
2
LI ET AL.
these parameters have been shown to affect cell responses such as adhesion, migration, and differentiation in the other cell types.9 –11 Matrix stiffness was controlled by the density of the network crosslinker; and specific cell-matrix interactions were promoted by the presentation of peptide sequences that mimicked the active cell adhesion domains on native ECM proteins. In the short term, hESCs that were cultured on the sIPN adhered to the surface, remained viable, maintained the morphology, and expressed the markers of undifferentiated hESCs. This was the first demonstration that a completely synthetic ECM can support short-term self-renewal of hESCs.
MATERIALS AND METHODS
polymerization scheme and a schematic representation of the polymerized sIPN network are depicted in Figure 1(a,b).
Characterization of the sIPNs The viscoelastic properties of the sIPNs were characterized by dynamic oscillatory shear measurements, using a parallel plate rheometer (Paar Physica MCR 300). The complex modulus, G* , and loss angle were determined by measuring the storage modulus (G⬘) and loss modulus (G⬙) over a frequency range of 0.001–10 Hz. The sIPNs were previously found to undergo a lower critical solution temperature (LCST) at 34°C.13 The LCST phase transition was determined using an UV–vis spectrophotometer by monitoring the transmittance of visible light ( ⫽ 500 nm) as a function of temperature.
Synthesis of the sIPNs Cell culture The sIPN consisted of poly(N-isopropylacrylamide-coacrylic acid) [p(NIPAAm-co-AAc)] that was loosely crosslinked with an acrylated peptide Gln-Pro-Gln-Gly-Leu-Ala-Lys-NH2 (QPQGLAK-NH2), a sequence designed to be cleaved by matrix metalloproteinase-13 (MMP-13) and other collagenases.8 The peptide crosslinker (QPQGLAK-NH2) (American Peptide Co.) was designed to match residues 904 –908 of human type II collagen with the addition of a glutamine residue to promote solubility and a lysine residue to provide amine functional groups for modification. Bifunctional acryl groups were introduced to the peptide via reaction with acryloyl chloride (Aldrich). To promote cell adhesion, the polymer network was interpenetrated by polyacrylic acid-graft-Ac-CGGNGEPRGDTYRAY-NH2 [p(AAc)-g-RGD] linear polymer chains. Linear p(AAc) chains (MW 450,000) (Polysciences) were modified with synthetic peptides (Ac-CGGNGEPRGDTYRAY-NH2) (American Peptide Co.).7 This RGD motif represents an active site in a number of extracellular proteins and binds to several integrin receptors, including ␣1, ␣v, 1, and ␣v3.12 Maleimide side groups were first grafted to the p(AAc) linear chains, and the RGD peptides were subsequently grafted to the maleimide side groups. The concentration of the grafted RGD was found to be 36.5 mol per gram of p(AAc)-g-RGD by fluorescence measurements using a FITC-conjugated RGD peptide. The sIPNs were synthesized by redox radical polymerization in an aqueous solution at room temperature, with molar ratios of 97:3:0.4 for NIPAAm:AAc:crosslinker, respectively. For polymer synthesis, nitrogen gas was bubbled through a mixture of NIPAAm (Polysciences), AAc (Polysciences), [p(AAc)-g-RGD], and peptide crosslinker in phosphate-buffered saline (PBS) for 15 min to remove dissolved oxygen. The sIPN polymerization was initiated using ammonium peroxydisulfate (AP) (Fisher) and N,N,N⬘,N⬘-tetramethylenediamine (TEMED) (Polysciences). The mixture was stirred vigorously for 15 s and allowed to polymerize at room temperature for 24 h. Prior to use, sIPNs were washed 3 times in ultrapure water to remove unreacted compounds, sterilized with 70% ethanol, and again washed 3 times to remove the ethanol.8 The
Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a
