HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ota, K. Articles by Ito, T. Search for Related Content PubMed Articles by Ota, K. Articles by Ito, T.

Department of Computational Biology and Center for Omics and Bioinformatics, Graduate School of Frontier Sciences, University of Tokyo, Kashiwa 277-8561, Japan

	Abstract

Top Abstract Introduction Results Discussion Experimental procedures References

 
Ubiquitination plays various critical roles in eukaryotic cellular regulation and is medated by a cascade of enzymes including ubiquitin protein ligase (E3). The Skp1–Cullin–F-box protein complex comprises the largest E3 family, in each member of which a unique F-box protein binds its targets to define substrate specificity. Although genome sequencing uncovers a growing number of F-box proteins, most of them have remained as "orphans" because of the difficulties in identification of their substrates. To address this issue, we tested a quantitative proteomic approach by combining the stable isotope labeling by amino acids in cell culture (SILAC), parallel affinity purification (PAP) that we had developed for efficient enrichment of ubiquitinated proteins, and mass spectrometry (MS). We applied this SILAC-PAP-MS approach to compare ubiquitinated proteins between yeast cells with and without over-expressed Mdm30p, an F-box protein implicated in mitochondrial morphology. Consequently, we identified the mitochondrial outer membrane protein Mdm34p as a target of Mdm30p. Furthermore, we found that mitochondrial defects induced by deletion of MDM30 are not only recapitulated by a mutant Mdm34p defective in interaction with Mdm30p but alleviated by ubiquitination-mimicking forms of Mdm34p. These results indicate that Mdm34p is a physiologically important target of Mdm30p.

	Introduction

Top Abstract Introduction Results Discussion Experimental procedures References

 
Ubiquitination is a common posttranslational modification in eukaryotes and known to affect many fundamental cellular processes, including cell cycle progression, signal transduction, DNA repair, membrane trafficking, and so forth (Pickart 2001b; Aguilar & Wendland 2003). Attachment of ubiquitin to a target protein is mediated by a cascade of enzymes composed of ubiquitin-activating enzyme (E1), ubiquitin-conjugating enzyme (E2), and ubiquitin protein ligase (E3) (Hershko & Ciechanover 1998). Although many cascades share E1 and E2 in common, each cascade defines its substrate specificity by using a unique E3 that physically interacts with target proteins (Pickart 2001a).

Among the known E3s, the Skp1–Cullin–F-box protein (SCF) complex comprises the largest and most well-characterized family. Although Skp1 and Cullin are the constant components shared among the complexes, F-box protein is the variable one unique to each complex to determine its substrate specificity (Patton et al. 1998). An F-box protein bears a conserved F-box motif and other protein-binding domains to interact with Skp1 and its target proteins, respectively, thereby recruiting specific substrates to the ubiquitination machinery (Patton et al. 1998). Although a growing number of F-box proteins have been discovered by genome sequencing, the vast majority of them have not been associated with any substrates and hence remained as "orphans" (Gagne 2002; Willems et al. 2004). Even in the budding yeast Saccharomyces cerevisiae, one of the most leading and tractable model organisms for ubiquitin research, most of their 21 F-box proteins have remained as orphans and their systematic investigation is still challenging. This is mainly due to the lack of universally effective strategy for comprehensive identification of substrates for E3s.

Recent progress in mass spectrometry (MS)-based proteomics has enabled cataloguing of proteins subject to various post-translational modifications, including phosphorylation (Ficarro et al. 2002), glycosylation (Zhang et al. 2003), and ubiquitination (Peng et al. 2003). Accordingly, it is, in principle, possible to identify substrates for an E3 through a quantitative comparison of ubiquitinated proteins between the cells with differential activities of the E3. It is obvious that such comparison requires a robust method for differential isotope labeling. A promising method would be the one termed stable isotope labeling by amino acids in cell culture (SILAC), in which cells to be compared are metabolically labeled with "light" and "heavy" isotopomers of an essential amino acid so that the two proteomes can be distinguished by means of MS (Ong et al. 2002). Because cognate peptides in the "light" and "heavy" proteomes share identical chemical characteristics except for their masses, one can combine the harvested cells to be compared before protein extraction, thereby eliminating the effects of any variability inherent to independent sample preparations to improve reliability of quantification.

Here we propose a strategy to identify substrates for a given E3 based on SILAC and a method termed parallel affinity purification (PAP), which we had developed for efficient enrichment of polyubiquitinated proteins (Ota et al. 2008), followed by MS. We applied this SILAC-PAP-MS approach to compare ubiquitinated proteins between yeast cells with and without over-expressed Mdm30p, an F-box protein implicated in mitochondrial morphology (Dimmer et al. 2002; Fritz et al. 2003; Escobar-Henriques et al. 2006; Cohen et al. 2008), to identify the mitochondrial outer membrane protein Mdm34p (Youngman et al. 2004) as a target of Mdm30p. Subsequent analyses demonstrated that Mdm30p-mediated ubiquitination of Mdm34p is critical to mitochondrial integrity.

