Molecular Biology of the Cell
Vol. 19, 2457–2464, June 2008
Ubiquitin–Proteasome-dependent Degradation of a
Mitofusin, a Critical Regulator of Mitochondrial Fusion
Mickael M.J. Cohen,* Guillaume P. Leboucher,*†Nurit Livnat-Levanon,†
Michael H. Glickman,†and Allan M. Weissman*
*Laboratory of Protein Dynamics and Signaling, National Cancer Institute, Frederick, MD 21702;
and†Department of Biology, Technion-Israel Institute of Technology, Haifa 32000, Israel
Submitted February 29, 2008; Accepted March 7, 2008
Monitoring Editor: Janet Shaw
The mitochondrion is a dynamic membranous network whose morphology is conditioned by the equilibrium between
ongoing fusion and fission of mitochondrial membranes. In the budding yeast, Saccharomyces cerevisiae, the transmem-
brane GTPase Fzo1p controls fusion of mitochondrial outer membranes. Deletion or overexpression of Fzo1p have both
been shown to alter the mitochondrial fusion process indicating that maintenance of steady-state levels of Fzo1p are
required for efficient mitochondrial fusion. Cellular levels of Fzo1p are regulated through degradation of Fzo1p by the
F-box protein Mdm30p. How Mdm30p promotes degradation of Fzo1p is currently unknown. We have now determined
that during vegetative growth Mdm30p mediates ubiquitylation of Fzo1p and that degradation of Fzo1p is an ubiquitin-
proteasome–dependent process. In vivo, Mdm30p associates through its F-box motif with other core components of
Skp1-Cullin-F-box (SCF) ubiquitin ligases. We show that the resulting SCFMdm30pligase promotes ubiquitylation of
Fzo1p at mitochondria and its subsequent degradation by the 26S proteasome. These results provide the first demonstra-
tion that a cytosolic ubiquitin ligase targets a critical regulatory molecule at the mitochondrial outer membrane. This
study provides a framework for developing an understanding of the function of Mdm30p-mediated Fzo1p degradation in
the multistep process of mitochondrial fusion.
Mitochondria are dynamic in nature and collectively all
mitochondria in a cell functionally constitute a single tubu-
lar network, the morphology of which is determined by an
equilibrium between fusion and fission (Bleazard et al., 1999;
Sesaki and Jensen, 1999; Fritz et al., 2003; Okamoto and
Shaw, 2005; Hoppins et al., 2007). Disruption of fission in the
budding yeast Saccharomyces cerevisiae drives the equilib-
rium toward fusion resulting in net-like mitochondrial struc-
tures that appear to collapse to one side of the cell (fused).
Conversely, abrogation of fusion shifts the equilibrium to-
ward fission, which is manifest as dot-like fragmented mi-
tochondria (fizzed; see examples in Figure 1A). Such alter-
ations are linked to apoptosis in mammals and compromise
cell life span in fungus (Okamoto and Shaw, 2005; Chan,
2006; Heath-Engel and Shore, 2006; Martinou and Youle,
2006; Scheckhuber et al., 2007). Maintaining the capacity of
mitochondria to fuse normally is essential to inheritance of
mitochondrial DNA and in turn for respiration (Hermann et
Mitochondrial outer membrane fusion is controlled by
mitofusins, a family of GTPases integral to the mitochon-
drial outer membrane. The yeast mitofusin, Fzo1p, is unsta-
ble during both vegetative and nonvegetative growth. Reg-
ulating levels of expression of Fzo1p is of critical importance
as either deletion or overexpression of Fzo1p alters the mi-
tochondrial fusion process resulting in fizzed or abnormal
aggregated mitochondria, respectively (Hermann et al., 1998;
Rapaport et al., 1998; Fritz et al., 2003; Escobar-Henriques et
The ubiquitin-proteasome system (UPS) constitutes the
major mechanism by which cells acutely alter levels of cy-
tosolic, nuclear, and endoplasmic reticulum (ER) proteins in
a highly regulated manner. This occurs generally, but not
exclusively, by conjugation with chains of ubiquitin linked
through lysine 48 (K48) of ubiquitin, which targets modified
proteins to the 26S proteasome for degradation (Glickman
and Ciechanover, 2002). Ubiquitin ligases (E3s) mediate the
transfer of ubiquitin from ubiquitin-conjugating enzymes
(E2s) to specific substrates and are the primary determinants
of specificity in ubiquitylation (Fang and Weissman, 2004).
During nonvegetative growth, corresponding to mating con-
ditions, a role for the 26S proteasome in degradation of
Fzo1p has been reported. However, evidence for ubiquity-
lation has surprisingly been lacking, and no ubiquitin ligase
has been implicated in this process (Neutzner and Youle,
2005; Escobar-Henriques et al., 2006).
During vegetative growth, the level of Fzo1p is regulated
by Mdm30p. Mdm30p is a member of the F-box family of
proteins. The F-box is a 50-amino acid protein interaction
motif encoded in ?15 genes in S. cerevisiae and ?70 genes in
mammals. F-box proteins are generally thought to serve as
substrate recognition elements of ubiquitin ligases of the
Skp1-Cullin-F-box (SCF) family (Cardozo and Pagano, 2004;
Willems et al., 2004; Petroski and Deshaies, 2005). In this
regard, Mdm30p has been shown to promote ubiquitylation
and proteasomal degradation of the Gal4p transcription fac-
tor (Muratani et al., 2005). Nonetheless, to date not all F-box
proteins have been positively linked to the SCF complex or
This article was published online ahead of print in MBC in Press
on March 19, 2008.
il) or Allan M. Weissman (email@example.com).
© 2008 by The American Society for Cell Biology 2457
even to a specific function in the UPS (Galan et al., 2001;
Frescas et al., 2007).
Mdm30p associates with mitochondria (Fritz et al., 2003),
appears to physically interact with Fzo1p (Escobar-Hen-
riques et al., 2006), and has been shown to target Fzo1p for
degradation (Fritz et al., 2003; Escobar-Henriques et al.,
2006). The importance of Mdm30p in mitochondrial function
is underscored by the finding that deletion of MDM30 ab-
rogates mitochondrial fusion and leads to aggregated mito-
chondria (Dürr et al., 2006; see examples in Figure 1A) and
consequently to defective mitochondrial DNA inheritance
and a failure to respire (Fritz et al., 2003). However, the
mechanism by which Mdm30p promotes Fzo1p degradation
has not been elucidated.
