A chaperone-assisted degradation pathway targets kinetochore proteins to ensure genome stability

Abstract

Cells are regularly exposed to stress conditions that may lead to protein misfolding. To cope with this challenge, molecular chaperones selectively target structurally perturbed proteins for degradation via the ubiquitin-proteasome pathway. In mammals the co-chaperone BAG-1 plays an important role in this system. BAG-1 has two orthologues, Bag101 and Bag102, in the fission yeast Schizosaccharomyces pombe. We show that both Bag101 and Bag102 interact with 26S proteasomes and Hsp70. By epistasis mapping we identify a mutant in the conserved kinetochore component Spc7 (Spc105/Blinkin) as a target for a quality control system that also involves, Hsp70, Bag102, the 26S proteasome, Ubc4 and the ubiquitin-ligases Ubr11 and San1. Accordingly, chromosome missegregation of spc7 mutant strains is alleviated by mutation of components in this pathway. In addition, we isolated a dominant negative version of the deubiquitylating enzyme, Ubp3, as a suppressor of the spc7-23 phenotype, suggesting that the proteasome-associated Ubp3 is required for this degradation system. Finally, our data suggest that the identified pathway is also involved in quality control of other kinetochore components and therefore likely to be a common degradation mechanism to ensure nuclear protein homeostasis and genome integrity.

Figure 1. Bag101 and Bag102 interact with 26S proteasomes and Hsp70.

(A) Domain organization (shown to scale) of human BAG-1S and the S. pombe homologs Bag101 and Bag102. The truncations used for the precipitation experiments are shown. (B) The indicated GST fusion proteins were used in pull down experiments with extract from S. pombe cells expressing ZZ-tagged Rpn11. The precipitated material was analyzed by SDS-PAGE and blotting for Hsp70 (upper panel), the ZZ-tagged 26S proteasome subunit Rpn11, and 20S particle α subunits (middle panels). Equal loading was checked by staining with Coomassie Brilliant Blue (CBB) (lower panel). (C) Immunoprecipitates from wild type S. pombe cells expressing Flag-tagged Bag101 and Bag102 were resolved by SDS-PAGE and analyzed by blotting, using antibodies to the proteasome subunit Mts4/Rpn1, and Flag. (D) Differential interference contrast (DIC) and fluorescence micrographs of wild type S. pombe transformed to express GFP-tagged Bag101 and Bag102. DAPI staining was used to mark the nucleus. (E) Lysates from S. pombe cells transformed to express Flag- and GFP-tagged Bag102 were separated into a soluble fraction and pellet. The pellet fraction was then treated with proteinase K and Triton X-100 as indicated, before the samples were analyzed by SDS-PAGE and blotting. Dph1 served as a control for a soluble protein. Bip1 served as a control for an ER luminal protein.

Figure 2. spc7-23 cells are temperature sensitive and defective in DNA segregation.

(A) The growth of wild type and spc7-23 cells on solid media was compared at the indicated temperatures. (B) Cells carrying the mutant spc7-23 mutation were incubated at 25°C or 30°C and stained to visualize DNA (blue) and tubulin (green). Note that the DNA segregation becomes unequal at the restrictive temperature.

Figure 3. Spc7-23 levels are regulated by Bag102.

(A) The growth on rich media of wild type, bag101Δ, bag102Δ, spc7-23 and double mutants was compared at the indicated temperatures. (B) The abundance and kinetochore localization of the Spc7-23-GFP protein or, as a control, wild type Spc7-GFP, was analyzed in cells at 25°C (left panel) or at 30°C for 6 hours (right panel). Each fluorescent signal represents the Spc7-23-GFP kinetochore signal of an individual interphase cell as kinetochores are clustered at the spindle pole body during interphase. Note that at 30°C the Spc7-23-GFP signal is reduced in a wild type (bag102 ⁺) or bag101Δ background. In a bag102Δ background the Spc7-23-GFP signal is increased at both temperatures. Scale bar: 5 µm. (C) Equal (wild type) or unequal DNA segregation was quantified in spc7-23-gfp and spc7-23-gfp bag102Δ cells at 30°C. ** p<0.01 (Welch test) for the spc7-23-gfp bag102Δ strain compared to the spc7-23-gfp strain. For spc7-23-gfp (n = 200) and spc7-23-gfp bag102Δ (n = 100) late anaphase cells. Note that wild type DNA segregation is re-established in the spc7-23bag102Δ double mutant. (D) Spc7-3HA or Spc7-23-3HA was precipitated using antibodies to HA or control antibodies of the same isotype against an irrelevant protein (α2-macroglobulin) at either 4°C or 30°C. The precipitated material was resolved by SDS-PAGE and analyzed by blotting for the presence of Hsp70 (upper panel). Blotting to HA (Spc7 and Spc7-23) served as a loading control (lower panel). Note that Spc7-23 interacts with Hsp70, especially at 30°C.