	Results

Top Abstract Introduction Results Discussion Experimental procedures References

 
SILAC-PAP-MS identifies Mdm34p as a target of Mdm30p

To identify potential targets of Mdm30p, we screened for ubiquitinated proteins whose amounts are affected by over-expression of Mdm30p using SILAC coupled with PAP, which we had developed for selective isolation of poly-ubiquitinated proteins (Ota et al. 2008), followed by MS. In the PAP, two forms of ubiquitin, each carrying a unique affinity tag (i.e. His₈ or Flag), are concurrently expressed in the same cell, and proteins are purified by sequential use of affinity chromatography specific to each tag. The first step is Ni-NTA-agarose chromatography in the presence of high concentration of urea, which enriches proteins containing His₈-tagged ubiquitin (His₈-Ubi) and eliminates nonubiquitinated proteins, mono-ubiquitinated proteins bearing Flag-Ubi, and free Flag-tagged ubiquitin (Flag-Ubi). The second step is immunoaffinity chromatography for Flag-epitope, which enriches proteins containing Flag-Ubi and eliminates free His₈-Ubi and mono-ubiquitinated proteins bearing His₈-Ubi, both comprising a substantial fraction of the Ni-NTA-purified fraction. Thus, this sequential procedure should retain, in principle, only protein molecules bearing both His₈-Ubi and Flag-Ubi, or poly- and multi-ubiquitinated proteins. Indeed, we demonstrated that PAP can effectively enrich poly-ubiquitinated proteins (Ota et al. 2008). Therefore, a combination of SILAC and PAP followed by MS (SILAC-PAP-MS) is expected to help one quantify relative change of ubiquitinated proteome between the cells that differ in the activity of an E3 to identify candidates for its substrates (Fig. 1A).

View larger version (15K): [in this window] [in a new window]   Figure 1  SILAC-PAP-MS identifies Mdm34p as a potential target of Mdm30p. (A) A SILAC-PAP-MS strategy to search potential targets of an F-box protein. The cells over-expressing an F-box protein, His8-tagged ubiquitin, and Flag-tagged ubiquitin were grown in SC media containing d7-Leu. On the other hand, the control cells with the tagged ubiquitins were grown in SC media containing d0-Leu. Equal amounts of these cells were combined and subjected to protein extraction. Ubiquitinated proteins were selectively enriched from the extract using the PAP procedure (Ota et al. 2008). The obtained proteins were digested with proteases and analyzed with LC-MS/MS. The ion chromatograms were extracted from mass spectra and the ratios of "light" and "heavy" peptides were calculated from each peak area to identify differentially ubiquitinated proteins. (B) Quantification of an Mdm34p-derived peptide. Ion chromatograms were extracted for light and heavy isotopomers of an Mdm34p-derived peptide DVLPSLIFNTSQNWFTNR. The results of two independent experiments are shown in the upper and lower panels. Each peak area was determined and calculated for the peptide quantification using Xcalibur software (ThermoElectron) as described previously (Kito et al. 2007). The experiments 1 and 2 showed 3.03- and 2.08-fold increase in the amount of the peptide DVLPSLIFNTSQNWFTNR, respectively.

These analyses reproducibly detected a significant increase in the amount of a peptide derived from Mdm34p, a mitochondrial outer membrane protein (Youngman et al. 2004) (Fig. 1B; Table S1). To confirm enhanced ubiquitination of Mdm34p by Mdm30p, we co-expressed T7-tagged Mdm34p and His₈-Ubi in the presence or absence of over-expressed Mdm30p, enriched ubiquitinated proteins using Ni-NTA agarose chromatography, and probed the obtained fractions with an anti-T7 antibody. We detected a significant increase in the amount of ubiquitinated, but not unmodified, Mdm34p on over-expression of Mdm30p (Fig. 2A). The increase was more prominent than the one observed in the SILAC-PAP-MS, presumably because we over-expressed not only Mdm30p but also Mdm34p in this experiment. By contrast, Gly1p, an enzyme involved in glycine biosynthesis that is known to be ubiquitinated but has no reported relation with Mdm30p (Liu et al. 1997; Peng et al. 2003), failed to show any enhancement in ubiquitination, consistent with the SILAC-PAP-MS data (Fig. 2A; Table S1).

View larger version (23K): [in this window] [in a new window]   Figure 2  Mdm30p affects ubiquitination of Mdm34p. (A) Validation of enhanced ubiquitination of Mdm34p by over-expressed Mdm30p. Mdm34p (black arrow) or Gly1p (white arrow) was tagged with T7-epitope and expressed by ADH1 promoter on pALeu in YPH499 cells. His8-Ubi was expressed by GAL10 promoter on pESC-URA in conjunction without or with Mdm30p expressed by GAL1 promoter on the same plasmid. An aliquot of whole cell extract (WCE) and eluate from Ni-NTA agarose chromatography (Ni-NTA) from each strain was analyzed by immunoblotting with an anti-T7 antibody. Mdm30p O/E; over-expressed Mdm30p. (B) Effect of MDM30 deletion on ubiquitination of endogenous Mdm34p. The BY4741 strain with or without MDM30 was modified so that its MDM34 encodes a protein C-terminally tagged with T7-epitope. These strains were transformed with pESC-URA to express His8-Ubi but not Mdm30p. Mitochondria-enriched fractions (Mito) from these cells were subjected to Ni-NTA agarose chromatography to obtain a fraction enriched for ubiquitinated proteins (Ni-NTA). Both Mito and Ni-NTA fractions were subjected to an immunoblot analysis with the anti-T7 antibody (upper panel). The same blot was re-probed with an anti-porin antibody (lower panel). Filled and open arrows indicate the positions of unmodified Mdm34p and porin, respectively.