In this study, we investigated the mechanism by which
Fzo1p is targeted for degradation during vegetative growth.
We establish that Fzo1p is ubiquitylated and targeted for
proteasomal degradation. This ubiquitylation is mediated
by an SCF ubiquitin ligase that includes Mdm30p as the
substrate recognition factor (SCFMdm30p). Thus, we have
identified a mechanism whereby a critical protein integral to
the mitochondrial outer membrane is targeted for destruc-
tion by cytosolic components of the UPS.
MATERIALS AND METHODS
Yeast Strains and Media
The S. cerevisiae strains used in this study are listed in Supplemental Data,
Table S1. Standard methods were used for growth, transformation and ge-
netic manipulation of S. cerevisiae. Complete (YPDextrose) and minimal
(SDextrose, SGlycerol) media supplemented with either 2% dextrose or 3%
glycerol were prepared as described (Sherman et al., 1986). In the indicated
strains (see Supplemental Data, Table S1), FZO1 was chromosomally tagged
with three copies of an HA epitope sequence, as previously described
(Longtine et al., 1998).
The plasmids used in this study are listed in Supplemental Data, Table S1.
Ubiquitin wild type, K48R, or K63R, expressed under the control of the CUP1
promoter were overproduced by growing cells for 1 h in the presence of 0.1
mM copper sulfate. The MDM30-hemagglutinin (HA) construct was gener-
ated as follows. The 500 base pairs upstream of the ATG and downstream of
the STOP codon of MDM30 were subcloned in the pRS316 vector, yielding the
pRS316 MDM30prom/ter vector. MDM30 lacking its STOP codon or both its
STOP codon and sequence encoding its F-box (amino acids 1-58) were cloned
into pRS316 MDM30prom/ter followed by insertion of the HA tag down-
stream of the MDM30 coding sequence, resulting in MDM30-HA and ?fbox-
HA, respectively. MDM30-HA was then subcloned in the p426TEF vector and
previously described mutations of amino acids in its F-box motif (fbox-HA;
Escobar-Henriques et al., 2006) that are critical for its association with Skp1p
(Galan and Peter, 1999) were generated by site-directed mutagenesis.
Antiserum raised against Fzo1p was generously provided by J. Nunnari
(University of California, Davis, CA). Monoclonal antibodies utilized were
against HA and Myc epitopes (12CA5 and 9E10; Santa Cruz Biotechnology,
Santa Cruz, CA) phosphoglucokinase (PGK; Monoclonal 22C5; Molecular
Probes, Eugene, OR), Tom20p (kindly provided by A. Azem, Tel Aviv Uni-
versity), Hexokinase1 (Hxk1p; kindly provided by A. Azem), or Cdc53p
(yC-17; Santa Cruz Biotechnology). Antiserum recognizing Cue1p was pro-
vided by Z. Kostova (National Cancer Institute).
Yeast Extracts and Cycloheximide Chase
Cells grown in YPD or SD were collected during the exponential growth
phase. Total protein extracts were prepared by the NaOH-trichloroacetic acid
(TCA) lysis method (Avaro et al., 2002). To monitor constitutive turnover of
Fzo1, cycloheximide (CHX) was added to yeast cultures growing at 37°C to a
final concentration of 100 ?g/ml. Thermosensitive strains were incubated for
2 h at 37°C before adding CHX. Total protein extracts were prepared at the
indicated time points after addition of CHX.
Pulse-Chase Metabolic Labeling
Cells grown in YPD were collected during the exponential growth phase,
incubated for 50 min at 23°C in 1 ml labeling media (SD-Met) lacking methi-
onine, and pulsed with 25 ?Ci of35S methionine (Perkin-Elmer Cetus, Nor-
walk, CT) per OD of cells for 15 min at 30°C followed by 15 min at 37°C. After
addition of 1 ml chase media (SD-Met supplemented with 6 mg/ml methio-
nine and 2 mg/ml BSA), cells were incubated at 37°C. After addition of chase
media ?2 ODs of cells were collected immediately (time ? 0) and at 30 and
60 min and treated as described previously (Moreau et al., 1997). Results were
quantified by Storm Phosphorimager and ImageQuant software (GE Health-
care, Waukesha, WI). Error bars were determined by calculating the SD from
three independent experiments.
For coimmunoprecipitation assays, cells were lysed at 4°C with glass beads in
immunoprecipitation (IP) buffer (50 mM HEPES, pH 7.5, 50 mM sodium
chloride, 0.6% Triton X-100, 10% glycerol, 20 mM iodoacetamide, and pro-
tease inhibitor; Complete mini, Roche, Indianapolis, IN). Insoluble material
was removed by centrifugation for 30 min at 13,000 ? g. An aliquot of the
supernatant was precipitated with 50% TCA (pre-IP lysate). The remaining
supernatant was incubated with Anti-HA Affinity Matrix (Roche) for 2 h at
4°C. Beads washed with IP buffer were heated in sample buffer before
resolution by SDS-PAGE and analysis by immunoblotting. For immunopre-
cipitation of Fzo1-HA or Mdm30-HA, yeast extracts were prepared using the
NaOH-TCA lysis method (Avaro et al., 2002). Extracts were then boiled for 10
min at 70°C in SDS loading buffer and insoluble material removed by centrifu-
gation for 5 min at 13,000 ?g. Resulting supernatants were diluted 10-fold in IP
buffer, and immunoprecipitations were performed as described above.
Cell fractionation was performed as described (Meisinger et al., 2000). In brief,
cell cultures were grown to midlog phase (OD6001-2), and spheroplasts were
prepared by treating the cells with Zymolyase-20T (MP Biomedicals, Solon,
OH). After gentle homogenization of spheroplasts and centrifugation at
1500 ? g, the supernatant (total fraction) was further subject to centrifugation
at 12,000 ? g for 10 min, yielding a supernatant (S) and a mitochondrial
enriched pellet fraction (P). Subcellular fractions were assayed for cytosolic
and mitochondrial proteins; Tom20 and Hexokinase1 were used as mitochon-
drial and cytosolic markers, respectively.
Glycerol Growth Analysis
For glycerol growth assays, cultures grown overnight in SD medium were
pelleted, resuspended at OD600? 1, and diluted 1:10 five times in water.
Three microliters of the dilutions were spotted on plates and grown for 2 d
(SD) or 4 d (SG) at 30°C or 37°C.
