Direct ubiquitin independent recognition and degradation of a folded protein by the eukaryotic proteasomes-origin of intrinsic degradation signals

Abstract

Eukaryotic 26S proteasomes are structurally organized to recognize, unfold and degrade globular proteins. However, all existing model substrates of the 26S proteasome in addition to ubiquitin or adaptor proteins require unstructured regions in the form of fusion tags for efficient degradation. We report for the first time that purified 26S proteasome can directly recognize and degrade apomyoglobin, a globular protein, in the absence of ubiquitin, extrinsic degradation tags or adaptor proteins. Despite a high affinity interaction, absence of a ligand and presence of only helices/loops that follow the degradation signal, apomyoglobin is degraded slowly by the proteasome. A short floppy F-helix exposed upon ligand removal and in conformational equilibrium with a disordered structure is mandatory for recognition and initiation of degradation. Holomyoglobin, in which the helix is buried, is neither recognized nor degraded. Exposure of the floppy F-helix seems to sensitize the proteasome and primes the substrate for degradation. Using peptide panning and competition experiments we speculate that initial encounters through the floppy helix and additional strong interactions with N-terminal helices anchors apomyoglobin to the proteasome. Stabilizing helical structure in the floppy F-helix slows down degradation. Destabilization of adjacent helices accelerates degradation. Unfolding seems to follow the mechanism of helix unraveling rather than global unfolding. Our findings while confirming the requirement for unstructured regions in degradation offers the following new insights: a) origin and identification of an intrinsic degradation signal in the substrate, b) identification of sequences in the native substrate that are likely to be responsible for direct interactions with the proteasome, and c) identification of critical rate limiting steps like exposure of the intrinsic degron and destabilization of an unfolding intermediate that are presumably catalyzed by the ATPases. Apomyoglobin emerges as a new model substrate to further explore the role of ATPases and protein structure in proteasomal degradation.

Figure 1. Purification and characterization of holo and apo myoglobin (Mb).

(A) Mb was purified by cation exchange chromatography and was adjudged to be pure by SDS-PAGE analysis. (B) UV-visible spectrum of the holo and apoMb were recorded from 500 nm to 240 nm. A distinct Soret peak (410 nm) was observed in the holo form the intensity of which was reduced to about 95% in the apoMb indicating successful removal of heme. (C) Quaternary structure of apoMb was assessed by gel permeation chromatography. ApoMb (dash line) eluted at the same retention volume as the holo protein (solid line) demonstrating similarity in the protein fold.

Figure 2. Purified yeast 26S proteasome recognizes and degrades apoMb in vitro in the absence of ubiquitin and any trans-acting element in an ATP dependent manner.

(A) Purified 26S proteasomes were to able degrade apoMb but not the holoprotein. Proteins were incubated with 26S proteasomes at 37°C. Rate of degradation was followed by SDS-PAGE (inset) and quantified as described in methods. (B) Degradation of apoMb is dependent on the 19S regulatory particle and ATP. Purified 20S core particles were not able to degrade apoMb. Degradation by 26S proteasome was inhibited by MG132, epoxomicin and Velcade but not PMSF. No significant degradation was observed in the presence of ATPγS, the non-hydrolysable analog of ATP. (C) ApoMb and not the holo form is recognized by the 26S proteasomes. ApoMb and holoMb were incubated with immobilized proteasome and detected using anti-Mb antibody. (D) ApoMb and not the holo form is able to stimulate the ATPase activity of the 26S proteasomes. Data represent the mean values of at least three independent experiments ±S.D. * Single experiment.

Figure 3. Floppy F-helix is crucial for the degradation of ApoMb.

(A) AGADIR prediction (parameter, pH 7.5, Temperature 273K, Ionic strength 0.15 M) of the helical propensity of floppy F-helix. Wt sequence (solid line), G80AP88AS92AH97E (dashed grey) and G80AP88AS92AH97N (dashed black). (B) MD simulation of wt apoMb and the F-helix mutant. The wt sequence melts immediately at 400 K while the F-helix mutant remains stable even at the end of 2.8 ns simulation. (C) To verify the role of floppy F-helix exposed upon removal of heme, helix stabilizing mutations were introduced. Far-UV CD spectrum shows that the Apo F-helix mutant has enhanced secondary structure as compared to the wt ApoMb. The difference spectra were obtained by subtracting the spectra of apowt from apoF-helix (MRE values are on Y2). (D) Apo wt and apo F-helix proteins were incubated with proteasome. The rate of degradation was followed by SDS-PAGE (inset) and quantified as described in methods. Data represent the mean values ±S.D of at least three independent experiments for wt apoMb and five independent experiments for F-helix mutant. Remarkably, stabilization of F-helix rendered ApoMb more resistant to degradation by the proteasome.

Figure 4. Limited proteolysis demonstrates the presence of an unstable and stable F-helix in wt and F-helix mutant respectively.

(A) Cleavage sites of trypsin and chymotrypsin on Mb are diagrammatically represented (in F-helix underlined amino acids are mutated). Trypsin (B) and chymotrypsin (C) were added to wt and F-helix mutant. Aliquots at various time intervals were analyzed by Tricine-SDS Page. Wt protein is cleaved by chymotrypsin as soon as the enzyme is added (0 min). These fragments are not contaminant in the preparation as can be seen from the purity of Mb in Figure 1. Substrate alone controls were stable (data not shown).

Figure 5. Floppy F-helix sensitizes the proteasome for the presence of substrate and the interaction is reinforced by N-terminal helices.

Overlapping peptides panning the entire length of myoglobin were tested for binding to the immobilized proteasome by ELISA. (A) A-helix peptide which bound strongly to the proteasome was tested for its ability to compete with apoMb. This peptide brought about 80% inhibition at 3 µM concentration. (B) The B-helix, CD-loop, E7 and F7 peptides which share sequences with the floppy F-helix also inhibited the binding of apoMb. However they were considerably less potent than the A-helix peptide. All incubations were done in the presence of 100 nM MG132. Data represent mean values of at least three independent ±S.D. For E7 S.D. is not plotted for clarity.

Figure 6. Mutation of buried Leu residues in the G-helix shortens the half-life of apoMb.

(A) Apo L115C and (B) apo L104C mutant proteins were incubated with proteasome in the presence or absence of MG132. Rate of degradation was followed by SDS-PAGE and quantified as described in methods. Data represent mean values of at least three independent experiments ±S.D.

Figure 7. A model for the mechanistic steps involved in the degradation of apoMb based on structural changes in mutations used as snap shots.

Removal of heme exposes a previously buried F-helix which is in a dynamic equilibrium between a partially folded and unfolded structure. This transition is a rate limiting step. Exposure of this floppy helix sensitizes the proteasome to the presence of the substrate. ApoMb is anchored to the proteasome by interactions primarily through the A-helix. Additional interactions stabilize the enzyme-substrate complex. Degradation is primed by the insertion of the floppy helix in the form of a loop into the central channel that runs across the proteasome. An intermediate composed of AGH helices is likely to be formed. Melting of this intermediate by the ATPases to generate an unstructured region long enough to reach the active site is a likely rate limiting step. Mutations can stabilize or destabilize this unfolding intermediate affecting the rate of degradation.