Yeast Pth2 is a UBL domain-binding protein that participates in the ubiquitin–proteasome pathway
2006; Springer Nature; Volume: 25; Issue: 23 Linguagem: Inglês
10.1038/sj.emboj.7601418
ISSN1460-2075
AutoresTakashi Ishii, M Funakoshi, Hideki Kobayashi,
Tópico(s)Fungal and yeast genetics research
ResumoArticle2 November 2006free access Yeast Pth2 is a UBL domain-binding protein that participates in the ubiquitin–proteasome pathway Takashi Ishii Takashi Ishii Department of Molecular Biology, Graduate School of Medical Sciences, Kyushu University, Higashi-ku, Fukuoka, Japan CREST, Japanese Science and Technology Agency, Kawaguchi, Saitama, Japan Search for more papers by this author Minoru Funakoshi Minoru Funakoshi Department of Molecular Biology, Graduate School of Medical Sciences, Kyushu University, Higashi-ku, Fukuoka, JapanPresent address: Department of Molecular Biophysics & Biochemistry, Yale University, 266 Whitney Avenue, New Haven, CT 06520-8114, USA Search for more papers by this author Hideki Kobayashi Corresponding Author Hideki Kobayashi Department of Molecular Biology, Graduate School of Medical Sciences, Kyushu University, Higashi-ku, Fukuoka, Japan CREST, Japanese Science and Technology Agency, Kawaguchi, Saitama, Japan Search for more papers by this author Takashi Ishii Takashi Ishii Department of Molecular Biology, Graduate School of Medical Sciences, Kyushu University, Higashi-ku, Fukuoka, Japan CREST, Japanese Science and Technology Agency, Kawaguchi, Saitama, Japan Search for more papers by this author Minoru Funakoshi Minoru Funakoshi Department of Molecular Biology, Graduate School of Medical Sciences, Kyushu University, Higashi-ku, Fukuoka, JapanPresent address: Department of Molecular Biophysics & Biochemistry, Yale University, 266 Whitney Avenue, New Haven, CT 06520-8114, USA Search for more papers by this author Hideki Kobayashi Corresponding Author Hideki Kobayashi Department of Molecular Biology, Graduate School of Medical Sciences, Kyushu University, Higashi-ku, Fukuoka, Japan CREST, Japanese Science and Technology Agency, Kawaguchi, Saitama, Japan Search for more papers by this author Author Information Takashi Ishii1,2, Minoru Funakoshi1 and Hideki Kobayashi 1,2 1Department of Molecular Biology, Graduate School of Medical Sciences, Kyushu University, Higashi-ku, Fukuoka, Japan 2CREST, Japanese Science and Technology Agency, Kawaguchi, Saitama, Japan *Corresponding author. Department of Molecular Biology, Graduate School of Medical Sciences, Kyushu University, Maidashi 3-1-1, Higashi-ku, Fukuoka 812-8582, Japan. Tel.: +81 92 642 6179; Fax: +81 92 642 6183; E-mail: [email protected] The EMBO Journal (2006)25:5492-5503https://doi.org/10.1038/sj.emboj.7601418 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Ubiquitin-like (UBL)-ubiquitin-associated (UBA) proteins such as Rad23 and Dsk2 mediate the delivery of polyubiquitinated proteins to the proteasome in the ubiquitin–proteasome pathway. We show here that budding yeast peptidyl-tRNA hydrolase 2 (Pth2), which was previously recognized as a peptidyl-tRNA hydrolase, is a UBL domain-binding protein that participates in the ubiquitin–proteasome pathway. Pth2 bound to the UBL domain of both Rad23 and Dsk2. Pth2 also interacted with polyubiquitinated proteins through the UBA domains of Rad23 and Dsk2. Pth2 overexpression caused an accumulation of polyubiquitinated proteins and inhibited the growth of yeast. Ubiquitin-dependent degradation was accelerated in the pth2Δ mutant and was retarded by overexpression of Pth2. Pth2 inhibited the interaction of Rad23 and Dsk2 with the polyubiquitin receptors Rpn1 and Rpn10 on the proteasome. Furthermore, Pth2 function involving UBL-UBA proteins was