Hola

Cell, Vol. 113, 853–865, June 27, 2003, Copyright 2003 by Cell Press

Anteroposterior Patterning in Hemichordates and the Origins of the Chordate Nervous System
Christopher J. Lowe,1,2,* Mike Wu,1 Adrian Salic,2 Louise Evans,2 Eric Lander,3 Nicole Stange-Thomann,3 Christian E. Gruber,4 John Gerhart,1,* and Marc Kirschner2 1 Department of Molecular and Cell Biology University of California,Berkeley Berkeley, California 94720 2 Department of Cell Biology Harvard Medical School Boston, Massachusetts 02115 3 Whitehead Institute MIT Center for Genome Research Cambridge, Massachusetts 02142 4 Express Genomics, Inc. 1306 Bailes Lane, Suite F Frederick, Maryland 21701 ical homologies among these phyla has been fraught with difficulties, as their adult body plans appear so divergent (Gee,1996). Yet, underlying the development of morphology are distinct gene networks that are expressed at specific locations within the bodies of all members of the chordate phylum. A comparison of such domains among the three extant deuterostome phyla could reveal similarities in axial organization obscured by divergent morphology and facilitate the reconstruction of the basic body plan of thehypothetical deuterostome ancestor. Hemichordates may hold more promise for these comparisons than do echinoderms since they have several proposed morphological affinities with chordates. Indeed, Bateson (1886b) originally classified hemichordates as chordates based on gill slits; a stomochord with uncertain homology to the chordate notochord (Balser and Ruppert, 1990); dorsal and ventral nerve cords, eachof which has been proposed as the homolog of the chordate dorsal hollow cord (Morgan, 1894; Nub¨ ler-Jung and Arendt, 1999); and a ventral post-anal extension, proposed as the homolog of the chordate dorsal tail (Burdon-Jones, 1952). However, the homologies are easily controverted and hemichordates were reclassified into their own phylum by the 1940s. The nervous system is the key element in mosthypotheses on chordate origins. Three hypotheses currently account for the origin of the chordate nervous system, all consistent with recent molecular phylogenies, yet all mutually incompatible. In the auricularian hypothesis of Garstang ([1894, 1928] see also Lacalli [1994] and Nielsen [1999]), the chordate nerve cord was thought to have originated from the nervous system of a motile auricularialarva of an ancestral sessile deuterostome adult. Bilateral rows of cilia and the associated nerves were said to have converged to the dorsal midline to form the nervous system. In the hemichordate hypothesis, Bateson (1886b) proposed that chordates evolved directly from a hemichordate-like adult ancestor, not a larva. According to him, the ancestor, like the extant adult hemichordate, had achordate-like dorsal hollow nerve cord (most prominent in the collar region, Figure 1A). In the inversion hypothesis (Geoffroy-St. Hilaire, 1822; revised by Arendt and Nubler-Jung, 1996; DeRob¨ ertis and Sasai, 1996), a worm-like ancestor of annelids, arthropods, and chordates already had a well-formed central nervous system, ventrally placed. An adult ancestor of deuterostomes inverted its bodydorsoventrally, placing the nerve cord dorsal in chordates. Recent support for this hypothesis comes from the inverse disposition of neurogenic and nonneurogenic ectoderm in the embryos of chordates and Drosophila, correlated with the inverse disposition of TGF- signals and their antagonists (De Robertis and Sasai, 1996). A more classical perspective of nervous system evolution that has not enjoyedmuch support from recent molecular analyses is the proposal that the bilaterian ancestor had a diffuse nervous system that was centralized independently in different bilaterian lineages (Brusca and Brusca, 1990). One reason for resistance

Summary The chordate central nervous system has been hypothesized to originate from either a dorsal centralized, or a ventral centralized, or a noncentralized...