Publisher Correction: A call for a better understanding of causation in cell biology.

In the above article, the name of the first author was spelled incorrectly. This has been corrected in the HTML and PDF versions of the article.

The extra-embryonic area opaca plays a role in positioning the primitive streak of the early chick embryo.

Classical studies have established that the marginal zone, a ring of extra-embryonic epiblast immediately surrounding the embryonic epiblast (area pellucida) of the chick embryo, is important in setting embryonic polarity by positioning the primitive streak, the site of gastrulation. The more external extra-embryonic region (area opaca) was thought to have only nutritive and support functions. Using experimental embryology approaches, this study reveals three separable functions for this outer region. First, juxtaposition of the area opaca directly onto the area pellucida induces a new marginal zone from the latter; this induced domain is entirely posterior in character. Second, ablation and grafting experiments using an isolated anterior half of the blastoderm and pieces of area opaca suggest that the area opaca can influence the polarity of the adjacent marginal zone. Finally, we show that the loss of the ability of such isolated anterior half-embryos to regulate (re-establish polarity spontaneously) at the early primitive streak stage can be rescued by replacing the area opaca by one from a younger stage. These results uncover new roles of chick extra-embryonic tissues in early development.

'Neighbourhood watch' model: embryonic epiblast cells assess positional information in relation to their neighbours.

In many developing and regenerating systems, tissue pattern is established through gradients of informative morphogens, but we know little about how cells interpret these. Using experimental manipulation of early chick embryos, including misexpression of an inducer (VG1 or ACTIVIN) and an inhibitor (BMP4), we test two alternative models for their ability to explain how the site of primitive streak formation is positioned relative to the rest of the embryo. In one model, cells read morphogen concentrations cell-autonomously. In the other, cells sense changes in morphogen status relative to their neighbourhood. We find that only the latter model can account for the experimental results, including some counter-intuitive predictions. This mechanism (which we name the 'neighbourhood watch' model) illuminates the classic 'French Flag Problem' and how positional information is interpreted by a sheet of cells in a large developing system.

The embryonic node behaves as an instructive stem cell niche for axial elongation.

In warm-blooded vertebrate embryos (mammals and birds), the axial tissues of the body form from a growth zone at the tail end, Hensen's node, which generates neural, mesodermal, and endodermal structures along the midline. While most cells only pass through this region, the node has been suggested to contain a small population of resident stem cells. However, it is unknown whether the rest of the node constitutes an instructive niche that specifies this self-renewal behavior. Here, we use heterotopic transplantation of groups and single cells and show that cells not destined to enter the node can become resident and self-renew. Long-term resident cells are restricted to the posterior part of the node and single-cell RNA-sequencing reveals that the majority of these resident cells preferentially express G2/M phase cell-cycle-related genes. These results provide strong evidence that the node functions as a niche to maintain self-renewal of axial progenitors.

Reflections on the past, present and future of developmental biology.

Developmental Biology embodies some of the most fundamental questions in Biology and can trace its roots back to several thousand years ago; the last 100 years have been particularly extraordinary. In part the advances have been fuelled by new technical advances and knowledge in many other areas, which have contributed to shaping the field as truly interdisciplinary. During those 100 years some of our predecessors identified some key questions and a few important principles especially by trying to find general rules that govern what cells are able to do and how they choose between different options, as well as principles of experimental design that can be used to uncover those rules even before we know their physicochemical underpinnings. But the field has been changing rapidly in the last two decades. Here I present a brief overview of some of the changes that have taken place over the last Century and a personal view of current directions. The picture that emerges is of some dark clouds on the horizon, so this is also a call to arms for our colleagues to try to regain what the field has been losing.

A niche for axial stem cells - A cellular perspective in amniotes.

