June 18, 2015;
Multi-site phosphorylation regulates NeuroD4 activity during primary neurogenesis: a conserved mechanism amongst proneural proteins.
BACKGROUND: Basic Helix Loop Helix (bHLH) proneural transcription factors are master regulators of neurogenesis that act at multiple stages in this process. We have previously demonstrated that multi-site phosphorylation of two members of the proneural protein family, Ngn2
, limits their ability to drive neuronal differentiation when cyclin-dependent kinase levels are high, as would be found in rapidly cycling cells. Here we investigate potential phospho-regulation of proneural protein NeuroD4
(also known as Xath3), the Xenopus homologue of Math3/NeuroM, that functions downstream of Ngn2
in the neurogenic cascade.
RESULTS: Using the developing Xenopus embryo
system, we show that NeuroD4
is expressed and phosphorylated during primary neurogenesis, and this phosphorylation limits its ability to drive neuronal differentiation. Phosphorylation of up to six serine/threonine-proline sites contributes additively to regulation of NeuroD4
proneural activity without altering neuronal subtype specification, and number rather than location of available phospho-sites is the key for limiting NeuroD4
activity. Mechanistically, a phospho-mutant NeuroD4
displays increased protein stability and enhanced chromatin binding relative to wild-type NeuroD4
, resulting in transcriptional up-regulation of a range of target genes that further promote neuronal differentiation.
CONCLUSIONS: Multi-site phosphorylation on serine/threonine-proline pairs is a widely conserved mechanism of limiting proneural protein activity, where it is the number of phosphorylated sites, rather than their location that determines protein activity. Hence, multi-site phosphorylation is very well suited to allow co-ordination of proneural protein activity with the cellular proline-directed kinase environment.
[+] show captions
References [+] :
Fig. 1 NeuroD4 is expressed and phosphorylated during primary neurogenesis (A-C). a Protein sequence alignment for human, mouse and Xenopus NeuroD4 homologues using Clustal W. The conserved bHLH domain is indicated in green and SP/TP sites are highlighted in red. A consensus line is also shown below the alignment to indicate the degree of conservation of amino acids at each position: (*) denotes identical residues in all three sequences; (:) denotes highly conserved residues; (.) denotes weakly conserved residues. b Endogenous expression of NeuroD4 at stage 14 (i) and neural-β-tubulin at stage 17 (ii) was determined by whole mount ISH (white arrows correspond to zones of primary neurogenesis; TG = trigeminal ganglia). Dorso-ventral views, anterior up, stages as indicated. c Western Blot analysis of extracts from stage 13 embryos injected with an HA-tagged wild type (WT) NeuroD4 mRNA, either with or without protein phosphatase treatment. Tubulin was used as a loading control
Fig. 2 Phospho-mutant 6T/S-A NeuroD4 has increased proneural activity relative to WT NeuroD4 (a-g). a Schematic representation of WT NeuroD4 and full phospho-mutant 6 T/S- A NeuroD4 protein sequences, indicating the relative positions of the six SP or TP sites that have been mutated to AP sites. b-g Two-cell stage embryos were unilaterally injected with 100 pg of either WT or 6T/S-A NeuroD4 mRNA, and at stage 18, gene expression was assayed by qRT-PCR (b), or by whole mount ISH (d-g) with representative embryo images shown in (c). For qRT-PCR analysis (b) significance is calculated as described in the methods section for phospho-mutant NeuroD4 relative to WT NeuroD4 (blue adjoining lines and stars) and relative to uninjected control embryos (shown with black stars) [n = 5]; (p <0.05) = *; (p <0.025) = **; (p <0.0125) = ***. For ISH analysis, embryos were scored according to the scale described in Additional file 1 for expression of neural-β-tubulin (D [n = 100–115]), p27Xic1 (E [n = 24-33]), xMyt1 (F [n = 26-32]), and xNeuroD1 (G [n = 23- 28]). Views in (C [i-iii]) are dorso-ventral (DV); views in (C [iv]) are rostro-caudal (RC) with dorsal surface facing up, trigeminal ganglia indicated by arrows. All images show injected side to the right.
