Introduction
Contractility underlies manifold processes in cell and tissue morphogenesis, including cell migration, cell shape changes, or junction collapse
1. In epithelial tissues, cell contractions impact neighboring cells by exerting forces on adherens junctions. This mechanical linkage may elicit specific responses and could thus positively or negatively affect contractility and cytoskeletal organization in neighboring cells, i.e., mediate non‐autonomous mechanical behaviors
5. Within a tissue, cellular contraction and cell–cell interactions based on such force transduction can contribute to emergent tissue behavior, such as the formation of folds and furrows. The function of mutual cell–cell interactions, however, is difficult to study by classical genetic approaches. What is needed are methods for acute noninvasive interventions with high temporal and spatial resolution, ideally on the scale of seconds and of single cells.
For controlling cell contractility, optogenetic approaches have recently been developed. Cell contractility can be inhibited by optically induced membrane recruitment of PI(4,5)P
2 leading to interference with phosphoinositol metabolism and subsequent suppression of cortical actin polymerization
6. Optical activation of contractility has been achieved by light‐induced activation of the Rho‐ROCK (Rho kinase) pathway, which controls myosin II‐based contractility
7. While functionally effective, such optogenetic methods require multiple transgenes driving the expression of modified proteins such as light‐sensitive dimerization domains, which restrict the application to genetically tractable organisms. In addition, chromophores used in optogenetic effectors are activated by light in the visible spectrum, which limits the choice of labels and reporters for concurrent cell imaging.
Optochemical methods represent an alternative to genetically encoded sensor and effector proteins
9. Intracellular calcium ions (Ca
2+) are known to be an important regulator of contractility in many cell types. Ca
2+ plays a central role not only in muscle contraction, but also in cultured epithelial cells
10, in amnioserosa cells during dorsal closure
11, during neural tube closure
12, and in the folding morphogenesis of the neural plate
14. In
Drosophila oogenesis, tissue‐wide increase in intracellular Ca
2+ activates myosin II and impairs egg chamber elongation
15. In
Xenopus, a transient increase in Ca
2+ concentration induces apical constriction in cells of the neural tube
16. Although the detailed mechanism of Ca
2+‐induced contraction in non‐muscle cells remains to be resolved, it conceivably offers a simple and temporally precise way to interfere with and control contractile activity. In neuroscience, optochemical methods for the release of intracellular Ca
2+ have been well established and widely employed
17. Here, we report an optochemical method to control epithelial cell contractility via Ca
2+‐mediated light activation of myosin (CaLM) on the scale of seconds and at single‐cell resolution during tissue morphogenesis in
Drosophila embryos. Optochemical control of contractility by Ca
2+ uncaging has minimal spectral overlap with fluorescent protein reporters and optogenetic activators. Our results provide evidence for a ROCK‐dependent effect of increased intracellular Ca
2+ on activating non‐muscle myosin II and its recruitment to the actomyosin cortex.
Discussion
We developed and validated a new method, which we designate CaLM to induce cell contraction in epithelial tissues with precise temporal and spatial control. The approach applies Ca2+ uncaging, which has been well established in neurobiology, for example, to epithelial cell and developmental biology. By inducing Ca2+ bursts in single or multiple cells, CaLM enabled us to induce contraction in selected cells to about half of the cross‐sectional area within a minute. The induced contraction did not damage cells or perturb tissue integrity. To our best knowledge, this is the first report for optically controlled cell contraction on the minute scale and at single‐cell resolution in vivo during epithelial tissue morphogenesis.
CaLM is based on UV laser‐induced photolysis of a Ca
2+ chelator that has been widely employed
18. The caged compound “NP‐EGTA, AM” is membrane‐permeant and thus allows convenient application on the tissue scale. The 355‐nm pulsed UV laser, which we employ in this study, is compatible with modern objectives and can be conveniently mounted on standard live imaging microscopes via the epiport, for example. The dose of UV light depends on factors such as light scattering by the tissue and thickness of the sample. The actual dose of light at the target site can only be estimated and needs to be carefully titrated for the specific experimental system. We employed a genetically encoded Ca
2+ sensor protein for setting up the experimental conditions and testing the scale and time course of the Ca
2+ burst. Alternatively, Ca
2+ indicator dyes may be applied, depending on the sample. Besides the 355‐nm pulsed UV laser, we tested the suitability of a continuous wave laser at 405 nm, which is often installed at standard confocal microscopes. Using point scan illumination similar to FRAP protocols, we did not detect any increased signal of the GCaMP reporter (Fig
EV5). The inefficiency of the 405‐nm laser is consistent with the absence of significant absorbance of NP‐EGTA at wavelengths longer than 400 nm
19. Since our focus is to use CaLM to control contractility at single‐cell resolution during tissue morphogenesis. In order to make the approach easy of handling, we only used 100× objective in all experiments. To stimulate contractility in multiple cells simultaneously, we applied CaLM in four amnioserosa cells (Fig
EV3). Technically, CaLM should be applicable also to even more cells (e.g., 15–20 cells). Such experimental schemes will be tested in future investigations.