The head-tail axis in birds and mammals develops from a growth zone in the tail-end, which contains the node. This growth zone then forms the tailbud. Labelling experiments have shown that while many cells leave the node and tailbud to contribute to axial (notochord, floorplate) and paraxial (somite) structures, some cells remain resident in the node and tailbud. Could these cells be resident axial stem cells? If so, do the node and tailbud represent an instructive stem cell niche that specifies and maintains these stem cells? Serial transplantation and single cell labelling studies support the existence of self-renewing stem cells and heterotopic transplantations suggest that the node can instruct such self-renewing behaviour. However, only single cell manipulations can reveal whether self-renewing behaviour occurs at the level of a cell population (asymmetric or symmetric cell divisions) or at the level of single cells (asymmetric divisions only). We combine data on resident cells in the node and tailbud and review it in the context of axial development in chick and mouse, summarising our current understanding of axial stem cells and their niche and highlighting future directions of interest.

Sequential changes in cellular properties accompanying amniote somite formation.

Somites are transient structures derived from the pre-somitic mesoderm (PSM), involving mesenchyme-to-epithelial transition (MET) where the cells change their shape and polarize. Using Scanning electron microscopy (SEM), immunocytochemistry and confocal microscopy, we study the progression of these events along the tail-to-head axis of the embryo, which mirrors the progression of somitogenesis (younger cells located more caudally). SEM revealed that PSM epithelialization is a gradual process, which begins much earlier than previously thought, starting with the dorsalmost cells, then the medial ones, and then, simultaneously, the ventral and lateral cells, before a somite fully separates from the PSM. The core (internal) cells of the PSM and somites never epithelialize, which suggests that the core cells could be 'trapped' within the somitocoele after cells at the surfaces of the PSM undergo MET. Three-dimensional imaging of the distribution of the cell polarity markers PKCζ, PAR3, ZO1, the Golgi marker GM130 and the apical marker N-cadherin reveal that the pattern of polarization is distinctive for each marker and for each surface of the PSM, but the order of these events is not the same as the progression of cell elongation. These observations challenge some assumptions underlying existing models of somite formation.

Exploring the roles of FGF/MAPK and cVG1/GDF signalling on mesendoderm induction and convergent extension during chick primitive streak formation.

During primitive streak formation in the chick embryo, cells undergo mesendoderm specification and convergent extension at the same time and in the same cells. Previous work has implicated cVG1 (GDF3) as a key factor for induction of primitive streak identity and positioning the primitive streak, whereas FGF signalling was implicated in regulating cell intercalation via regulation of components of the WNT-planar cell polarity (PCP) pathway. FGF has also been reported to be able to induce a primitive streak (but lacking the most axial derivatives such as notochord/prechordal mesendoderm). These signals emanate from different cell populations in the embryo, so how do they interact to ensure that the same cells undergo both cell intercalation and acquire primitive streak identity? Here we begin to address this question by examining in more detail the ability of the two classes of signals in regulating the two developmental events. Using misexpression of inducers and/or exposure to inhibitors and in situ hybridisation, we study how these two signals regulate expression of Brachyury (TBXT) and PRICKLE1 as markers for the primitive streak and the PCP, respectively. We find that both signals can induce both properties, but while FGF seems to be required for induction of the streak by cVG1, it is not necessary for induction of PRICKLE1. The results are consistent with cVG1 being a common regulator for both primitive streak identity and the initiation of convergent extension that leads to streak elongation.

Neural induction by the node and placode induction by head mesoderm share an initial state resembling neural plate border and ES cells.

Around the time of gastrulation in higher vertebrate embryos, inductive interactions direct cells to form central nervous system (neural plate) or sensory placodes. Grafts of different tissues into the periphery of a chicken embryo elicit different responses: Hensen's node induces a neural plate whereas the head mesoderm induces placodes. How different are these processes? Transcriptome analysis in time course reveals that both processes start by induction of a common set of genes, which later diverge. These genes are remarkably similar to those induced by an extraembryonic tissue, the hypoblast, and are normally expressed in the pregastrulation stage epiblast. Explants of this epiblast grown in the absence of further signals develop as neural plate border derivatives and eventually express lens markers. We designate this state as "preborder"; its transcriptome resembles embryonic stem cells. Finally, using sequential transplantation experiments, we show that the node, head mesoderm, and hypoblast are interchangeable to begin any of these inductions while the final outcome depends on the tissue emitting the later signals.

A 3D molecular atlas of the chick embryonic heart.

