Three-dimensional morphogenesis of MDCK cells induced by cellular contractile forces on a viscous substrate

Abstract

Substrate physical properties are essential for many physiological events such as embryonic development and 3D tissue formation. Physical properties of the extracellular matrix such as viscoelasticity and geometrical constraints are understood as factors that affect cell behaviour. In this study, we focused on the relationship between epithelial cell 3D morphogenesis and the substrate viscosity. We observed that Madin-Darby Canine Kidney (MDCK) cells formed 3D structures on a viscous substrate (Matrigel). The structures appear as a tulip hat. We then changed the substrate viscosity by genipin (GP) treatment. GP is a cross-linker of amino groups. Cells cultured on GP-treated-matrigel changed their 3D morphology in a substrate viscosity-dependent manner. Furthermore, to elucidate the spatial distribution of the cellular contractile force, localization of mono-phosphorylated and di-phosphorylated myosin regulatory light chain (P-MRLCs) was visualized by immunofluorescence. P-MRLCs localized along the periphery of epithelial sheets. Treatment with Y-27632, a Rho-kinase inhibitor, blocked the P-MRLCs localization at the edge of epithelial sheets and halted 3D morphogenesis. Our results indicate that the substrate viscosity, the substrate deformation, and the cellular contractile forces induced by P-MRLCs play crucial roles in 3D morphogenesis.

Figure 1. Measurement of the matrigel viscosity after genipin treatment.

(a) Schematic representation of matrigel viscosity measurement conditions. The matrigel used for the measurement was poured into a cylinder built from a plastic tube and a cover glass prior to solidification. After solidification, the gel surface was filled with medium. The radius of the cylinder is 1.5 mm, and its length is approximately 50 mm. The radius of the stainless ball is 0.75 × 10⁻³ m. The cylinders were sealed with oil clay to prevent changes in the medium pH. (b) Kymographs of the stainless balls in the matrigel. The genipin concentrations in the matrigel are indicated above each image: 0 mM (No-treatment; NT), 0.25 mM, and 0.50 mM. Numbers under the images represent the time relative to the start (0 h) and the end (48 h) of the observation. White arrowhead shows the initial position of the stainless ball. (c) Time-displacement plot of stainless balls through the matrigel. The displacement in the plot corresponds to the result obtained from the time-lapse observations in Fig. 1b. The horizontal axis is the relative time and the vertical axis is the relative displacement of stainless balls. (d) Mean velocities of stainless balls and the matrigel viscous moduli. The mean velocities are the gradients of the linear approximate equations. This experiment was repeated three times. Each value represents the mean value ± standard error of the mean (s.e.m.). Statistical significance between NT and other two samples: n.s., p ≥ 0.05, *p < 0.05.

Figure 2. MDCK cell morphological change on genipin-treated matrigel.

(a–c) Immunofluorescence images of MDCK cells on genipin-treated matrigel. The concentrations of genipin are (a) 0 mM (NT), (b) 0.25 mM, and (c) 0.5 mM. The cells and matrigel surface structure are detected by F-actin and laminin-111 staining, respectively. Cross-sectional views are shown together. Scale bar (white) = 50 μm. Scale bar (yellow) = 20 μm. (d) Relationship between cell morphology and substrate viscosity. The images of 3D morphology were reconstructed using the results presented in (a–c). White arrowhead shows the protrusion-like structure in the 3D structure. Scale bar = 30μm.

Figure 3. The matrigel surface is deformed during the 3D morphogenesis.

(a) Trajectories of latex beads embedded in the matrigel. The beads are shown as magenta spheres and the trajectories of the beads are shown as green lines. The green lines represent the displacement for 1.5 h. NT: No reagent control, Y-27632: ROCK inhibitor (10 μM). For this analysis, more than 50 latex beads were randomly chosen from each condition. Scale bar = 50 μm. (b) The box plot shows the displacement of embedded latex beads. Error bars represent s.e.m. ****p < 0.0001.

Figure 4. Morphological change of MDCK cells treated with inhibitors.

(a–e) Immunofluorescence staining of the cells and matrigel surface. The conditions are the following: (a) No-treatment (NT), (b) DMSO as a control for the following reagents except for Y-27632, (c) Blebbistatin, (d) Y-27632, and (e) ML-7. Orthogonal views are shown together. Scale bar (white) = 50 μm. Scale bar (yellow) = 20 μm. Three colonies were randomly chosen in each experiment (N = 3 (at least)). (f) The graph indicates the proportion of cell structures. Cell structures are categorized into three morphology types: tulip hat-like structure, monolayer sheet, and others. Thirty colonies were randomly chosen in three independent experiments (N = 3). Results are shown as mean value ± s.e.m. *p < 0.05, ***p < 0.001, ****p < 0.0001.

Figure 5. Localization and expression of phosphorylated-MRLC in the cells.

(a–d) F-actin, 1P-MRLC, and 2P-MRLC staining on the matrigel surface. Each sample was treated with (a) no reagent (NT), (b) DMSO as a control, (c) Y-27632, and (d) ML-7. Cross-sectional views, X-Z, and Y-Z plane are shown together. White arrowheads in the images represent 1P-MRLC and 2P-MRLC localization. Three colonies were randomly chosen in each experiment (N = 3 (at least)). Scale bar (white) = 50 μm and scale bar (yellow) = 20μm. (e,f) Immunoblot and statistical analysis of 1P-MRLC and 2P-MRLC protein expression in MDCK cells cultured on matrigel-coated dishes. The cells cultured on non-treated matrigel were used as control. Mean values are calculated from three independent experiments. Error bars represent s.e.m. *p < 0.05, **p < 0.01.

Figure 6. A proposed model for epithelial cell 3D morphogenesis induced by the substrate viscosity and cellular contractile forces.

(a) The morphological change involved the substrate viscosity. The substrate viscosity was altered by the extent of cross-links between peptide chains contained in the matrigel. Pink lines represent the peptide chains. The blue dots represent the cross-linking sites in the peptide chains. Red dots indicate the localization of the cellular contractile force and their sizes indicate the strength of the cellular contractile force (b) 3D morphology on the viscous substrate changes depending on the strength of the cellular contractile force.