For visualization of mitochondria, yeast strains were transformed with plasmid
pYX232-mtGFP, encoding mitochondria-targeted GFP (mtGFP; Westermann and
Neupert, 2000). Cultures in logarithmic growth were fixed with 3.7% formalde-
hyde (Sigma, St. Louis, MO) for 10 min, washed in KPi buffer (0.02 M KH2PO4,
0.08 M K2HPO4, 1 M sorbitol, pH 7.5) and mounted on Superfrost microscope
slides (Esco Products, Oak Ridge, NJ) in phosphate-buffered saline. Cells were
then analyzed by epifluorescence microscopy on an Axiovert 200M microscope
(Carl Zeiss MicroImaging, Thornwood, NY) using a 100? oil-immersion objec-
tive. Images were recorded with a Hamamatsu ORCA-ER camera (Hamamatsu
Photonics, Hamamatsu City, Japan). For each field of cells, between 30 and 40
pictures were taken in the Z-coordinate, and cells were deconvoluted using
Improvision Openlab 4.0.2 software (Improvision, Lexington, MA). The mor-
Quantification was confirmed by independent counting by a second individual
blinded to the identity of the strains. Results are displayed in Supplemental Data
Table S2 and Figure 1B.
The F-Box Motif of Mdm30p Is Essential for
Mitochondrial Fusion and Respiration as well as for
To assess the mechanism by which Mdm30p regulates mi-
tochondrial morphology and particularly the importance of
its F-box domain, we analyzed mitochondria from mdm30?
cells expressing either wild-type Mdm30p or a form mutated
in its F-box (Figure 1, B and C, and Supplemental Data,
Table S1). This mutation has previously been shown to
result in increased steady-state levels of Fzo1p (Escobar-
Henriques et al., 2006). About 70% of wild-type cells dis-
played mitochondria characterized as being tubular. The
remaining 30% were partitioned between those scored as
having fused, fizzed, and aggregated mitochondria (see ex-
M.M.J. Cohen et al.
Molecular Biology of the Cell2458
amples in Figure 1A). In agreement with previous reports
(Fritz et al., 2003; Dürr et al., 2006; Escobar-Henriques et al.,
2006), among mdm30? cells almost 70% displayed aggre-
gated mitochondria (Figure 1C, vector transformation). Re-
introducing MDM30 into mdm30? cells (MDM30-HA), un-
der a constitutive TEF promoter, restored mitochondrial
morphology to a distribution similar to wild-type cells and
completely reversed the marked increase in aggregated mi-
tochondria seen in mdm30? cells (Figure 1C). However, cells
expressing the F-box mutant (fbox-HA) retained a mitochon-
drial distribution similar to that observed in the mdm30?
mutant, pointing to an essential role for the F-box in main-
taining a normal distribution of mitochondrial morphology.
To confirm the functional significance of these findings,
we evaluated the impact on respiration of mutating the
F-box of Mdm30p. Loss of mitochondrial fusion has been
shown to manifest itself as a failure of mdm30? cells to
respire properly and a decreased capacity to grow on media
containing only a nonfermentable carbon source (Fritz et al.,
2003; Dürr et al., 2006). Strains from Figure 1C were therefore
tested for growth on selective media containing either dex-
trose (fermentable) or glycerol (nonfermentable) as the sole
carbon source (Figure 1D). As shown previously (Fritz et al.,
2003), mdm30? cells are defective for growth on glycerol
media. Consistent with the morphological findings, this
growth defect was rescued by expression of Mdm30-HA but
not the F-box mutant Mdm30-HA (fbox-HA), once again
pointing to an essential role for the F-box motif in mitochon-
drial function as well as morphology. Similar findings for
both mitochondrial morphology and growth on glycerol
were obtained when wild-type and truncated Mdm30p lack-
ing its F-box (?f-box-HA) were expressed from the endoge-
nous MDM30 promoter (Supplemental Data, Figure S1).
Mitochondrial aggregation in mdm30? cells has been
shown to correlate with accumulation of Fzo1p (Fritz et al.,
2003). More recently, Mdm30p was shown to be essential for
Fzo1p degradation during vegetative growth (Escobar-Hen-
riques et al., 2006). In agreement with this latter study, we
observed that endogenous Fzo1p, which is naturally turned
over in cells, was completely stabilized upon genomic dele-
tion of MDM30 (Figure 1E, left panels). Given our findings
that the F-box of Mdm30p is required for normal mitochon-
drial morphology and respiration (Figure 1, C and D), we
asked whether the F-box is also required for Fzo1p degra-
dation. mdm30? strains expressing Mdm30-HA or its F-box
mutant (fbox-HA) were used to monitor Fzo1p turnover
(Figure 1E, right panels). Although Fzo1p was degraded
upon expression of Mdm30-HA, it remained stable upon
expression of the F-box mutant (fbox-HA), consistent with
previous findings (Escobar-Henriques et al., 2006). This dif-
ferential stability was also observed with wild-type and
F-box–deleted Mdm30p expressed from its endogenous pro-
moter (Supplemental Data, Figure S1, C and D). Together,
results from Figure 1 indicate that the F-box of Mdm30p is
essential for efficient mitochondrial fusion and respiration as
well as for Fzo1p degradation.
Critical Components of the SCF Complex Participate in
Most F-box proteins are thought to act as substrate recogni-
tion subunits of SCF ubiquitin ligase complexes (Cardozo
and Pagano, 2004; Willems et al., 2004; Petroski and De-
% of cells
0 90 180 0 90 180
0 90 180
0 90 180
CHX (min) :
mitochondrial fusion and respiration as well as deg-
radation of Fzo1p. (A) Typical images of fused
(green), tubular (yellow), fizzed (red), and aggre-
gated (blue) mitochondria quantified in Figures 1C
and Supplemental Data Table S2. (B) Amino acids
19, 20, 22, and 23 conserved in F-box motifs were
mutated as indicated in Mdm30-HA to yield the
fbox-HA mutant. (C) Mitochondrial morphology was
assessed in wild-type (vector; W303 background),
mdm30? (vector), or mdm30? cells expressing either
Mdm30-HA or an F-box mutant of Mdm30-HA
(fbox-HA) under control of the TEF promoter. (D)
The same strains as described in C were grown at
37°C on selective media containing either dextrose
or glycerol as the only carbon source. (E) Rate of
Fzo1p degradation was analyzed in the indicated
strains (W303 background) after shift to 37°C and
treatment with CHX. Yeast extracts were prepared
at the indicated times and remaining Fzo1p or
Mdm30-HA evaluated by immunoblotting. Levels of a
stable protein, phosphoglucokinase (PGK), are shown
as a loading control.
