By using live imaging of an
By using live imaging of an ERK fluorescence resonance energy transfer (FRET) sensor, the authors show that ERK activity propagates in a stepwise manner from the center to the periphery of the tracheal placode. This row-by-row propagation of ERK activity fits a relay model in which each row activates the row next to it. The authors show evidence that the relay works via a feedback mechanism in which ERK activates Rho, which is an endopeptidase that activates the EGF ligand Spitz, which in turn binds and activates EGF receptors (EGFRs) in neighboring cells. Activated EGFR is known to induce ERK activity closing the feedback loop (Figure 1A). It is also shown that mutants that interfere with this feedback (trh and vvl) lead to a graded rather than stepwise ERK activity. So what is the role of the relay mechanism in the tracheal placode? It is suggested that it plays an important role in the invagination process of the tracheal placode. It is shown that concentric myosin cables are sequentially formed on the cell boundaries as the wavefront of ERK activity propagates. Moreover, it is shown that myosin accumulation correlates with local differences in ERK activity, namely that myosin accumulates on cell boundaries that separate between Oligomycin A with high ERK activity and cells with low ERK activity (Figure 1B). The myosin cables that are formed exert tension that drives the invagination process in the placode. The authors further suggest that Sons of Sevenless (Sos), a guanine nucleotide exchange factor (GEF), is the molecular link between EGFR signaling and myosin accumulation. The mechanism that identifies the differences in ERK activity in neighboring cells still remains to be elucidated. To better understand why a relay mechanism is required for the invagination process, Ogura et al. developed a mathematical model that couples differences in ERK activity to the tension on the boundary between cells (Figure 1B). One part of the model describes the feedback between the EGF ligand and the ERK activity in neighboring cells using coupled dynamic equations. The second part of the model describes the mechanics of the cells in the tracheal placode using a 2D vertex model (Farhadifar et al., 2007) that calculates the morphological configuration with the minimal mechanical energy. Critically, the differences in the levels of ERK activity between adjacent cells control the boundary tensions (by accumulating myosins). Ogura et al. use the mathematical model to compare the relay mechanism, where a feedback between ERK activity and signaling occurs, and a morphogen gradient mechanism, where ERK responds to ligands that are secreted from the central cells but that do not feed back on ligand activity (Figure 1A). It is shown that the sharp front of the ERK activity wave in the relay mechanism is required to explain the sequential formation of the myosin rings and for the proper bending of the placode (Figure 1B). The gradient model, on the other hand, does not capture the observed timing and shape of the invagination. This type of “domino effect” relay mechanism has been previously observed in other developmental systems and is not restricted to EGFR-ERK signaling. For example, the morphogenetic furrow that sets up the Drosophila ommatidia relies on a complex relay mechanism that involves feedback between hedgehog, Dpp, and EGF signaling (Greenwood and Struhl, 1999). Another example is the process of lateral induction that defines the prosensory region in the vertebrate inner ear, in which Notch signaling in each cell promotes the expression of the Notch ligand Jag1, which in turn activates Notch in the next row of cells (Hartman et al., 2010, Petrovic et al., 2014). One thing that is shared between these different processes is the need to generate an organized pattern or a field of cells in a stepwise manner. A relay mechanism has two main features that are important for achieving that: a sharp front and a constant propagation speed. In comparison, an increasing morphogen gradient would not have either of these properties and hence would not work well for defining a stepwise process (Figure 1A). The invagination of the tracheal placode fits well in this picture because it requires the sharp front in ERK activity to generate myosin-mediated tension, as well as the constant propagation speed for the sequential appearance of the concentric myosin cables. It would be interesting to see whether the relay mechanism identified by this work may also be relevant in other processes controlled by the EGFR-ERK pathway.