We present a detailed analysis of gene expression in the 2-day (HH12) embryonic chick heart. RNA-seq of 13 micro-dissected regions reveals regionalised expression of 15,570 genes. Of these, 132 were studied by in situ hybridisation and a subset (38 genes) was mapped by Optical Projection Tomography or serial sectioning to build a detailed 3-dimensional atlas of expression. We display this with a novel interactive 3-D viewer and as stacks of sections, revealing the boundaries of expression domains and regions of overlap. Analysis of the expression domains also defines some sub-regions distinct from those normally recognised by anatomical criteria at this stage of development, such as a previously undescribed subdivision of the atria into two orthogonal sets of domains (dorsoventral and left-right). We also include a detailed comparison of expression in the chick with the mouse and other species.

Snail2 and Zeb2 repress P-cadherin to define embryonic territories in the chick embryo.

Snail and Zeb transcription factors induce epithelial-to-mesenchymal transition (EMT) in embryonic and adult tissues by direct repression of E-cadherin transcription. The repression of E-cadherin transcription by the EMT inducers Snail1 and Zeb2 plays a fundamental role in defining embryonic territories in the mouse, as E-cadherin needs to be downregulated in the primitive streak and in the epiblast, concomitant with the formation of mesendodermal precursors and the neural plate, respectively. Here, we show that in the chick embryo, E-cadherin is weakly expressed in the epiblast at pre-primitive streak stages where it is substituted for by P-cadherin We also show that Snail2 and Zeb2 repress P-cadherin transcription in the primitive streak and the neural plate, respectively. This indicates that E- and P-cadherin expression patterns evolved differently between chick and mouse. As such, the Snail1/E-cadherin axis described in the early mouse embryo corresponds to Snail2/P-cadherin in the chick, but both Snail factors and Zeb2 fulfil a similar role in chick and mouse in directly repressing ectodermal cadherin genes to contribute to the delamination of mesendodermal precursors at gastrulation and the proper specification of the neural ectoderm during neural induction.

The role of the notochord in amniote vertebral column segmentation.

The vertebral column is segmented, comprising an alternating series of vertebrae and intervertebral discs along the head-tail axis. The vertebrae and outer portion (annulus fibrosus) of the disc are derived from the sclerotome part of the somites, whereas the inner nucleus pulposus of the disc is derived from the notochord. Here we investigate the role of the notochord in vertebral patterning through a series of microsurgical experiments in chick embryos. Ablation of the notochord causes loss of segmentation of vertebral bodies and discs. However, the notochord cannot segment in the absence of the surrounding sclerotome. To test whether the notochord dictates sclerotome segmentation, we grafted an ectopic notochord. We find that the intrinsic segmentation of the sclerotome is dominant over any segmental information the notochord may possess, and no evidence that the chick notochord is intrinsically segmented. We propose that the segmental pattern of vertebral bodies and discs in chick is dictated by the sclerotome, which first signals to the notochord to ensure that the nucleus pulposus develops in register with the somite-derived annulus fibrosus. Later, the notochord is required for maintenance of sclerotome segmentation as the mature vertebral bodies and intervertebral discs form. These results highlight differences in vertebral development between amniotes and teleosts including zebrafish, where the notochord dictates the segmental pattern. The relative importance of the sclerotome and notochord in vertebral patterning has changed significantly during evolution.

Calreticulin is a secreted BMP antagonist, expressed in Hensen's node during neural induction.

Hensen's node is the "organizer" of the avian and mammalian early embryo. It has many functions, including neural induction and patterning of the ectoderm and mesoderm. Some of the signals responsible for these activities are known but these do not explain the full complexity of organizer activity. Here we undertake a functional screen to discover new secreted factors expressed by the node at this time of development. Using a Signal Sequence Trap in yeast, we identify several candidates. Here we focus on Calreticulin. We show that in addition to its known functions in intracellular Calcium regulation and protein folding, Calreticulin is secreted, it can bind to BMP4 and act as a BMP antagonist in vivo and in vitro. Calreticulin is not sufficient to account for all organizer functions but may contribute to the complexity of its activity.

Multiple roles of timing in somite formation.

