During gastrulation the notochord is formed. This long thin structure marks the midline within the mesoderm and inherits organizer function. Measuring the distance from this line (and not earlier from the patch-like organizer) can provide positional information for the mediolateral or DV axis. At early stages, the prospective notochord is a regulating pattern forming system: a removal leads to a re-expression of corresponding markers in chick and in Xenopus (Psychoyos and Stern, 1996; Yuan et al, 1995; Levin and Mercola, 1998). This suggests that the cells next to the notochord are competent, but that the notochord suppresses its own further extension by some sort of lateral inhibition. However, if lateral inhibition is involved, why does the notochord (or more generally speaking the midline pattern) not decay into separated patches? Why does the lateral inhibition work only to the sides and not along the long extension? According to the model, a stripe-like activation is stable if the self-enhancing reaction has an upper bound (see Periodic Structures ). However, it is not possible to generate a single straight and long-extended structure using this mechanism alone since stripe formation requires a restricted lateral inhibition and other nearby stripes would not be suppressed. This is in contrast to the formation of a single spot-like activation in which the lateral inhibition need not be restricted by an upper bound of the activator concentration. A “hot stripe” system can be generated by a unique hot spot system (organizer) if the spot system induces the stripe system or its elongation while, in turn, the hot stripe repels and induces a shift of the hot spot.
The scheme above shows a possible mechanism: cells close to the blastoporus (red) move from both sides towards the organizer (blue). After being close to the organizer once, the cells change behavior. They are no longer attracted to it and leave the organizer as a unified band. In this way, a spot-like organizer causes the formation of a stripe-like organizer (like an air plane that sucks in air and leaves behind a vapor trail). Such a movement is compatible with the well-known convergence-extension (Keller et al., 1985). Stem cells residing in the organizer may add cells to this band to form the most central elements of the midline (green). The passing through or becoming close to the organizer may stop a countdown-like process in the cell, for instance, the sequential activation of more 5’ Hox genes. This fixes their actual AP positional specification along the midline.
As shown in the simulation above, the following simple model (Meinhardt, 2000) captures essential elements of this hot spot – hot stripe conversion: a system that is tuned to make stripes (green) is triggered by the organizer, i.e., a system that is activated in a spot-like manner (blue). Since the stripe system (notochord) also repels the spot system (organizer), the spot system is shifted in front of the tip of the stripe, causing its straight elongation. Therefore, cells obtain temporarily organizer quality before participating in midline formation in agreement with observations by Joubin and Stern (1999). Due to a saturation in the self-enhancement, the stripe system does not disintegrate into individual patches and establishes the midline. This stripe (e.g., chordin, noggin), in turn, generates positional information for the DV or mediolateral axis by acting as a sink for a ubiquitously produced substance (pink, e.g., BMP-4). The local concentration of the latter increases with distance from the midline and provides positional information (Dosch et al., 1997). In this simulation, lines of new cells are added next to the blastoporus (yellow). A more realistic model would require incorporation of the actual cell movement toward the organizer.