independent of its peptidyl-tRNA hydrolase activity. These results suggest that Pth2 negatively regulates the UBL-UBA protein-mediated shuttling pathway in the ubiquitin–proteasome system. Introduction Recruitment of ubiquitinated proteins to the proteasome is a crucial step in the ubiquitin–proteasome pathway. The integrity of this pathway is ensured by a sequential mode of action: capture of ubiquitin conjugates, delivery to the proteasome and release from the proteasome ubiquitin receptors. Ubiquitinated proteins can be recognized directly by the 19S proteasomal subunit Rpn10 (van Nocker et al, 1996; Saeki et al, 2002; Elsasser et al, 2004; Verma et al, 2004). In addition to this pathway, UBL-UBA proteins help to deliver proteins to be degraded to the proteasome. UBL-UBA proteins belong to a family of proteins that contain a UBL domain at the N terminus and a UBA domain at the C terminus. Through the ability of the UBL domain to interact with polyubiquitin receptors on the 19S proteasome regulatory particle and the UBA domain to bind to polyubiquitin, UBL-UBA proteins such Rad23 and Dsk2 in Saccharomyces cerevisiae serve as shuttle factors delivering ubiquitinated substrates to the proteasome (Schauber et al, 1998; Lambertson et al, 1999; Wilkinson et al, 2001; Chen and Madura, 2002; Funakoshi et al, 2002; Rao and Sastry, 2002). These shuttle factors interact with the 19S proteasomal subunit Rpn1, which acts as a scaffold on the proteasome (Elsasser et al, 2002). A recently developed cell-free system has demonstrated that UBL-UBA shuttle proteins promote protein degradation by the proteasome (Verma et al, 2004). These accumulated findings have advanced our understanding of the molecular basis of recognition of polyubiquitinated proteins by the proteasome (reviewed by Hartmann-Petersen et al, 2003; Madura, 2004; Elsasser and Finley, 2005). However, how the shuttle proteins regulate the flux of polyubiquitinated proteins to the proteasome receptors is not well understood. To investigate the regulation of UBL-UBA proteins in shuttling pathways, we searched for interacting factors of UBL-UBA proteins in a two-hybrid screen, and identified peptidyl-tRNA hydrolase 2 (Pth2) from S. cerevisiae as an interacting protein of Rad23 and Dsk2. Pth is an enzyme that cleaves peptidyl-tRNA to tRNA and peptides during translation elongation (Menninger, 1976) and is essential for growth in bacteria. In archaea, a second Pth enzyme named Pth2 has been identified, whose amino-acid sequence is completely different from bacterial Pth (Rosas-Sandoval et al, 2002). Eukaryotic cells encode both Pth and Pth2, and curiously, the two eukaryotic enzymes are dispensable for yeast viability (Menez et al, 2002; Rosas-Sandoval et al, 2002; Fromant et al, 2003). Pth2 has two domains: a conserved C-terminal half is the catalytic domain (de Pereda et al, 2004), but no function has been ascribed to the highly divergent N-terminal half, suggesting that this region may confer species specificity to Pth2. As we show in this report, budding yeast Pth2 is a UBL domain-binding protein that inhibits ubiquitin-mediated protein degradation. In addition, the C-terminal domain of Pth2 is required for the interaction with Rad23 and Dsk2, but this Pth2 function is independent of its C-terminal peptidyl-tRNA hydrolase activity. Based on our findings, we discuss a regulatory role of Pth2 in substrate delivery to the proteasome via an interaction with the UBL domain of UBL-UBA proteins. Budding yeast Pth2 may have a role in the ubiquitin–proteasome pathway and as a peptidyl-tRNA hydrolase. Results Yeast Pth2 interacts with Dsk2 and Rad23 in vivo and in vitro We screened for proteins that interact with a UBL-UBA protein