Fig. 3 Similar levels of proneural activity are seen amongst single site mutants (a-d). a Schematic representation of the single site phospho-mutant NeuroD4 constructs demonstrating the relative position of the SP or TP site mutated to AP in each. Two-cell stage embryos were unilaterally injected with 100 pg mRNA of the respective NeuroD4 construct and assayed at stage 18 for expression of neural-β-tubulin by qRT-PCR (B[n = 3]), or whole embryo ISH (C[n = 51-77]) with representative images shown in (d). Views are dorso-ventral with injected side to the right; (p <0.05) = *; (p <0.025) = **; (p <0.0125) = ***
Fig. 4 Similar levels of proneural activity are seen amongst paired site mutants (a-d). a Schematic representation of the paired site phospho-mutant NeuroD4 constructs. Two-cell stage embryos were unilaterally injected with 100 pg mRNA of the respective NeuroD4 construct and assayed at stage 18 for expression of neural-β-tubulin by qRT-PCR (B[n = 3]), or whole embryo ISH (C[n = 62-79]) with representative images shown in (d). Views are dorso-ventral with injected side to the right; (p <0.05) = *; (p <0.025) = **; (p <0.0125) = ***
Fig. 5 Cumulative mutation of phosphorylation sites creates step-wise increases in proneural activity (a-d). a Schematic representation of the phospho-mutant series of constructs demonstrating the SP and/or TP sites that are mutated to AP sites in each. The series consists of cumulative mutation of sites, working from N to C termini. Two-cell stage embryos were unilaterally injected with 100 pg mRNA of the respective NeuroD4 construct and assayed at stage 18 for expression of neural-β-tubulin by qRT-PCR (B[n = 3]), or whole embryo ISH (C[n = 59-77]) with representative images shown in (d). Views are dorso-ventral with injected side to the right; (p <0.05) = *; (p <0.025) = **; (p <0.0125) = ***
Fig. 6 Both WT and 6T/S-A NeuroD4 generate predominantly sensory neurons (a-e). Embryos were unilaterally injected at the two-cell stage with 100 pg of either WT or 6T/S-A NeuroD4 mRNA and at stage 18, embryos were assayed by qRT-PCR for expression of pan neuronal gene neural-β-tubulin, sensory neuron marker xHox11L2, motor neuron marker xHb9 or interneuron marker xVsx1 (a, b). Data is shown on two separate graphs due to differences in scale [n = 3]. Whole embryo ISH was also conducted to compare the pattern of expression of neural-β-tubulin and xHox11L2 (c) [n = 30–38]. Views in (c) are all dorso- ventral with injected side to the right; (p <0.05) = *; (p <0.025) = **; (p <0.0125) = ***
Fig. 7 6T/S-A NeuroD4 shows enhanced protein stability and enhanced chromatin binding relative to WT NeuroD4 (a-d). Embryos were injected at the one cell stage with 200 pg mRNA encoding HA-tagged versions of either WT or 6T/S-A NeuroD4. a Western blot analysis on extracts prepared from stage 13 embryos, with tubulin as a loading control. b The density of the protein band for WT or 6T/S-A NeuroD4 was quantified and expressed relative to the tubulin loading control of each sample. Mean values are shown from independent duplicate samples with the standard error of the mean. c At stage 14, cross-linking and chromatin isolation were performed prior to western blot analysis as before. No HA-tagged NeuroD4 protein was detected in the uninjected embryos confirming antibody specificity, and no tubulin protein was detected in the chromatin fraction. d NeuroD4 protein bands were again quantified relative to tubulin or histone H3 bands for cytoplasmic and chromatin fractions respectively. Mean values are shown from independent duplicate samples with the standard error of the mean.
Additional file 1. Representative embryo images to demonstrate semi-quantitative scoring system used for in situ hybridisation data. Neurogenesis was graded by comparing the extent and pattern of neural β-tubulin expression following in situ hybridisation on the injected side of the embryo relative to the uninjected side and uninjected control embryos. Scores were assigned as: 0, no difference; +1, mild increase in neurogenesis within the neural plate, with or without occasional ectopic neurons on the injected side; +2, moderate increase in neurogenesis with ectopic expression of neural β-tubulin occurring in patches on the injected side and sometimes bilaterally; +3, marked increase in neurogenesis with extensive ectopic expression of neural β-tubulin in a more homogenous pattern on the injected side and sometimes bilaterally (1396kb)
neurod4 (neuronal differentiation 4) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage 14, dorsal view, anterior up.
Requirement of multiple basic helix-loop-helix genes for retinal neuronal subtype specification.