The HSF-6 hESC cell line, a federally approved line derived at UCSF,14 was obtained from UCSF under a Materials Transfer Agreement. The standard hESC culture conditions were used as the positive control. The hESC cultures were incubated at 37°C, in 10% CO2,14 and cultured on MEFs derived from the CF-1 strain mice (Charles River). NIH guidelines for the care and use of laboratory animals have been observed. The MEFs were mitotically inactivated using gamma irradiation and cultured on gelatin (collagen derivative) adsorbed to tissue culture polystyrene15 (Falcon). Complete culture medium (KSR) consisted of the following: Knockout-DMEM (Gibco), 20% Knockout Serum Replacement (Gibco), 2 mM Glutamine (Gibco), 0.1 mM nonessential amino acids (NEAA) (Gibco), 0.1 mM -Mercaptoethanol (Sigma), and 4 ng/mL basic fibroblast growth factor (FGF)-2 (R&D Systems). To test the ability of synthetic ECMs to maintain hESCs, hESCs were cultured without MEFs on sIPNs. The hESC colonies were maintained in conditioned KSR media, which consisted of incubating KSR on MEFs for 24 h so that the secreted signaling molecules from MEFs could be transferred to the hESCs. We chose to use the conditioned KSR medium so that we could focus on the cell-matrix interactions, while providing the yet unknown set of soluble factors necessary for self-renewal. Ultimately, the goal of this work is to identify the media constituents once the ECM conditions are defined.
Immunofluorescence staining of hESCs The hESC samples were washed, fixed with 2% paraformaldehyde, and permeabilized with 0.1% Triton X-100. The samples were blocked with serum, and incubated with the primary antibodies against OCT-4 (Santa Cruz) and SSEA-4 (Chemicon) overnight. Then the cultures were washed incubated with FITC-conjugated secondary antibodies (Santa Cruz).14 Images were acquired using a Nikon Eclipse TS100 microscope with a Pixera 600 CL-CU camera. Photoshop was used for image preparation.
HYDROGELS FOR HESC SELF-RENEWAL
3
Figure 1. sIPN synthesis: poly(N-isopropylacrylamide-co-acrylic acid) [p(NIPAAm-co-AAc)], interpenetrated by polyacrylic acid-graft-Arg-Gly-Asp [p(AAc)-g-RGD] linear polymer chains; (a) polymerization scheme, (b) schematic representation of the polymerized sIPN network, and (c) representative measurement of complex shear modulus ( G* ) as a function of frequency at 22°C and 37°C.
RESULTS AND DISCUSSION Experiments designed to assess hESC self-renewal employed sIPNs with RGD ligand concentrations ranging from 0 to 150 M, while holding all other conditions constant. Rheological measurements were repeated three times for each sIPN, with a total of n ⫽ 32 when the measurements for all the sIPNs were combined. At 1 Hz and 22°C, the mean complex shear modulus ( G* ) was 70 Pa ⫾ 27 (SD); at 1 Hz and 37°C, the mean G* was 139 Pa ⫾ 71 (SD). Figure 1(c) shows a representative G* measurement. To compare the ability of sIPNs versus MEFs to support the self-renewal of hESCs, several characteristics were assessed: colony attachment, colony morphology, cell viability, and the presence of hESC markers. Morphological changes were one of the early indicators of differentiation. Undifferentiated hESC colonies that were cultured on MEFs (positive control) are shown in Figures 2(a) and 3(a). Undifferentiated hESCs exhibited high nucleus to cytoplasm ratio, formed tightly packed colonies with defined colony
borders, and expressed embryonic stem cell markers such as the transcription factor OCT-4, and surface carbohydrate moieties SSEA-3 and SSEA-4. Figures 2(e) and 3(e) show hESC colonies that were cultured on gelatin-adsorbed tissue-culture polystyrene (negative control). Under these conditions, very few colonies were able to attach to the culture substrate. Once attached, the cells underwent spontaneous differentiation. Morphologically, differentiated hESC colonies had indistinct colony borders, with larger cells that migrated away from the colony. These cells often took on spindle-like fibroblastic shapes or developed long processes. In contrast, hESCs cultured on the sIPNs [Figs. 2(c) and 3(c)] exhibited morphologies similar to those of undifferentiated hESCs cultured on MEFs [Figs. 2(a) and 3(a)], where colonies had distinct borders with small (⬃10 m diameter) and tightly packed cells. Immunofluorescence staining was conducted to assess whether cells retained markers of undifferentiated hESCs. The POU family transcription factor OCT-4 is a highly specific and necessary marker for Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a
4
LI ET AL.