in a two-hybrid screen using Dsk2 as bait. Among 2 × 105 transformants from an S. cerevisiae cDNA library, three clones interacted with Dsk2 strongly, and each of these clones was identified as PTH2 (Figure 1A), which encodes a full-length peptidyl-tRNA hydrolase. First, we tested for a direct interaction in vitro between Pth2 and the UBL-UBA proteins (Figure 1B). His6-T7-tagged Pth2 bound directly to GST-Dsk2 (lane 5) and GST-Rad23 (lane 6), but not to the DNA damage-inducible UBL-UBA protein, Ddi1 (lane 7). Next, we coexpressed the T7-Pth2 with GST-Dsk2 or GST-Rad23 in yeast, and in vivo binding was tested using a GST pull-down assay followed by immunoblotting with anti-T7 (Figure 1C). T7-Pth2 co-precipitated with GST-Dsk2 (lane 10) and GST-Rad23 (lane 11) and, to a lesser extent, with GST-Ddi1 (lane 12). The weak binding to Ddi1 was abolished in the dsk2Δrad23Δ mutant (lane 16), suggesting that the interaction of Pth2 is meditated by Dsk2 and Rad23. Therefore, Pth2 binds selectively to the UBL-UBA proteins Rad23 and Dsk2 in vivo. Next, we confirmed in vivo interaction of Pth2 with UBL-UBA protein at endogenous expression level (Figure 1D). FLAG-tagged Pth2 was expressed from PTH2 own promoter in pth2Δ cells (middle panel, lane 4) and the extracts were immunoprecipitated with anti-FLAG beads. FLAG-Pth2 expressing at endogenous levels or even less in pth2Δ co-precipitated with endogenous Dsk2 (bottom panel, lane 4). Figure 1.Pth2 binds to Dsk2 and Rad23 but not to Ddi1. (A) Identification of PTH2 in a two-hybrid assay. Pth2 was obtained as a Dsk2-interacting clone. Interaction of Pth2 with Dsk2 was tested by histidine-prototrophic growth, using DSK2 as bait and PTH2 as prey (upper panel) or vice versa (lower panel). (B) Binding of Pth2 to Dsk2 and Rad23 in vitro. His6-T7-Pth2 (1 μg) was mixed with 1 μg of GST-Dsk2, GST-Rad23, GST-Ddi1 or GST alone in lysis buffer and incubated with 20 μl glutathione-Sepharose (GSH beads) for 1 h. Material bound to GSH beads was immunoblotted with the indicated antibodies. (C) Binding of Pth2 to Dsk2 and Rad23 in vivo. T7-Pth2 was coexpressed with GST-Dsk2, GST-Rad23, GST-Ddi1 or GST alone in wild-type yeast or in dsk2Δrad23Δ. The extracts (lanes 1–6) and the GST precipitates from wild-type (YPH499) (lanes 7–12) and dsk2Δrad23Δ (lanes 13–16) were immunoblotted as indicated. (D) Interaction of Pth2 with Dsk2 at endogenous expression level in yeast. FLAG-Pth2 was expressed from PTH2 own promoter in pth2Δ, and was immunoprecipitated with anti-FLAG M2 beads. The extracts (top and middle panels) and the FLAG-precipitates from the extracts (bottom panel) were immunoblotted with anti-Dsk2 or anti-Pth2 antibodies as indicated. An asterisk indicates FLAG-Pth2. Because the sensitivity of our anti-Rad23 is lower than that of anti-Dsk2 (ca. 1/10, see Materials and methods), we could not show a Rad23 signal under this condition. Also, we could not test co-immunoprecipitation of Dsk2/Rad23 with anti-Pth2 antibody, probably because our Pth2 antibody was not effective for immunoprecipitations. Download figure Download PowerPoint Pth2 interacts with polyubiquitinated proteins indirectly via Dsk2 and Rad23 in vivo To further investigate the interaction of Pth2 with UBL-UBA proteins, yeast cell extracts were fractionated by gel filtration, and each fraction was immunoblotted (Figure 2A). Pth2 reproducibly eluted in fractions containing high molecular weight proteins (bottom panel) but not at the expected monomer size (∼25 kDa). The Pth2-containing fractions contained Dsk2, Rad23 and polyubiquitinated proteins in cells. Also, GST-Pth2 expressed in yeast co-precipitated with polyubiquitinated proteins (Figure 2B, lane 7) together with Dsk2 and Rad23 in