Harnessing traction-mediated manipulation of the cell/matrix interface to control stem-cell fate

Abstract

Stem cells sense and respond to the mechanical properties of the extracellular matrix. However, both the extent to which extracellular-matrix mechanics affect stem-cell fate in three-dimensional microenvironments and the underlying biophysical mechanisms are unclear. We demonstrate that the commitment of mesenchymal stem-cell populations changes in response to the rigidity of three-dimensional microenvironments, with osteogenesis occurring predominantly at 11-30 kPa. In contrast to previous two-dimensional work, however, cell fate was not correlated with morphology. Instead, matrix stiffness regulated integrin binding as well as reorganization of adhesion ligands on the nanoscale, both of which were traction dependent and correlated with osteogenic commitment of mesenchymal stem-cell populations. These findings suggest that cells interpret changes in the physical properties of adhesion substrates as changes in adhesion-ligand presentation, and that cells themselves can be harnessed as tools to mechanically process materials into structures that feed back to manipulate their fate.

Fig. 1. Matrix Compliance Alters Mesenchymal Stem Cell Fate in 3D Matrix Culture

(A–C). In-situ staining of encapsulated clonally derived mMSC (D1) for Alkaline Phosphatase (ALP) activity (Fast blue; osteogenic biomarker, blue) and neutral lipid accumulation (Oil Red O; adipogenic biomarker, red) after 1 week of culture in the presence of combined osteogenic and adipogenic chemical supplements within encapsulating matrices comprised of (A) RGD-modified alginate, (B) RGD-modified agarose, or (C) RGD-modified PEGDM hydrogels presenting 754µM RGD with varying E. (D). Immunofluorescence staining for OCN (green) and the nuclear counterstain DAPI (blue) in cryosectioned alginate matrices of varying E containing mMSC. (E). Western analysis of osteogenic (Cbfa-1, OPN) and adipogenic (PPAR-γ, Adn) protein expression in mMSC cultured in RGD-alginate hydrogels for 1 week. (F). Western analysis of Cbfa-1 and PPAR-γ protein expression in primary hMSC after 1 week of 3D culture within alginate matrices in which both E and RGD density were varied in parallel. E values shown are for hydrogels after 1 day in culture, after which point no changes in E occur. Scale bars: (A); 100µm, (B,C); 50µm, (D); 20µm.

Fig. 2. Cell and Nuclear Morphology are not Strongly Correlated to Mechanics of 3D Matrices

(A–C). Representative micrographs showing cross-sections of mMSC 2 hr after encapsulation into 3D alginate matrices with varying E and constant (754 µM) RGD density, visualized by (A) Differential-Interference Contrast (DIC), (B) F-Actin staining (Alexa Fluor 568-Phalloidin), or (C) Nuclear staining (Ethidium Homodimer). E values shown are for hydrogels at the time of cell encapsulation. Scale bars: 10µm.

Figure 3. Mechanically-Controlled α₅-integrin-RGD bond Formation Correlates with Stem Cell Osteogenic Lineage in 3D

(A). Immunofluorescence analysis of α₅ and α_V integrins bound to ECM in mMSC adherent to FN or VN coated glass (2D culture). (B). Localization of EGFP-α₅-integrins or naked EGFP in mMSC encapsulated into 3D alginate matrices with or without RGD. Note, α₅-integrins appear clustered within cells and localized to the cell-matrix interface at the periphery of confocal cross-sections in RGD-modified matrices (arrowhead). (C). ELISA quantification of α₅ () and α_V () integrin binding to RGD-biotin presented at varying density by either 2D or 3D alginate matrices (* p < 0.01, t-test). (D). α₅-integrin-RGD bond formation in matrices with varying stiffness presenting either 37 µM () or 754 µM () RGD-biotin (* p < 0.01, t-test). α₅-integrin binding to matrices presenting 754µM RGE-biotin was negligible. Error bars are SEM (n = 4–5). Scale bars: (A), 20µm; (B), 5µm.

Figure 4. Cell-RGD Bond Formation Exhibits a Biphasic Dependence on Matrix Stiffness

(A). Representative confocal images of fluorescein (green) and TAMRA-RGD emission from mMSC encapsulated into hydrogels presenting 377 µM RGD, RGD-TAMRA or RGE-TAMRA. The cell-matrix interface is shown at high resolution in insets i and ii. (B). Calculated N_b for mMSC in matrices where available RGD density and elastic modulus were varied in parallel. (C). Response surface depicting N_b as a function of E and N_RGD / cell revealed significant effects of both RGD density and the interaction between RGD density and elastic modulus (2-way ANOVA; p < 0.01). (D). Curve of N_b versus E generated from FRET studies using matrices formed from various alginate polymers and crosslinking agents presenting a constant density (37 µM) of RGD. (E). Curve of N_b versus E for untreated cells (), or cells treated with either 20mM BDM () or 20µg/mL cyclohexamide (▲) encapsulated into matrices presenting 37µM RGD. FRET analyses of cell-RGD bonds were performed 2 hr after encapsulating cells, and E values shown are for hydrogels at the time of cell encapsulation. Scale bars: (B); 10µm.

Figure 5. Cell-Traction Mediated Reorganization of Ligands Presented by Synthetic ECM

(A). Schematic of the method used to calculate the minimum depth of protrusion of integrin receptors into RGD-modified alginate. As shown, cells were assumed to be spherical with R_cell equal to 10µm. The equation used to calculate the minimum receptor penetration depth, h, is shown. (B). Measurements of mMSC-RGD bond number (N_b) in 3D matrices presenting either 15 µM () or 150 µM () RGD were used to calculate h. (C). Schematic depicting enhanced cell-RGD bond formation due to nanoscale RGD clustering mediated by cell traction forces. (D). Schematic of FRET assay to monitor cell-traction mediated nanoscale RGD-clustering of RGD-CFsC and RGD-TAMRA attached to different alginate chains. (E). FRET measurements of nanoscale RGD-clustering by encapsulated mMSC (* p < 0.01 compared to other conditions, Holm-Bonferonni test). FRET analyses of integrin ligation and nanoscale-RGD reorganization were performed 2 hr after encapsulating cells, and E values shown are for hydrogels at the time of cell encapsulation. Schematic drawings are not meant to be to scale. Error bars are SEM for calculated protrusion depth calculations (n = 3–5) and SD (n = 3) for clustering measurements.

Figure 6. Long Term Regulation of Osteogenic Commitment and Role of Specific Integrins in Stem Cell Fate in 3D Matrices

(A). Western analysis of matrix synthesis over the time-course of mMSC culture in matrices with varied rigidity which present 754µM RGD. (B). Normalized osteocalcin (OCN) secretion by mMSC after 3 weeks of 3D matrix culture. (C). Western analysis of Cbfa-1 expression in mMSC after 3 weeks of matrix culture. (D). Histologic analysis of encapsulated mMSC cultured for 1 week in matrices of varying stiffness but constant RGD density (754µM) in which α₅-RGD bonds or α_V-RGD bonds were inhibited with 50µg/mL function blocking antibodies: In-situ staining for Alkaline Phosphatase activity (Fast blue) and Neutral Lipid accumulation (Oil Red O) (left) or OCN immunofluorescence (green) and DAPI nuclear counterstain (blue) of cells in cryosectioned matrices (right). E values shown are for hydrogels after 1 day in culture. Error bars are SD (n = 3). Scale bars: ALP/ORO stain: 100µm; immunofluorescence: 20µm.