The F-box of Mdm30p is required for
Ubiquitin-dependent Mitochondrial Fusion
Vol. 19, June 20082459
shaies, 2005), thereby promoting substrate ubiquitylation
followed by proteasomal degradation. In S. cerevisiae Skp1p
serves as an adaptor between F-box proteins and the cullin,
Cdc53p. The Skp1p-Cdc53p core together with a small RING
finger protein serves as a molecular scaffold that also re-
cruits a specific ubiquitin-conjugating enzyme (E2), Cdc34p.
Our finding that the F-box of Mdm30p is essential for
Fzo1p degradation raises the possibility that Mdm30p func-
tions as part of an SCF E3 ubiquitin ligase (SCFMdm30p),
potentially providing a mechanism by which Mdm30p tar-
gets Fzo1p for degradation. To directly assess whether
Mdm30p associates with Skp1p-Cdc53p in cells, HA-tagged
Mdm30p was immunoprecipitated from whole cell extracts
prepared from mdm30? cells expressing wild-type Mdm30p
(Mdm30-HA) or Mdm30p mutated in its F-box (fbox-HA).
Immunoprecipitates were resolved on SDS-PAGE and immu-
noblotted with Cdc53p antibody (Figure 2A, middle panel).
Cdc53p coimmunoprecipitated with wild-type Mdm30-HA
but not the F-box mutant. This result establishes that Mdm30p
associates with core components of the SCF and that this in-
teraction is dependent on an intact Mdm30p F-box.
Having established that Mdm30p is a bona fide compo-
nent of an SCF E3 (SCFMdm30p), we evaluated the require-
ments for Cdc34p and Cdc53p on Fzo1p degradation by
CHX chase (Figure 2B). Compared with the wild-type strain,
a marked stabilization of Fzo1p was observed in the condi-
tional mutants, cdc53ts and cdc34ts, at the restrictive temper-
ature (37°C). To assess effects on protein turnover without
inhibiting protein synthesis, Fzo1p half-life in cdc53ts and
cdc34ts strains was assessed by metabolic labeling using
35S methionine (Figure 2C). Although Fzo1p was degraded
with a half-life of ?30 min in wild-type cells, it was mark-
edly stabilized in thermosensitive mutants of either CDC53
or CDC34 (cdc53ts and cdc34ts) with half-lives of greater than
60 min. These results establish that, in addition to Mdm30p,
critical components of the SCF complex participate in Fzo1p
Fzo1p Is Modified with Ubiquitin at the Mitochondria
The observation that the ubiquitin ligase SCFMdm30pis re-
quired for Fzo1p degradation strongly suggests that Fzo1p
is a substrate for ubiquitylation. Ubiquitylated intermediates
0 90 180 0 90 180
0 90 180
CHX (min) :
20 50 60
60 0 30 60
Chase (min) :
60 0 30 60
Chase (min) :
required for degradation of Fzo1p. (A) Epitope-tagged Mdm30p
(Mdm30-HA) was immunoprecipitated (IP) from the indicated cell
lysates (W303 background) with antibody recognizing HA. Whole
cell lysates (pre-IP; bottom panel) and immunoprecipitates from
equal amounts of lysates from each strain were analyzed by immu-
noblotting (IB) with antibody recognizing either Cdc53p or HA.
Excess starting material (lysate) was utilized in IPs to maximize
visualization of endogenous Cdc53. (B) Degradation of Fzo1p in
wild-type, cdc53ts, or cdc34ts strains was evaluated by CHX chase as
in Figure 1E. (C) Fzo1-HA turnover was analyzed by35S pulse-chase
metabolic labeling in wild-type (triangle), cdc53ts (square, left), and
cdc34ts (square, right) strains. Graphs represent quantification of
three independent experiments. Representative experiments are
shown below. Data are plotted relative to the amount at the begin-
ning of the chase. Asterisks indicate nonspecific band.
Core SCF components interact with Mdm30p and are
IB: HA IB: Fzo1p
IB: MycIB: HA
Short and long exposures of an identical anti-Fzo1p immunoblot
probing equal amounts of wild-type (BY4742) and fzo1? whole cell
extracts. Unmodified Fzo1p is indicated by a thick arrow; modified
forms are marked with thin arrows. Nonspecific bands (also de-
tected in fzo1? extracts) are indicated by asterisks. (B) Crude extracts
from FZO1 (W303 background) and FZO1–3HA strains were pro-
cessed for immunoblotting with antibodies that recognize either HA
or Fzo1p. (C) Crude extracts from Ub (SUB280) and Myc-His6-Ub
(SUB595) strains were processed for immunoblotting to detect
Fzo1p. (D) Immunoprecipitates with HA antibody from Ub and
Myc-His6-Ub strains having chromosomally HA-tagged (FZO1-HA)
or untagged (FZO1) FZO1 were immunoblotted with either HA
antibody or Myc antibody. (E) Distribution of Fzo1p between cyto-
solic (S) and mitochondrially enriched (P) fractions prepared from
wild-type cells (W303 background). Hexokinase1p (Hxk1p) and
Tom20p were used as cytosolic and mitochondrial markers, respec-
tively. Asterisks correspond to nonspecific bands observed also in
fzo1? (data not shown).
Mitochondrial Fzo1p is modified with ubiquitin. (A)
M.M.J. Cohen et al.
Molecular Biology of the Cell2460
are frequently difficult to detect as they are rapidly degraded
by the proteasome. Using a polyclonal antibody directed
against Fzo1p, we observed multiple immunoreactive spe-
cies above the major Fzo1p band on long exposure of im-
munoblots (Figure 3A, right panel). These higher molecular
weight bands were reproducibly and specifically observed
in extracts prepared from wild-type yeast but not from fzo1?
cells. The same pattern was observed whether monitoring
endogenous Fzo1p or chromosomally tagged FZO1-HA us-
ing either anti-Fzo1p or anti-HA (Figure 3B). A slight retar-
dation in the migration pattern of Fzo1p and the higher
molecular weight bands was observed in the FZO1-HA
strain when detected by Fzo1p antibody, consistent with
increased mass conferred by the three copies of an HA tag.
