Pex2 and pex12 function as protein-ubiquitin ligases in peroxisomal protein import

Abstract

The PTS1-dependent peroxisomal matrix protein import is facilitated by the receptor protein Pex5 and can be divided into cargo recognition in the cytosol, membrane docking of the cargo-receptor complex, cargo release, and recycling of the receptor. The final step is controlled by the ubiquitination status of Pex5. While polyubiquitinated Pex5 is degraded by the proteasome, monoubiquitinated Pex5 is destined for a new round of the receptor cycle. Recently, the ubiquitin-conjugating enzymes involved in Pex5 ubiquitination were identified as Ubc4 and Pex4 (Ubc10), whereas the identity of the corresponding protein-ubiquitin ligases remained unknown. Here we report on the identification of the protein-ubiquitin ligases that are responsible for the ubiquitination of the peroxisomal protein import receptor Pex5. It is demonstrated that each of the three RING peroxins Pex2, Pex10, and Pex12 exhibits ubiquitin-protein isopeptide ligase activity. Our results show that Pex2 mediates the Ubc4-dependent polyubiquitination whereas Pex12 facilitates the Pex4-dependent monoubiquitination of Pex5.

FIG. 1.

Ubiquitination of Pex5 depends on the presence of zinc ions and the RING finger peroxins. (A) Pex5 was isolated by immunoprecipitation with anti-Pex5 antibodies from whole-cell membranes of oleic acid-induced wild-type cells in the presence or absence of TPEN, zinc chloride, and NEM to inhibit deubiquitination enzymes. (B) Wild-type cells were broken by glass bead treatment in the presence or absence of TPEN, zinc chloride, and MG132 to inhibit proteasomal degradation. Proteins were subsequently precipitated by TCA, and samples were subjected to immunoblot analysis. Detection of the mitochondrial porin served as a loading control. (C) Pex5 was isolated from whole-cell membranes of the indicated strains in the presence of NEM and subjected to immunoblot analysis. Deletion of either one of the RING peroxins resulted in a loss of Pex5 ubiquitination. (D) Whole-cell TCA lysates of indicated strains were prepared and analyzed by immunoblot analysis. Whereas deletion of PEX4 stabilized the polyubiquitinated form of Pex5, no modification was detectable when in addition one of the RING peroxins was deleted, indicating that each of the ring finger peroxins is required for Pex5 polyubiquitination. Detection of Pex5 and ubiquitinated Pex5 was performed with antibodies against Pex5 or anti-Ub antibodies as indicated.

FIG. 2.

The RING complex exhibits Ub ligase activity. (A) The membrane-bound protein import complex was isolated from digitonin-solubilized membranes of wild-type cells expressing Pex2-TEV-ProtA as well as HA-tagged Pex4 via IgG-Sepharose and subsequent TEV protease cleavage. As a control, wild-type cells expressing no ProtA fusion protein were treated equally. The corresponding digitonin-solubilized membranes (totals) are shown in the right column. Indicated proteins of the TEV protease eluates and totals were detected by immunoblot analysis. (B and C) Recombinant Pex5 was incubated in the presence or absence of recombinant E1, zinc ions, and the Pex2 complex isolated from pex5Δ membranes. Reactions were stopped by the addition of SDS sample buffer after 45 min (B) or indicated time points (C), and modified Pex5 species were detected by immunoblot analysis. Ubiquitinated Pex5 was clearly detectable only when all constituents were present. Lack of either zinc or E1 led to significantly smaller amounts of modified Pex5.

FIG. 3.

Autoubiquitination of the recombinant RING domains of Pex2, Pex10, and Pex12. Isolated GST fusions of the RING finger domains of the three RING peroxins, Pex2 (aa 215 to 271), Pex10 (aa 238 to 337), and Pex12 (aa 293 to 399), were incubated with the indicated E2 enzymes in the presence or absence of E1, Ub, and methylated Ub (meUb) (A) or with Ub in the presence or absence of E1 (B). Autoubiquitination of the proteins was monitored by immunoblot analysis with antibodies against GST (A) or Ub (B). Ubc4 and Pex4 but not Ubc7 facilitated the modifications of the three RING domains.

