Chewing the Fat over PCP and growth

The Drosophila protocadherin Fat (Ft) affects planar cell polarity (PCP) but also inhibits the overgrowth of imaginal discs by activating the Hippo pathway. The intracellular domain (ICD) mediates much of Ft activity, but are its PCP and Hippo pathway activities separable? To answer this question, Hitoshi Matakatsu and Seth Blair have undertaken a detailed structure-function analysis of the Ft ICD (see p. 1498). They identify PCP and Hippo pathway-specific domains in the Ft ICD and, surprisingly, report that these domains do not map to any previously identified protein interaction domains in the Ft ICD, nor, with one exception, to the regions that are highly conserved in mammalian Fat4. They also confirm and extend evidence showing that the extracellular domain of Ft has activities in both the PCP and Hippo pathways that are mediated by the protocadherin Dachsous. Based on their results, the authors conclude that the PCP and Hippo pathway activities of Fat and Dachsous are largely separable.

Robo-Slit signals regulate CNS motoneuron axon exit

In mammals, signals transmitted from the central nervous system (CNS) to muscles via motoneurons control movement. To form these circuits, motoneurons extend their axons out of the CNS at specialised exit points. Here (p. 1435), Zaven Kaprielian and colleagues use mouse spinal accessory motoneurons (SACMNs) to investigate how this essential phase of motor axon pathfinding is controlled. SACMNs, which innervate neck and back muscles, leave the spinal cord at lateral exit points (LEPs). In mice lacking the homeodomain transcription factor Nkx2.9, the researchers report, SACMN axons project normally to the LEP but fail to exit the CNS. Robo2 expression in SACMNs is downregulated in Nkx2.9 null mice, they report, and SACMN axons fail to exit the spinal cord in Robo2-deficient animals. Finally, the Robo2 ligands Slit1-3 are present at the LEP and SACMN axons fail to exit the CNS in Slit-deficient mice. Together, these results suggest that Nkx2.9 controls SACMN axon exit from the CNS by regulating Robo2-Slit signalling.

sRNA paths to plant female gamete development

During the first phase of Arabidopsis female gamete formation (megasporogenesis), a somatic ovule cell differentiates into a megaspore mother cell and divides to generate four haploid megaspores. In the next phase (megagametogenesis), one of these megaspores undergoes syncytial mitosis and differentiates to form the female gametophyte. It’s known that a somatic small RNA (sRNA) pathway restricts reproductive potential to this functional megaspore but what controls the megasporogenesis to megagametogenesis transition? Here (p. 1399), Matthew Tucker and co-workers examine gene expression patterns in ovule tissues and show that an sRNA pathway is also involved in this phase of female gamete formation. The researchers report that ARGONAUTE5 (AGO5), a putative sRNA pathway effector, is expressed around reproductive cells during megasporogenesis and show that a unique semi-dominant ago5-4 insertion allele disrupts the initiation of megagametogenesis. Expression of a viral RNAi suppressor protein in the somatic cells flanking the megaspores produces a similar phenotype. Thus, the researchers conclude, at least two somatic sRNA pathways contribute to female gametophyte development in Arabidopsis.

PHD-fingers point to meristem initiation

Apical meristems, which are indispensable for plant growth and development, contain stem cells and the organizer, both of which are specified during early embryogenesis. During root meristem initiation, specification of hypophysis (the precursor of the organizer) requires the transcription factor MONOPTEROS (MP), which functions in part by activating the expression of TARGET of MP (TMO) transcription factors. But how does MP activate the expression of these genes in the context of chromatin to regulate meristem initiation? On p. 1391, Yoshibumi Komeda and colleagues report that the plant homeodomain (PHD)-finger proteins OBERON (OBE) and TITANIA (TTA) are essential for MP-dependent embryonic root meristem initiation in Arabidopsis. They show that these PHD-finger proteins interact with each other and that, although MP expression is unaltered by mutations in OBE/TTA genes, the expression of TMO5 and TMO7 is locally lost in obe1 obe2 embryos. These and other data indicate that PHD-finger protein complexes control the activation of transcription factor genes during root meristem initiation.

