Distinct cis elements in the 3' UTR of the C. elegans cebp-1 mRNA mediate its regulation in neuronal development

Abstract

The 3' untranslated regions (3' UTRs) of mRNAs mediate post-transcriptional regulation of genes in many biological processes. Cis elements in 3' UTRs can interact with RNA-binding factors in sequence-specific or structure-dependent manners, enabling regulation of mRNA stability, translation, and localization. Caenorhabditis elegans CEBP-1 is a conserved transcription factor of the C/EBP family, and functions in diverse contexts, from neuronal development and axon regeneration to organismal growth. Previous studies revealed that the levels of cebp-1 mRNA in neurons depend on its 3' UTR and are also negatively regulated by the E3 ubiquitin ligase RPM-1. Here, by systematically dissecting cebp-1's 3' UTR, we test the roles of specific cis elements in cebp-1 expression and function in neurons. We present evidence for a putative stem-loop in the cebp-1 3' UTR that contributes to basal expression levels of mRNA and to negative regulation by rpm-1. Mutant animals lacking the endogenous cebp-1 3' UTR showed a noticeable increased expression of cebp-1 mRNA and enhanced the neuronal developmental phenotypes of rpm-1 mutants. Our data reveal multiple cis elements within cebp-1's 3' UTR that help to optimize CEBP-1 expression levels in neuronal development.

Keywords: Axon regeneration; Mechanosensory neurons; Stem-loop; Translation regulation; mRNA secondary structure; mRNA stability; rpm-1.

Fig. 1

Identification of two cis elements in cebp-1 3′ UTR regulating reporter gene expression. A) Illustration of cebp-1 3′ UTR and transgene expression constructs. Top illustration shows full length 3′ UTR (1–600 nucleotides), with color labeling of cis elements examined in this study. Color scheme continues into other figures. cebp-1 3′ UTR reporter constructs GFP driven by Prgef-1 pan-neuronal promoter; and a control (CTRL) construct is with unc-54 3′ UTR. Prgef-1 and GFP are not drawn to scale. All transgenes are inserted at ttTi5606 on Chromosome II. FL represents full length cebp-1 3′ UTR, and Δ followed by a letter to denote the cis element deleted in the expression construct. B) Representative projection images of confocal Z-stack of GFP expression in the nerve ring for cebp-1 3′ UTR FL and ΔE reporters, and the control 3′ UTR of unc-54. Dotted red line through cebp-1 3′ UTR shows example of line scan used for quantification of GFP using ImageJ. Scale bar, 20 μm. C) Quantification of GFP intensity in the nerve ring of L4 stage animals, n ≥ 10 for each genotype. AU: artificial unit; Error bars represent standard error of the mean (SEM). Statistics, One-way ANOVA with Tukey’s post hoc tests; ns, not significant; **** p < 0.0001. D) qRT-PCR for GFP mRNA in L4 animals, ribosomal gene rps-25 used as a reference. 3 biological replicates for each strain. Error bars, SEM; ns, not significant, * p < 0.5 (One-way ANOVA with Tukey’s post hoc tests).

Fig. 2

cis element D likely has a putative stem-loop secondary structure and may mediate translation of GFP reporter. A) Schematic of expression constructs for ΔD, ΔD1-3 and Dsr cebp-1 mutant 3′ UTR reporters. Prediction for presence (+) or absence (0) for RNA secondary or hairpin structure in each 3′ UTR mutant is shown on the right column. B) Sample z-stack images of the nerve ring of cebp-1 3′ UTR (FL) and ΔD3 reporter in wild type background. Scale bar, 20 μm. C) Quantification of GFP in the nerve ring of L4 stage animals. n ≥ 10 for each genotype. D) qRT-PCR for GFP mRNA in L4 animals; rps-25 used as a reference. 3 biological replicates for each strain. C-D: Error bars, SEM. Statistics, One-way ANOVA with Tukey’s post hoc tests, **** p < 0.0001; ns, not significant.

