Slippery sequences stall the 26S proteasome at multiple points along the translocation pathway

Abstract

In eukaryotes, the ubiquitin-proteasome system is responsible for intracellular protein degradation. Proteins tagged with ubiquitin are recognized by ubiquitin receptors on the 19S regulatory particle (RP) of the 26S proteasome, unfolded, routed through the translocation channel of the RP, and are then degraded in the 20S core particle (CP). Aromatic paddles on the pore-1 loops of the RP's Rpt subunits grip the substrate and pull folded domains into the channel, thereby unfolding them. The sequence that the aromatic paddles grip while unfolding a substrate is therefore expected to influence the extent of unfolding, and low complexity sequences have been shown to interfere with grip. However, the detailed spatial requirements for grip while unfolding proteins, particularly from the N-terminus, remain unknown. We determined how the location of glycine-rich tracts relative to a folded domain impairs unfolding. We find that, in contrast to a previous report, inserting glycine-rich sequences closer to the folded domain reduced unfolding ability more than positioning them further away. Locations that have the biggest effect on unfolding map onto the regions where the aromatic paddles are predicted to interact with the substrate. Effects on unfolding from locations up to 67 amino acids away from the folded domain suggest that there are additional interactions between the substrate and the proteasome beyond the aromatic paddles that facilitate translocation of the substrate. In sum, this study deepens understanding of the mechanical interactions within the substrate channel by mapping the spacing of interactions between the substrate and the proteasome during unfolding.

Phase separation of polyubiquitinated proteins in UBQLN2 condensates controls substrate fate

“…Since sGFP is quite stable, and, depending on ubiquitination state, is not always fully degraded by the proteasome (31, 32), we used a Cy5-labeled model substrate containing the same degron followed by two domains – the unstructured activation domain from the p160 transcriptional co-activator for thyroid hormone and retinoid receptors (ACTR) (33) followed by folded E. coli dihydrofolate reductase (DHFR) (34). Although DHFR is stably folded, the proteasome can easily unfold and degrade the entire substrate in the absence of UBQLN2 ( Figure 3 ), as indicated by the Cy5-labeled peptides that appear at the bottom of the gel (34). UBQLN2 protected both K63-linked and mixed substrates, but not K48-linked substrates, against proteasomal degradation ( Figure 3A,B) .…”

“…Constructs expressing His-SUMO-R-Neh2Dual-sGFP, His-SUMO-R-Neh2Dual-eGFP were cloned into previously existing His-SUMO-R-Neh2Dual constructs (26) using Gibson assembly and verified using Sanger sequencing. A construct expressing His-SUMO-R-Neh2Dual-ACTR-DHFR (which is lysine-free except for the degron region and contains a single cysteine between ACTR and DHFR) was described previously (34). An S. cerevisiae strain with TagRFP-T-Rpn6 (yDAK56) and a 3X-Flag-tag on Rpn11 was created using CRISPR as described previously (48).…”

“…UBQLN2 proteins were expressed, purified and, as indicated, labeled with Alexa Fluor 647, Alexa Fluor 555, or DyLight-488 as described previously (15). Substrates were purified by Ni-NTA chromatography and SUMO protease cleavage as described previously, and R-Neh2Dual-ACTR-DHFR was labeled with sulfo-cyanine5 maleimide (Lumiprobe) on a single unique cysteine immediately N-terminal to DHFR, and repurified by gel filtration as described previously (26, 34). 26S Proteasome was purified from S. cerevisiae via a 3X-FLAG tag on Rpn11 as described previously (26).…”