Protein Synthesis in the Developing Neocortex at Near-Atomic Resolution Reveals Ebp1-Mediated Neuronal Proteostasis at the 60S Tunnel Exit

Abstract

Protein synthesis must be finely tuned in the developing nervous system as the final essential step of gene expression. This study investigates the architecture of ribosomes from the neocortex during neurogenesis, revealing Ebp1 as a high-occupancy 60S peptide tunnel exit (TE) factor during protein synthesis at near-atomic resolution by cryoelectron microscopy (cryo-EM). Ribosome profiling demonstrated Ebp1-60S binding is highest during start codon initiation and N-terminal peptide elongation, regulating ribosome occupancy of these codons. Membrane-targeting domains emerging from the 60S tunnel, which recruit SRP/Sec61 to the shared binding site, displace Ebp1. Ebp1 is particularly abundant in the early-born neural stem cell (NSC) lineage and regulates neuronal morphology. Ebp1 especially impacts the synthesis of membrane-targeted cell adhesion molecules (CAMs), measured by pulsed stable isotope labeling by amino acids in cell culture (pSILAC)/bioorthogonal noncanonical amino acid tagging (BONCAT) mass spectrometry (MS). Therefore, Ebp1 is a central component of protein synthesis, and the ribosome TE is a focal point of gene expression control in the molecular specification of neuronal morphology during development.

Keywords: Ebp1; cryo-EM; development; neocortex; neurons; pSILAC/BONCAT mass spectrometry; proteostasis; ribosome; ribosome profiling; selective ribosome profiling.

Figure 1.. Ebp1 Is a Highly Associated Cofactor of the Neocortex Ribosome across Development

(A) Schematic of the experimental system to measure the architecture of active protein synthesis (polysomal ribosomes) from the neocortex across embryonic and early postnatal neurogenesis. (B) Analytic density gradient fractionation of A260-normalized neocortex lysates, measuring the relative abundance of ribosomal subunits, 80S ribosomes, and polysomes. A260 curves plotted as mean ± SD across replicate fractionations (n = 2–3) for each stage, baseline (1.0) centered at onset of 40S peak. (C) Statistical comparison of ∑A260 within gray marked regions in (B), shown as mean ± SD with significance testing by one-way ANOVA and Dunnett’s post hoc test versus E12.5. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. (D) MS analysis (n = 3) of neocortex polysomal complexes across development, scatterplots comparing E12.5 with each subsequent stage for enrichment of Ebp1, ribosomal proteins (RPs) of the large (Rpl) and small (Rps) subunits, and translation-associated proteins (GO: 0006417). See also Figures S1 and S2A. (E) Neocortex expression of Ebp1, Rpl, Rps, and translation-associated genes measured in total steady-state levels by RNA-seq (left, n = 2) and MS (right, n = 3) across developmental stages. Median expression is plotted ± SD; one-way ANOVA and Bonferroni corrected post hoc test versus E12.5, p < 0.05. Significantly changing levels versus E12.5 (true) are shown as filled circles and non-significant (false) values as empty circles. (F) Western blot probing for Ebp1 (Ebp1^CT, Ebp1^NT) in total neocortex lysates compared to full-length recombinant Ebp1-His, along with the RP uL30 and Gapdh; full blots are shown in Figures S3A and S3B. (G) Jitter plots comparing the median stoichiometry of Rpl and Rps (centered at 0) with Ebp1 and translation-associated proteins in total, 80S, and polysomes at E12.5. Other stages are shown in Figure S2B. (H) Western blot analysis (top) of Ebp1 enrichment in free, 80S, and polysome fractions across development compared to Gapdh and uL30. Quantification (bottom, n = 2 blots) of Ebp1 and uL30 levels versus E12.5 is shown, and values represent mean ± SD (t test for significance: *p < 0.05; ***p < 0.001). See also Figures S3C-S3E.