This indicates that these higher molecular weight species
represent modified forms of Fzo1p.
The ladder-like pattern of these higher molecular weight
forms of Fzo1p is highly suggestive of ubiquitylation. To
confirm this possibility, extracts from yeast strains express-
ing ubiquitin tagged with both Myc and His6epitopes as the
sole source of ubiquitin were processed for immunoblotting
with anti-Fzo1p and the migration pattern compared with
similar extracts prepared from a strain expressing untagged
ubiquitin (Figure 3C). We observed that although migration
of unmodified Fzo1p band was similar in both strains, all of
the higher molecular weight species were shifted upward in
the strain expressing Myc-His6-Ub. This upward shift is
consistent with differential incorporation of tagged Myc-
His6-Ub in one strain and untagged Ub in the other. To
unequivocally and directly demonstrate ubiquitylation of
Fzo1p, lysates from chromosomally tagged FZO1-HA cells
expressing Myc-His6-Ub as the sole ubiquitin source were
immunoprecipitated with HA antibody and immunoblotted
with Myc antibody. This led to the specific detection of
high-molecular-weight species that comigrated with those
seen when replicate blots were probed with HA antibody
(Figure 3D). These results conclusively establish for the first
time that Fzo1p, expressed at endogenous levels, undergoes
As Fzo1p is an integral mitochondrial outer membrane
protein, we wanted to determine the cellular location of
ubiquitylated Fzo1p. Whole cell extract was fractionated and
cytosolic (S) and mitochondria-enriched fractions (P) were
tested for Fzo1p content by immunoblotting with Fzo1p
antibody (Figure 3E). Both unmodified and ubiquitylated
forms of Fzo1p were found almost exclusively in the mito-
chondria-enriched fraction. The mitochondrial localization
of ubiquitylated Fzo1p was further confirmed by sucrose
gradient analysis (Supplemental Data, Figure S2). These
data indicate that ubiquitylated Fzo1p is localized to mito-
chondria and does not represent mis-targeted or mis-local-
SCFMdm30pMediates K48-linked Ubiquitylation of Fzo1p
Having established that the higher molecular weight species
of Fzo1p correspond to Fzo1p-ubiquitin conjugates, we
asked whether ubiquitylation of Fzo1p is dependent on
Mdm30p and other SCF components. As is apparent, the
ubiquitylated forms of Fzo1p that were detected in wild-
type cells were undetectable in the mdm30? mutant (Figure
4A, cf. left two lanes). Moreover, although MDM30-HA re-
stored Fzo1p ubiquitylation in mdm30? cells, the F-box mu-
tant (fbox-HA) did not. Similarly, Fzo1p ubiquitylation was
decreased in both conditional SCF mutants, cdc34ts and
cdc53ts, when grown at the restrictive temperature com-
pared with isogenic wild-type controls (Figure 4B). We con-
clude that Mdm30p, Cdc34p, and Cdc53p are essential for
the Fzo1p ubiquitylation that is observed in wild-type cells.
The best characterized function of ubiquitylation, target-
ing of proteins to the 26S proteasome for degradation, occurs
largely, although perhaps not exclusively, as a consequence
of covalent modification of substrates with chains of ubiq-
uitin that contain ubiquitin linked together through K48
(Fang and Weissman, 2004). Other cellular functions are
known to involve monoubiquitylation or polyubiquitylation
through other lysines of ubiquitin, especially K63 (Mukho-
padhyay and Riezman, 2007). To gain insight into the link-
ages being generated on Fzo1p, wild-type and fzo1? cells
were transformed with a high copy vector for overexpres-
sion of ubiquitin (wild type) or ubiquitin in which either K48
or K63 was mutated to arginine (K48R or K63R). Overex-
linked ubiquitylation of Fzo1p. (A) Cellular levels of Fzo1p,
Mdm30p (Mdm30-HA), or a version of Mdm30p with a mutated
F-box incapable of binding to Skp1p (fbox-HA) were analyzed in
whole cell extracts prepared from the indicated strains (W303 back-
ground) grown at 30°C. Unmodified and ubiquitylated forms of
Fzo1p are indicated by thick and thin arrows, respectively. PGK is
shown as a loading control. (B) Ubiquitylated forms of Fzo1p were
analyzed in whole cell extracts prepared from wild-type, cdc53ts, or
cdc34ts cells grown at 23°C and shifted to 37°C. PGK serves as a
loading control. (C) Anti-Fzo1p immunoblot of whole cell extracts
prepared from fzo1? and wild-type cells (BY4742) transformed with
vectors overexpressing either ubiquitin (WT), K48R ubiquitin, K63R
ubiquitin, or empty vector as control.
Mdm30p, Cdc34p, and Cdc53p are required for K48-
Ubiquitin-dependent Mitochondrial Fusion
Vol. 19, June 20082461
pression of wild-type ubiquitin is known to enhance the
steady-state level of substrate ubiquitylation. Incorporation
of overexpressed K48R or K63R ubiquitins into chains pre-
cludes elongation of polyubiquitin chains linked through
these residues (Galan and Haguenauer-Tsapis, 1997). Strik-
ingly, we observed that overexpression of K48R ubiquitin
resulted in a downward shift in ubiquitylated forms of
Fzo1p, reflecting either multiple mono-ubiquitylation events
or short chains “capped” with unextendable K48R ubiquitin
(Figure 4C). In contrast, overexpression of wild-type or
K63R ubiquitin resulted in an upward shift in ubiquitylated
species of Fzo1p, consistent with increased availability of
ubiquitin competent for K48 chain elongation. Taken to-
gether, these data not only confirm that Fzo1p is ubiquity-
lated, but also provide strong evidence that the higher mo-
lecular weight forms include modification with K48-linked
Fzo1p Is Degraded by the 26S Proteasome during
The results presented thus far establish that Fzo1p is ubiq-
uitylated at the mitochondria in a manner that is dependent
on an intact SCFMdm30pubiquitin ligase, that this ubiquity-
lation appears to be largely K48-linked in nature, and that
this modification strongly correlates with degradation of
Fzo1p, because deletion of MDM30 or mutation of its F-box
abolishes both ubiquitylation and degradation of Fzo1p.
Collectively these observations raise the possibility that
Fzo1p turnover during vegetative growth is a consequence
of proteasomal degradation.
