HSP70-binding motifs function as protein quality control degrons

Abstract

Protein quality control (PQC) degrons are short protein segments that target misfolded proteins for proteasomal degradation, and thus protect cells against the accumulation of potentially toxic non-native proteins. Studies have shown that PQC degrons are hydrophobic and rarely contain negatively charged residues, features which are shared with chaperone-binding regions. Here we explore the notion that chaperone-binding regions may function as PQC degrons. When directly tested, we found that a canonical Hsp70-binding motif (the APPY peptide) functioned as a dose-dependent PQC degron both in yeast and in human cells. In yeast, Hsp70, Hsp110, Fes1, and the E3 Ubr1 target the APPY degron. Screening revealed that the sequence space within the chaperone-binding region of APPY that is compatible with degron function is vast. We find that the number of exposed Hsp70-binding sites in the yeast proteome correlates with a reduced protein abundance and half-life. Our results suggest that when protein folding fails, chaperone-binding sites may operate as PQC degrons, and that the sequence properties leading to PQC-linked degradation therefore overlap with those of chaperone binding.

Keywords: Chaperone; Proteasome; Protein degradation; Protein quality control; Protein stability; Protein unfolding.

Fig. 1

Assessing degron activity with the Ura3-HA-GFP reporter system. a A schematic representation of the Ura3-HA-GFP reporter system. In-frame fusion of degrons to the C-terminus of the reporter will lead to proteasomal degradation (Pac-Man) leading to a 5-FOA resistant phenotype. b As a proof of principle for the reporter system, the growth of wild-type cells expressing the reporter vector with and without fusion to the CL1 degron was compared by dilutions on solid media. Note that the growth defect of the CL1 strain on media without uracil is suppressed by the addition of the proteasome inhibitor bortezomib (BZ), while in the presence of 5-FOA, the CL1 strain is able to form colonies. c Analysis of protein levels in vector or CL1 cells after 4 h of bortezomib (BZ) treatment by SDS-PAGE and Western blotting, using antibodies to the HA-tag in the reporter. Blotting against Pgk1 served as loading control

Fig. 2

Investigation of the degron capacity of the canonical Hsp70-binding motif, RLLL. a Structure of the Hsp70 substrate-binding domain (SBD) (blue) in complex with the Hsp70-binding peptide, NRLLLTG (red) (PDB: 4PO2) [79]. b Illustration of APPY-variants fused in one, two or three copies to the Ura3-HA-GFP-reporter. Note that the entire 22-residue APPY peptide was used. From the literature [67], we predicted strong (RLLL), intermediate (RAAA) and weak (DAAA) degron motifs. c Cells carrying the indicated fusions were analyzed by microscopy for GFP fluorescence. Bar, 10 µm. d Cell lysates of cells carrying the constructs listed in panel b were prepared and compared by SDS-PAGE and Western blotting using antibodies to the HA-tag. Tubulin served as loading control. e Analysis of the indicated protein levels by SDS-PAGE and Western blotting from cells treated with the translation inhibitor cycloheximide (CHX) (left panel) or the proteasome inhibitor bortezomib (BZ) (right panel). Pgk1 served as loading control. f Growth assay of cells carrying the constructs listed in panel b

Fig. 3

Characterization of the RLLL degron. a The solubility of the reporter proteins was analyzed by fractionating cell extracts into soluble supernatant (S) and insoluble pellet (P) fractions by centrifugation. The protein levels were analyzed by SDS-PAGE and Western blotting using antibodies to the HA-tag in the reporter. Pgk1 served as a loading control for the soluble protein fractions, while Pma1 served as loading control for the insoluble protein fractions. b Myc-tagged Ssa1 (Hsp70) was co-immunoprecipitated (co-IP) with the indicated GFP fusions using GFP-trap resin from cultures treated with bortezomib. ATP was not added to the buffers used for the co-IPs. The precipitated protein was analyzed by SDS-PAGE and blotting with antibodies to myc and GFP. c The protein levels of the RLLL degron were compared in the indicated yeast strains by SDS-PAGE and blotting for the HA-tag on the reporter. A Ponceau S staining of the membrane is included as loading control. d The dependence of E3 ligases for targeting the RLLL degron was analyzed by growth assays on solid media using the indicated null mutants. Wild-type cells transformed with the reporter vector alone were included for comparison