Figure 2.. Ebp1 Is Enriched in Early-Born NSCs and Localizes Throughout the Neuronal Cytoplasm

(A) Expression heatmaps of Ebp1 compared to averaged Rpl and Rps family mRNA enrichment in scRNA-seq analysis of the developing mouse neocortex, derived from (Telley et al., 2019). Relative expression in apical progenitor (AP) NSCs during differentiation into mature neurons (N4d) is shown on the y axis, corresponding to NSC birthdates E12, E13, E14, and E15 on the x axis. (B) Immunohistochemistry analysis of Ebp1 the developing neocortex ventricular zone (VZ) and cortical plate (CP). Early-born NSCs in the VZ generate lower layer (LL) neurons, while later-born NSCs in the VZ generate upper layer (UL) neurons. Axonal white matter (WM); DAPI staining (gray). Zoomed images (inset, left) correspond to the VZ and leading-edge of the CP at each stage, quantified signal/area (n = 5-7) in each region of interest (inset, right heatmap). AU, arbitrary units. See also Figures S4A and S4B. (C) Immuno-electron microscopy with anti-Ebp1^NT immunogold labeling (black dots) in the neocortex at E12.5, E15.5, and P0. Neural stem cells (NSCs; blue nuclei) and neurons (N, red nuclei). Nucleoli (n), mitochondria (m, green), endoplasmic reticulum (er), dendrite (D), plasma membrane (arrows). (D) Quantification of (C), comparing the cytoplasmic versus nuclear distribution of Ebp1 (n = 5–64 cells per condition) in VZ stem cells and CP neurons, with 1° antibody leave-out control. See also Figures S4C and S4D. Significance testing by Welch ANOVA, *p < 0.001. (E) Primary neuronal cultures from the E12.5 neocortex, immunocytochemistry at div 0, 2, 4, and 5 for Nestin, Nex:Cre;tdTomato, and Ebp1. Growing neurites and distal growth cones are indicated (arrows).

Figure 3.. Ebp1 Binds the 60S Tunnel Exit (TE) in Actively Translating and Inactive 80S Complexes

(A) Cryo-electron micrograph of pooled monosome and polysome complexes from P0 mouse neocortical lysates ex vivo. (B) Cryo-EM maps of (A) with extra-ribosomal density conforming to mouse Ebp1 (PDB: 2V6C) over the 60S TE (side view, top image; aerial view, bottom image). N-terminal Ebp1 residues (NT, black ribbon) corresponding to full-length “p48” Ebp1. See also Figures S5-S7. (C) Actively translating (left: classical state with A/A and P/P tRNAs) and non-translating (right: rotated state with eEF2) 80S-Ebp1 complexes. (D) Model of the Ebp1 binding surface at the 60S peptide TE, including 60S rRNA helices H24, H53, H59, and 60S RPs eL19, uL23, uL24, and uL29. (E) Aerial view of the Ebp1 footprint (red outline) over the 60S peptide TE, with rRNA helices and RP model surfaces colored as in (D); residues/nucleosides making electrostatic interactions with Ebp1 are highlighted (yellow). (F) 2D structure diagram of Ebp1 domains adapted from Kowalinski et al. (2007), orienting Ebp1 on the ribosome surface, with binding domains highlighted (yellow).

Figure 4.. Ebp1-60S Binding Utilizes a Conserved H59 Latch Mechanism and Is Incompatible with Simultaneous Binding of Other TE Cofactors

(A) 60S rRNA H59 and H53 models in conformations with and without Ebp1, adjacent to the Ebp1 insert domain. See also Figures S8A and S8B. (B–D). Ebp1-60S binding interface in detail, with interacting residues highlighted for Ebp1 (gray) and the 60S (yellow). See also Figures S8C-S8E. (E–G) Global alignment of Ebp1, Metap2, and Arx1 (top, ribbon), likewise when viewed from within the 60S tunnel (bottom, electrostatic potential map) from the perspective of emerging nascent chain. See also Figure S8F. (H) Aerial view with overlapping footprints of eukaryotic TE binding factors superimposed on the neocortex 60S. Accession numbers are as follows: Metap2, PDB: 1KQ9; Arx1, PDB: 5APN; Sec61, PDB: 3J7R; SRP, PDB: 6FRK; Ltn1, PDB: 3J92; NatA, PDB: 6HD7; Ttc5, PDB: 6T59; RAC, EMDB: 6105; NAC, EMDB: 4938. (I) Jitter plots comparing the median stoichiometry of Rpl and Rps (centered at 0) with Ebp1 and other TE cofactors in total, 80S, and polysomes at E12.5 and P0. See also Figure S9. (J) Ebp1-60S binding affinity assay (Figure S10C), with independent replicate experiments (white and gray circles) and curve best fit to the data. 60S concentration (blue line) maintained at a constant 100 nM. (K) Ebp1-60S binding dynamics assessed by pelleting assay and western blot. Binding pellet signal for (1) super-saturating Ebp1-His, (2) native Ebp1 in RRL, and (3) competition between added Ebp1-His and native Ebp1 in RRL. Arrow, native Ebp1; star, Ebp1-His signal.

