Role of Pax genes in eye evolution: a cnidarian PaxB gene uniting Pax2 and Pax6 functions.

PaxB from Tripedalia cystophora, a cubomedusan jellyfish possessing complex eyes (ocelli), was characterized. PaxB, the only Pax gene found in this cnidarian, is expressed in the larva, retina, lens, and statocyst. PaxB contains a Pax2/5/8-type paired domain and octapeptide, but a Pax6 prd-type homeodomain. Pax2/5/8-like properties of PaxB include a DNA binding specificity of the paired domain, activation and inhibitory domains, and the ability to rescue spa(pol), a Drosophila Pax2 eye mutant. Like Pax6, PaxB activates jellyfish crystallin and Drosophila rhodopsin rh6 promoters and induces small ectopic eyes in Drosophila. Pax6 has been considered a "master" control gene for eye development. Our data suggest that the ancestor of jellyfish PaxB, a PaxB-like protein, was the primordial Pax protein in eye evolution and that Pax6-like genes evolved in triploblasts after separation from Cnidaria, raising the possibility that cnidarian and sophisticated triploblastic eyes arose independently.

Amnioserosa cell constriction but not epidermal actin cable tension autonomously drives dorsal closure.

Tissue morphogenesis requires coordination of multiple force-producing components. During dorsal closure in fly embryogenesis, an epidermis opening closes. A tensioned epidermal actin/MyosinII cable, which surrounds the opening, produces a force that is thought to combine with another MyosinII force mediating apical constriction of the amnioserosa cells that fill the opening. A model proposing that each force could autonomously drive dorsal closure was recently challenged by a model in which the two forces combine in a ratchet mechanism. Acute force elimination via selective MyosinII depletion in one or the other tissue shows that the amnioserosa tissue autonomously drives dorsal closure while the actin/MyosinII cable cannot. These findings exclude both previous models, although a contribution of the ratchet mechanism at dorsal closure onset remains likely. This shifts the current view of dorsal closure being a combinatorial force-component system to a single tissue-driven closure event.

Formation of the bicoid morphogen gradient: an mRNA gradient dictates the protein gradient.

The Bicoid (Bcd) protein gradient is generally believed to be established in pre-blastoderm Drosophila embryos by the diffusion of Bcd protein after translation of maternal mRNA, which serves as a strictly localized source of Bcd at the anterior pole. However, we previously published evidence that the Bcd gradient is preceded by a bcd mRNA gradient. Here, we have revisited and extended this observation by showing that the bcd mRNA and Bcd protein gradient profiles are virtually identical at all times. This confirms our previous conclusion that the Bcd gradient is produced by a bcd mRNA gradient rather than by diffusion. Based on our observation that bcd mRNA colocalizes with Staufen (Stau), we propose that the bcd mRNA gradient forms by a novel mechanism involving quasi-random active transport of a Stau-bcd mRNA complex through a nonpolar microtubular network, which confines the bcd mRNA to the cortex of the embryo.

A key role of Pox meso in somatic myogenesis of Drosophila.

The Pax gene Pox meso (Poxm) was the first and so far only gene whose initial expression was shown to occur specifically in the anlage of the somatic mesoderm, yet its role in somatic myogenesis remained unknown. Here we show that it is one of the crucial genes regulating the development of the larval body wall muscles in Drosophila. It has two distinct functions expressed during different phases of myogenesis. The early function, partially redundant with the function of lethal of scute [l(1)sc], demarcates the ;Poxm competence domain', a domain of competence for ventral and lateral muscle development and for the determination of at least some adult muscle precursor cells. The late function is a muscle identity function, required for the specification of muscles DT1, VA1, VA2 and VA3. Our results led us to reinterpret the roles of l(1)sc and twist in myogenesis and to propose a solution of the 'l(1)sc conundrum'.

The toposome, essential for sea urchin cell adhesion and development, is a modified iron-less calcium-binding transferrin.

We describe the structure and function of the toposome, a modified calcium-binding, iron-less transferrin, the first member of a new class of cell adhesion proteins. In addition to the amino acid sequence of the precursor, we determined by Edman degradation the N-terminal amino acid sequences of the mature hexameric glycoprotein present in the egg as well as that of its derived proteolytically modified fragments necessary for development beyond the blastula stage. The approximate C-termini of the fragments were determined by a combination of mass spectrometry and migration in reducing gels before and after deglycosylation. This new member of the transferrin family shows special features which explain its evolutionary adaptation to development and adhesive function in sea urchin embryos: (i) a protease-inhibiting WAP domain, (ii) a 280 amino acid cysteine-less insertion in the C-terminal lobe, and (iii) a 240 residue C-terminal extension with a modified cystine knot motif found in multisubunit external cell surface glycoproteins. Proteolytic removal of the N-terminal WAP domain generates the mature toposome present in the oocyte. The modified cystine knot motif stabilizes cell-bound trimers upon Ca-dependent dissociation of hexamer-linked cells. We determined the positions of the developmentally regulated cuts in the cysteine-less insertion, which produce the fragments observed previously. These fragments remain bound to the hexameric 22S particle in vivo and are released only after treatment of the purified toposome with reducing agents. In addition, some soluble smaller fragments with possible signal function are produced. Sequence comparison of five sea urchin species reveals the location of the cell-cell contact site targeted by the species-specific embryo dissociating antibodies. The evolutionary tree of 2-, 1-, and 0-ferric transferrins implies their evolution from a basic cation-activated allosteric design modified to serve multiple functions.

An urbilaterian origin of the tripartite brain: developmental genetic insights from Drosophila.

Studies on expression and function of key developmental control genes suggest that the embryonic vertebrate brain has a tripartite ground plan that consists of a forebrain/midbrain, a hindbrain and an intervening midbrain/hindbrain boundary region, which are characterized by the specific expression of the Otx, Hox and Pax2/5/8 genes, respectively. We show that the embryonic brain of the fruitfly Drosophila melanogaster expresses all three sets of homologous genes in a similar tripartite pattern. Thus, a Pax2/5/8 expression domain is located at the interface of brain-specific otd/Otx2 and unpg/Gbx2 expression domains anterior to Hox expression regions. We identify this territory as the deutocerebral/tritocerebral boundary region in the embryonic Drosophila brain. Mutational inactivation of otd/Otx2 and unpg/Gbx2 result in the loss or misplacement of the brain-specific expression domains of Pax2/5/8 and Hox genes. In addition, otd/Otx2 and unpg/Gbx2 appear to negatively regulate each other at the interface of their brain-specific expression domains. Our studies demonstrate that the deutocerebral/tritocerebral boundary region in the embryonic Drosophila brain displays developmental genetic features similar to those observed for the midbrain/hindbrain boundary region in vertebrate brain development. This suggests that a tripartite organization of the embryonic brain was already established in the last common urbilaterian ancestor of protostomes and deuterostomes.

Headless flies produced by mutations in the paralogous Pax6 genes eyeless and twin of eyeless.

The two Pax6 gene homologs eyeless and twin of eyeless play decisive early roles in Drosophila eye development. Strong mutants of twin of eyeless or of eyeless are headless, which suggests that they are required for the development of all structures derived from eye-antennal discs. The activity of these genes is crucial at the very beginning of eye-antennal development in the primordia of eye-antennal discs when eyeless is first activated by the twin of eyeless gene product. This activation does not strictly depend on the Twin of eyeless protein, but is temperature-dependent in its absence. Twin of eyeless acts also in parallel to the eyeless gene and exerts functions that are partially redundant with those of Eyeless, while Eyeless is mainly required to prevent early cell death and promote eye development in eye-antennal discs.