Multicellular organisms, as the name suggests, are composed of many cells that work together. Cells need to stick together to ensure efficient cell coordination. Cadherins, which are calcium-dependent cell adhesion molecules, do just that: they make sure that neighboring cells maintain contact. Being as old as multicellularity itself, these ancient proteins have been studied by generations of scientists in many different experimental systems. One might therefore wonder if there are aspects of cadherin biology that are still unexplored.We understand how cadherins are post-translationally modified and trafficked to the cell membrane, where they bind to their cytoplasmic interactors and link-up with cadherins from neighboring cells; in what context are they influenced by mechanical forces; and, at the end, what triggers their eventual endocytosis for recycling or degradation (Leckband and de Rooij, 2014; Biswas and Zaidel-Bar, 2017; Zhang et al., 2023). Much of this understanding stems from studying E-cadherin – the epithelial variant – which is expressed at early stages of embryonic development. E-cadherin maintains tissue-specific expression throughout life and is also a key regulator of gastrulation movements in complex multicellular organisms. If we know so much already, then what's missing?The main function of E-cadherin is linked to its localization on the cell membrane, where it engages in trans-homophilic interactions, i.e. linking with E-cadherin molecules from the neighboring cell. As it turns out, one of the key requirements for establishing this link is cis-clustering, i.e. aggregation of E-cadherin molecules from the same cell, at the nanometer scale (Wu et al., 2015; Bertocchi et al., 2017; Padmanabhan et al., 2017; Chandran et al., 2021). Presumably, this cis-clustering slows down E-cadherin diffusion sufficiently to create hotspots for engagement with other E-cadherin molecules or cis-clusters from a neighboring cell to form stable spot adhesions. Naturally, researchers have shown that breaking these cis-clusters weakens E-cadherin-based cell adhesion, demonstrating the necessity of cis-clustering.One might expect that overexpression of E-cadherin should be sufficient to form larger E-cadherin clusters. As it turns out, however, such overexpression of E-cadherin also increases its endocytic removal, preventing the formation of larger clusters. Indeed, blocking endocytosis allows the formation of larger E-cadherin clusters (Truong Quang et al., 2013). Although blocking endocytosis might be a solution, one can imagine this would create more problems than it would solve. To demonstrate the importance of E-cadherin clustering, researchers have sometimes used chimeric constructs to produce cis-dimers (Yap et al., 1997). But creating and studying native E-cadherin cis-clusters has been challenging, until now.In their recent preprint, Lerchbaumer and colleagues describe a new method to induce E-cadherin cis-clustering (Lerchbaumer et al., 2026 preprint). They have managed something the community has been longing for: a method for E-cadherin clustering that forms native clusters that don't seem to get endocytosed. For the first time in the long history of experiments on E-cadherin, we now seem to have a handle on a key aspect of its biology: even though artificial, we now know how to make meaningful E-cadherin cis-clusters. The preprint authors achieve this not through overexpression of E-cadherin, but rather through its redistribution, using an optogenetic technique.Optogenetics, using light to activate/inhibit gene function, are a suite of techniques that's maturing fast. They use a variety of ways to create loss-/gain-of-function effects, exploiting the known function of what is targeted, altering its localization and/or activity. The authors used one such variant called LARIAT (light-activated reversible inhibition by assembled trap), which uses the CIBN-CRY2 system that, when coupled to a nanobody against GFP, clusters anything tagged with GFP. LARIAT could thus be used to produce cis-clusters of GFP-tagged E-cadherin.Through various experiments, the authors show that the artificially generated cis-clusters of E-cadherin are functional. They quantified the localization of known components of the E-cadherin-based adhesion complex that assembled normally. In the past, researchers have shown that a drastic change in E-cadherin levels can impact the distribution of the non-muscle myosin-II (MyoII) (Levayer and Lecuit, 2013), which is a key driver of morphogenetic movements. The authors tested whether LARIAT-mediated E-cadherin clustering also produced such changes and found MyoII distribution to be rather normal. Naturally, they also found that the tissue-scale tension distribution, driven by MyoII distribution, was normal, despite the observation that E-cadherin in these artificial clusters is more stable: it doesn't diffuse as it normally would and shows increased stability on the membrane. Consistent with these changes in E-cadherin turnover, the T1 transitions (i.e. type-1 cell neighbor exchanges) and convergent-extension movements were also affected. This indicated that junction remodeling, i.e. changes in the length of a cell-cell contact, was compromised.Previously, standard vertex models of epithelial tissue predicted that enhanced cell adhesion should promote T1 transitions (Bi et al., 2015), which in turn supports the convergent-extension movements. The authors decided to push these models further, building on their results. Given that T1 transitions were dramatically reduced despite near-normal MyoII levels and tension distribution, the authors hypothesized that an additional parameter should be incorporated into the mechanical model, namely dissipative cell adhesion, to account for the friction-like effects of having stronger cell-cell adhesion. While addition of this parameter didn't affect the tension distribution itself, the model now had a new prediction: increased E-cadherin clustering will significantly hamper morphogenetic movements that involve neighbor exchange. To test this prediction, the authors focused on two other morphogenetic movements: mesoderm invagination that mostly progresses without neighbor exchange; and delamination of neuroblasts that requires local neighbor exchange. As the model predicted, neuroblast delamination was hampered when LARIAT was activated to cluster E-cadherin, while mesoderm invagination progressed normally.E-cadherin-based cell-cell adhesion is a key regulator of tissue integrity and is often dysregulated in pathological scenarios, such as tumor metastasis. Thus, it is natural to equate homeostasis with stable adhesion, and to assume that this is also what E-cadherin must be doing during dynamic morphogenetic movements. It is easy to imagine how actomyosin contractility drives cell shape changes and junction remodeling, while E-cadherin-based adhesion merely makes sure that tissue integrity is maintained, and to think that any other roles that E-cadherin might play must be secondary to this key function. The authors challenge this notion with their experiments, where they show that E-cadherin-mediated cell-cell adhesion is also a key modulator of morphogenetic movements, especially those that involve junction remodeling and neighbor exchange. Perhaps normal E-cadherin regulation is so ‘nimble’ that it effortlessly responds to contractile forces and facilitates morphogenesis. It is only when it finds the ‘strength in unity’ that the ‘friction’ with the established norms becomes apparent.I thank Steffen Lemke and the members of the Zoology department at the University of Hohenheim for an exciting scientific environment, and FlyBase for its crucial support that makes Drosophila research possible.
Girish Kale (Wed,) studied this question.