wild-type cells but did not co-precipitate with polyubiquitinated proteins in dsk2Δrad23Δ (lane 10). Like the wild-type strain, GST-Pth2 co-precipitated polyubiquitin in either dsk2Δ (lane 8) or rad23Δ cells (lane 9). It thus seems that the Pth2–polyubiquitin interaction seen in wild-type cells depends on Dsk2 and Rad23. Pth2 alone did not bind directly to polyubiquitin in vitro (Figure 2C). Figure 2.Pth2 associates with polyubiquitinated proteins via Dsk2 and Rad23. (A) Gel filtration of yeast extracts on a Superdex 200 column. Each fraction was immunoblotted with the indicated antibodies. The bar at the bottom indicates the position of monomeric Pth2 (∼25 kDa) (Rosas-Sandoval et al, 2002). (B) Binding of Pth2, Dsk2, Rad23 and polyubiquitin conjugates. GST-Pth2 was expressed in wild-type (YPH499), dsk2Δ, rad23Δ or dsk2Δrad23Δ strains. The GSH beads precipitates were immunoblotted as indicated. (C) In vitro GST pull-down assays with Pth2 and polyubiquitin. In total, 1 μg of tetraubiquitin (lanes 1–5) or polyubiquitin (lanes 6–10) was incubated with GST-Pth2, GST-Dsk2, GST-Rad23, GST-Ddi1 or GST alone (1 μg each). The samples were precipitated with GSH beads and immunoblotted as indicated. (D) In vivo GST pull-down assays with Pth2 and polyubiquitin. GST-Pth2, GST-Dsk2, GST-Rad23, GST-Ddi1 or GST alone was expressed in wild-type cells. Cell extracts were precipitated with GSH beads and immunoblotted as indicated. (E) Growth of yeast overexpressing Pth2. A single copy of PTH2 was expressed from a galactose-inducible pGAL1-YEplac181 ( × 1), or two copies of PTH2 were expressed from both pGAL1-YEplac181 and pGAL1-YEplac195 ( × 2) in wild-type, dsk2Δ, rad23Δ or dsk2Δrad23Δ strains. Cells were incubated for 2–5 days at 30°C, the permissive-temperature for growth of dsk2Δrad23Δ (Biggins et al, 1996). Expression levels were determined by Pth2 immunoblotting (on the right). Download figure Download PowerPoint Overexpression of Pth2 caused accumulation of polyubiquitinated proteins (Figure 2D, lane 5; and B, lane 2); therefore, we also tested the growth of cells overexpressing Pth2 (Figure 2E). Growth was inhibited in wild-type cells overexpressing Pth2 (top left panel), whereas Pth2-induced growth inhibition was not observed in the double-deletion mutant dsk2Δrad23Δ (bottom left panel). However, growth of the single-deletion mutants, dsk2Δ and rad23Δ, was inhibited when PTH2 was overexpressed (right top and bottom panels). These genetic analyses suggest that the Pth2-induced growth inhibition we observed is also mediated by Dsk2 and Rad23. Pth2 inhibits ubiquitin-mediated degradation We next investigated whether Pth2 affects ubiquitin-mediated protein degradation. Degradation of the N-end rule substrate Leu-β-gal was accelerated in pth2Δ (Figure 3A) compared with wild-type cells. As a control, degradation of Ala-β-gal was not affected by PTH2 deletion. The in vivo substrate Sic1, a cyclin-dependent kinase inhibitor, was also examined (Figure 3B; Supplementary Figure 1). Sic1-HA was expressed in cells arrested by α-factor and released synchronously into the cell cycle. Degradation of Sic1-HA was slightly accelerated in pth2Δ (Figure 3B, left panels). To confirm this Pth2 effect, we also tested Sic1 degradation in cells overexpressing PTH2. Supporting the acceleration by the PTH2 deletion, overexpression of PTH2 retarded degradation of Sic1-HA (right panels). Consistent with these effects of Pth2 on degradation, PTH2 disruption suppressed the growth defect of the temperature-sensitive proteasome mutant pre9Δ (Figure 3C). Furthermore, the cells’ response to amino-acid analogs was affected by PTH2 disruption (Figure 3D); that is, pth2Δ became resistant to the amino-acid analogs