Role of BMP signaling during early development of the annelid Capitella teleta

Abstract

The mechanisms regulating nervous system development are still unknown for a wide variety of taxa. In insects and vertebrates, bone morphogenetic protein (BMP) signaling plays a key role in establishing the dorsal-ventral (D-V) axis and limiting the neuroectoderm to one side of that axis, leading to speculation about the conserved evolution of centralized nervous systems. Studies outside of insects and vertebrates show a more diverse picture of what, if any role, BMP signaling plays in neural development across Bilateria. This is especially true in the morphologically diverse Spiralia (~Lophotrochozoa). Despite several studies of D-V axis formation and neural induction in spiralians, there is no consensus for how these two processes are related, or whether BMP signaling may have played an ancestral role in either process. To determine the function of BMP signaling during early development of the spiralian annelid Capitella teleta, we incubated embryos and larvae in BMP4 protein for different amounts of time. Adding exogenous BMP protein to early-cleaving C. teleta embryos had a striking effect on formation of the brain, eyes, foregut, and ventral midline in a time-dependent manner. However, adding BMP did not block brain or VNC formation or majorly disrupt the D-V axis. We identified three key time windows of BMP activity. 1) BMP treatment around birth of the 3rd-quartet micromeres caused the loss of the eyes, radialization of the brain, and a reduction of the foregut, which we interpret as a loss of A- and C-quadrant identities with a possible trans-fate switch to a D-quadrant identity. 2) Treatment after birth of micromere 4d induced formation of a third ectopic brain lobe, eye, and foregut lobe, which we interpret as a trans-fate switch of B-quadrant micromeres to a C-quadrant identity. 3) Continuous BMP treatment from late cleavage (4d + 12h) through mid-larval stages resulted in a modest expansion of Ct-chrdl expression in the dorsal ectoderm and a concomitant loss of the ventral midline (neurotroch ciliary band). Loss of the ventral midline was accompanied by a collapse of the bilaterally-symmetric ventral nerve cord, although the total amount of neural tissue did not appear to be greatly affected. Our results compared to those from other annelids and molluscs suggest that BMP signaling was not ancestrally involved in delimiting neural tissue to one region of the D-V axis. However, the effects of ectopic BMP on quadrant-identity during cleavage stages may represent a non-axial organizing signal that was present in the last common ancestor of annelids and mollusks. Furthermore, in the last common ancestor of annelids, BMP signaling may have functioned in patterning ectodermal fates along the D-V axis in the trunk. Ultimately, studies on a wider range of spiralian taxa are needed to determine the role of BMP signaling during neural induction and neural patterning in the last common ancestor of this group. Ultimately, these comparisons will give us insight into the evolutionary origins of centralized nervous systems and body plans.