Models of biological pattern formation: from elementary steps to the organization of embryonic axes.

An inroad into an understanding of the complex molecular interactions on which development is based can be achieved by uncovering the minimum requirements that describe elementary steps and their linkage. Organizing regions and other signaling centers can be generated by reactions that involve local self-enhancement coupled to antagonistic reactions of longer range. More complex patterns result from a chaining of such reactions in which one pattern generates the prerequisites for the next. Patterning along the single axis of radial symmetric animals including the small freshwater polyp hydra can be explained in this way. The body pattern of such ancestral organisms evolved into the brain of higher organisms, while trunk and midline formation are later evolutionary additions. The equivalent of the hydra organizer is the blastopore, for instance, the marginal zone in amphibians. It organizes the anteroposterior axis. The Spemann organizer, located on this primary organizer, initiates and elongates the midline, which is responsible for the dorsoventral pattern. In contrast, midline formation in insects is achieved by an inhibitory signal from a dorsal organizer that restricts the midline to the ventral side. Thus, different modes of midline formation are proposed to be the points of no return in the separation of phyla. The conversion of the transient patterns of morphogenetic signaling into patterns of stable gene activation can be achieved by genes whose gene products have a positive feedback on the activity of their own gene. If several such autoregulatory genes mutually exclude each other, a cell has to make an unequivocal decision to take a particular pathway. Under the influence of a gradient, sharply confined regions with particular determinations can emerge. Borders between regions of different gene activities, and the areas of intersection of two such borders, become the new signaling centers that initiate secondary embryonic fields. As required for leg and wing formation, these new fields emerge in pairs at defined positions, with defined orientation and left-right handedness. Recent molecular-genetic results provide strong support for theoretically predicted interactions. By computer simulations it is shown that the regulatory properties of these models correspond closely to experimental observations (animated simulations are available at www.eb.tuebingen.mpg.de/meinhardt).