azetidine and canavanine. Contrary to our expectations, however, the triple mutant dsk2Δrad23Δpth2Δ was more sensitive to azetidine than dsk2Δrad23Δ, suggesting that Pth2 functions in a distinct pathway in the ubiquitin–proteasome system (see Discussion). In contrast to the drug response results, pth2Δ did not affect survival after UV irradiation (Figure 3D). Collectively, these data indicate that Pth2 is involved negatively in the ubiquitin-mediated degradation pathway. Figure 3.Pth2 inhibits the ubiquitin–proteasome pathway. (A) Degradation of the N-end rule substrate was accelerated in pth2Δ. Leu-β-gal and Ala-β-gal (indicated by arrowheads) were induced with galactose in wild-type (YPH499) or pth2Δ cells. Cell samples were collected at the indicated times after the induction was terminated, and degradation was assessed by immunoblotting with anti-β-gal. The quantification of the experiments is shown on the right. Note that, despite our expectations, Pth2 overexpression barely inhibited degradation of Leu-β-gal (data not shown), which might be indicative that Pth2 is involved in substrate specificity for degradation (cf Figure 3B). (B) Degradation of Sic1 in pth2Δ and PTH2-overexpressing cells. Wild-type (YPH499) and pth2Δ cells were arrested by α-factor treatment and then Sic1-HA was expressed for 1 h by galactose induction. For overexpression, Sic1-HA and GST-Pth2 were coexpressed in synchronized wild-type cells for 2 h by galactose induction. After release into the cell cycle, cell samples were collected at the indicated times and Sic1-HA protein was followed by immunoblotting with anti-HA. (C) Rescue of the proteasome defect of pre9Δ by PTH2 deletion. Growth was tested for 2–3 days at 30°C, at which temperature the pre9Δ mutant shows impaired growth. (D) Effects of PTH2 deletion on sensitivity to amino-acid analogs and UV irradiation. Wild-type (YPH499) and pth2Δ were spotted onto SD plates containing 0.5 mM AZC or 3 μg/ml canavanine in serial dilutions (upper panels). Wild-type and the mutants (dsk2Δrad23Δ and dsk2Δrad23Δpth2Δ) were spotted onto YPD plates containing 5 mM AZC in serial dilutions (bottom left panel). Cells were also spotted on YPD plates and exposed to 50 J/m2 UV irradiation (bottom right panel). The plates were incubated at 30°C for 2–4 days. Download figure Download PowerPoint Pth2 inhibits interactions of Rad23/Dsk2 with polyubiquitin receptors on the proteasome To further delineate Pth2's role in delivery of ubiquitinated proteins to the proteasome, we examined the effects of Pth2 on the interaction between the proteasome ubiquitin receptors Rpn1 and Rpn10 and UBL-UBA proteins Rad23 and Dsk2. First, we confirmed that Rpn1 interacts with Rad23 (Figure 4A, lane 6) but not with Dsk2 (lane 5) in vitro (cf; Elsasser et al, 2002). Then, to test whether Pth2 alters the Rad23–Rpn1 interaction, GST pull-down experiments were carried out using purified GST-Rpn1 and His6-T7-Rad23 in the presence of His6-T7-Pth2. Figure 4B shows that increasing amounts of Pth2 decreased binding of Rad23 to Rpn1 in vitro (lanes 4–6). Similarly, we confirmed that Rpn10 interacts with Dsk2 (Figure 4A, lane 12) but not with Rad23 (lane 13) in vitro. Again, increasing amounts of Pth2 competed with Dsk2 for Rpn10 binding (Figure 4C). Figure 4.Pth2 inhibits the interaction between UBL-UBA proteins and ubiquitin receptors. (A) Rpn1–Rad23 and Rpn10–Dsk2 interactions in vitro. GST-Dsk2, GST-Rad23, GST-Ddi1 or GST alone (1 μg each) was incubated with 1 μg of either His6-T7-Rpn1 (lanes 1–7) or His6-T7-Rpn10 (lanes 8–14) for 1 h, followed by precipitation with GSH beads and immunoblotting as indicated. Inputs represent 5% of the amounts used for the assay. (B) Effect of Pth2 on the