Protein C-terminal variations impact proteostasis

Figures

• 

  Fig. 1. Systematic evaluation of protein half-lives affected by disease-associated nonstop mutations. A Comparison of protein stability between GFP alone and GFP fused to the C-terminal 20 residues of the indicated wild-type (WT) proteins, as measured by GPS assays. B Analysis of GFP stability when tagged with WT or readthrough (RT) C-termini, as indicated on the left. C Schematic of the high-throughput platform used to assess the impact of WT and disease-relevant RT C-termini on protein stability. Oligonucleotides encoding 58-residue WT and RT C-termini were cloned into the GPS reporter and introduced into HEK293T cells to generate stable GPS reporter lines. Cells were sorted into eight bins based on GFP/RFP ratios, and oligonucleotides from each bin were sequenced to calculate a Protein Stability Index (PSI) for each peptide, with lower PSI values indicative of stronger degradation signals. D Correlation of PSI values across two independent biological replicates. E Violin plots showing the distributions of mean PSI for WT and RT peptides across two biological replicate screens. Solid and dashed lines indicate the median and the first/third quartiles, respectively. Asterisks indicate…

• 

  Fig. 2. Disease-relevant nonstop mutations can either increase or decrease protein half-lives. A Correlation between the length of RT extensions and the magnitude of changes in protein stability (∆PSI). Box plots indicate medians (central line), first and third quartiles (box edges), and 10th and 90th percentiles (whiskers). B Comparison of PSI values between WT and RT peptides for nonstop mutations resulting in ≤ 20-residue extensions. Blue triangles denote degradation or stabilization caused by the introduction or disruption of known C-degrons. C C-terminal sequences corresponding to blue triangles in (B), with known degron motifs highlighted in blue. D Validation of protein stability changes using full-length WT and RT proteins. The type of nonstop mutation is labeled on the blots. E C-terminal sequences of proteins analyzed in (F). F Validation of stability changes caused by nonstop mutations using full-length constructs. G Impact of nonstop mutations on the stability of oncoproteins and tumor suppressors identified in the screen, based on TSGene and ONGene annotations54,55. H, I Validation of changes in nonstop mutation-induced protein stability in full-length tumor suppressors,…

• 

  Fig. 3. C-terminal variations from alternative splicing and programmed RT modulate protein half-lives. A Analysis of isoform-specific C-terminal variants (last 20 residues) generated by alternative splicing across 21,784 human genes. B Comparison of protein stability between full-length isoforms of the indicated genes. C C-terminal sequences of the isoforms analyzed in (B). Differences between isoforms are highlighted in red. D, E Comparison of stability between full-length canonical proteins and their corresponding programmed RT variants. Source data are provided as a Source Data file.

• 

  Fig. 4. Characterization of the impact of C-terminal amino acid composition and arrangement on protein stability using a random peptide platform. A Comparison of GFP stability with or without C-terminal fusion to various epitope tags. B Schematic of peptide templates designed to examine the influence of amino acid identity and position on protein stability. C Random peptide assays conducted in biological triplicate to assess degron activity across libraries with varying amino acid compositions and arrangements. Data are shown as mean \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\pm $$\end{document}±SD. Amino acids are color-coded by physicochemical properties. The gray bar represents the control template, with fully randomized 12-mer (terminal) or 15-mer (composition) peptides. Asterisks indicate statistical significance compared to control (*, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001; unpaired two-tailed Students’ t test). D Relative degradation levels of peptide templates with dispersed…

• 

  Fig. 5. Amino acid compositional analysis of the C-terminal region of canonical cytosolic proteins. A PCA of the last 150 residues of canonical human cytosolic proteins. B Principal component loading plots corresponding to (A), with amino acids color-coded according to physicochemical properties. C Analysis of consecutive hydrophobic or acidic residues in the last 50 residues of human cytosolic proteins. Frequencies of contiguous stretches of varying lengths (x-axis) are normalized against a background generated from 10,000 random iterations. Central line in box plots indicates medians; box edges indicate first and third quartiles; whiskers represent min/max values; asterisks indicate statistical significance (****, p < 0.0001; two-sided permutation test). All of the p-values are provided in the Source Data file. D PCA of amino acid usage in the last 150 residues of canonical cytosolic proteins across species, without (left) and with (right) normalization to each species’ overall proteomic amino acid usage. Data points from different species are represented by different shapes, and amino acid positions are color-coded. E PCA of C-terminal amino acid usage across canonical cytosolic…

• 

  Fig. 6. Cysteine and tryptophan enrichment promotes degradation of proteins with RT and frameshift extensions. A PCA of the C-terminal regions of canonical and aberrant proteins. Analyzed regions include the last 20 residues of canonical and aberrant proteins. B Fold changes in relative frequencies of the 20 amino acids at the C-termini of canonical, RT and frameshift proteins, grouped into four categories based on similarity. C Fold changes in Cys, Trp, Asp and Glu frequencies at the C-termini of canonical and aberrant proteins across multiple species. D Stability comparison between full-length canonical proteins and their RT variants containing C-terminal Cys/Trp-enriched extensions. C-terminal sequences are shown above, with Cys and Trp highlighted in blue. E Stability comparison between canonical proteins and RT variants with or without Cys/Trp-to-Ala substitutions. F PSI comparison for RT peptides before and after substituting Cys/Trp (top) or Asp/Glu (bottom) with Ala. Source data are provided as a Source Data file.

• 

  Fig. 7. RNF139 and MARCH6 ubiquitin ligases target hydrophobic C-termini to promote protein degradation. (A) Amino acid sequences and ∆PSI values of the RT regions analyzed in (B). Hydrophobic residues are highlighted by orange boxes. B Stability comparison of GFP fused to the indicated RT peptides, with or without Bag6 or RNF126 knockout via sgRNAs. Knockdown of APPBP2 serves as a control. C CRISPR screens identifying proteolytic machinery components involved in degradation driven by the indicated RT peptides. Each dot in the plot represents a gene identified from the screening results, and it is positioned according to its confidence and effect size, as quantified by the casTLE score (see “Methods”). D Stability comparison for GFP fused to the indicated RT peptides, with or without sgRNA-mediated knockout of APPBP2, Bag6, RNF126, RNF139 or MARCH6. E Amino acid sequences of RGS18-RT and its mutant variants. First and second hydrophobic stretches are highlighted by orange or red boxes, respectively; the scrambled forms of the stretches are marked by dashed boxes; the TN motif is marked in blue. F Stability comparison between GFP alone and GFP fused to the peptides indicated on the right.…