Because Mdm30p was demonstrated to function as a component of SCF^Mdm30 to ubiquitinate Fzo1 (Cohen et al. 2008), it is likely that SCF^Mdm30 also ubiquitinates Mdm34p. Notably, it would not be the sole E3 for Mdm34p, because ubiquitination of Mdm34p is not completely eliminated in mdm30 cells (Fig. 2B). However, Mdm30p is critically involved in physiologically important ubiquitination of Mdm34p as shown below.

Mdm30p binds a region in the C-terminal portion of Mdm34p for ubiquitination

To learn whether Mdm34p is a direct target of Mdm30p, we examined a physical interaction between Mdm34p and Mdm30p using a sensitized yeast two-hybrid system (see Experimental procedures). We detected a weak but significant two-hybrid interaction between the two proteins (data not shown). Previous studies demonstrated that Mdm34p is anchored to mitochondrial outer membrane through its N-terminal region and leaves its C-terminal portion exposed to the cytoplasm (Youngman et al. 2004). Indeed, the C-terminal half of Mdm34p (a.a. 219–459) displayed a stronger interaction with Mdm30p than its full-length form (a.a. 1–459) did (data not shown). Thus, we constructed a series of Mdm34p mutants with various deletions in the C-terminal half and examined them for binding to Mdm30p. Consequently, it turned out that a C-terminal tail region (a.a. 219–365) is sufficient to mediate the interaction (Fig. 3A). Furthermore, we found that deletion of a region (a.a. 289–336) makes the C-terminal half incapable of binding to Mdm30p (Fig. 3A). Thus, we designated this indispensable region as ID.

View larger version (26K): [in this window] [in a new window]   Figure 3  A physical interaction between Mdm30p and Mdm34p is critical to mitochondrial function and morphology. (A) Two-hybrid interaction between Mdm30p and Mdm34p. Various C-terminal fragments of Mdm34p (left panel) were fused with a tandem dimer of Gal4p activation domain and examined for two-hybrid interaction with Mdm30p fused with Gal4p DNA-binding domain using HIS3 and ADE2 reporter genes. Five-fold serial dilutions of each transformant were spotted onto SC–Trp/Leu and SC–Trp/Leu/His/Ade plates and incubated for 3 and 8 days, respectively (right panel). (B) Mdm30p-mediated ubiquitination of Mdm34p is dependent on the interaction between Mdm30p and Mdm34p. We expressed Mdm12p or either of the two Mdm12p-Mdm34p chimeras (see below) as a T7-tagged form by ADH1 promoter on pALeu in YPH499 cells, in which His8-Ubi was co-expressed with or without Mdm30p from pESC-URA. An aliquot of whole cell extract (WCE) and eluate from Ni-NTA agarose chromatography (Ni-NTA) from each strain was analyzed by immunoblotting with an anti-T7 antibody. Wild; wild-type Mdm12p. +34C; Mdm12p fused with the C-terminal fragment of Mdm34p (a.a. 219–459), which is required for interaction with Mdm30p as shown in (A). +34CID; Mdm12p fused with the C-terminal fragment of Mdm34p lacking the ID region (a.a. 219–288, 337–459), which is incapable of interacting with Mdm30p as shown in (A). Mdm30p O/E; over-expressed Mdm30p. (C) Mdm30p–Mdm34p interaction is critical for mitochondrial respiratory function. Five-fold serial dilutions of wild-type cells (YPH499), mdm34 cells with episomally expressed Mdm34p and Mdm34p-ID, mdm34 cells, and mdm30 cells were spotted and grown on synthetic glucose and synthetic glycerol plates, and incubated for 3 and 12 days, respectively. (D) Mdm30p–Mdm34p interaction is critical for mitochondrial morphology. Fluorescence microscopic images are presented for the indicated strains. Each strain was grown to late log phase in SC medium, and its mitochondria were visualized with MitoFluor Red 589 staining and fluorescence microscopy.

We thus took an alternative approach that uses an Mdm12p-Mdm34p chimera protein. We attached the C-terminal tail region of Mdm34p (a.a. 219–365), which is sufficient to interact with Mdm30p (Fig. 3A), to the C-terminal end of Mdm12p, another mitochondrial outer membrane protein, and examined its ubiquitination in vivo. Although Mdm12p and Mdm34p belong to the same biological pathway (Youngman et al. 2004; Dimmer et al. 2005), the former was ubiquitinated much less efficiently than the latter was, even in the presence of over-expressed Mdm30p (Fig. 3B). By contrast, the Mdm12p-Mdm34p chimera was efficiently modified with ubiquitin and, more importantly, this modification was largely dependent on Mdm30p (Fig. 3B). Intriguingly, deletion of the ID region from the Mdm12p-Mdm34p chimera eliminated the Mdm30p-dependent strong ubiquitination, while leaving the basal level of modification unaffected (Fig. 3B).

Thus, Mdm30p likely interacts with a region in the C-terminal portion of Mdm34p, and this interaction is critically involved in the ubiquitination of Mdm34p. In this context, it is intriguing to note that Mdm30p may not direct ubiquitination of the C-terminal tail of Mdm34p unless it is recruited to mitochondria, Because Mdm30p-dependent ubiquitination was not observed when it is attached to green fluorescent protein (GFP) that localizes in the cytoplasm (data not shown).

Mdm30p–Mdm34p interaction is critical to mitochondrial integrity

To examine a role for the Mdm30p–Mdm34p interaction in respiratory function and morphology of mitochondria, we made mdm34 strains bearing a low-copy plasmid that expresses either Mdm34p or Mdm34p-ID from MDM34 promoter. Although we failed to detect these proteins due to their low abundance, we assumed it unlikely that Mdm34p-ID is intrinsically less stable than Mdm34p is, because both could be expressed to a comparable level from ADH1 promoter to share a similar half-life (Figure S1 in Supporting Information).

