© 2006 Nature Publishing Group Despite individual random variations, the external bilateral symmetry of the vertebrate body plan is almost perfect. However, on the interior things are not sym-metrical1: the left side contains most of the heart, the stomach, the pancreas and the spleen, whereas the right side contains most of the liver and the gall bladder. In addition, the left lung has fewer lobes than the right lung, and the gut coils anticlockwise. These asymmetries are uniform among individuals within a species, and departures from this situation in humans result in severe medical conditions.In addition to its relevance to human health, how vertebrate left–right (LR) asymmetry is established is an important question for developmental biologists, and several insights have been gained over the past 10 years from studies of chick, Xenopus laevis, mouse and zebrafish embryos2–6. Although they are not necessarily universally applicable, the development of LR asymmetry can be broken down into the following main steps (FIG. 1). First, an initial event breaks the bilateral symmetry of the embryo, probably by converting cues encoded in the already established anterior–posterior and dorsal–ventral axes into LR information7. Laterality cues are then transferred to the embryo node (Spemann’s organizer in X. laevis or Hensen’s node in birds and mammals) or its derivatives (Kupffer’s vesicle in teleost fish). Subsequently, LR asymmetries are established in and/or around the node, after which LR information is conveyed from the node to the left lateral plate mesoderm (LPM). Side-specific gene-expression domains are then established and stabilized along wide areas of both the left and right LPM. Finally, LR information is trans-ferred to the organ primordia, so that left- and right-side specific morphogenetic programmes are executed.It has become clear that the establishment of LR asymmetries is controlled by robust genetic and epige-netic mechanisms, some of which show a remarkable degree of evolutionary conservation, whereas others seem to be species-specific. Although our knowledge of some of the steps involved is still fragmentary, many insights have been gained recently into the molecular and cellular mechanisms that underlie and regulate these processes, especially those that occur at the early stages of LR asymmetry determination. In addition to these advances, exciting new results show how LR asym-metrical patterning is integrated within the context of a bilaterally symmetrical body plan.Breaking the bilateral symmetry of the embryoHow is the initial bilateral symmetry of the embryo con-sistently broken in one direction? One way to investigate this is to search for the earliest sign of laterality infor -mation, which might provide clues as to how this information is generated. Genetic, pharmacological and microsurgical approaches have identified progressively earlier requirements for the correct establishment of LR asymmetries in frog, chick and zebrafish embryos. Nevertheless, despite years of intensive research, no satisfactory explanation has been provided for the initial symmetry-breaking event in any of these species.*Center of Regenerative Medicine in Barcelona and Institució Catalana de Recerca i Estudis Avançats (ICREA), Doctor Aiguader 80, 08003 Barcelona, Spain.‡Gene Expression Laboratory, Salk Institute for Biological Studies, 10010 North Torrey Pines Road, La Jolla, California 92037, USA.Correspondence to J.C.I.B. e-mail: [email protected]:10.1038/nrg1830Embryo nodeA transient structure located at the anterior tip of the primitive streak in embryos of amniotes (birds, reptiles and mammals); also known as Hensen’s node in birds and mammals. The embryo node functions as the gastrula organizer and is therefore functionally equivalent to the dorsal lip of the blastopore (Spemann’s organizer) in amphibians and the shield of teleost fishes.Left–right asymmetry in the vertebrate embryo: from early information to higher-level integrationÁngel Raya* and Juan Carlos Izpisúa Belmonte‡Abstract | Although vertebrates seem to be essentially bilaterally symmetrical on the exterior, there are numerous interior left–right asymmetries in the disposition and placement of internal organs. These asymmetries are established during embryogenesis by complex epigenetic and genetic cascades. Recent studies in a range of model organisms have made important progress in understanding how this laterality information is generated and conveyed to large regions of the embryo. Both commonalities and divergences are emerging in the mechanisms that different vertebrates use in left–right axis specification. Recent evidence also provides intriguing links between the establishment of left–right asymmetries and the symmetrical elongation of the anterior–posterior axis.NATURE REVIEWS | GENETICS VOLUME 7 | APRIL 2006 | 283REVIEWS© 2006 Nature Publishing Group VDRLeft LPMMidlineAPRight LPMNodeRNodalexpressionLRVDAPhgLbaLdcfeThe mouse embryo might represent an exception to this. Unlike in frog, chick or zebrafish, no LR asym-metrical features, or requirements for LR asymmetry, have been identified in the mouse embryo before node formation. Furthermore, a mechanism that is solely based on the function of the node (the ‘nodal flow’) has been shown to generate laterality information de novo and therefore to be potentially sufficient to break the initial bilateral symmetry of the embryo. The nodal flow refers to the leftward flow of extracellular fluid that is generated in the mouse Hensen’s node by cilia that line the ventral side of this structure. Recent insights into this mechanism, and related hypotheses, are presented in detail below. Here we note that although the nodal flow is crucial for LR patterning in the mouse, whether it represents the initial symmetry-breaking event is a mat-ter of debate, and several alternative possibilities can be proposed2,4,6,8–13.Recent evidence casts doubt on two extreme hypoth-eses that have been previously proposed — namely that either nodal cilia provide a general mechanism for breaking the initial symmetry of all vertebrate embryos14, or nodal cilia are themselves not impor-tant for LR asymmetry determination15. There are currently two equally plausible models. In the first, the nodal flow provides the symmetry-breaking event in the mouse embryo and is likely to represent an innovation of rodents (or, more generally, of