Figure 4. Spc7-23 is degraded via the ubiquitin-proteasome pathway.

(A) The growth on solid media of wild type (lower panel) and spc7-23 cells (upper panel) transformed with either a control vector (vector) or PstI digested genomic DNA construct encoding mts2-219X was compared at different temperatures. (B) The growth of wild type, mts2-1, spc7-23 and the mts2-1spc7-23 double mutant was compared at the indicated temperatures. (C) The growth on solid media of wild type, nas6Δ, spc7-23 and the nas6Δspc7-23 double mutant was compared at the indicated temperatures. (D) Equal (wild type) or unequal DNA segregation was quantified in spc7-23-gfp and spc7-23-gfp nas6Δ cells at 30°C. ** p<0.01 (Welch test) for the spc7-23-gfp nas6Δ strain compared to the spc7-23-gfp strain. For spc7-23-gfp (n = 200) and spc7-23-gfp nas6Δ (n = 100) late anaphase cells. Note that wild type DNA segregation is re-established in the spc7-23nas6Δ double mutant. (E) The amount of Spc7-23 protein was followed in cultures at 27°C and 30°C where protein synthesis was inhibited with 100 µg/mL cycloheximide (CHX) for 4 hours. To some cultures 1 mM of the proteasome inhibitor Bortezomib (BZ) was also added. Equal loading was checked using antibodies to tubulin. (F) The growth on solid media of wild type and spc7-23 cells was compared at the indicated temperatures in the absence (control) or presence of 100 µM BZ. (G) Strains with the indicated genetic backgrounds and transformed to express 6His-tagged ubiquitin were lysed and used for precipitation experiments with a Ni²⁺ resin in 8 M urea. The precipitated material was analyzed by blotting with antibodies to the HA-tag on Spc7-23 or to the 6His tag on ubiquitin. The arrowhead marks the position where non-ubiquitylated Spc7-23 migrates. Note that ubiquitylated Spc7-23 species are visible in proteasome and bag102Δ mutants.

Figure 5. Spc7-23 degradation depends on Ubc4, Ubr11 and San1.

(A) The growth of the indicated strains on solid media was compared at 25°C (upper panel) and 30°C (lower panel). (B) The growth of the indicated strains on solid media was compared at 25°C (left panel) and 30°C (right panel). (C) Equal (wild type) or unequal DNA segregation was quantified in spc7-23-gfp ubr11Δ and spc7-23-gfp san1Δ cells at 30°C. ** p<0.01 (Welch test) for the spc7-23-gfp ubr11Δ and spc7-23-gfp san1Δ strains compared to the spc7-23-gfp strain. For spc7-23-gfp (n = 200), spc7-23-gfp ubr11Δ (n = 100) and spc7-23-gfp san1Δ (n = 100) late anaphase cells. Note that wild type DNA segregation is re-established in the spc7-23ubr11Δ and spc7-23san1Δ double mutants. (D) V5-tagged Ubr11 was immunoprecipitated from cells treated with Bortezomib (BZ) using antibodies to V5 or control antibody. The precipitated material was analyzed by SDS-PAGE and blotting for V5 (Ubr11) and HA (Spc7). (E) Flag-tagged San1 was immunoprecipitated from cells treated with Bortezomib (BZ) using antibodies to Flag or control antibody. The precipitated material was analyzed by SDS-PAGE and blotting for Flag (San1) and HA (Spc7). (F) Strains with the indicated genetic backgrounds and transformed to express 6His-tagged ubiquitin were either not treated or treated with 1 mM of the proteasome inhibitor Bortezomib (BZ) overnight. 6His-tagged ubiquitin was precipitated with a Ni²⁺ resin in 8 M urea. The precipitated material was analyzed by SDS-PAGE and blotting with antibodies to the HA-tag on Spc7-23. The arrowhead marks the position where non-ubiquitylated Spc7-23 migrates. Note that ubiquitylated Spc7-23 species are less abundant in ubr11Δ and san1Δ mutants.