The budding yeast grows by respiration when carbon sources are limited to nonfermentable ones (e.g. glycerol), whereas it does by fermentation in the presence of fermentable ones (e.g. glucose). Accordingly, respiration-defective yeast shows a growth defect in glycerol media but not in glucose media. Thus, we compared growth of the two strains in glucose and glycerol media. In glucose medium, both grew well to show a comparable growth rate, although their parental mdm34 strain easily fell into rho^– status and had a severe growth defect (Fig. 3C). In glycerol medium, the two strains substantially differ in growth; the cells with Mdm34p-ID grew less efficiently than those with Mdm34p did (Fig. 3C).

We also examined these strains for their mitochondrial morphology. Expression of Mdm34p made mdm34 cells to harbor thick string-like mitochondria, which are similar to those observed in wild-type cells cultivated in conventional glucose liquid medium (Fig. 3D). By contrast, expression of the interaction-defective Mdm34p-ID made the cells to harbor short and collapsed mitochondria, which are close to those typically observed in mdm30 cells (Dimmer et al. 2002; Fritz et al. 2003; Escobar-Henriques et al. 2006; Cohen et al. 2008) (Fig. 3D).

Taken together, the yeast with Mdm34p defective in interaction with Mdm30p displayed growth defect on nonfermentable carbon source (Fig. 3C) and aberrant mitochondrial morphology (Fig. 3D). These results suggest that the interaction between Mdm30p and Mdm34p, which is required for efficient ubiquitination of the latter (Fig. 3B), is critical to mitochondrial respiratory function and morphology. Note that the cells lacking Mdm30p displayed severer defects than those with an impaired Mdm30p–Mdm34p interaction did (Fig. 3C,D), presumably because Mdm30p has additional mitochondrial targets, such as Fzo1p (Fritz et al. 2003; Escobar-Henriques et al. 2006; Cohen et al. 2008) (see below).

Mdm34 is a physiologically important target of Mdm30p

It is usual that an F-box protein has multiple targets involved in different biological processes (Patton et al. 1998). Indeed, Mdm30p has been reported to act on two proteins other than Mdm34p, namely Fzo1p (Fritz et al. 2003; Escobar-Henriques et al. 2006; Cohen et al. 2008), a regulator of mitochondrial fission and fusion, and Gal4p (Muratani et al. 2005), a transcription factor to activate genes involved in galactose metabolism, although peptides from these proteins escaped the detection with our LC-MS/MS system to make us incapable of quantifying them (see Discussion).

To examine whether Mdm34p is a physiologically important one among the targets of Mdm30p, we constructed Mdm34p variants that would mimic its ubiquitinated form and tested their ability to suppress mitochondrial defects of mdm30 cells. If Mdm34p is a functionally important target of Mdm30p, the variants may suppress the defects substantially. We prepared two different mimics of ubiquitinated Mdm34p by C-terminally attaching an ubiquitin moiety to the full-length protein (a.a. 1–459; Mdm34p-Ubi) or its N-terminal half (a.a. 1–288; Mdm34p-C-Ubi). Note that the latter lacks the cytoplasmic tail portion that mediates the interaction with Mdm30p (Fig. 3A). We examined whether these mimics can suppress growth defect on nonfermentable carbon sources and aberrant mitochondrial morphology of mdm30 cells.

Expression of Mdm34p, Mdm34p-Ubi, Mdm34-C-Ubi, or Mdm30p in mdm30 cells barely affected growth on fermentable carbon sources (i.e. glucose) (Fig. 4A). By contrast, their expression differentially affected growth on nonfermentable carbon sources (i.e. glycerol/ethanol). Although the wild-type Mdm34p failed to improve growth in glycerol/ethanol medium, both mimics alleviated the growth defect (Fig. 4A). Importantly, the alleviation appeared to be largely independent of the C-terminal half of Mdm34p that mediates the interaction with Mdm30p, because Mdm34p-Ubi and Mdm34p-C-Ubi had comparable effects. Thus, the ubiquitination is epistatic to the interaction: the interaction is likely required before and for ubiquitination of Mdm34p.

View larger version (24K): [in this window] [in a new window]   Figure 4  Ubiquitination-mimicking variants of Mdm34p alleviate mitochondrial defects of cells lacking Mdm30p. (A) Ubiquitination-mimicking variants of Mdm34p partially restore growth of mdm30 cells on nonfermentable carbon sources. To calculate the doubling time, mdm30 cells expressing the indicated proteins from MDM30 promoter on pRS415 were inoculated into synthetic glucose and glycerol/ethanol liquid media, and OD600 was measured at every 2 and 4 h, respectively. Mdm34p-Ubi; full-length Mdm34p (a.a. 1–459) C-terminally tagged with ubiquitin. Mdm34p-C-Ubi; the N-terminal half of Mdm34p (a.a. 1–288) C-terminally tagged with ubiquitin. (B) Ubiquitination-mimicking variants of Mdm34p partially restore mitochondrial morphology of mdm30 cells. Each strain grown to mid log phase in SC–Leu liquid medium was subjected to MitoFluor Red 589 staining for fluorescence microscopy. (C) Quantitative summary of microscopic inspections in (B). For each strain, more than 200 cells were inspected and classified into "wild type," "intermediate," or "dot" based on mitochondrial morphology.