Rad23–Rpn1 interaction in vitro. GST-Rpn1 (1 μg) was incubated with His6-T7-Rad23 (1 μg) in the presence of increasing amounts of His6-T7-Pth2 ( × 1=1 μg) for 1 h, followed by precipitation with GSH beads and immunoblotting as indicated. (C) Effect of Pth2 on the Dsk2–Rpn10 interaction in vitro. GST-Rpn10 (1 μg) was incubated with Dsk2-His6 (1 μg) in the presence of increasing amounts of His6-T7-Pth2 ( × 1=1 μg) for 1 h, followed by precipitation with GSH beads and immunoblotting as indicated. Download figure Download PowerPoint Next, we tested the effect of Pth2 on the interaction between the proteasome and UBL-UBA proteins (Figure 5A). To address this issue, an intact proteasome 19S regulatory particle containing FLAG-Rpt1 and polyubiquitin receptors Rpn1 and Rpn10 (data not shown, Elsasser et al, 2002) was immunoprecipitated from yeast cells with anti-FLAG beads and then incubated with the UBL domain fragment (1–77) of Rad23-GST or Dsk2-GST in the presence of His6-T7-Pth2. We used the 19S particle instead of the 26S proteasome because the FLAG-Pre1-tagged 26S proteasome was prone to dissociate in our buffer conditions in the GST pull-down assay (data not shown). Figure 5A shows that increasing amounts of Pth2 decreased the binding of 19S regulatory particle to Dsk2 (lanes 3–5) and to Rad23 (lanes 6–8), indicating that Pth2 inhibits the interaction of Dsk2 and Rad23 with the 19S particle in vitro. In the case of Rad23, a higher dose of Pth2 was required to compete with the proteasome–Rad23 interaction in vitro (lanes 6–8) (Supplementary Figure 2). Figure 5.Pth2 inhibits the interaction between UBL-UBA proteins and the proteasome. (A) Effect of Pth2 on the interaction between the proteasome and Dsk2/Rad23 in vitro. 19S regulatory particle was affinity-purified with FLAG-Rpt1 from yeast cells in which endogenous RPT1 had been replaced with FLAG-RPT1. Rad23-GST UBL domain (1–77) protein or Dsk2-GST UBL domain (1–77) protein (1 μg) was mixed with the 19S proteasome (19S RP) and incubated with GSH beads for 1 h in the presence of increasing amounts of His6-T7-Pth2 ( × 1=1 μg) followed by immunoblotting of the precipitates with anti-Rpt3, anti-FLAG, anti-T7 and anti-GST. (B) Effect of Pth2 on the interaction between the proteasome and Dsk2/Rad23 in yeast. GST or GST-Pth2 was overexpressed (O/E) by the galactose-inducible pGAL1-YEplac112-GST or pGAL1-YEplac112-GST-PTH2 for 4 h in yeast in which endogenous RPT1 was replaced with FLAG-His6-tagged RPT1. Extracts were precipitated with anti-FLAG M2 beads, followed by immunoblotting with anti-Rpt3, anti-GST, anti-FLAG, anti-Dsk2 or anti-Rad23. Extracts were immunoblotted as controls (lanes 1–2). (C) Effect of Pth2 on the interaction between the proteasome and Rad23 in yeast extracts. Lanes 1–4: GST or GST-Pth2 was overexpressed in yeast cells as in (B). Cell extracts were precipitated in the presence of purified GST-Pth2 (lane 4, 15-fold in molar excess; asterisk) with anti-FLAG M2 beads, followed by immunoblotting with anti-Rpt3, anti-GST, anti-FLAG and anti-Rad23 (lanes 1–4). Extracts were immunoblotted as a control (lanes 1–2). Lanes 5–8: experimental conditions were the same as (A), except that cell extracts were used instead of a purified 19S proteasome. Purified His6-T7-Pth2 (30-fold excess) was added to the cell extract (lane 8). (D) The proteasome does not contain Pth2. Endogenous PRE1 was replaced with FLAG-His6-tagged PRE1. Pth2 was overexpressed either in wild-type (YPH499) or the FLAG-His6-tagged PRE1-integrated strain (lanes 2 and 4, respectively). Extracts were precipitated with anti-FLAG M2 beads, followed by immunoblotting with anti-Rpt1, anti-20S, anti-FLAG and anti-Pth2 (lanes 5–8). Extracts were immunoblotted