LAUSR

During development in metazoan embryos, the fundamental embryonic axes are established by organizing centers that influence the fates of nearby cells. Among the spiralians, a large and diverse branch of protostome metazoans, studies have shown that an organizer sets up the dorsal-ventral axis, which arises from one of the four basic cell quadrants during development (the dorsal, D quadrant). Studies in a few species have also revealed variation in terms of how and when the D quadrant and the organizer are established. In some species the D quadrant is specified conditionally, via cell-cell interactions, while in others it is specified autonomously, via asymmetric cell divisions (such as those involving the formation of polar lobes). The third quartet macromere (3D) typically serves as the spiralian organizer; however, other cells born earlier or later in the D quadrant lineage can serve as the organizer, such as the 2d micromere in the annelid Capitella teleta or the 4d micromere in the mollusc Crepidula fornicata. Here we present work carried out in the snail C. fornicata to show that establishment of a single D quadrant appears to rely on a combination of both autonomous (via inheritance of the polar lobe) and conditional mechanisms (involving induction via the progeny of the first quartet micromeres). Through systematic ablation of cells, we show that D quadrant identity is established between 5th and 6th cleavage stages, as it is in other spiralians that use conditional specification. Subsequently, following the next cell cycle, organizer activity takes place soon after the birth of the 4d micromere. Therefore, unlike the case in other spiralians that use conditional specification, the specification of the D quadrant and the activity of the dorso-ventral organizer are temporally and spatially uncoupled. We also present data on organizer function in naturally-occurring and experimentally-induced twin embryos, which possess multiple D quadrants. We show that supernumerary D quadrants can arise in C. fornicata (either spontaneously or following polar lobe removal); when multiple D quadrants are present these do not exhibit effective organizer activity. We conclude that the polar lobe is not required for D quadrant specification, though it could play a role in effective organizer activity. We also tested whether the inheritance of the small polar lobe by the D quadrant is associated with the ability to laterally inhibit neighboring quadrants by direct contact in order to normally prevent supernumerary organizers from arising. Finally, we discuss the variation of spiralian organizers in a phylogenetic context.