The detailed mechanism for the induced Ca
2+ burst and profile remains unclear. At this point, we do not know the origin and fate of Ca
2+ ions measured by the GCaMP sensor protein. A proportion of the Ca
2+ ions will be released from the photolyzed cage. It is conceivable, that in addition to this, intra‐ or extracellular Ca
2+ reservoirs are opened by Ca
2+‐gated Ca
2+ channels, comparable to SERCA in muscle cells
34. As the Ca
2+ levels return to low levels within minutes after uncaging, calcium ions may be exported from the cytoplasm to internal reservoirs such as ER or to the outside by Ca
2+ transporters.
The detailed mechanism of how Ca
2+ is functionally linked to contractile actomyosin also remains unclear, although there is no doubt that Ca
2+ is involved in regulation of contractility in many cell types
10. It is clear that Ca
2+ does not directly act on actomyosin similar to the contractile system involving troponin C, given the time lag between Ca
2+ burst and contractility in the range of many seconds. The delayed response may indicate an indirect link via a signaling cascade.
In non‐muscle cells, contractility is mediated by non‐muscle myosin II, which is largely controlled by Rho‐ROCK pathway
4. In the cells we tested, we find that Ca
2+ is linked to this pathway at the position of ROCK. CaLM induces contractility by activating the medial pool of non‐muscle myosin II, at least. Whether other pools of myosin II, such as junctional or basal myosin, are also activated remains unclear.
An expected consequence of a contracting cell within an epithelial tissue is a mechanical pull on its neighbors, which should be mediated by junctional complexes. This is an important issue, because an immediate application of CaLM is in tissue morphogenesis with one of its central questions of how the temporal–spatial distribution of forces leads to changes in visible morphology. We tested the potential mechanical pull of target cells on its neighbors in two ways. Firstly, we applied a Vinculin‐derived reporter, which preferentially binds to the open conformation of α‐Catenin. α‐Catenin undergoes a force‐dependent conformational change, which opens a Vinculin binding site under mechanical pull
27. Secondly, we directly assayed junctional tension in neighboring cells by measuring the initial recoil velocity after ablation. This experiment nicely shows the versatility of CaLM. The pulsed UV laser is employed for two tasks: firstly, the controlled uncaging in a single‐target cell and secondly, shortly afterward the precise ablation of a single junction, all recorded in a movie of the tissue. CaLM will be, in principle, useful in many types of experiments concerning tissue morphogenesis. For example, intercellular coupling between neighboring cells poses a challenge to experimental design in studies of tissue morphogenesis. Here, cause and consequence cannot be easily distinguished without targeted activation of cellular contractility and precise external control of cellular behaviors. Thus, acute interference is mandatory for dissecting causal functional dependencies.
Taken together, CaLM allows us to control rapid cell contractility and generates forces within the tissue during morphogenesis. CaLM can be applied to a wide range of processes and organisms and should greatly improve our ability to study the causality of cell contractility in tissue mechanics and mechanotransduction in vivo. Importantly, CaLM does not require any genetically encoded protein and can be readily applied to any stock and genetic background. The independence from genetic constitution should vastly accelerate analysis and enable screening of mechanobiological cellular pathways and components, e.g., by comparing wide arrays of mutants to wild‐type behavior. In addition, Ca2+ uncaging is likely to open applications in manifold experimental systems with low genetic tractability. Importantly, UV‐induced Ca2+ uncaging leaves the entire visible spectrum available for optical interfacing with florescent protein indicators and opsin‐based effectors. This in particular increases the options for simultaneously recording of cell and tissue behavior with the large palette of available fluorescent protein tags from CFP to RFP.