Figure 2. Morphology and OCT-4 immunofluorescence of hESCs at Day 5. (a, b) hESCs cultured on MEFs exhibited small, tightly packed cells with distinct colony borders. (c, d) hESCs cultured on sIPN ( G* ⬃70 Pa, 150 M RGD) exhibited similar morphologies when compared with (a, b). (e, f) hESCs cultured on gelatin-adsorbed polystyrene exhibited morphologies of spontaneously differentiating cells, with spindle-shaped cells and indistinct colony borders. OCT-4 was present in some cells under all three conditions. However, note that in hESCs cultured on polystyrene (f), white arrows point to cells beyond the colony edge which were not positive for OCT-4.
undifferentiated hESCs, and SSEA-4 is a glycolipid cell surface antigen strongly expressed in undifferentiated hESCs.2 Results showed the presence of OCT-4 and SSEA-4 in cultures of all three conditions at day 5 (Figs. 2 and 3). This indicated that even under suboptimal culture conditions, some hESCs did not completely lose their undifferentiated characteristics after 5 days.16 However, for the hESCs cultured to gelatinadsorbed polystyrene [Fig. 2(f)], cells beyond the edge of the colony were not positive for OCT-4, indicating that they had spontaneously differentiated. By comparison, the hESCs cultured on sIPNs [Fig. 2(d)] were within a tight border and were positive for OCT-4.
Interestingly, the OCT-4 fluorescence appeared somewhat diffuse in the center region of the colony. We attributed this result to competing fluorescence from out-of-focus cell layers in the colony. In addition, cell viability was examined using calcein-AM stain (Molecular Probes) and the hESCs cultured on the sIPN were found to be viable (data not shown). This indicated that the substrate material was not toxic to hESCs. Finally, Figure 4 shows the morphology of hESC colonies that were cultured on the sIPN with various ligand concentrations. At 0 M RGD concentration, very low hESC adhesion was observed. At 45 M RGD concentration, colony mor-
Figure 3. Morphology and SSEA-4 immunofluorescence of hESCs at Day 5. (a, b) hESCs cultured on MEFs. (c, d) hESCs cultured on sIPN ( G* ⬃70 Pa, 45 M RGD). (e, f) hESCs cultured on gelatin-adsorbed polystyrene. SSEA-4 was present in colonies under all three conditions.
Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a
HYDROGELS FOR HESC SELF-RENEWAL
5
References 1.
2.
3.
4.
5.
Figure 4. hESCs cultured on sIPN of various RGD adhesion ligand concentrations. (a, b, c, d) ⫽ 0, 45, 105, 150 M, respectively. At 0 M RGD concentration, very low hESC adhesion was observed. At 45 M RGD concentration, colony morphology was highly variable, where some colonies exhibited tight borders while other did not. Qualitatively, hESCs cultured on sIPNs of higher RGD concentrations (105 and 150 M) exhibited morphologies most similar to undifferentiated hESCs. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]
phology was highly variable, where some colonies exhibited tight borders while other did not [Figs. 3(c) and 4(b)]. Qualitatively, hESCs cultured on sIPNs of higher RGD concentrations (105 and 150 M) [Figs. 4(c,d)] exhibited morphologies most similar to undifferentiated hESCs.
6.
7.
8.
9.
10.
11.
CONCLUSION In summary, we have demonstrated the first completely synthetic ECM environment that can support short-term hESC self-renewal. These systems have advantages of ease of scale-up and decreased risks of disease transmission from unknown pathogens and foreign oligosaccharide residues picked up from mouse feeder cells.17 Furthermore, the sIPN provides a three-dimensional ECM environment where the matrix stiffness and ligand density can be independently tuned. Given that long-term self-renewal is paramount to myriad applications using hESCs, our next step is to optimize the system to support hESC selfrenewal in the long-term and controlled differentiation into specific cell types for applications in regenerative medicine.
