Msp1 cooperates with the proteasome for extraction of arrested mitochondrial import intermediates

Abstract

The mitochondrial AAA ATPase Msp1 is well known for extraction of mislocalized tail-anchored ER proteins from the mitochondrial outer membrane. Here, we analyzed the extraction of precursors blocking the import pore in the outer membrane. We demonstrate strong genetic interactions of Msp1 and the proteasome with components of the TOM complex, the main translocase in the outer membrane. Msp1 and the proteasome both contribute to the removal of arrested precursor proteins that specifically accumulate in these mutants. The proteasome activity is essential for the removal as proteasome inhibitors block extraction. Furthermore, the proteasomal subunit Rpn10 copurified with Msp1. The human Msp1 homologue has been implicated in neurodegenerative diseases, and we show that the lack of the Caenorhabditis elegans Msp1 homologue triggers an import stress response in the worm, which indicates a conserved role in metazoa. In summary, our results suggest a role of Msp1 as an adaptor for the proteasome that drives the extraction of arrested and mislocalized proteins at the mitochondrial outer membrane.

FIGURE 1:

Genetic interaction analysis of MSP1 and genes encoding the subunits of the TOM complex. (A) indicated strains were grown on YPD medium at 30°C and 37°C. All drops were spotted in a serial dilution of 1:10. (B) Analysis of the levels of precursor and mature form of the chaperonin Hsp60 at 37°C and 30°C in tom mutants by SDS–PAGE, Western blot and immunodecoration. (C) Stability of the TOM complex in the indicated mutants was analyzed by Blue Native PAGE and immunodecoration against Tom40. (D) Msp1 expression levels in wild type and the ∆tom5 mutant. Protein levels of Msp1 and cytosolic Pgk1 were analyzed in wild type and the ∆tom5 mutant, and quantification of the Msp1 expression levels relative to Pgk1 was performed. Error bars indicate standard deviations. The difference between the two strains is statistically significant by the one sample t test (p = 0.0067, n = 5). (E) The strains were grown on glucose-containing medium at 30°C. Whole cell extracts were analyzed for levels of the outer membrane proteins Msp1, Por1, Fzo1, Ugo1, Tom20, Tom40, the inner membrane proteins Tim23 and Mia40, and the matrix proteins Aco1 and Hsp60. Pgk1 was decorated as cytosolic control.

FIGURE 2:

Analysis of tom5 and msp1 mutants. (A, B) Complementation analysis with an Msp1 overexpression plasmid (A) and Msp1 expression from a plasmid under the endogenous promoter (B) on glucose-containing selective medium at 30° and 37°C. The dots were spotted in a serial dilution of 1:10. (C) Protein levels of the precursor (p) and mature (m) form of the chaperonin Hsp60 at 30° and 37°C in ∆tom5 mutants with various Msp1 expression levels and (D) quantification of three independent experiments. Error bars indicate standard deviations. Statistical significance was tested by two-way ANOVA with the Tukey’s multiple comparison test. (E) In organello import of unfolded Cytb2ΔTM-DHFR into isolated mitochondria. After indicated time points, mitochondria were washed and analyzed by SDS–PAGE, Western blot and immunodecoration. (F) Western blot analysis of endogenous protein levels showing proteins affected in Msp1 overexpressing cells.

FIGURE 3:

Msp1-dependent chase of folded precursor from the TOM complex and its functional conservation in C. elegans. (A) Recombinant Cytb2-ΔTM-DHFR was prefolded in the presence of methotrexate and bound to mitochondria of the indicated strains. After indicated time points mitochondria were reisolated, washed and analyzed by SDS–PAGE, Western blot and immunodecoration and quantified (n = 3). Unprocessed precursor was loaded as Total (T). Error bars indicate standard deviations. (B) Cytb2-ΔTM-DHFR was expressed under the GAL1 promoter and prefolded in the presence of aminopterin in the indicated strains. After indicated time points, aliquots were taken from the cultures. Whole cell extracts were prepared and analyzed by SDS–PAGE, Western blot and immunodecoration. Cytb2-ΔTM-DHFR was quantified and chase and precursor/mature form ratio at the beginning of the chase were plotted (n = 3). Statistical significance was tested by two-way ANOVA with the Tukey’s multiple comparison test. Error bars indicate standard deviations. (C) Wild-type (+/+) and mspn-1(tm3831) animals carrying the P_hsp-6GFP transcriptional reporter (strains MD4432 and MD4430, respectively) were analyzed by brightfield and fluorescence microscopy (scale bar indicates 0.5 mm; intensity scale 1–1000). (D) The P_hsp-6GFP fluorescence intensities were quantified. Error bars indicate standard deviations. The difference between the two strains is statistically significant by Mann–Whitney test (p = 0.0022; n = 6).

FIGURE 4:

Identification of ubiquitin/proteasome system components in the quality control pathway at the TOM complex. (A) Cytb2-ΔTM-DHFR-His was expressed under the GAL1 promoter and prefolded in the presence of aminopterin in the cultures. NiNTA affinity purification was performed with the isolated mitochondria and copurified proteins were identified by SDS–PAGE, Western blot and immunodecoration. (T, 10% total; FT, 10% flow through; E, 100% eluate) (B) Interacting proteins were analyzed by mass spectrometry-based proteomics after NiNTA purification and differences are depicted in the volcano plot after statistical comparison in Limma(R). The bait protein Msp1-His is most abundantly enriched with an enrichment factor (log2 fold change) of 11 in the top right corner. The strongest interactor is the proteasomal protein Rpn10, which was exclusively found in the Msp1-His interactome and not in the wild type control. (C) Isolated mitochondria were analyzed by SDS–PAGE, Western blot, and decorated against ubiquitin. (D) Strains expressing chromosomally integrated His-tagged ubiquitin and Cytb2-ΔTM-DHFR were grown in the presence and absence of aminopterin. Cells were collected and lysed. The cell lysate was subjected to NiNTA affinity purification and 0.5% total (T) and 100% bound fraction (Eluate [E]) were analyzed by SDS–PAGE, Western blot, and immunodecoration against DHFR and ubiquitin.

FIGURE 5:

Proteasomal activity is required for extraction of arrested precursors. (A) Recombinant Cytb2-ΔTM-DHFR was prefolded in the presence of methotrexate and bound to wild-type mitochondria in the presence or absence of proteasome inhibitor MG-132. After indicated time points, mitochondria were reisolated and mitochondrial pellet and TCA-precipitated supernatant were analyzed by SDS–PAGE, Western blot, and immunodecoration against DHFR. (T, 10% of unprocessed Cytb2-ΔTM-DHFR precursor of each sample; a degradation product was labeled with an asterisk). (B) A growth analysis of the indicated yeast strains was performed on YPGal medium at 30°C or 37°C. Drops were spotted in serial dilution of 1:10. (C) Analysis of the levels of precursor and mature form of the chaperonin Hsp60 at 37° and 30°C by SDS–PAGE, Western blot, and immunodecoration.

FIGURE 6:

Model for the concerted action of Msp1 and the proteasome (similar to Cdc48 and proteasome function during ERAD) during extraction of Msp1 substrates. TOM subunits showing a genetic interaction with Msp1 are depicted in dark gray.