Taken together, these results indicate that Mdm34p is a physiologically important target of Mdm30p.

	Discussion

Top Abstract Introduction Results Discussion Experimental procedures References

 
We developed and applied SILAC-PAP-MS to successfully identify Mdm34p as a novel target of Mdm30p (Figs 1, 2). Furthermore, we provide evidence that Mdm34p is a physiologically critical substrate of Mdm30p. First, similar mitochondrial defects were shared among mdm30 cells and those expressing the wild-type Mdm30p and a mutant Mdm34p defective in interaction with Mdm30p (Fig. 3). Second, mitochondrial defects of mdm30 cells are alleviated by expression of ubiquitination-mimicking variants of Mdm34p (Fig. 4).

Mdm34p was suggested to form a membrane protein scaffold or mitochore, which is implicated in maintenance of mitochondrial shape, attachment of mitochondria to the actin cytoskeleton, and proper segregation of mitochondrial DNA (Boldogh et al. 2003; Meisinger et al. 2007). Accordingly, mdm34 cells displayed severe phenotypes including round-shaped mitochondria, growth defect on nonfermentable carbon sources, and an increased rate of mitochondrial DNA loss. Because the precise mode of action is not well understood for Mdm34p, it remains largely elusive how its ubiquitination contributes to mitochondrial integrity. Nevertheless, we can assume, at least, that the ubiquitination is not simply involved in proteasomal degradation, because the level of unmodified Mdm34p was barely affected by over-expression or depletion of Mdm30p (Fig. 2). In consistent with these findings, MS analysis of affinity purified Mdm34p indicated that its ubiquitin chain is extended mainly via Lys-63, but not Lys-48 (unpublished data). Furthermore, we also found that the ubiquitin variant lacking all Lys residues except for Lys-63, but not the one retaining only Lys-48, is incorporated into Mdm34p as efficiently as the one carrying both Lys-48 and Lys-63 is (unpublished data). Further efforts are currently underway to reveal a precise role for Mdm30p-mediated ubiquitination of Mdm34p in mitochondrial integrity.

The results reported here would also serve as a proof-of-principle for the SILAC-PAP-MS approach to identify substrates for E3s. This approach can be flexibly applied to a variety of comparison among mutants in the ubiquitin system, such as those with constitutive over-expression of an F-box protein, transient over-expression of an F-box-deleted protein, and defective ubiquitination-related genes. Indeed, we have successfully applied SILAC-PAP-MS to identify substrates for deubiquitinating enzymes (in preparation). In addition, it can be also used for the search of proteins that are differentially ubiquitinated in response to various stimuli.

The coverage of current analysis was, however, not satisfactory; we failed to detect known substrates such as Fzo1p (Fritz et al. 2003; Escobar-Henriques et al. 2006; Cohen et al. 2008) and Gal4p (Muratani et al. 2005). Use of two-dimensional LC would extend the coverage (Ota et al. 2008). It is also desirable to include a step to effectively eliminate peptides derived from ubiquitin per se, which comprise a large fraction of the PAP sample. In addition, Because most substrates, including Fzo1p and Gal4p, are degraded on ubiquitination, inhibition of proteolysis would be necessary to enhance detection of such short-lived substrates. Otherwise, the screen may be skewed toward substrates that are not targeted for degradation or those bearing non-Lys-48 chains, such as Mdm34p (see above).

With these refinements, the SILAC-PAP-MS approach would contribute to identify further substrates for each E3, similarly to other approaches that were recently reported to use protein chip (Gupta et al. 2007), high-throughput quantitative microscopic screening of a GFP-tagged strain library (Benanti et al. 2007), and a modified two-hybrid system (Kishi et al. 2007). Cumulative efforts with these approaches would eventually uncover specificity and redundancy of substrate recognition in ubiquitination at a proteome-wide level, thereby considerably improving our understanding of this complex post-translational modification system essential to eukaryotic cellular life.

	Experimental procedures

Top Abstract Introduction Results Discussion Experimental procedures References

 
Yeast strains and their manipulations

We used YPH499 (MATa ade2-101 leu2-1 trp1-63 ura3-52 lys2-801 his3-200) in all experiments but the one shown in Fig. 2B, in which we used BY4741 (MATa his31 leu20 met150 ura30). Yeast cells were transformed using the standard lithium-acetate method (Gietz & Woods 2002). For strain modification, we adopted the PCR-based one-step gene disruption method (Güldener et al. 1996). We used the KanMX cassette (Güldener et al. 1996) as a selection marker in all experiments but the one shown in Fig. 2B, in which we used Candida glabrata URA3 instead. For the expression of tagged ubiquitin and/or other proteins, yeast cells were grown in synthetic complete (SC) medium (Adams et al. 1997) containing 2.0% raffinose at 30 °C to an OD₆₀₀ of 0.8–1.0 before the addition of galactose to a final concentration of 2.0%.