To test the effect of altered proteasome function on Fzo1p
degradation, Fzo1p turnover was assayed by CHX chase in
two different proteasome mutant strains (cim3-1 and pre1-1
pre2-2), which we compared with their wild-type isogenic
controls. The conditional proteasome mutations either
slowed (pre1-1 pre2-2) or completely inhibited (cim3-1) deg-
radation of Fzo1p (Figure 5A). These results strongly sug-
gest that Fzo1p is a target for the ubiquitin-proteasome
To confirm the CHX chase results, pulse-chase metabolic
labeling was performed (Figure 5B). We chose to use the
pre1-1 pre2-2 strain, which carries thermosensitive mutations
in two of the catalytic subunits of the 20S catalytic core of the
proteasome, because it showed only a partial effect on Fzo1p
degradation by CHX chase and therefore would be most
important to confirm through a different approach. As is
evident, mutation of these two core proteasome units dou-
bled the half-life of Fzo1p (30 min in wild-type cells vs. 60
min in pre1-1 pre2-2 cells).
Finally, steady-state levels of unmodified and ubiquity-
lated Fzo1p were assayed in yeast strains bearing thermo-
sensitive mutations in 26S proteasome subunits from the 19S
lid (mpr1-1), 19S base (rpt2RF, cim3-1) or 20S core (pre1-1
pre2-2). In each of these four examples, steady-state levels of
unmodified as well as ubiquitylated Fzo1p increased in
proteasome mutants relative to the four different wild-type
isogenic control strains at 37°C but not at 23°C (Figure 5C, cf.
right and left panels). These results confirm the importance
of the proteasome in Fzo1p degradation and provide further
evidence that ubiquitylated forms of this protein are tar-
geted for degradation. Together, the results presented in
Figure 5 establish a clear role for proteasomes in regulating
the constitutive turnover of Fzo1 during vegetative growth.
The major mechanism for acutely regulating levels of cellu-
lar proteins is the UPS. In addition to myriad cytoplasmic
and nuclear proteins, the UPS is implicated in the degrada-
tion of proteins from other organelles, most notably the ER.
ER-associated degradation (ERAD) consists of a complex set
of processes that are responsible for ubiquitylating and de-
grading many transmembrane proteins as well as ER lumi-
nal proteins. In contrast to ERAD, a role for the UPS in
degradation of outer mitochondrial membrane proteins re-
mains surprisingly obscure.
A number of reports have provided indirect links between
the UPS and mitochondria (Fisk and Yaffe, 1999; Sutovsky et
al., 1999; Hitchcock et al., 2003; Peng et al., 2003; Thompson et
al., 2003; Rinaldi et al., 2004; Altmann and Westermann, 2005;
Dürr et al., 2006). The best evidence so far for involvement of
the UPS in directly regulating mitochondrial outer membrane
proteins is in mammals where a specific E3, MARCHV/
MITOL, is implicated in ubiquitylating two components of
degradation was assessed in pre1-1 pre2-2 and cim3-1 proteasome
thermosensitive strains and their corresponding isogenic wild-type
strains (see Supplemental Data, Table S1) by treating with CHX.
Yeast extracts were prepared at the indicated times and remaining
Fzo1p evaluated by immunoblotting. (B) Fzo1-HA turnover was
analyzed by35S pulse-chase metabolic labeling in wild-type and
pre1-1 pre2-2 strains. Graph on the left represents quantification of
three experiments. A representative experiment is shown on the
right. (C) Fzo1p levels were analyzed in four proteasome mutant
strains and their isogenic wild-type controls grown at the permis-
sive temperature (23°C) or after a 3-h shift to the restrictive temper-
ature (37°C). Long (top panels) and short (middle panels) exposures
of identical anti-Fzo1p blots are displayed. Note relative increase in
levels in mutant strains at the restrictive temperature. PGK is uti-
lized as a loading control.
Fzo1p is degraded by the 26S proteasome. (A) Fzo1p
M.M.J. Cohen et al.
Molecular Biology of the Cell2462
the mitochondrial fission apparatus, DRP1 and FIS1, which
are the human orthologues of yeast Dnm1p and Fis1p, re-
spectively. However, there is a lack of consensus as to
whether this ubiquitylation serves to target these factors for
proteasomal degradation or facilitates other nonproteolytic
functions (Nakamura et al., 2006; Yonashiro et al., 2006;
Karbowski et al., 2007).
The literature regarding the mitochondrial mitofusins is
even more complicated. During nonvegetative (mating type)
growth, degradation of the single yeast mitofusin, Fzo1p,
has been suggested to be dependent on the proteasome.
However, there is no direct evidence for ubiquitylation or
involvement of a specific E3, and a role for Mdm30p has
been excluded (Neutzner and Youle, 2005).
The groups of Westermann and Langer have demon-
strated, and we confirm herein, that during vegetative
growth, degradation and maintenance of normal levels of
Fzo1p is dependent on the F-box protein Mdm30p (Fritz et
al., 2003), with a specific requirement for an intact F-box
(Escobar-Henriques et al., 2006). However, Escobar-Henriques et
al. concluded that this degradation of Fzo1p is independent
of ubiquitin, the SCF and proteasomes (Escobar-Henriques
et al., 2006). This led the authors to conclude that Fzo1p is
degraded during vegetative growth by a novel UPS-inde-
pendent proteolytic pathway that was still dependent on
Mdm30p having an intact F-box. The findings presented in
the current study lead to a very different conclusion. We
provide direct evidence of ubiquitylation of endogenously
expressed Fzo1p that is highly suggestive of K48-linked
ubiquitin chains. This ubiquitylation is dependent on
Mdm30p capable of assembling with other SCF components
through its F-box motif. Moreover, both CHX and35S pulse-
chase metabolic labeling experiments implicate the UPS and
particularly SCFMdm30pin Fzo1p degradation. Our inter-
nally consistent positive findings unequivocally establish an
important role for the UPS in determining the fate of Fzo1p.
