Introduction: A large body of studies have highlighted that cells are sensitive to nanotopographies or geometrical cell-scale structures. However, natural biotopes also exhibit much larger topographical cues that are often curved and smooth, such as walls of blood vessels, bone cell cavities, or other cell bodies. Very little is known about how isolated cells and tissues read and integrate cell-scale curvatures, and the mechanisms leading to the integration of such physical cues. Objective: Herein, our aim was to develop new model surfaces with controlled edge-free cell-scale anisotropic and isotropic sinusoidal patterns to investigate the mesenchymal stem cell and epithelial cell layers" response to cell-scale curvature variations. Materials and methods: Herein we develop a two-step fabrication method to produce a series of sinusoidal landscapes with very low micro roughness. We combine live imaging, biochemistry and modeling approaches to decipher integration mechanisms at the cellular and tissue levels using respectively human mesenchymal stem cells (hMSCs) and MDCK epithelial cells. Results: First, we report a new cellular sense which we term "curvotaxis" that enables the isolated hMSCs to react to cell-scale curvature variations, a ubiquitous trait of cellular biotopes. We show that hMSCs avoid convex regions during their migration and position themselves in concave valleys. Computational modeling, pharmacological assays and live imaging show that curvotaxis relies on a dynamic interplay between the nucleus and the cytoskeleton - the nucleus acting as a curvature sensor that guides cell migration towards concave curvatures (1). Further, we report the curvature-modulated anisotropic growth of unconfined epithelia over cell-scale grooves and ridges of various transversal curvature. Curved regions of the substrate work as "topographical barriers", causing heterogeneity and reorientation of the nuclei and F-actin position. As a result, the epithelium displays a spatial bias in various morphogenetic processes such as migration or mitosis (2). Conclusion: Altogether, this work establishes cell-scale curvature as a major tuning parameter to regulate the growth of cells and epithelia and opens new possibilities for tissue engineering research.