Joining forces in the neural tube

In developing vertebrates, neural tube closure (NTC) – the formation of the neural tube from a sheet of neural ectoderm – requires both neural ectoderm and non-neural ectoderm. But, whereas cell shape changes, cell rearrangement and cell division in the neural ectoderm are essential for NTC, the cellular changes in the non-neural ectoderm that contribute to NTC are unclear. Now, on p. 1417, Naoto Ueno and co-workers use digital scanned laser light sheet fluorescence microscopy to examine the movements of non-neural ectoderm cells in Xenopus embryos during NTC. The researchers show that the collective movement of deep-layer non-neural ectoderm cells towards the dorsal midline helps to drag along the superficial non-neural ectoderm during NTC. Inhibition of this movement by deletion of integrin β1 function, they report, blocks NTC completion. By contrast, oriented cell division, cell shape changes and cell rearrangement in the non-neural ectoderm have little or no role in NTC. Together, these results suggest that a global reorganisation of embryonic tissues is involved in NTC.

Neurula rotation breaks ascidian LR symmetry

In many vertebrate embryos, monocilia-generated fluid flow in the node establishes left-sided mesodermal expression of nodal and breaks embryonic symmetry, which leads to the stereotypical left-right (LR) asymmetry seen in most animals. Tadpole larvae of the ascidian Halocynthia roretzi also show LR asymmetry but ascidian embryos have no cavity in which fluid flow can be generated. Now, Kazuhiko Nishide and colleagues show that in H. roretzi left-sided expression of nodal, triggered by neurula rotation, in the epidermis at the late neurula stage is required for LR asymmetry (p. 1467). Shortly before the onset of nodal expression, they report, the neurula rotates within the vitelline membrane until its left side is orientated downwards. Monocilia in the epidermis probably generate the driving force for this rotation. Moreover, experimental perturbation of neurula rotation indicates that contact between the left epidermis and the vitelline membrane generated through neurula rotation promotes left-sided nodal expression. Neurula rotation could also be involved in the establishment of LR asymmetry in other ascidian species, suggest the researchers.

Plus…

Diverse roles for VEGF-A in the nervous system

Recent studies, reviewed by Francesca Mackenzie and Christiana Ruhrberg, have shown that vascular endothelial growth factor A, which is best known for its roles in blood vessel growth, promotes a wide range of neuronal functions.

miR-125 seals hESC neural fate

MicroRNAs, small non-coding RNAs, have recently emerged as key regulators of embryonic development. In particular, they can coordinate cell fate determination by blocking alternative cell fate choices. Here (p. 1247), Alexandra Benchoua and colleagues report that the microRNA miR-125 contributes to the neural specification of pluripotent human embryonic stem cells (hESCs). By using a culture system that promotes hESC neuralization, the researchers show that miR-125 is expressed in a time window compatible with a role in neural commitment in vitro. They show that miR-125 promotes the conversion of pluripotent cells into SOX1-positive neural precursors by, at least in part, blocking the expression of SMAD4, a key regulator of pluripotent stem cell lineage commitment that promotes non-neural cell fates. Finally, the researchers show that expression of miR-125 is responsive to the level of TGFβ-like molecules. Together, these results identify a central role for miR-125 in the irreversible neural lineage commitment of pluripotent stem cells in response to external stimuli.

GLE1: keeping neural precursors alive

Lethal congenital contracture syndrome 1 (LCCS1) is a prenatally fatal, autosomal recessive human disorder. Affected foetuses have multiple defects, including limb deformities (contractures), loss of voluntary muscle movement and a distinct neuropathology. Mutations in GLE1, which encodes a protein involved in mRNA export and translation, have been implicated in LCCS1 and, on p. 1316, Susan Wente and co-workers investigate the link between Gle1 function and LCCS1 pathology using zebrafish as a model system. They report that disruption of Gle1 function produces phenotypes in zebrafish embryos that parallel those of human LCCS1 foetuses, including a reduction of motoneurons and aberrant arborization of motor axons. Surprisingly, the researchers report, apoptosis of neural precursors, rather than degeneration of differentiated neurons, as previously suggested, causes the motoneuron deficiency. The researchers propose, therefore, that rapidly dividing cells, including organ precursors in both neuronal and non-neuronal tissue, have a high demand for Gle1 activity, and that apoptosis of these precursors because of Gle1 deficiency produces the pleiotropic abnormalities seen in LCCS1 foetuses.