The discrepancy between the conclusions reached in Esco-
bar-Henriques et al. and our findings are difficult to recon-
cile. However, we certainly cannot discount the possibility
that in addition to the UPS other means of degrading Fzo1p
could exist including through mitochondrial and other cy-
tosolic proteases as well as by autophagy. Regardless, the
clear involvement of the UPS in Fzo1p degradation, estab-
lished herein, should lay the groundwork for further inves-
tigation of the signals that target Fzo1p for degradation and
the significance of this degradation in mitochondrial func-
The findings presented in this study are in accord with a
paradigm based on cells lacking Mdm30p (Fritz et al., 2003):
increased levels of Fzo1p due to failure to regulate the level
of this protein results in mitochondrial aggregation and a
failure to respire. Despite the requirement for the Mdm30p
F-box in Fzo1p degradation found in Escobar-Henriques et
al., the same group found that Mdm30 lacking its F-box
overexpressed from the CUP promoter surprisingly restored
normal mitochondrial morphology in mdm30? cells (Dürr et
al., 2006). In contrast, we find that, in these cells, expression
of F-box mutants of Mdm30p from either the TEF or MDM30
promoter results in a failure to reverse the abnormal aggre-
gated mitochondrial morphology seen in mdm30? cells, a
finding confirmed by the failure to restore growth on a
nonfermentable carbon source (glycerol). As suggested
(Dürr et al., 2006), the restoration of mitochondrial morphol-
ogy obtained with the CUP promoter in Dürr et al. could be
a consequence of overexpression of truncated Mdm30 that
binds to and inactivates excess Fzo1p. If this is the case, it is
unlikely of physiological significance.
The determination that an integral membrane protein of
the mitochondrial outer membrane is ubiquitylated while
still mitochondria-associated and unambiguously degraded
by the proteasome leads us to posit a general UPS-depen-
dentprocess of mitochondrial-associated
(MAD). Analogous to ERAD, ubiquitin ligases intrinsic to
the mitochondrial outer membrane, such as MARCHV/
MITOL (Nakamura et al., 2006; Yonashiro et al., 2006; Kar-
bowski et al., 2007) and the newly described MULAN (Li et
al., 2008), as well as E3s that are recruited to the mitochon-
drial outer membrane, such as the SCF, will be involved in
this process. As with ERAD there is likely to be a high
degree of complexity, and we predict that a number of
mitochondrial and cytosolic proteins will be implicated in
playing roles either in protecting proteins or facilitating their
targeting to the UPS and in retro-translocation from mito-
chondrial membranes through as yet to be established mech-
anisms. With definitive proof for involvement of the UPS at
the mitochondria now established, a number of exciting
questions arise including how individual substrates are rec-
ognized; the extent to which the UPS might be involved in
the fate of proteins in the intramembranous space, the inner
mitochondrial membrane, and the mitochondrial matrix;
and the degree to which MAD targets misfolded proteins as
well as highly regulated normal proteins involved in critical
mitochondrial functions such as fusion and fission.
We are grateful to Drs. Abdussalam Azem (Tel Aviv University), Jodi Nun-
nari (University of California, Davis), Maurits Kleijnen (Harvard Medical
School), and Benedikt Westermann (University of Bayreuth, Germany) for
their generous gifts of antibodies, expression plasmids, and yeast strains. We
thank Dr. Zlatka Kostova, other members of the LPDS, and Dr. Catherine
Dargemont for invaluable discussions and comments. This research was
supported by the Intramural Research Program of the National Institutes of
Health, National Cancer Institute, Center for Cancer Research, and a grant
from the USA-Israel Binational Science Foundation (BSF) to M.H.G. and
Altmann, K., and Westermann, B. (2005). Role of essential genes in mitochon-
drial morphogenesis in Saccharomyces cerevisiae. Mol. Biol. Cell 16, 5410–5417.
Avaro, S., Belgareh-Touze, N., Sibella-Arguelles, C., Volland, C., and
Haguenauer-Tsapis, R. (2002). Mutants defective in secretory/vacuolar path-
ways in the EUROFAN collection of yeast disruptants. Yeast 19, 351–371.
Bleazard, W., McCaffery, J. M., King, E. J., Bale, S., Mozdy, A., Tieu, Q.,
Nunnari, J., and Shaw, J. M. (1999). The dynamin-related GTPase Dnm1
regulates mitochondrial fission in yeast. Nat. Cell Biol. 1, 298–304.
Cardozo, T., and Pagano, M. (2004). The SCF ubiquitin ligase: insights into a
molecular machine. Nat. Rev. Mol. Cell Biol. 5, 739–751.
Chan, D. C. (2006). Mitochondria: dynamic organelles in disease, aging, and
development. Cell 125, 1241–1252.
Dürr, M., Escobar-Henriques, M., Merz, S., Geimer, S., Langer, T., and
Westermann, B. (2006). Nonredundant roles of mitochondria-associated F-box
proteins Mfb1 and Mdm30 in maintenance of mitochondrial morphology in
yeast. Mol. Biol. Cell 17, 3745–3755.
Escobar-Henriques, M., Westermann, B., and 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.
Fang, S., and Weissman, A. M. (2004). A field guide to ubiquitylation. Cell
Mol. Life Sci. 61, 1546–1561.
Fisk, H. A., and Yaffe, M. P. (1999). A role for ubiquitination in mitochondrial
inheritance in Saccharomyces cerevisiae. J. Cell Biol. 145, 1199–1208.
Frescas, D., Guardavaccaro, D., Bassermann, F., Koyama-Nasu, R., and
Pagano, M. (2007). JHDM1B/FBXL10 is a nucleolar protein that represses
transcription of ribosomal RNA genes. Nature 450, 309–313.
Ubiquitin-dependent Mitochondrial Fusion
Vol. 19, June 20082463
Fritz, S., Weinbach, N., and Westermann, B. (2003). Mdm30 is an F-box Download full-text
protein required for maintenance of fusion-competent mitochondria in yeast.
Mol. Biol. Cell 14, 2303–2313.
Galan, J. M., and Haguenauer-Tsapis, R. (1997). Ubiquitin lys63 is involved in
ubiquitination of a yeast plasma membrane protein. EMBO J. 16, 5847–5854.
Galan, J. M., and Peter, M. (1999). Ubiquitin-dependent degradation of mul-
tiple F-box proteins by an autocatalytic mechanism. Proc. Natl. Acad. Sci.
USA 96, 9124–9129.
Galan, J. M., Wiederkehr, A., Seol, J. H., Haguenauer-Tsapis, R., Deshaies,
R. J., Riezman, H., and Peter, M. (2001). Skp1p and the F-box protein Rcy1p
form a non-SCF complex involved in recycling of the SNARE Snc1p in yeast.
Mol. Cell. Biol. 21, 3105–3117.