Plasmids

We used pESC-URA, a derivative of pESC-LEU vector (Stratagene; Genbank AF063849 [GenBank] ) that bears URA3 instead of LEU2, for transient expression of tagged ubiquitin and Mdm30p. Because we found that a tandem dimer of tagged-ubiquitin (i.e. Flag-ubiquitin-His₈-ubiquitin) is processed into tagged-ubiquitin monomers with high efficiency when expressed in yeast (unpublished observation), we cloned a DNA fragment encoding the dimer into the multiple cloning site 1 downstream of GAL10 promoter of pESC-URA. The multiple cloning site 2 downstream of GAL1 promoter of pESC-URA was either occupied by MDM30 ORF or kept empty. All Lys residues but Lys-48 and Lys-63 in the tagged ubiquitins were replaced by Arg, because this variant increased the yield of ubiquitnated proteins for unknown reason (unpublished observation). For constitutive expression of T7-tagged candidate substrates, we used a derivative of pAS2–1 (Clontech; Genbank U30497 [GenBank] ) termed pALeu, which bears LEU2 instead of TRP1 and lacks both CYH2 and GAL4. A DNA fragment encoding each T7-tagged candidate protein was cloned into the multiple cloning site between ADH1 promoter and ADH1 terminator of pALeu. For the experiments to complement mdm30 cells (Fig. 4), we used MDM30 promoter to express Mdm34p and its ubiquitination-mimicking variants from a low-copy vector pRS415 (GenBank U03449 [GenBank] ) to recapitulate temporal expression pattern of Mdm30p.

SILAC and PAP

Cells to be compared were grown in SC medium containing 100 mg/L of either "light" d0-Leu or "heavy" d7-Leu (Cambridge Isotope Laboratories). Equal amounts of the cells, as determined by OD₆₀₀, were mixed at a 1:1 ratio and harvested. The PAP was conducted as described previously (Ota et al. 2008). Briefly, ~5 g wet weight of the mixed cells was suspended in 15 mL of lysis buffer (8 M urea/0.1 mM NaH₂PO₄/10 mM Tris, pH 8.0) supplemented with a protease inhibitor mixture (Sigma), and lysed by vigorous shaking with glass beads. The obtained lysate was diluted by 10-fold with wash buffer A (8 M urea/0.1 mM NaH₂PO₄/1.0% Triton X-100/0.5% SDS/10 mM Tris, pH 8.0). Unbroken cells and debris were removed by centrifugation at 4500 x g for 15 min. The cleared supernatant was incubated with 1 mL of Ni-NTA agarose (Qiagen) overnight at room temperature. The resin was packed into a column and sequentially washed with 10 mL of wash buffer A, 10 mL of wash buffer B (8 M urea/500 mM NaCl/0.1 mM NaH₂PO₄/2.0% Triton X-100/10 mM Tris, pH 8.0), 10 mL of wash buffer C (6 M guanidine-HCl/0.1 mM NaH₂PO₄/0.5% Triton X-100/10 mM Tris, pH 6.3), and 10 mL of wash buffer D (300 mM NaCl/50 mM NaH₂PO₄/15 mM imidazole/2.0% Triton X-100/10 mM Tris, pH 8.0). Bound proteins were eluted with 1 mL of elution buffer (100 mM NaCl/50 mM NaH₂PO₄/500 mM imidazole/1.0% Triton X-100/10 mM Tris, pH 8.0). The pH of each buffer was adjusted to the indicated value using either HCl or NaOH. The eluate from the Ni-NTA agarose column was incubated with 0.1 mL of Anti-Flag M2 agarose (Sigma) for 1 h at 4 °C. The agarose beads were collected by brief centrifugation and washed 4 times with buffer F (phosphate-buffered saline [PBS] containing 10% glycerol and 1.0% Triton X-100) in the same way. Bound proteins were eluted from the beads with 0.2 mL buffer F containing 0.25 mg/mL Flag peptide (Sigma).

LC-MS/MS

Proteins obtained by the PAP procedure were precipitated with trichloroacetic acid, dissolved in 8 M urea and 0.1 M Tris-HCl (pH 8.5), and S-alkylated with iodoacetamide as described previously (Kito et al. 2007). After a 4-fold dilution with 0.1 M Tris-HCl (pH 8.5), the proteins were digested with sequence grade modified trypsin (Promega) at 37 °C for 15–18 h.

LC-MS/MS was carried out using DiNa nano-LC Systems (KYA technologies) coupled on-line to LCQ Deca XP Plus ion-trap mass spectrometer (Thermo Fisher Scientific). Reverse-phase LC was carried out using a capillary column (5 cm x 150 µm i.d.) packed with C₁₈ particles, and the peptides were eluted using 360-, 45-, and 45-min gradients from 0 to 32%, 32 to 40%, and 40 to 80% of acetonitrile in 0.1% formic acid, respectively, at a flow rate of 200 nL/min. The column eluate was sprayed directly into the electrospray ionization source of the mass spectrometer, and MS and MS/MS data were acquired in a data-dependent mode with 2-min dynamic exclusion.

The MS/MS spectra were subjected to database search against the yeast ORF database from Saccharomyces Genome Database (http://www.yeastgenome.org/) with SEQUEST program as previously described (Kito et al. 2007). The search results that met the previously reported criteria (Kito et al. 2007) were defined as positive identifications.

Relative abundance of each peptide was determined using ASAPRatio program (Li et al. 2003) (Supporting data Table S1). For peptides escaping the detection by ASAPRatio and/or those of particular interest, each ion chromatogram was manually extracted and peak area ratio between light and heavy peptides was determined using Xcalibur software (ThermoElectron) as previously described (Kito et al. 2007).

In vivo ubiquitination assay

We co-transformed YPH499 cells with a pALeu plasmid that constitutively expresses each T7-tagged candidate protein and a pESC-LEU plasmid that expresses His₈- and Flag-ubiquitin with or without Mdm30p. The transformants were grown to mid log phase in raffinose medium and induced for 2.5 h with galactose. Cell lysis and Ni-NTA purification were carried out as described earlier. For each strain, 1% of cleared whole cell extract and purified fraction were used for an immunoblot analysis with an anti-T7 antibody (Novagen).