Separating karrikin and strigolactone effects

Karrikins are smoke-derived butenolides that, by stimulating seed germination and enhancing seed responses to light, allow plants to exploit reduced competition for light, water and nutrients after wildfires. By contrast, strigolactones are plant-derived butenolides that regulate shoot and root architecture. In Arabidopsis, responses to both classes of molecule require the F-box protein MAX2, so how are the physiologically distinct responses to karrikins and strigolactones achieved? Here, Mark Waters and colleagues suggest that the answer to this puzzle lies with evolutionary specialization within the DWARF14 superfamily of α/β hydrolases (see p. 1285). In rice, strigolactone-dependent inhibition of shoot branching requires DWARF14. The researchers show that, in Arabidopsis, the DWARF14 orthologue AtD14 is necessary for normal strigolactone responses, whereas the AtD14 paralogue KAI2 is required for karrikin responses. Notably, the expression patterns of AtD14 and KAI2 are consistent with the plant’s capacity to respond to specific growth regulators at different developmental stages. Thus, AtD14 and KAI2 define proteins that permit the separate regulation of strigolactone and karrikin signalling by MAX2.

Foxj1: Not(o) enough for node function

During embryonic patterning in amniotes, cilia in the node generate a leftward flow of extra-embryonic fluid that establishes left-right asymmetry. Now, on p. 1276, Achim Gossler and co-workers report that Noto and Foxj1 (two key transcription factors that are expressed in the node) differentially regulate mouse node formation, nodal ciliogenesis and cilia positioning. Noto, which controls node morphogenesis, nodal ciliogenesis and left-right asymmetry in mice, acts upstream of Foxj1 but the role of Foxj1 in nodal ciliogenesis is unclear. To address this issue, the researchers analyse mouse embryos deficient for Foxj1 and embryos in which the Foxj1-coding sequence replaces the Noto-coding sequence. Foxj1 expressed from the Noto locus is functional and restores the formation of motile nodal cilia. However, in the absence of functional Noto, Foxj1 is insufficient for the correct positioning of these cilia and cannot restore normal node morphology. Thus, the researchers conclude, Noto regulates nodal ciliogenesis through Foxj1 but regulates node morphogenesis and cilia localization independently of Foxj1.

Engrailed 1 shapes the skull

Bones in the vertebrate skull are connected by cranial sutures: flexible fibrous joints that are essential for normal postnatal brain growth. The coronal suture separates the parietal and frontal bones, which are derived from cephalic paraxial mesoderm (Mes) cells and neural crest (NeuC) mesenchyme, respectively. But where do the mesenchymal precursors that generate the coronal suture come from? Ron Deckelbaum, Cynthia Loomis and colleagues (see p. 1346) use genetic fate mapping to show that, in mice, these precursors originate from hedgehog-responsive Mes cells that migrate into a supraorbital domain to establish a lineage boundary with NeuC mesenchyme. Importantly, the researchers show that the transcription factor Engrailed 1 (En1) regulates cell movement and NeuC/Mes lineage boundary positioning during coronal suture formation, and also prevents premature osteogenic conversion of the sutural mesenchyme by controlling the level of fibroblast growth factor receptor 2. Further investigation of these molecular mechanisms could lead to cell-based therapies for craniosynostosis (premature suture closure), which can cause craniofacial deformities and impaired brain development.