Figure 5.. Ebp1-Ribosome Complexes Engage in Translation Initiation and Elongation, with High Occupancy prior to N-Terminal Membrane Targeting

(A) Correlation between the neuronal Ebp1-ribosome interactome and total translatome measured by SeRP (n = 2); mRNAs with RPKM enrichment ≥1.5-fold are highlighted. See also Figures S11 and S12. (B) Gene Ontology (GO) analysis of mRNAs enriched in the Ebp1-ribosome interactome versus total translatome from (A). (C) Proteome-wide metagene read density of the Ebp1-ribosome interactome versus total translatome over the coding sequence, aligned to the start (left) or stop (right) codon, plotted as mean with 95% confidence intervals (CIs). See also Figure S12C. (D) Metagene plots as in (C), separated by subcellular protein localization. (E) Cytoplasmic and signal peptide-containing protein metagene plots aligned to the start or stop codon (left figure), highlighting the relative enrichment at 70 codons (gray dashed line). Metagene plot for signal peptide-containing proteins aligned to the C-terminal codon of the signal sequence (right figure), with 60S tunnel transit region 40 codons downstream (gray box). (F) Metagene plot for transmembrane domain (TMD)-containing proteins with (left figure) and without (right figure) an upstream translocon signal peptide, aligned to the C-terminal codon of the first TMD. (G) Metagene read density distribution comparing Ebp1 knockdown versus control neuronal ribosome profiling (n = 3), separated by subcellular protein localization, and aligned to the start or stop codon. See also Figure S13. (H) Ebp1 knockdown and control ribosome P-site count metagene plots (95% CI), aligned to the start or stop codon, for cytoplasmic and signal peptide-containing mRNAs. Inset right: scaled to highlight relative differences at the start codon P-site. (I) Ebp1 knockdown/control fold change ribosome P-site counts at the start, stop, and adjacent codons (±3 nt) in the CDS of cytoplasmic and signal peptide-containing mRNAs. All ribosome positions shown for both mRNA groups are significantly different than control conditions (hypothesis siEbp1/control ≠ 1; 95% CI; p < 0.001), in addition to the significant difference annotated in the figure with *p = 0.03. Significance testing by t test.

Figure 6.. Ebp1 Regulates Neocortical Neuronal Morphology during Development and the Synthesis of Membrane-Targeted Cell Adhesion Molecules (CAMs)

(A) E12 in utero electroporation (IUE) of NSCs followed by neuronal analysis at E16, comparing shEbp1 and scrambled shRNA control, and rescue by co-electroporation with Ebp1 overexpression (oeEbp1). Co-electroporation with CAG-GFP visualizes transfected cells, shown magnified (bottom), including basally projecting axons (arrows) forming white matter tracts below upper (UL) and lower (LL) layers. (B) Morphology tracing GFP labeled neurons in control, shEbp1, and rescue sh+oeEbp1 conditions from (A). (C) Sholl analysis of (B), comparing branching per unit distance from the soma (top figure) and sum total (bottom figure) (n = 15 cells per condition). Values represent mean ± SD, with significance testing by one-way ANOVA with Bonferroni corrected post hoc test versus control (*p < 0.01). (D) Schematic of the strategy to measure both chronic proteostasis and acute protein synthesis responses to Ebp1 knockdown in Neuro2a cells with pSILAC and BONCAT MS, respectively. AHA pulsed for 4 h. (E) pSILAC- and pSILAC-AHA-labeled protein levels in siEbp1 relative to non-targeting siRNA control in biological replicates (n = 2) with label swab. Ebp1 levels were below the MS quantification threshold in siEbp1 conditions and thus not plotted. The number of significantly changing proteins at ≥2-, 1.5-, and 1.25-fold change thresholds are shown (dotted lines), in addition to the total number of proteins measured (top left). See also Figure S14. (E′) Significantly changing proteins measured in common between pSILAC and pSILAC-AHA datasets at 1.25-fold-change thresholds. (F) GO pathway analysis of proteins in (E) with ≥ 2-fold change in Ebp1 knockdown conditions. (G) Metagene enrichment plots of the Ebp1-interactome (Ebp1-IP) and Ebp1 knockdown ribosome distribution translating L1cam mRNA, aligned to the start and stop codons, plotted as mean with 95% CIs.

Figure 7.. Model of Ebp1 Function in Protein Synthesis and Neurodevelopment

For details, see text. ER, endoplasmic reticulum.