MAPK signaling is necessary for neurogenesis in Nematostella vectensis

Abstract

Background: The nerve net of Nematostella is generated using a conserved cascade of neurogenic transcription factors. For example, NvashA, a homolog of the achaete-scute family of basic helix-loop-helix transcription factors, is necessary and sufficient to specify a subset of embryonic neurons. However, positive regulators required for the expression of neurogenic transcription factors remain poorly understood.

Results: We show that treatment with the MEK/MAPK inhibitor U0126 severely reduces the expression of known neurogenic genes, Nvath-like, NvsoxB(2), and NvashA, and known markers of differentiated neurons, suggesting that MAPK signaling is necessary for neural development. Interestingly, ectopic NvashA fails to rescue the expression of neural markers in U0126-treated animals. Double fluorescence in situ hybridization and transgenic analysis confirmed that NvashA targets represent both unique and overlapping populations of neurons. Finally, we used a genome-wide microarray to identify additional patterning genes downstream of MAPK that might contribute to neurogenesis. We identified 18 likely neural transcription factors, and surprisingly identified ~40 signaling genes and transcription factors that are expressed in either the aboral domain or animal pole that gives rise to the endomesoderm at late blastula stages.

Conclusions: Together, our data suggest that MAPK is a key early regulator of neurogenesis, and that it is likely required at multiple steps. Initially, MAPK promotes neurogenesis by positively regulating expression of NvsoxB(2), Nvath-like, and NvashA. However, we also found that MAPK is necessary for the activity of the neurogenic transcription factor NvashA. Our forward molecular approach provided insight about the mechanisms of embryonic neurogenesis. For instance, NvashA suppression of Nvath-like suggests that inhibition of progenitor identity is an active process in newly born neurons, and we show that downstream targets of NvashA reflect multiple neural subtypes rather than a uniform neural fate. Lastly, analysis of the MAPK targets in the early embryo suggests that MAPK signaling is critical not only to neurogenesis, but also endomesoderm formation and aboral patterning.

Fig. 1

Dose-dependent effects of the MEK inhibitor U0126 on embryonic gene expression. Control blastula stages at 24 hours post fertilization (a, g, m) and embryos treated with increasing concentrations of U0126 (b–f, h–l, n–r). In situ hybridization on blastula stages using Nvsprouty (a–f), Nvbrachyury (g–l), or NvfgfA1 (m–r) antisense probes. All images are lateral views with the presumptive endomesoderm (animal pole, future oral pole) to the top. The insets correspond to animal pole views. Numbers in lower left corner correspond to the number of animals with phenotype pictured / total number of animals assayed

Fig. 2

U0126 blocks ERK activation, endomesoderm formation, and gastrulation. a–e Control embryos. f–j U0126-treated embryos. a, f Confocal z-sections using anti-phospho-ERK (cyan) to visualize activated ERK. The gray dotted lines indicate the outline of the embryo and yellow arrows the accumulation of phosphorylated ERK (pERK)-positive cells above background levels. b–e, g–j Confocal z-sections using phalloidin (green) to show f-actin filaments and propidium iodide (red) to visualize the nuclei. a, b, f, g Blastula stages [24 hours post fertilization (hpf)]. c, h Late gastrula stages (48 hpf). d, i Early planula (72 hpf). e, j Late planula (96 hpf). All images are lateral views with the animal/oral pole to the top. The insets in a and f correspond to animal pole views. Ratios in g–j indicate the number of embryos displaying the phenotype shown in the image to the total number of analyzed embryos

Fig. 3

U0126 treatment results in a global decrease in neurogenic transcription factors. mRNA in situ expression of Nvath-like (A), NvsoxB(2) (B), and NvashA (C) in control embryos treated with in 1/3× artificial seawater (ASW) with 0.1 % dimethyl sulfoxide (DMSO). Expression of Nvath-like (A’), NvsoxB(2) (B’), and NvashA (C’) in animals treated with 15 μm U0126 in 1/3× ASW with 0.1 % DMSO. Embryos were classified and quantified as the percent having normal expression, weak expression, or no expression. Refer to key in figure for classification of phenotypes. The phenotypic class with the highest percentage of embryos is indicated. All embryo images are of early gastrula stage. The main figure panels are ectodermal focal planes of lateral views with the presumptive oral side to the left. The insets show deeper focal planes used to confirm embryonic stage