Fig. 3

rpm-1 regulates cebp-1 3′ UTR reporters through cis elements C and D2. A) Sample z-stack images of GFP expression in nerve ring from cebp-1 3′ UTR (FL) and cebp-1 3′ UTR(ΔE) reporters in wild type and rpm-1(lf) background. Scale bar, 20 μm. B) Quantification of GFP in the nerve ring of L4 stage animals. Strains grown at 25 °C. n ≥ 10 for each genotype. Error bars, SEM. **** p < 0.0001 (One-way ANOVA with Tukey’s post hoc tests). C) qRT-PCR for GFP mRNA in L4 animals; rps-25 used as a reference gene. 3 biological replicates for each strain, except that data for CTRL; rpm-1(lf) were from 2 biological replicates. Error bars, SEM. ns, not significant, *** p < 0.001 (One-way ANOVA with Tukey’s post hoc tests).

Fig. 4

cis elements D and E in 3′ UTR are required for proper function of CEBP-1 in mechanosensory neuron development. A) Sample z-stack images of PLM neuron axon morphology in wild type, rpm-1(lf), cebp-1(0), rpm-1(lf); cebp-1(0). Anterior is to the left, arrow denotes the “hook” phenotype (Schaefer et al., 2000), red asterix denotes the synaptic branch from the PLM to the nerve cord, and yellow asterix denotes the ALM cell body. Scale bar, 20 μm. B) Schematic of ALM and PLM mechanosensory neuron in wild type animals. Red asterix denotes the synaptic branch from PLM to nerve cord (labeled in black) C) Schematic of pan-neural expression of CEBP-1 protein with full-length 3′ UTR or deletion of cis elements D2 or E. Single-copy transgene is inserted on Chromosome II. D) Percentage of synaptic branch loss in strains expressing various MosSci transgenes of CEBP-1 in rpm-1(lf); cebp-1(0). WT represents wild-type. Quantification was from 3 biological replicates, total number of animals are shown in each column. E) qRT-PCR for cebp-1 mRNA from L4 animals of each single-copy CEBP-1 transgene in the rpm-1(lf); cebp-1(0). rps-25 used as a reference gene for qRT-PCR. 3 biological replicates for each strain used. D-E: Error bars, SEM. ns, not significant, **** p < 0.0001 * p < 0.05 (One-way ANOVA with Dunn’s multiple comparisons test).

Fig. 5

Deletion of endogenous cebp-1 3′ UTR results in enhanced touch neuron developmental phenotypes of rpm-1(lf). A) Genome alignment of cebp-1 with black arrow showing direction of transcription/translation. Deletions ju1451 and ju1452 shown in relation to cebp-1 3′ UTR with dotted lines. B) Sample z-stack images of PLM axon in wild type, cebp-1(ju1452), and rpm-1(lf); cebp-1(ju1452). Anterior is on the left, red asterix denotes the synaptic branch from the PLM to the nerve cord, yellow asterix denotes the ALM cell body, and white hashtag represents the AVM. Scale bar, 20 μm. C) Percentage of synaptic branch loss in different strains. Experiments from 3 biological replicates with total number of animals are shown for each genotype. D) Quantification of PLM overextension in wild type, cebp-1(ju1451), cebp-1(ju1452). Experiments from 3 biological replicates with total number of animals are shown for each genotype. E) Normalized regrowth 24 h after axotomy in L4 worms, n > 25 worm each group. C-E: Error bars, SEM. Statistics, One-way ANOVA with Tukey’s post hoc tests; ns, not significant; * p < 0.05, *** p < 0.001.

Fig. 6

Model of cebp-1 3′ UTR’s role. A) In wild type background or with rpm-1 signaling, cebp-1 levels are kept low under negative regulation by RPM-1, which partly depends on its 3′ UTR (labeled in orange) through a putative stem-loop and element C. B) In rpm-1(lf) background, negative regulation is absent, leading to increased cebp-1 mRNA levels and abnormal neural development. C) rpm-1(lf), in conjunction with deletion of cebp-1’s 3′ UTR, leads to a further increase in cebp-1 mRNA leading to enhancement of abnormal neural development.