as a control (lanes 1–4). Download figure Download PowerPoint We also examined whether Pth2 prevents the interaction of Rad23 and Dsk2 with the proteasome in vivo. As shown in Figure 5B, overexpression of Pth2 in yeast decreased the amount of Dsk2 associated with the proteasome (lanes 3 and 4). Under the same conditions, we could not detect a decreased amount of Rad23 associated with the proteasome, consistent with the results in Figure 5A (lanes 6–8). Therefore, we examined the effect of greater amounts of Pth2 on the proteasome–Rad23 interaction. Figure 5C shows that a higher dose of Pth2 inhibited the interaction of Rad23 with the proteasome in the cell extracts (lanes 3 and 4). This result was confirmed by a GST pull-down assay of the proteasome from extracts with Rad23-GST, followed by immunoblotting for the proteasomal subunits, FLAG-tagged Rpt1 and Rpt3 (lanes 7 and 8). Pth2 itself did not associate with the proteasome; no Pth2 signal was detected in the 26S proteasome-precipitated fraction, even in cells overexpressing Pth2 (Figure 5D). Taking these results together, it appears that Pth2 competitively inhibits the interaction between UBL-UBA proteins and polyubiquitin receptors on the proteasome. Pth2 binds directly to the UBL domains of Rad23 and Dsk2 We tested which domain of UBL-UBA proteins is required for binding to full-length Pth2. A series of deletion mutants of Rad23 and Dsk2 were constructed (Figure 6A), and their binding to Pth2 was tested in vitro and in vivo. Figure 6B shows that the N-terminal UBL domain of Rad23 (lane 5) bound to Pth2 in vitro, but the N-terminal Rad23 truncation (lane 6) did not. The requirement of the UBL domain of Dsk2 for Pth2 binding was also shown by deletion analysis of Dsk2 in vitro (Figure 6B, lanes 12–16). Thus, Pth2 binds directly to the UBL domains of Rad23 and Dsk2. Figure 6.Pth2 binds to the UBL domains of Rad23 and Dsk2. (A) A diagram of Rad23 and Dsk2 constructs used for in vitro and in vivo binding assays. Summary of Pth2 binding to the mutants are shown on the right. Asterisks indicate an indirect interaction detected in cells (see text). (B) Binding of Pth2 to Rad23 and Dsk2 mutants in vitro. His6-T7-Pth2 (1 μg) was incubated with various GST-Rad23 mutants (lanes 1–6) or Dsk2 mutants (lanes 7–16) for 1 h, precipitated with GSH beads, and immunoblotted with the indicated antibody. Because Dsk2 contains only a single C-terminal UBA domain, we constructed and used a series of Dsk2 mutants for domain analysis. (C) Binding of Pth2 to Rad23 and Dsk2 mutants in vivo. T7-Pth2 was coexpressed with GST-Rad23 (lanes 1–6) or GST-Dsk2 (lanes 7–15) mutants in wild-type (YPH499) cells, and the cell extracts were precipitated with GSH beads followed by immunoblotting with the indicated antibodies. (D) Binding of Pth2 to Rad23 and Dsk2 mutants in vivo in dsk2Δrad23Δ. The in vivo assay was carried out as in (C), using cell extracts of dsk2Δrad23Δ. Download figure Download PowerPoint Pth2 bound in vivo to the UBL domain of Rad23 (Figure 6C, lane 5) and Dsk2 (lane 11); Pth2 also bound weakly to the N-terminal truncation mutants of Rad23 and Dsk2 (lanes 6, 13, 15). However, when a double deletion dsk2Δrad23Δ mutant was tested in the in vivo binding assay (Figure 6D), the Rad23 mutant fragment (78–398) did not interact with Pth2 (lane 4), indicating that the weak signal is due to an indirect interaction that polyubiquitin chains bridge between the GST-UBA domain fusion proteins and endogenous Rad23 or Dsk2, which then binds to Pth2. Notably, Pth2 also interacted with the Dsk2 UBA domain in vivo but not with Rad23 UBA domain (Figure 6C and D). Pth2 appeared to bind both the Dsk2 UBL and the UBA domains at equivalent levels (Figure 