Well wrapped: vimentin regulates myelination

Myelination of peripheral nerves, which is essential for the rapid and efficient propagation of electrical messages along axons, has to be carefully controlled during development. However, the molecular mechanisms that determine myelin sheath thickness are only partly understood. Here (p. 1359), Stefano Carlo Previtali and colleagues report that the intermediate filament protein vimentin negatively regulates peripheral nerve myelination. Vimentin is highly expressed in Schwann cells and neurons during embryonic development and during nerve regeneration. The researchers show that loss of neuronal vimentin results in peripheral nerve hypermyelination in transgenic mice and in a myelinating co-culture system. This increased myelin sheath thickness occurs, they report, due to an increase in the levels of axonal neuregulin1 (NRG1) type III, a key regulator of myelin formation in peripheral nerves. Finally, they show that vimentin acts synergistically with the protease TACE to regulate NRG1 type III levels. Together, these results provide new insights into peripheral nerve myelination and identify potential targets for the treatment of peripheral neuropathies.

Plus…

Fluid flows and forces in development: functions, features and biophysical principles

Cells are subjected to various forces during morphogenesis, including those resulting from microscopic fluid flows. Here, Julien Vermot and colleagues review the biomechanical features and the physiological functions of biological fluid flows during development. See the Review article on p. 1229.

Neural circuit building

During development, sensory neurons form neural circuits with motoneurons. Although the anatomical details of these circuits are well described, less is known about the molecular mechanisms underlying their formation. To investigate the involvement of motoneurons in sensory neuron development, Hirohide Takebayashi and colleagues analyse sensory neuron phenotypes in the dorsal root ganglia (DRG) of Olig2 knockout mouse embryos, which lack motoneurons (see p. 1125). These embryos, they report, also have reduced numbers of sensory neurons but increased numbers of apoptotic cells in the DRG. In addition, the axonal projections of the sensory neurons in these embryos are abnormal. Because neurotrophin 3 (Ntf3) and its receptors are strongly expressed in motoneurons and sensory neurons, respectively, the researchers also investigate whether Ntf3 is one of the motoneuron-derived factors that regulate sensory neuron development. Notably, the sensory neuron phenotypes in Ntf3 conditional knockout embryos resemble those observed in Olig2 knockout embryos. Thus, the researchers propose, motoneuron-derived Ntf3 is a pre-target neurotrophin that is essential for survival and axonal projection of sensory neurons.

SIK3 bones up on chondrocyte hypertrophy

Most vertebrate bones develop through endochondral ossification. During this process, proliferating chondrocytes form a cartilage scaffold, differentiate into hypertrophic chondrocytes and die. The cartilage scaffold is then degraded and replaced by bone. Chondrocyte hypertrophy is, therefore, crucial for endochondral ossification. On p. 1153, Noriyuki Tsumaki and colleagues identify salt-inducible kinase 3 (SIK3) as an essential factor for chondrocyte hypertrophy in mice. SIK3-deficient mice, the researchers report, exhibit dwarfism, bone malformation and accumulation of chondrocytes in various bones. These phenotypes, they suggest, are due to impaired chondrocyte hypertrophy. Consistent with this suggestion, SIK3 is expressed in prehypertrophic and hypertrophic chondrocytes in the embryonic bones and postnatal growth plates of wild-type mice. Other experiments show that SIK3 anchors histone deacetylase 4 (HDAC4) in the cytoplasm, thereby releasing MEF2C, a transcription factor that facilitates chondrocyte hypertrophy, from suppression by HDAC4 in the nucleus. These results suggest that the regulation of HDAC4 by SIK3 is important for the progression of chondrocyte hypertrophy during skeletal development.

PRMT5 and stem cell function

Stem cells are essential for growth, development, gamete production and tissue homeostasis but what regulates their maintenance and function in vivo? On p. 1083, Phillip Newmark and colleagues report that the conserved protein arginine methyltransferase PRMT5 promotes stem cell function in planarian flatworms. These organisms contain a population of adult stem cells called neoblasts that can regenerate all the worm’s tissues. Neoblasts characteristically contain chromatoid bodies, large cytoplasmic ribonucleoprotein (RNP) granules similar to structures that are present in the germline of many organisms. The researchers show that, like germline RNP granules, chromatoid bodies contain proteins bearing symmetrical dimethylarginine (sDMA) modifications, probably including the PIWI family member SMEDWI-3. PRMT5 is responsible for sDMA modification of these proteins, they report, and PRMT5 depletion results in fewer chromatoid bodies, fewer neoblasts, and defects in regeneration, growth and homeostasis. Together, these results identify new chromatoid body components that are involved in neoblast function and add to the evidence that suggests that sDMA modification of proteins stabilises RNP granules.