During development, vertebrate embryos produce serially repeated elements, the somites, on each side of the midline. These generate the vertebral column, skeletal musculature and dermis. They form sequentially, one pair at a time, from mesenchymal tissue near the tail. Somite development is a complex process. The embryo must control the number, size, and timing of somite formation, their subdivision into functional regions along three axes, regional identity such that somites develop in a region-specific way, and interactions with neighbouring tissues that coordinate them with nearby structures. Here we discuss many timing-related mechanisms that contribute to set up the spatial pattern.

A resegmentation-shift model for vertebral patterning.

Segmentation of the vertebrate body axis is established in the embryo by formation of somites, which give rise to the axial muscles (myotome) and vertebrae (sclerotome). To allow a muscle to attach to two successive vertebrae, the myotome and sclerotome must be repositioned by half a segment with respect to each other. Two main models have been put forward: 'resegmentation' proposes that each half-sclerotome joins with the half-sclerotome from the next adjacent somite to form a vertebra containing cells from two successive somites on each side of the midline. The second model postulates that a single vertebra is made from a single somite and that the sclerotome shifts with respect to the myotome. There is conflicting evidence for these models, and the possibility that the mechanism may vary along the vertebral column has not been considered. Here we use DiI and DiO to trace somite contributions to the vertebrae in different axial regions in the chick embryo. We demonstrate that vertebral bodies and neural arches form by resegmentation but that sclerotome cells shift in a region-specific manner according to their dorsoventral position within a segment. We propose a 'resegmentation-shift' model as the mechanism for amniote vertebral patterning.

Gata2 provides an early anterior bias and uncovers a global positioning system for polarity in the amniote embryo.

The first axis to be specified during vertebrate development is that between the site where gastrulation will begin and the opposite pole of the embryo (dorsoventral axis in amphibians and fish, anteroposterior in amniotes). This relies on Nodal activity, but different vertebrates differ in how this activity is positioned. In chick, the earliest known asymmetry is posterior expression of the TGFß-related factor Vg1, close to the future Nodal expression domain. Here we show that the transcription factor Gata2 is expressed anteriorly before this stage. Gata2 influences the site of primitive streak formation and its role is independent from, and upstream of, Vg1 and Wnt. However, although Vg1 is required for streak formation, Gata2 does not act as an absolute anterior specifier, but provides an anterior bias. These findings point to previously unsuspected global determinants of polarity of the early amniote embryo.

The hypoblast (visceral endoderm): an evo-devo perspective.

When amniotes appeared during evolution, embryos freed themselves from intracellular nutrition; development slowed, the mid-blastula transition was lost and maternal components became less important for polarity. Extra-embryonic tissues emerged to provide nutrition and other innovations. One such tissue, the hypoblast (visceral endoderm in mouse), acquired a role in fixing the body plan: it controls epiblast cell movements leading to primitive streak formation, generating bilateral symmetry. It also transiently induces expression of pre-neural markers in the epiblast, which also contributes to delay streak formation. After gastrulation, the hypoblast might protect prospective forebrain cells from caudalizing signals. These functions separate mesendodermal and neuroectodermal domains by protecting cells against being caught up in the movements of gastrulation.

Sharpening of the anterior neural border in the chick by rostral endoderm signalling.

The anterior border of the neural plate, presumed to contain the prospective peripheral portion (roof) of the prospective telencephalon, emerges within a vaguely defined proneural ectodermal region. Fate maps carried out at HH4 in the chick reveal that this region still produces indistinctly neural, placodal and non-neural derivatives; it does not express neural markers. We examined how the definitive anterior border domain of the rostral forebrain becomes established and comes to display a neural molecular profile, whereas local non-neural derivatives become separated. The process, interpreted as a border sharpening mechanism via intercalatory cell movements, was studied using fate mapping, time-lapse microscopy and in situ hybridization. Separation of neural and non-neural domains proceeds along stages HH4-HH4+, is well advanced at HH5, and is accompanied by a novel dorsoventral intercalation, oriented orthogonal to the border, that distributes transitional cells into molecularly distinct neural and non-neural fields. Meanwhile, neuroectodermal Sox2 expression spreads peripherally from the neighbourhood of the node, reaching the nascent anterior border domain at HH5. We also show that concurrent signals from the endodermal layer are necessary to position and sharpen the neural border, and suggest that FGF8 might be a component of this signalling.