12.
13.
14.
15.
16.
17.
The authors thank Megan Bodnar and Rina Seerke for their technical advice.
Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, Jones JM. Embryonic stem cell lines derived from human blastocysts. Science 1998;282(5391):1145– 1147. Reubinoff BE, Pera MF, Fong CY, Trounson A, Bongso A. Embryonic stem cell lines from human blastocysts: Somatic differentiation in vitro. Nat Biotechnol 2000;18(4):399 – 404. Richards M, Fong CY, Chan WK, Wong PC, Bongso A. Human feeders support prolonged undifferentiated growth of human inner cell masses and embryonic stem cells. Nat Biotechnol 2002;20(9):933–936. Xu C, Inokuma MS, Denham J, Golds K, Kundu P, Gold JD, Carpenter MK. Feeder-free growth of undifferentiated human embryonic stem cells. Nat Biotechnol 2001;19(10):971–974. Cheng L, Hammond H, Ye Z, Zhan X, Dravid G. Human adult marrow cells support prolonged expansion of human embryonic stem cells in culture. Stem Cell 2003;21(2):131–142. Semler EJ, Ranucci CS, Moghe PV. Mechanochemical manipulation of hepatocyte aggregation can selectively induce or repress liver-specific function. Biotechnol Bioeng 2000;69(4): 359 –369. Stile RA, Burghardt WR, Healy KE. Synthesis and characterization of injectable poly(n-isopropylacrylamide)-based hydrogels that support tissue formation in vitro. Macromolecules 1999;32(22):7370 –7379. Kim S, Healy KE. Synthesis and characterization of injectable poly(N-isopropylacrylamide-co-acrylic acid) hydrogels with proteolytically degradable cross-links. Biomacromolecules 2003;4(5):1214 –1223. Engler AJ, Griffin MA, Sen S, Bonnemann CG, Sweeney HL, Discher DE. Myotubes differentiate optimally on substrates with tissue-like stiffness: Pathological implications for soft or stiff microenvironments. J Cell Biol 2004;166(6):877– 887. Lo CM, Wang HB, Dembo M, Wang YL. Cell movement is guided by the rigidity of the substrate. Biophys J 2000;79(1): 144 –152. Yeung T, Georges PC, Flanagan LA, Marg B, Ortiz M, Funaki M, Zahir N, Ming W, Weaver V, Janmey PA. Effects of substrate stiffness on cell morphology, cytoskeletal structure, and adhesion. Cell Motil Cytoskeleton 2005;60(1):24 –34. Rezania A, Healy KE. Integrin subunits responsible for adhesion of human osteoblast-like cells to biomimetic peptide surfaces. J Orthop Res 1999;17(4):615– 623. Kim S, Chung EH, Gilbert M, Healy KE. Synthetic MMP-13 degradable ECMs based on poly(N-isopropylacrylamide-coacrylic acid) semi-interpenetrating polymer networks. I. Degradation and cell migration. J Biomed Mater Res A 2005;75(1): 73– 88. Bodnar MS, Meneses JJ, Rodriguez RT, Firpo MT. Propagation and maintenance of undifferentiated human embryonic stem cells. Stem Cells Dev 2004;13(3):243–253. BD Falcon Cell Culture Products. Falcon uses a vacuum gasplasma process to create hydrophilic surfaces on tissue-culture polystrene. www.bdbiosciences.com. Abeyta MJ, Clark AT, Rodriguez RT, Bodnar MS, Pera RA, Firpo MT. Unique gene expression signatures of independently-derived human embryonic stem cell lines. Hum Mol Genet 2004;13(6):601– 608. Martin MJ, Muotri A, Gage F, Varki A. Human embryonic stem cells express an immunogenic nonhuman sialic acid. Nat Med 2005;11(2):228 –232.
Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a