BR(in)G1 on male meiosis

Mammalian germ cell development and gametogenesis involve several genome-wide changes in epigenetic modifications and chromatin structure. Here (p. 1133), Terry Magnuson and co-workers explore the role of the mammalian SWI/SNF chromatin-remodelling complex during spermatogenesis in mice. The researchers report that levels of the SWI/SNF catalytic subunit brahma-related gene 1 (BRG1) peak during the early stages of meiosis. Consistent with this expression pattern, germline ablation of Brg1 produces germ cells that arrest during prophase 1, the stage of meiosis during which the induction and repair of DNA double-strand breaks generates recombination between homologous chromosomes. In line with the timing of their meiotic arrest, BRG1-depleted spermatocytes accumulate unrepaired DNA and fail to complete synapsis. They also exhibit global alterations to histone modifications and chromatin structure, including alterations that are associated with DNA damage and heterochromatin. The researchers propose, therefore, that BRG1 has an essential role in spermatogenesis and that BRG1-containing complexes function in the programmed recombination and repair events that occur during meiosis.

Binary route to (non)-neural competence

During gastrulation, neural crest and cranial placodes originate at the neural plate border and from an adjacent territory, respectively. But do these ectodermal tissues arise from a common precursor or from neural and non-neural ectoderm (the binary competence model)? On p. 1175, Gerhard Schlosser and colleagues use tissue grafting in Xenopus embryos to tackle this controversy. They show that, at neural plate stages, competence for induction of neural plate, border and crest markers is restricted to neural ectoderm, whereas competence for induction of panplacodal markers is confined to non-neural ectoderm. The homeobox protein Dlx3 and the transcription factor GATA2 are both required cell-autonomously for panplacodal and epidermal marker expression in non-neural ectoderm, they report. Moreover, the ectopic expression of Dlx3 (but not GATA2) in the neural plate is sufficient to induce non-neural markers, whereas the overexpression of Dlx3 or GATA2 suppresses neural plate, border and crest markers in the neural plate. Together, these results support the binary competence model and implicate Dlx3 in the regulation of non-neural competence.

Morphogen-based simulation of fin development

One of the greatest challenges in developmental biology is to understand how shape and size are controlled during development. Interactions between growth and pattern formation mechanisms are key drivers of morphogenesis but are difficult to study experimentally because of the highly dynamic nature of development in space and time. Here (p. 1188), Anne-Gaëlle Rolland-Lagan and co-workers use simulation modelling to explore how mobile signals, such as morphogens, might coordinate growth and patterning during zebrafish caudal fin development and regeneration. The zebrafish fin comprises 16 to 18 bony rays, each of which contains multiple joints along its proximodistal axis that give rise to segments. The researchers propose a model in which the interaction of three postulated morphogens can account for the available experimental data on fin growth and joint patterning and for the regeneration of a properly shaped fin following amputation. This simple, plausible model provides a theoretical framework that could guide future searches for the molecular regulators of fin growth and regeneration.

Plus…

CTCF: insights into insulator function during development

The nuclear protein CTCF when bound to insulator sequences can prevent undesirable crosstalk between genomic regions and can shield genes from enhancer function. Here, Rainer Renkawitz and colleagues discuss the mechanisms underlying developmentally regulated CTCFdependent transcription. See the Primer on p. 1045

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

Claudio Stern and Karen Downs discuss the function and evolution of the chick hypoblast and the visceral endoderm in mouse, highlighting the common roles played by these tissues. See the Review on p. 1059