A dynamically localized protease complex and a polar speciﬁcity factor control a cell cycle master regulator. Cell 124, 535–547. 15. Matroule, J.Y., Lam, H., Burnette, D.T., and Jacobs-Wagner, C. (2004). Cytokinesis monitoring during development: rapid pole-to-pole shuttling of a signaling protein by localized kinase and phosphatase in Caulobacter. Cell 118, 579–590. 16. Holtzendorff, J., Hung, D., Brende, P., Reisenauer, A., Viollier, P.H., McAdams, H.H., and Shapiro, L. (2004). Oscillating global regulators control the genetic circuit driving a bacterial cell cycle. Science 304, 983–987. 17. Bloom, J., and Cross, F.R. (2007). Novel role for Cdc14 sequestration: Cdc14 dephosphorylates factors that promote DNA replication. Mol. Cell. Biol. 27, 842–853. 1 Department of Biology, San Francisco State University, 1600 Holloway Avenue, San Francisco, California 94132, USA. 2 Biology Department, Santa Clara University, 500 El Camino Real, Santa Clara, California 95053, USA. E-mail: cstephens@scu.edu DOI: 10.1016/j.cub.2007.01.031 Vertebrate Gastrulation: The BMP Sticker Shock BMPs are essential regulators of cell fate during early embryonic development. Molecular genetics and in vivo imaging of cell behaviors in zebraﬁsh now demonstrate a role for BMPs in the control of cell adhesion. The work reveals an important new mechanism governing cell movements during gastrulation. John B. Wallingford 1, * and Richard M. Harland 2 Even though embryos of diverse vertebrate groups develop a similar body plan after neurulation, the most cursory inspection of early developmental stages shows enormous differences in how these animals reach this phylotypic stage. Many of the differences in early development reﬂect constraints of embryo nutrition, such as the need for a large yolk supply in egg-laying vertebrates versus the need to implant in the uterine wall in mammals [1]. But even with those obvious constraints, one of the shocking features of early vertebrate development is that completely different cell behaviors are used during gastrulation [2–4]. For example, the chick and the mouse show large-scale epithelial-to-mesenchymal transitions, whereby cells ingress from the epiblast to form the mesoderm. In contrast, sheets of cells remain coherent as they move into the embryo in the frog Xenopus. Do these differences illustrate deep divergence in the mechanism of gastrulation, or are we as yet too ignorant to see the underlying similarities? Of course any understanding of such issues will require us to know much more about the mechanisms that control cell behaviors during gastrulation. A new paper from Hammerschmidt and colleagues in this issue of Current Biology makes a very welcome contribution [5]. A popular hallmark of frog and ﬁsh gastrulation is the movement of convergence and extension, during which tightly packed cells converge and intercalate to lengthen and narrow the anterior-posterior axis. However, in the bony ﬁshes, the ventral and lateral cells initially migrate as loose cells towards the dorsal midline in a luxuriant extracellular matrix [6–8] — a behavior that is conspicuously absent from Xenopus [2–4] (Figure 1A,B). Previous analyses of mutant zebraﬁsh had indicated a negative role for bone morphogenetic proteins (BMPs) in this dorsal migration of lateral mesoderm cells [9]. However, as ventral identities are also speciﬁed by BMP signaling and ventral cells do not engage in robust cell movements, it can be difﬁcult to deconstruct how immediately BMPs affect morphogenesis (e.g. [10–12]). Are BMPs directly involved in the cell movements, or do BMPs simply specify a cell fate that then displays a certain morphogenetic property? In the new work, von der Hardt et al. [5] reveal that BMPs have direct effects on cell adhesion and thereby affect lateral cell movements during zebraﬁsh gastrulation. Not only do the authors separate cell fate and cell movement, but they also exploit high-resolution imaging of the cells to gain insight into the underlying cellular mechanisms. Global morphogenesis is something that can be disrupted all too easily, but cellular behaviors often respond to molecular manipulations with great speciﬁcity, so this type of analysis provides additional strength to the conclusions. To modulate BMP signals the authors used an array of mutants and morpholino oligonucleotide-mediated knockdowns — either of the genes for the ligands, the Alk8 receptor, or of the cytoplasmic transducer Smad5. BMP signaling was then restored locally by application of BMP beads. Ventral placement of such beads restored a normal dorsally directed migration of lateral mesoderm cells. Strikingly, however, dorsal placement of such beads in bmp2 morphants was sufﬁcient to drive cell migration ventrally, leading to a piling up of mesodermal cells on the ventral side (Figure 2A–C). One key experiment was to dissociate the effects of BMP signaling on cell movement from its effects on cell identity. The authors found that the migratory properties of cells were radically changed between embryos that otherwise showed the same expression patterns of markers of cell identity. These experiments exploited the observation that lamellipodial activity and migration require the hyaluron synthase Has2 [13]. In embryos in which has2 expression was Current Biology Vol 17 No 6 R206