FIG. 4.

RING peroxin-dependent ubiquitination of a synthetic model substrate. Indicated combinations of the recombinant model substrate GST-S peptide-RING finger fusion proteins, the S protein, and either Ub or Ub fused to a His tag (His₆-Ub) were incubated with Pex4 (A) or Ubc4 (B) and analyzed by immunoblotting with antibodies against Ub. The appearance of ubiquitinated S protein depends on the presence of any of the three RING domains and Ub.

FIG. 5.

RING peroxin-dependent ubiquitination of the PTS1 receptor Pex5. Purified GST fusions of the RING finger domains were incubated with recombinant His-Pex5 and either Pex4 (A) or Ubc4 (B) as E2 enzyme. Pex5 ubiquitination was monitored by immunoblot analysis with Pex5- and Ub-specific antibodies. Slower-migrating His-Pex5 forms representing the ubiquitinated receptor were detected when Pex4 was combined with Pex12-RING and when Ubc4 was combined with Pex2-RING. These data demonstrate that these two RING peroxins harbor Pex5-specific protein-Ub ligase activity. The data also indicate that Pex4 and Pex12 as well as Ubc4 and Pex2 form cooperating pairs in the Pex5-specific ubiquitination.

FIG. 6.

Phenotypic analysis of RING truncation mutants. (A) The growth behavior of indicated S. cerevisiae wild-type and mutant strains on oleic acid as the sole carbon source was investigated using serial dilutions. These were spotted and incubated for 7 days at 30°C. (B) Strains expressing GFP-PTS1 were examined for GFP fluorescence by bright-field microscopy as indicated. Bar = 10 μm. Analysis of live cells for GFP fluorescence was performed with a Zeiss Axioplan microscope and AxioVision 4.1 software (Zeiss, Jena, Germany). The RING domains of each of the three RING peroxins were found to be essential for the import of GFP-PTS1 into peroxisomes.

FIG. 7.

Pex12-RING is required for Pex5 monoubiquitination while Pex2-RING is essential for Pex5 polyubiquitination. The influence of genomic truncations on the monoubiquitination (A and B) and polyubiquitination (C) of Pex5 was analyzed in vivo. Pex5 was isolated by immunoprecipitation from total cell membranes prepared from the indicated oleic acid-induced wild-type strains, null mutants, or strains harboring the genomic truncation of the RING peroxins. The pex8Δ strain served as a negative control. Samples were analyzed by immunoblot analysis with antibodies against Pex5 and Ub. (A and B) To monitor monoubiquitination, NEM was applied to the cells prior to breakage. Genomic truncation of the Pex12-RING prevents the receptor monoubiquitination whereas this modification is still present upon truncation of either the RING domain of Pex2 or that of Pex10. This result indicates that Pex12 is responsible for the Pex5 monoubiquitination. (C) Polyubiquitination occurs in mutants affected in late stages of the peroxisomal protein import pathway represented here by the pex4Δ strain. Truncation of the RING domain of Pex2 but not of Pex10 or Pex12 prevents the receptor polyubiquitination in a pex4Δ background. This result demonstrates that the RING domain of Pex2 is required for Pex5 polyubiquitination.

FIG. 8.

The ubiquitination cascade of the peroxisomal protein import machinery. Pex5, the import receptor for peroxisomal matrix proteins harboring a type I signal sequence (PTS1), is monoubiquitinated and polyUb. Polyubiquitination of Pex5 requires the Ub-conjugating enzyme Ubc4 and the RING peroxin Pex2 as protein-Ub ligase (E3). polyUb Pex5 is supposed to be recognized by the AAA peroxins Pex1 and Pex6 and directed to proteasomal degradation as a part of a quality control system. Monoubiquitination of the receptor is catalyzed by the peroxisomal E2 enzyme Pex4 and Pex12 as the responsible protein-Ub ligase (E3). Monoubiquitination of Pex5 can be regarded as an export signal supposedly recognized by the AAA peroxins, which displace the receptor from the membrane to the cytosol, where it is then available for another round of import.