6C, lanes 11 and 13). Even in dsk2Δrad23Δ (Figure 6D), Pth2 co-precipitated with the UBA domain alone (lane 9) in vivo. By contrast, a Dsk2 mutant fragment (78–335) that lacks both the UBL and UBA domains did not bind to Pth2 (lane 8). Therefore, in addition to the UBL domain, Pth2 interacts with the Dsk2 UBA domain in vivo. However, this Pth2–UBA interaction seems to be mediated by an additional cellular factor(s) in yeast cells because Pth2 did not bind the UBA domain in vitro (Figure 6B) (see Discussion). Pth2 function in the ubiquitin–proteasome pathway is independent of its C-terminal peptidyl-tRNA hydrolase activity The Pth2 C-terminal domain, which contains the peptidyl-tRNA hydrolase activity, is conserved across species; however, the N-terminal domain is not conserved, and its function is unknown (Rosas-Sandoval et al, 2002). We investigated whether the peptidyl-tRNA hydrolase activity of Pth2 is required for its binding to UBL-UBA proteins. Guided by a structural analysis of the catalytic domain (de Pereda et al, 2004), we produced a series of Pth2 peptidyl-tRNA hydrolase mutants by site-directed mutagenesis of the putative active center (Figure 7A). The peptidyl-tRNA hydrolase activity of these mutants was tested by complementation of the Escherichia coli mutant, pthts (Rosas-Sandoval et al, 2002). The yeast Pth2 mutants D171A, D171N, DRTQ171AAAA, DRTQ171NLVL and RTQ174LVL failed to rescue the temperature-sensitive growth of pthts (summarized in Figure 7A), indicating a lack of peptidyl-tRNA hydrolase activity in the Pth2 mutants. When we examined the ability of these Pth2 mutants to bind to Rad23 (Figure 7B, lanes 1–9) and Dsk2 (lanes 10–18), we found that these Pth2 mutants could bind to Rad23 (lanes 5–9) and to Dsk2 (lanes 14–18). Overexpression of mutant Pth2 (D171A) inhibited growth to a similar degree as wild-type Pth2 (Figure 7C). We also confirmed this result by complementation test for cell sensitivity to amino-acid analog: pth2Δ expressing Pth2 mutant (D171A) became sensitive to azetidine compared with pth2Δ alone (Figure 7D, see also Figure 3D). These data, therefore, demonstrated that Pth2 function in the ubiquitin-mediated pathway is independent of its peptidyl-tRNA hydrolase activity. In addition, this complementation assay also shows that FLAG-tagged Pth2 expressing from PTH2's own promoter in pth2Δ made more sensitive to azetidine than pth2Δ, supporting that the tagged Pth2 is functional in yeast cells (Figure 7D). Like FLAG-Pth2, GST-Pth2 expressing in pth2Δ at endogenous level rescued the azetidine-resistant phenotype of pth2Δ (data not shown). As shown in Figure 7E, the C-terminal half (83–208) of Pth2 bound to Rad23 (lane 12) and Dsk2 (lane 24) in vivo; however, the N-terminal domain (1–82) did not bind these proteins (lanes 11 and 23). Furthermore, overexpression of the N-terminal domain (1–82), which lacks the Rad23/Dsk2-binding site, did not inhibit the growth of yeast, unlike wild-type Pth2 (Supplementary Figure 3). Therefore, the C-terminal domain of Pth2 is necessary for binding to Rad23 and Dsk2. Figure 7.The function of Pth2 in the ubiquitin–proteasome pathway is independent of its peptidyl-tRNA hydrolase activity. (A) Pth2 mutants and the sequences of the peptidyl-tRNA hydrolase active site. Conserved residues of the active site (D171, R174, T175 and Q176) were changed by site-directed mutagenesis as indicated in italics. A summary of the hydrolase activity and Rad23/Dsk2 binding of these mutants is shown on the right. (B) Binding of Pth2 hydrolase mutants to Rad23 and Dsk2. Each T7-tagged Pth2 mutant was coexpressed with GST-Rad23 (lanes 1–9) or with GST-Dsk2 (lanes 10–18) in yeast, and cell extracts were pre
Referência(s)