Constricting cells and bending tissues
Two-dimensional sheets of polarized cells called epithelia can be folded by collective constriction of cells on their apical side, a process called apical constriction. Apical constriction is associated with diverse morphogenetic events, such as tube formation (i.e. neural tube) or cell invagination and epithelial-mesenchymal transition (EMT). During Drosophila gastrulation, apical constriction drives the invagination of mesoderm cells into the embryo interior, resulting in the formation of a “ventral furrow”. We have found that contractile forces that drive ventral furrow cell apical constriction result from dynamic networks of actin filaments and the molecular motor myosin II (myosin). Cells constrict via a series of contraction pulses, which are interrupted by pauses in which the constricted state of the cell is maintained. Pulses result from periodic contraction and remodeling of an actin and myosin network that spans the apical cortex. Watch the ventral furrow video. Between contraction pulses, actin and myosin fibers generate tension and prevent the cell from relaxing. Importantly, mutants that fail to assemble actin-myosin fibers, lack epithelial tension, and fail to stabilize cell shape between contraction pulses. Thus, tension across the cell apex is required for cells to constrict incrementally, like a ratchet. How does a cell actively remodel its force-generating machinery to generate contractile pulses while maintaining tension to constrict like a ratchet? We are visualizing the dynamics and cytoskeletal structures that accompany apical constriction and testing how cytoskeletal dynamics generate force. In addition, we are interested in analyzing the function of diverse cytoskeletal proteins that can lead to different types of cytoarchitecture.
Signals that tell the cell/tissue to contract
Cytoskeletal dynamics and apical constriction in ventral furrow cells are triggered by expression of the transcription factors Snail and Twist. We have found that Snail and Twist regulate distinct steps of the contractile ratchet. Snail is required to initiate contraction pulses. In contrast, contraction pulses occur in normally in twist mutants, but Twist is required to stabilize cell shape between contractile pulses. We are interested in signaling events downstream of Snail and Twist that regulate cell shape change.
Genetic screens identified heterotrimeric G-proteins and the Rho1 (RhoA) GTPase as downstream signals that are activated by Twist expression. We discovered that the Rho1 effector, Rho-associated kinase (Rok), is polarized to the “center” of the apical domain during apical constriction. Given that Rok levels are highest in this medioapical domain and lowest at junctions, we propose that the cell exhibits a “radial cell polarity” that distinguishes these distinct apical domains. Rok likely regulates contractility by phosphorylating and activating myosin. We are investigating how radial cell polarity is established and how it mediates myosin activation and apical constriction. Our ultimate goal is to visualize the dynamic flow of information from transcription factors to the cytoskeleton.
Coordinating cell shape change across a tissue
Approximately 1,200 cells undergo apical constriction during Drosophila mesoderm invagination. The timing of apical constriction must be coordinated across the entire tissue, otherwise uneven force generation will cause groups of cells to become stretched. Apical constriction could be coordinated by biochemical and mechanical signals. Indeed, mutants that affect signaling downstream of Twist result in uneven apical constriction, cell stretching, and abnormal furrow formation. How are contractile ratchets coordinated in different cells across the tissue? We are using computational techniques to analyze how contractile signals, cytoskeletal dynamics, and cell shape change are coordinated across the tissue to collectively result in tissue contraction and invagination.
Force transmission and mechanical feedback
For contractile forces to result in tissue invagination, cytoskeletal structures in individual cells must be integrated between cells to transmit forces to the tissue level. Adherens junctions (AJs), which contain the transmembrane adhesion molecule E-cadherin, link neighboring epithelial cells to each other and the cell surface to the actin cortex. Reducing AJ levels by RNAi results in tissue-wide tears across the ventral furrow that result from actomyosin generated tension, suggesting that AJs are critical to transmit epithelial tension. Watch the armadillo mutant embryo video. We are investigating how a dynamic cytoskeletal network is coupled to AJs and how this coupling transmits cellular forces to the tissue-level. Forces that are transmitted between cells in a tissue could feed-back on cells to regulate their behavior. Importantly, we have used laser cutting experiments to show that epithelial tension is anisotropic in the ventral furrow, which causes cells to preferentially constrict in a specific direction. This demonstrates that ventral furrow cells “feel” and “respond” to tissue-level forces. The resulting tissue mechanics could result from constraints imposed by the neighboring tissue. We are using genetic and mechanical perturbations to alter the landscape of forces across the tissue to determine how tissue mechanics influences cell behavior.