Fig. 4

The RLLL degron sequence-dependence. a Fusion of the Ura3-HA-GFP-reporter to the listed APPY variants induced growth phenotypes that were assessed on solid media. The RLLL motif is highlighted (bold) and amino acid substitutions are shown in red. b The abundance of variants from panel A was compared by SDS-PAGE and Western blotting of whole-cell lysates using antibodies to HA. Pgk1 served as a loading control

Fig. 5

Screening reveals a large sequence space of functional degrons with the RLLL motif. a Schematic overview of the library construction and screen. Three primer pairs were designed to cover the pTR1412 vector, encoding the Ura3-HA-GFP-reporter. Four amino acid positions in the shown C-terminal peptide were randomized by using a primer carrying TriMix20 at the bold-faced positions marked X. PCR amplification yielded three partially overlapping fragments, which were assembled by gap repair following transformation into yeast. Transformants were selected on media without leucine (–LEU) and on media without leucine but with 5-FOA (–LEU + 5-FOA) to select for degrons. Finally, DNA was extracted and sequenced. b Enrichment and depletion of amino acid types in degrons. The plot shows the log₂ of the ratio of the fractions of each amino acid found in the top 100 peptides relative to the full set of peptides seen in the screen. Residues with a positive value are enriched in degrons and those with negative values are depleted. c The ten most prevalent sequences in the 5-FOA-treated library are listed along with the enrichment scores. d The growth of the most prevalent sequences from the 5-FOA library was compared on solid media

Fig. 6

The number of exposed Hsp70-binding sites correlates with protein turnover. a The Limbo predictor for peptide binding to E. coli Hsp70 (DnaK) [17] was used to analyze all 18,706 sequences (red), or the top 100 sequences (blue) or the ten most prevalent sequences (green) from the 5-FOA resistant library. A high Limbo score indicates a greater likelihood for binding to Hsp70. b Proteome wide analysis of predicted Hsp70-binding sites. The number of residues that are both predicted to be Hsp70-binding sites and disordered, and thus solvent exposed, is different for proteins with different levels of abundance and half-life. Very low abundance proteins are an exception to this trend which has been suggested to reflect the absence of selection against exposed degrons when the protein concentrations are lower than the binding affinity of harmful interactions [16]

Fig. 7

The APPY peptide functions as a degron in human cells. a Upon transfection of the HEK293T landing pad cells, BxbI catalyzes site-specific recombination between BxbI attachment sites in the vector and the landing pad, leading to single-copy expression of GFP-APPY from the Tet-on promoter. As the vector also contains an internal ribosomal entry site (IRES) followed by mCherry, the GFP-APPY levels can be normalized to the mCherry levels. 2A signifies self-cleaving peptides. b GFP:mCherry ratio distributions of cells expressing the indicated GFP-APPY variants, obtained by flow-cytometry of ~ 10,000 cells. c GFP:mCherry ratio distributions of cells expressing the GFP-APPY (RLLL) peptide or the DAAA variant of APPY after 5 h treatment with 15 μM bortezomib (BZ) or DMSO (solvent control), obtained by flow-cytometry of ~ 10,000 cells. d GFP:mCherry ratio distributions of cells expressing the APPY (RLLL) peptide or the DAAA variant of APPY after 22 h treatment with 5 μM YM01 or no treatment, obtained by flow-cytometry of ~ 10,000 cells. e Relative expression of the HSPA1A and HSPA1B genes derived from qPCR. The results were normalized to the expression of non-recombinant cells. n = 3, error bars show the standard deviation, asterisks indicate significant differences (p < 0.05) based on a two-tailed unpaired Student’s t test

Fig. 8

Schematic illustration of the proposed degradation pathway for misfolded proteins. Mutations or stress conditions increase the risk for protein misfolding leading to degron exposure. Hsp70 binds the degron motif in the misfolded protein, thus facilitating protein refolding (top) or degradation (bottom) in the presence of Ubr1 and JDPs. Ubr1-mediated ubiquitylation targets the protein for proteasomal degradation in collaboration with NEF proteins Hsp110 and Fes1