Glickman, M. H., and Ciechanover, A. (2002). The ubiquitin-proteasome
proteolytic pathway: destruction for the sake of construction. Physiol. Rev. 82,
Heath-Engel, H. M., and Shore, G. C. (2006). Mitochondrial membrane dy-
namics, cristae remodelling and apoptosis. Biochim. Biophys. Acta 1763,
Hermann, G. J., Thatcher, J. W., Mills, J. P., Hales, K. G., Fuller, M. T.,
Nunnari, J., and Shaw, J. M. (1998). Mitochondrial fusion in yeast requires the
transmembrane GTPase Fzo1p. J. Cell Biol. 143, 359–373.
Hitchcock, A. L., Auld, K., Gygi, S. P., and Silver, P. A. (2003). A subset of
membrane-associated proteins is ubiquitinated in response to mutations in
the endoplasmic reticulum degradation machinery. Proc. Natl. Acad. Sci.
USA 100, 12735–12740.
Hoppins, S., Lackner, L., and Nunnari, J. (2007). The machines that divide and
fuse mitochondria. Annu. Rev. Biochem. 76, 751–780.
Karbowski, M., Neutzner, A., and Youle, R. J. (2007). The mitochondrial E3
ubiquitin ligase MARCH5 is required for Drp1 dependent mitochondrial
division. J. Cell Biol. 178, 71–84.
Li, W., Bengtson, M. H., Ulbrich, A., Matsuda, A., Reddy, V. A., Orth, A.,
Chanda, S. K., Batalov, S., and Joazeiro, C. A. (2008). Genome-wide and
functional annotation of human E3 ubiquitin ligases identifies MULAN, a
mitochondrial E3 that regulates the organelle’s dynamics and signaling. PLoS
ONE 3, e1487.
Longtine, M. S., McKenzie, A., 3rd, Demarini, D. J., Shah, N. G., Wach, A.,
Brachat, A., Philippsen, P., and Pringle, J. R. (1998). Additional modules for
versatile and economical PCR-based gene deletion and modification in Sac-
charomyces cerevisiae. Yeast 14, 953–961.
Martinou, J. C., and Youle, R. J. (2006). Which came first, the cytochrome c
release or the mitochondrial fission? Cell Death Differ. 13, 1291–1295.
Meisinger, C., Sommer, T., and Pfanner, N. (2000). Purification of Saccharo-
myces cerevisiae mitochondria devoid of microsomal and cytosolic contamina-
tions. Anal. Biochem. 287, 339–342.
Moreau, V., Galan, J. M., Devilliers, G., Haguenauer-Tsapis, R., and Winsor, B.
(1997). The yeast actin-related protein Arp2p is required for the internaliza-
tion step of endocytosis. Mol. Biol. Cell 8, 1361–1375.
Mukhopadhyay, D., and Riezman, H. (2007). Proteasome-independent func-
tions of ubiquitin in endocytosis and signaling. Science 315, 201–205.
Muratani, M., Kung, C., Shokat, K. M., and 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.
Nakamura, N., Kimura, Y., Tokuda, M., Honda, S., and Hirose, S. (2006).
MARCH-V is a novel mitofusin 2- and Drp1-binding protein able to change
mitochondrial morphology. EMBO Rep. 7, 1019–1022.
Neutzner, A., and Youle, R. J. (2005). Instability of the mitofusin Fzo1 regu-
lates mitochondrial morphology during the mating response of the yeast
Saccharomyces cerevisiae. J. Biol. Chem. 280, 18598–18603.
Okamoto, K., and Shaw, J. M. (2005). Mitochondrial morphology and dynam-
ics in yeast and multicellular eukaryotes. Annu. Rev. Genet. 39, 503–536.
Peng, J., Schwartz, D., Elias, J. E., Thoreen, C. C., Cheng, D., Marsischky, G.,
Roelofs, J., Finley, D., and Gygi, S. P. (2003). A proteomics approach to
understanding protein ubiquitination. Nat. Biotechnol. 21, 921–926.
Petroski, M. D., and Deshaies, R. J. (2005). Function and regulation of cullin-
RING ubiquitin ligases. Nat. Rev. Mol. Cell Biol. 6, 9–20.
Rapaport, D., Brunner, M., Neupert, W., and Westermann, B. (1998). Fzo1p is
a mitochondrial outer membrane protein essential for the biogenesis of func-
tional mitochondria in Saccharomyces cerevisiae. J. Biol. Chem. 273, 20150–
Rinaldi, T., Pick, E., Gambadoro, A., Zilli, S., Maytal-Kivity, V., Frontali, L.,
and Glickman, M. H. (2004). Participation of the proteasomal lid subunit
Rpn11 in mitochondrial morphology and function is mapped to a distinct
C-terminal domain. Biochem. J. 381, 275–285.
Scheckhuber, C. Q., Erjavec, N., Tinazli, A., Hamann, A., Nystrom, T., and
Osiewacz, H. D. (2007). Reducing mitochondrial fission results in increased
life span and fitness of two fungal ageing models. Nat. Cell Biol. 9, 99–105.
Sesaki, H., and Jensen, R. E. (1999). Division versus fusion: Dnm1p and Fzo1p
antagonistically regulate mitochondrial shape. J. Cell Biol. 147, 699–706.
Sherman, F., Fink, G. R., and Hicks, J. B. (1986). Methods in Yeast Genetics, ed.
C.S.H., Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1986.
Sutovsky, P., Moreno, R. D., Ramalho-Santos, J., Dominko, T., Simerly, C., and
Schatten, G. (1999). Ubiquitin tag for sperm mitochondria. Nature 402, 371–
Thompson, W. E., Ramalho-Santos, J., and Sutovsky, P. (2003). Ubiquitination
of prohibitin in mammalian sperm mitochondria: possible roles in the regu-
lation of mitochondrial inheritance and sperm quality control. Biol. Reprod.
Westermann, B., and Neupert, W. (2000). Mitochondria-targeted green fluo-
rescent proteins: convenient tools for the study of organelle biogenesis in
Saccharomyces cerevisiae. Yeast 16, 1421–1427.
Willems, A. R., Schwab, M., and Tyers, M. (2004). A hitchhiker’s guide to the
cullin ubiquitin ligases: SCF and its kin. Biochim. Biophys. Acta 1695, 133–170.
Yonashiro, R. et al. (2006). A novel mitochondrial ubiquitin ligase plays a
critical role in mitochondrial dynamics. EMBO J. 25, 3618–3626.
M.M.J. Cohen et al.
Molecular Biology of the Cell 2464