Mitochondria-enriched fraction was obtained from BY4741 cells chromosomally T7-tagged at MDM34 locus by gentle homogenization and differential centrifugation. After an immunoblot analysis with the anti-T7 antibody, the same blot was re-probed with an anti-porin antibody (Invitrogen).

Yeast two-hybrid assay

Yeast two-hybrid assay was carried out as described previously (Ito et al. 2000) except for the use of pTAD, which expresses the cloned protein as a fusion with a tandem dimer of Gal4p activation domain, to sensitize the detection of weak interactions (unpublished results).

Fluorescence microscopy

Yeast cells were washed twice with PBS and suspended in PBS containing 250 nM MitoFluor Red 589 (Molecular Probes). After incubation for 15–30 min, cells were washed with and suspended in PBS. Fluorescence signals were visualized with Axioskop 2 plus fluorescent microscope (Carl Zeiss) and the images were captured with AxioCam color CCD camera and AxioVision 3.0.6 software (Carl Zeiss).

	Acknowledgements

 
This work was supported by Grant-in-Aid for Scientific Research on Priority Areas "Systems Genomics" and Grant-in-Aid for Young Scientists (B) from Ministry of Education, Culture, Sports, Science and Technology of Japan and by Bioinformatics Research and Development (BIRD) and Core Research for Evolutional Science and Technology (CREST) from Japan Science and Technology Agency (JST).

	Footnotes

 
Communicated by: Keiichi Nakayama

^* Correspondence: ito{at}k.u-tokyo.ac.jp

	References

Top Abstract Introduction Results Discussion Experimental procedures References

 
Adams, A., Gottschling, D.E., Kaiser, C.A. & Steams, T. (1997) Methods in Yeast Genetics: A Cold Spring Harbor Laboratory Course Manual, 1997 Ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.

Aguilar, R.C. & Wendland, B. (2003) Ubiquitin: not just for proteasomes anymore. Curr. Opin. Cell Biol. 15, 184–190.[CrossRef][Medline]

Benanti, J.A., Cheung, S.K., Brady, M.C. & Toczyski, D.P. (2007) A proteomic screen reveals SCF^Grr1 targets that regulate the glycolytic-gluconeogenic switch. Nat. Cell Biol. 9, 1184–1191.[CrossRef][Medline]

Boldogh, I.R., Nowakowski, D.W., Yang, H.C., Chung, H., Karmon, S., Royes, P. & Pon, L.A. (2003) A protein complex containing Mdm10p, Mdm12p, and Mmm1p links mitochondrial membranes and DNA to the cytoskeleton-based segregation machinery. Mol. Biol. Cell 14, 4618–4627.[Abstract/Free Full Text]

Cohen, M.M., Leboucher, G.P., Livnat-Levanon, N., Glickman, M.H. & Weissman, A.M. (2008) Ubiquitin-proteasome-dependent degradation of a mitofusin, a critical regulator of mitochondrial fusion. Mol. Biol. Cell 19, 2457–2464.[Abstract/Free Full Text]

Dimmer, K.S., Fritz, S., Fuchs, F., Messerschmitt, M., Weinbach, N., Neupert, W. & Westermann, B. (2002) Genetic basis of mitochondrial function and morphology in Saccharomyces cerevisiae. Mol. Biol. Cell 13, 847–853.[Abstract/Free Full Text]

Dimmer, K.S., Jakobs, S., Vogel, F., Altmann, K. & Westermann, B. (2005) Mdm31 and Mdm32 are inner membrane proteins required for maintenance of mitochondrial shape and stability of mitochondrial DNA nucleoids in yeast. J. Cell Biol. 168, 103–115.[Abstract/Free Full Text]

Escobar-Henriques, M., Westermann, B. & Langer, T. (2006) Regulation of mitochondrial fusion by the F-box protein Mdm30 involves proteasome-independent turnover of Fzo1. J. Cell Biol. 173, 645–650.[Abstract/Free Full Text]

Ficarro, S.B., McCleland, M.L., Stukenberg, P.T., Burke, D.J., Ross, M.M., Shabanowitz, J., Hunt, D.F. & White, F.M. (2002) Phosphoproteome analysis by mass spectrometry and its application to Saccharomyces cerevisiae. Nat. Biotechnol. 20, 301–305.[CrossRef][Medline]

Fritz, S., Weinbach, N. & Westermann, B. (2003) Mdm30 is an F-box protein required for maintenance of fusion-competent mitochondria in yeast. Mol. Biol. Cell 14, 2303–2313.[Abstract/Free Full Text]

Gagne, J.M. (2002) The F-box subunit of the SCF E3 complex is encoded by a diverse superfamily of genes in Arabidopsis. Proc. Natl. Acad. Sci. USA 99, 11519–11524.[Abstract/Free Full Text]

Gietz, R.D. & Woods, R.A. (2002) Transformation of yeast by lithium acetate/single-stranded carrier DNA/polyethylene glycol method. Methods Enzymol. 350, 87–96.[CrossRef][Medline]

Güldener, U., Heck, S., Fielder, T., Beinhauer, J. & Hegemann, J.H. (1996) A new efficient gene disruption cassette for repeated use in budding yeast. Nucleic Acids Res. 24, 2519–2524.[Abstract/Free Full Text]