Fig. 4

NvashA is insufficient to rescue neuronal loss resulting from U0126 treatment. mRNA in situ expression of each gene is indicated in control embryos raised in 1/3× artificial seawater (ASW) with 0.1 % dimethyl sulfoxide (DMSO) (A–H), compared to embryos treated with 15 μm U0126 in 1/3× ASW with 0.1 % DMSO (A’–H’), and compared to embryos injected with NvashA:venus mRNA and treated with 15 μm U0126 in 1/3× ASW with 0.1 % DMSO (A”–H”). Genes are referred to by either name or protein ID number used in the Nematostella genome database v1.0 (http://genome.jgi.doe.gov/Nemve1/Nemve1.home.html). Embryos were classified and quantified as the percent having normal expression, weak expression, or no expression. The phenotypic class with the highest percentage of embryos is indicated. Treatment with U0126 strongly reduced all neural gene expression (compare A’–H’ with A–H). Misexpression of NvashA in U0126-treated animals was not sufficient to rescue neuronal loss induced by U0126 treatment (compare A”–H” to both A–H and A’–H’). All animals in panels A–G, A’–G’, and A–-G” are aboral ectodermal focal planes of 48 hours post fertilization embryos (late gastrula stage). Embryos in panels H, H’, and H” are a lateral view with oral side facing towards the left

Fig. 5

NvashA regulates multiple neuronal subtypes. Shown are three-dimensional projections of two juvenile polyps shown with the oral side towards the top of the image. a Double fluorescent in situ hybridization of NvLWamide-like (red) and Nvserum amyloid A-like (green) in a late gastrula stage embryo. b The transgenic line for NvLWamide::mcherry expression is shown. Neural soma and neurites are observed throughout the body column and tentacles. Ectodermal neurons with three projections are observed in the body column (white arrow), which are not found in the other transgenic line. c The transgenic line for Nvserum amyloid A-like::mcherry is shown. There are many fewer neurons compared to NvLWamide::mcherry, but characteristic neurons are present. The U-shaped neuron that has two orally projecting neurites (pink arrow) is specific to this transgenic line. Both lines have neurons that are located just over the mesenteries and send projections orally and aborally in neural tracts overlaying the mesenteries (yellow arrows). In all images, asterisks indicate relative position of mouth

Fig. 6

U0126 targets expressed in individual cells throughout the ectoderm. Wild type salt-and-pepper gene expression analysis by in situ hybridization of genes differentially regulated by U0126 treatments. All animals are either blastula [24 hours post fertilization (hpf) – A–F, M–R, Y-Zd] or gastrula (48 hpf – G–L, S–X, Ze–Zj) stages. All images are lateral views with the animal pole to the top. The insets correspond to surface views. Antisense probes used as indicated. Green stars in Zd and Zj indicate that this gene was upregulated under U0126 conditions. All other genes were downregulated

Fig. 7

NvfgfRa does not regulate U0126-dependent salt-and-pepper gene expression. Relative fold change calculated from quantitative polymerase chain reaction analysis of triplicate injections of the NvfgfRa morpholino (MO) or a control MO is graphed. Broadly expressed genes, Nvsfrp1/5, NvotxC, and NvfgfRa, are included as controls. The remaining genes are the salt-and-pepper expressed genes identified by the U0126 microarray. The red box indicates a region (1.5× to −1.5× fold change) that was defined as the cut off for a significant change in expression. Error bars represent standard error

Fig. 8

NvashA regulates a subset of U0126-dependent salt-and-pepper expressed genes. a Relative fold change calculated from quantitative polymerase chain reaction analysis of triplicate injections of the NvashA morpholino (MO) versus a control MO (dark gray bars), or NvashA:venus-injected versus control venus-injected animals (light gray bars). Broadly expressed NvfgfRa, Nvsfrp1/5, Nvsix3/6, and NvotxC regional patterning genes were included as controls. Two positive control genes, 12533 and 242283, are included. The remaining genes are the salt-and-pepper expressed genes identified by the U0126 microarray. The red box indicates a region between 1.5× and −1.5× fold change that corresponds to an insignificant change in expression. The reciprocal phenotypes observed in NvashA MO and NvashA:venus mRNA-injected animals for Nvvsx-like (b–d) and Nvath-like (e–g) were confirmed by mRNA in situ hybridization. Embryos were classified and quantified as the percent having normal expression, weak expression, or no expression. The phenotypic class with the highest percentage of embryos is indicated. All embryo images are at the early gastrula stage. All images are ectodermal focal planes of aboral views. Error bars represent standard error

Fig. 9

Model of NvashA-dependent embryonic neurogenesis. a Model describing NvashA-dependent neurogenesis at early gastrula stage. Boxes represent indicted cell types. Solid regulatory lines represent published observations, and dashed lines represent likely regulatory interactions. b Biotapestry diagram of GRN