Figure 6. Spc7-23 degradation depends on the proteasome-associated DUB Ubp3.

(A) The growth on solid media of spc7-23 cells transformed with either a control plasmid (vector) or expression constructs for ubp3 ⁺ and ubp3-W466X was compared at different temperatures. (B) The growth on solid media of wild type, ubp3Δ, spc7-23 and the ubp3Δspc7-23 double mutant was compared at the indicated temperatures. (C) Equal (wild type) or unequal DNA segregation was quantified in spc7-23-gfp and spc7-23-gfp ubp3Δ cells at 30°C. ** p<0.01 (Welch test) for the spc7-23-gfp ubp3Δ strain compared to the spc7-23-gfp strain. For spc7-23-gfp (n = 200) and spc7-23-gfp ubp3Δ (n = 100) late anaphase cells. Note that wild type DNA segregation is re-established in the spc7-23ubp3Δ double mutant. (D) Cells transformed with vector (control) or Ubp3-Flag were used for immunoprecipitation experiments using antibodies to the Flag epitope. The precipitated material was analyzed by blotting for the proteasome subunit Mts4 (upper panel) or as a loading control Ubp3 (lower panel). (E) Wild type or ubp3Δ cells transformed to express 6His-tagged ubiquitin, were lysed and used for precipitation experiments with a Ni²⁺ resin in 8 M urea. The precipitated material was analyzed by blotting with antibodies to the HA-tag on Spc7-23 or the 6His tag on ubiquitin. The arrowhead marks the position where non-ubiquitylated Spc7-23 migrates.

Figure 7. Other kinetochore mutants are also suppressed by ubr11Δ and san1Δ.

(A) The growth on solid media of wild type, ubr11Δ, mis6-302 and the ubr11Δmis6-302 double mutant was compared at the indicated temperatures. (B) The growth on solid media of wild type, ubr11Δ, mal2-1 and the ubr11Δmal2-1 double mutant was compared at the indicated temperatures. (C) The growth on solid media of wild type, san1Δ, mis6-302 and the san1Δmis6-302 double mutant was compared at the indicated temperatures. (D) The growth on solid media of wild type, san1Δ, mal2-1 and the san1Δmal2-1 double mutant was compared at the indicated temperatures. (E) The growth on solid media of wild type, ubc4-1, mis6-302 and the ubc4-1mis6-302 double mutant was compared at the indicated temperatures. (F) The growth on solid media of wild type, ubc4-1, mal2-1 and the ubc4-1mal2-1 double mutant was compared at the indicated temperatures. (G) The growth on solid media of wild type, bag102Δ, mis6-302 and the bag102Δmis6-302 double mutant was compared at the indicated temperatures. (H) The growth on solid media of wild type, bag102Δ, mal2-1 and the bag102Δmal2-1 double mutant was compared at the indicated temperatures.

Figure 8. A chaperone-assisted degradation pathway for nuclear proteins.

The data presented here are compatible with a model where a nuclear protein becomes structurally perturbed to a degree where it is still functional, but molecular chaperones detect it as being misfolded. The protein is then ubiquitylated by the E2 and E3 enzymes Ubc4, Ubr11 and San1, and directed to the 26S proteasome via Bag102. Finally, at the 26S proteasome, the protein is deubiquitylated by the DUB Ubp3 and degraded. Ubiquitin is shown as grey spheres. The structural perturbation is shown as a star.