The cells of our bodies are exposed to a range of mechanical forces — including compression, shear and stretching — that they must resist to maintain tissue integrity and function. For example, skin responds to stretching forces by expanding. Physicians have exploited this particular response for more than 60 years1, implanting stretching devices in the skin to cause tissue expansion for plastic surgery or to repair birth defects2. But exactly how mechanical strain creates extra tissue in a living organism has not been known. Writing in Nature, Aragona et al.3 now provide compelling insights (at the molecular, single-cell and cell-population level) into how stem cells in the skin of mice sense and communicate stretch to make new tissue.
The surface of the skin — a multi-layered tissue called the epidermis — protects organisms against dehydration and environmental stresses, including mechanical challenges. To ensure lifelong protection, the epidermis is constantly renewed through the generation of new stem cells in its basal layer. This renewal is balanced with differentiation and the movement of stem cells to generate the upper, barrier-forming layers of the epidermis. Ultimately, the barrier-forming cells are shed from the surface, to be replaced by new cells.
Aragona et al. set out to examine how the epidermis responds to strain. The group positioned a device used in human surgeries — a self-inflating gel — under the skin of mice. They then examined indicators of force perception, including changes in cell shape, the structure of a mechanosensitive protein called α-catenin, and a network of keratin proteins that provides cells with mechanical resilience. This analysis revealed that epidermal stem cells do indeed sense and respond to strain. The authors observed a temporary increase in stem-cell division, followed by thickening of the epidermis. Thus, increased stem-cell renewal fuels stem-cell differentiation. The two effects combine to maintain a functional barrier at the same time as extra skin is generated.
The researchers next genetically engineered cells in the basal epidermal layer such that the stem cells and their descendants were fluorescently marked. Tracking of these cell lineages over time confirmed that stretching tips the renewal–differentiation balance in favour of making more stem cells. This explains why the epidermis expands in response to stretching.
Aragona et al. demonstrated that force changes stem cells at the molecular level in several ways. First, stretching increased the expression of genes involved in cell–cell adhesion, which have been shown to communicate force in vitro4. Second, expression of components of the actomyosin cytoskeleton — a network of protein filaments that generates contractile forces in cells5 — was increased. Third, stretching promoted signalling through the EGF–Map kinase–ERK pathway (a cascade of proteins that promotes growth). The researchers also assessed changes in chromatin, the DNA–protein complex that parcels up the genome in cells; such changes can lead to altered gene expression. This analysis revealed that stretch induced the expression of a network of regulatory genes that links stem-cell proliferation to skin maintenance.
The authors then examined how strain alters gene-expression profiles of single epidermal stem cells, by sequencing the cells’ RNA. This revealed that only a subpopulation of stem cells undergoes the molecular changes associated with a stretched state. Why might this be? Perhaps those that take on the stretched state experience greater force. Alternatively, maybe stem cells exist in varying biochemical states, and thus are more or less sensitive to force. Or local differences in stem-cell shape and mechanics could determine how each cell responds to stretch. Answering this question will require measurements of cellular forces and stiffness in vivo, which is still a major challenge. In addition, it remains unclear whether the stretched stem cells alone are driven to proliferate — or whether these cells then induce expansion of surrounding stem cells.
Aragona et al. next genetically engineered mice to lack Diaph3 and Myh9, genes involved in regulation of the actomyosin cytoskeleton. Without these genes, stem-cell responses to stretch were absent, leading to a barrier defect in the animals. The group observed similar effects in animals engineered to lack the genes encoding YAP and TAZ, and/or in which MAL was inhibited — these three transcription factors normally move to the nucleus to regulate gene expression in response to mechanical signals6. Next, the authors examined YAP, TAZ and MAL in mice lacking Diaph3 and Myh9. The transcription factors did not move to the nucleus in response to stretch in these animals. Thus, in normal skin, stretch reorganizes the actomyosin cytoskeleton to promote entry of YAP, TAZ and MAL into the nucleus. These proteins then coordinate transcriptional programs that promote skin growth and barrier formation (Fig. 1).
Finally, the researchers inhibited MAL or the EGF-pathway component ERK in their animals. Inhibition of either protein blocked stem-cell proliferation, but only MAL inhibition led to loss of the ‘stretched’ molecular state in a subset of stem cells. Thus, MAL regulates variable cell response to strain, whereas both ERK and MAL are necessary to promote the self-renewal of stem cells. Whether ERK is downstream of YAP, TAZ and MAL, or directly activated by the cytoskeleton7, and whether EGF–ERK signalling promotes adaptation to strain in the upper epidermal layers to maintain barrier function during skin expansion8, remain open questions.
Overall, Aragona and colleagues’ data support a model in which stretch is initially sensed by a subset of stem cells. These cells, through cytoskeletal reorganization and changes in gene expression, coordinate stem-cell renewal and differentiation with adaptation to the mechanical force being experienced. This response guarantees that the skin can maintain its protective function while expanding.
The research opens several avenues for future research. First, what is the contribution of other compartments of the skin (such as the upper, barrier-forming epidermal layers, or the thick dermal layer that underlies the epidermis) in sensing and communicating stretch? The authors’ analysis of mice lacking YAP or MAL suggests that stretch also induces a cytoskeletal response in differentiated cells of the upper epidermal layers. Stem-cell differentiation and upward movement can trigger renewal of neighbouring stem cells9, thus begging the question of whether the cells immediately above the basal layer are also required for stem-cell responses to strain.
Second, in vitro experiments have demonstrated10 that chromatin regulation in the nucleus is key to maintaining stem-cell identity and genome integrity under mechanical stress. Aragona and co-workers’ skin-expander model will now allow us to explore these mechanisms in vivo.
Third, current models of stem-cell renewal postulate that a single stem cell is equally capable of undergoing renewal or differentiation. However, Aragona and colleagues’ lineage-tracing experiments revealed that the number of cells derived from one stem cell (called basal-cell clones) tended to be even. This bias towards clones that have even numbers of cells became much more pronounced on stretching. How the stretched state promotes even-numbered clones is unclear.
The authors propose that this bias can be explained by a model in which stem cells exist in two-progenitor units, in which one stem cell is committed to renewal and the other to differentiation. Communication within and between units would balance the loss of cells through differentiation with renewal. The group performed a mathematical comparison, which indicated that the even-numbered-cell bias and clone dynamics they observed were more consistent with a two-progenitor than with a single-progenitor model. However, the jury on this is still out, because a recent study has provided fresh evidence for the one-progenitor model11.
The current work provides a major step forward in our understanding of how force is interpreted at the single-cell level in living organisms. Furthermore, it should encourage others to explore the use of mechanical signals to generate extra skin — not only for reconstructive surgery, but also for diseases associated with impaired regeneration.
Nature 584, 196-198 (2020)