Surface topography induces 3D self-orientation of cells and extracellular matrix resulting in improved tissue function

Abstract

The organization of cells and extracellular matrix (ECM) in native tissues plays a crucial role in their functionality. However, in tissue engineering, cells and ECM are randomly distributed within a scaffold. Thus, the production of engineered-tissue with complex 3D organization remains a challenge. In the present study, we used contact guidance to control the interactions between the material topography, the cells and the ECM for three different tissues, namely vascular media, corneal stroma and dermal tissue. Using a specific surface topography on an elastomeric material, we observed the orientation of a first cell layer along the patterns in the material. Orientation of the first cell layer translates into a physical cue that induces the second cell layer to follow a physiologically consistent orientation mimicking the structure of the native tissue. Furthermore, secreted ECM followed cell orientation in every layer, resulting in an oriented self-assembled tissue sheet. These self-assembled tissue sheets were then used to create 3 different structured engineered-tissue: cornea, vascular media and dermis. We showed that functionality of such structured engineered-tissue was increased when compared to the same non-structured tissue. Dermal tissues were used as a negative control in response to surface topography since native dermal fibroblasts are not preferentially oriented in vivo. Non-structured surfaces were also used to produce randomly oriented tissue sheets to evaluate the impact of tissue orientation on functional output. This novel approach for the production of more complex 3D tissues would be useful for clinical purposes and for in vitro physiological tissue model to better understand long standing questions in biology.

Fig. 1

Schematic illustration of the self-assembly method (a). Cells are plated in a single cell seeding procedure (a1) and their number increase through cell proliferation. They are cultured in the presence of sodium ascorbate to stimulate ECM synthesis (a2). Cells are maintained in culture until their neosynthesized ECM proteins have self-assembled into an adherent living tissue sheet comprised of cells and ECM (a3). The tissue sheet can be manipulated with tweezers and do not contain any exogenous biomaterials, only cells and their secreted ECM as shown in a magnified view illustration (a4). Process for the fabrication of the microstructured TPE substrates. The master was fabricated in Si wafer by standard photolithography. Etching of the gratings into the Si master was made by Cr deposition, liftoff of the photoresist and reactive ion etching using CF₄–O₂ gas with Cr as a mask (b). After an etching procedure, the silicon master is used to replicate the same structure in TPE. A flat piece of TPE is placed on the master and then heated to allow the TPE to flow in the gratings of the silicon master (c). After an oxygen plasma treatment and sterilization, TPE substrates are used for cell culture (d), samples are placed in Petri dishes before cell seeding at step (a1) to induce cell orientation in culture instead of using regular Petri dishes that result in randomly distributed cells in culture. AFM image of the silicon master after the etching procedure (e). SEM image of a microstructured substrate replica created by hot embossing (f).

Fig. 2

Immunofluorescence staining of actin filaments of corneal fibroblasts, smooth muscle cells and dermal fibroblasts cultured on SEBS substrates. Corneal fibroblasts grown on microstructured SEBS show that cells are confluent in the bottom monolayer and aligned with the gratings, the second cell layer is oriented with a characteristic angle shift from the bottom layer as we can observe at day 9 and day 13, which can not be found in control samples (a–f). Smooth muscle cells are also aligned with the gratings and have a specific orientation within their second cell layer at day 13, as opposed to the control where we can observe random cell distribution (g–l). For dermal fibroblasts, the orientation is unidirectional on the bottom layer and shows no specific orientation on the second layer, dermal fibroblasts grown on control sample show no sign of orientation (m–r). Gratings have a 4 µm period and a 1 µm linewidth. Normal distribution of angle shift measured for each cell type has been performed. The second layer of corneal fibroblasts displayed a 53.07 ± 8.1 degree shift from the first cell layer (s), whereas smooth muscle cells’ second layer angle shift was 38.55 ± 4.24 degrees (t). The dermal fibroblasts’ angle shift was 76.06 ± 32.07 degrees (u), suggesting that this cell type does not organize in a specific configuration. Results showed a significant difference (p < 0.05) between each groups, clearly demonstrating that angle shift is cell type dependant. * indicates that corneal fibroblasts’ angle shift is significantly different than the one of smooth muscle cells and dermal fibroblasts. ** indicates that smooth muscle cells’ angle shift is significantly different than the one of dermal fibroblasts.

Fig. 3

Confocal imaging of corneal fibroblasts, smooth muscle cells and dermal fibroblasts csECM on microstructured and flat SEBS. Type I collagen fibers are shown in red, cells nuclei in blue, the arrows represent collagen from the bottom layer and the arrowheads collagen fibers from the top layer. Human corneal fibroblasts grown on flat SEBS substrates (control) show no signs of csECM structural orientation. Same cells grown on microstructured (MS) SEBS samples reveal that csECM have an organization similar to the one observed in cells (a). Immunofluorescence staining of type I collagen in SMC sheets revealed the same observation (b). Immunofluorescence staining of dermal fibroblast csECM shows type I collagen fibers. The control shows very little cross-linked type I collagen fibers and MS samples do not indicates the presence of collagen fibers orientation (c).

Fig. 4

Electron microscopy images of corneal fibroblasts on flat (control) and microstructured SEBS. The x–y plane SEM images of a control sample (a) shows different layers of collagen without any preferred orientation as opposed to the microstructured sample (b) where the csECM has many layers of oriented collagen. Insets in (a) and (b) show a higher magnification of the anisotropic and oriented superposed collagen layers of a single cell sheet. Arrows and arrowheads are showing different layers of oriented collagen within the reconstructed stroma. The x–y plane of TEM images of the control sample at 15 000× (c) and 30 000× (e) show no organization of the collagen layers. In the samples grown on microstructured SEBS we can see in (d) at 15 000× that many collagen layers are organized as in a native corneal stroma and in (f) at 30 000× we see 2 collagen layers and a corneal fibroblast. In TEM images, the dots are representing collagen fibers perpendicular to the x–y plane and striated lines are collagen fibers running in a parallel direction of the x–y plane.

Fig. 5

Tissue functionality analyses. Corneal substitutes have been cultured on control and microstructured substrates. Microstructured samples exhibit a better transparency as we can observe on the macroscopic view and on the transmission spectrum measurement, scale bar 5 mm (a). Stress–strain curves of TEVM show that microstructured samples have improved mechanical properties when they are circumferentially aligned inside tissue-engineered substitutes compare to the non-organized substitutes, strain is expressed as a percentage of deformation (b). The tensile strength of dermal fibroblast sheets are slightly improved when they are cultured on microstructured substrates compare to flat substrates (c). Representative curves of transmission and stress–strain with a statistically significant difference for each condition.