Gupta, R., Kus, B., Fladd, C., Wasmuth, J., Tonikian, R., Sidhu, S., Krogan, N.J., Parkinson, J. & Rotin, D. (2007) Ubiquitination screen using protein microarrays for comprehensive identification of Rsp5 substrates in yeast. Mol. Syst. Biol. 3, 116.[Medline]

Hershko, A. & Ciechanover, A. (1998) The ubiquitin system. Annu. Rev. Biochem. 67, 425–479.[CrossRef][Medline]

Ito, T., Tashiro, K., Muta, S., Ozawa, R., Chiba, T., Nishizawa, M., Yamamoto, K., Kuhara, S. & Sakaki, Y. (2000) Toward a protein-protein interaction map of the budding yeast: A comprehensive system to examine two-hybrid interactions in all possible combinations between the yeast proteins. Proc. Natl. Acad. Sci. USA 97, 1143–1147.[Abstract/Free Full Text]

Kishi, T., Ikeda, A., Nagao, R. & Koyama, N. (2007) The SCF^Cdc4 ubiquitin ligase regulates calcineurin signaling through degradation of phosphorylated Rcn1, an inhibitor of calcineurin. Proc. Natl. Acad. Sci. USA 104, 17418–17423.[Abstract/Free Full Text]

Kito, K., Ota, K., Fujita, T. & Ito, T. (2007) A synthetic protein approach toward accurate mass spectrometric quantification of component stoichiometry of multiprotein complexes. J. Proteome Res. 6, 792–800.[CrossRef][Medline]

Li, X.J., Zhang, H., Ranish, J.A. & Aebersold, R. (2003) Automated statistical analysis of protein abundance ratios from data generated by stable-isotope dilution and tandem mass spectrometry. Anal. Chem. 75, 6648–6657.[Medline]

Liu, J.Q., Nagata, S., Dairi, T., Misono, H., Shimizu, S. & Yamada, H. (1997) The GLY1 gene of Saccharomyces cerevisiae encodes a low-specific L-threonine aldolase that catalyzes cleavage of L-allo-threonine and L-threonine to glycine: expression of the gene in Escherichia coli and purification and characterization of the enzyme. Eur. J. Biochem. 245, 289–293.[Medline]

Meisinger, C., Pfannschmidt, S., Rissler, M., Milenkovic, D., Becker, T., Stojanovski, D., Youngman, M.J., Jensen, R.E., Chacinska, A., Guiard, B., Pfanner, N. & Wiedemann, N. (2007) The morphology proteins Mdm12/Mmm1 function in the major β-barrel assembly pathway of mitochondria. EMBO J. 26, 2229–2239.[CrossRef][Medline]

Muratani, M., Kung, C., Shokat, K.M. & Tansey, W.P. (2005) The F box protein Dsg1/Mdm30 is a transcriptional coactivator that stimulates Gal4 turnover and cotranscriptional mRNA processing. Cell 120, 887–899.[CrossRef][Medline]

Ong, S.E., Blagoev, B., Kratchmarova, I., Kristensen, D.B., Steen, H., Pandey, A. & Mann, M. (2002) Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics. Mol. Cell. Proteomics 1, 376–386.[Abstract/Free Full Text]

Ota, K., Kito, K., Iemura, S., Natsume, T. & Ito, T. (2008) A parallel affinity purification method for selective isolation of polyubiquitinated proteins. Proteomics 8, 3004–3007.[CrossRef][Medline]

Patton, E.E., Willems, A.R. & Tyers, M. (1998) Combinatorial control in ubiquitin-dependent proteolysis: don't Skp the F-box hypothesis. Trends Genet. 14, 236–243.[CrossRef][Medline]

Peng, J., Schwartz, D., Elias, J.E., Thoreen, C.C., Cheng, D., Marsischky, G., Roelofs, J., Finley, D. & Gygi, S.P. (2003) A proteomics approach to understanding protein ubiquitination. Nat. Biotechnol. 21, 921–926.[CrossRef][Medline]

Pickart, C.M. (2001a) Mechanisms underlying ubiquitination. Annu. Rev. Biochem. 70, 503–533.[CrossRef][Medline]

Pickart, C.M. (2001b) Ubiquitin enters the new millennium. Mol. Cell 8, 499–504.[CrossRef][Medline]

Willems, A.R., Schwab, M. & Tyers, M. (2004) A hitchhiker's guide to the cullin ubiquitin ligases: SCF and its kin. Biochim. Biophys. Acta 1695, 133–170.[Medline]

Youngman, M.J., Hobbs, A.E., Burgess, S.M., Srinivasan, M. & Jensen, R.E. (2004) Mmm2p, a mitochondrial outer membrane protein required for yeast mitochondrial shape and maintenance of mtDNA nucleoids. J. Cell Biol. 164, 677–688.[Abstract/Free Full Text]

Zhang, H., Li, X.J., Martin, D.B. & Aebersold, R. (2003) Identification and quantification of N-linked glycoproteins using hydrazide chemistry, stable isotope labeling and mass spectrometry. Nat. Biotechnol. 21, 660–666.[CrossRef][Medline]

Received: 2 July 2008
Accepted: 23 July 2008

This article has been cited by other articles:

				W. Jonkers and M. Rep Lessons from Fungal F-Box Proteins Eukaryot. Cell, May 1, 2009; 8(5): 677 - 695. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ota, K. Articles by Ito, T. Search for Related Content PubMed Articles by Ota, K. Articles by Ito, T.