In an oxa1D mutant, cellular and mitochondrial levels of phosphatidylethanolamine were lowered similar to a mutant with PSD1 deleted, and the rate of phosphatidylethanolamine synthesis b
Trang 1is affected by the mitochondrial components Oxa1p and Yme1p
Ruth Nebauer1, Irmgard Schuiki1, Birgit Kulterer2, Zlatko Trajanoski2and Gu¨nther Daum1
1 Institute of Biochemistry, Graz University of Technology, Austria
2 Institute for Genomics and Bioinformatics and Christian-Doppler Laboratory for Genomics and Bioinformatics,
Graz University of Technology, Austria
Phosphatidylserine decarboxylases (PSDs) catalyze the
formation of phosphatidylethanolamine (PtdEtn) from
phosphatidylserine (PtdSer) These enzymes play a key
role in phospholipid metabolism from bacteria to
humans In the yeast Saccharomyces cerevisiae there
are two different PSDs, Psd1p, which is associated
with the inner mitochondrial membrane (IMM) [1],
and Psd2p, which is a component of the Golgi [2]
Unlike bacteria, yeast and other eukaryotes can also
synthesize PtdEtn via a pathway that is independent
of PSDs and uses cytidine diphosphate-ethanolamine
and diacylglycerol as substrates [3,4]
PtdEtn is an essential component of yeast mitochon-drial membranes Depletion of PtdEtn in mitochondria leads to dysfunctions in respiration, defects in the assembly of mitochondrial protein complexes and loss
of mitochondrial DNA [5–7] Deletion of the major PtdEtn-synthesizing enzyme, Psd1p, causes a substan-tial decrease in PtdEtn in cellular and mitochondrial membranes, thereby conferring a petite phenotype characterized by a loss of respiratory capacity [5] The link between cell respiration and PtdEtn homeostasis
in mitochondria tempted us to speculate that: (a) other defects resulting in the depletion of mitochondrial
Keywords
mitochondria; Oxa1p;
phosphatidylethanolamine;
phosphatidylserine decarboxylase; yeast
Correspondence
G Daum, Institute of Biochemistry, Graz
A-8010 Graz, Austria
Fax: +43 316 873 6952
Tel: +43 316 873 6462
E-mail: guenther.daum@tugraz.at
(Received 27 August 2007, revised 10
Octo-ber 2007, accepted 11 OctoOcto-ber 2007)
doi:10.1111/j.1742-4658.2007.06138.x
The majority of phosphatidylethanolamine, an essential component of yeast mitochondria, is synthesized by phosphatidylserine decarboxylase 1 (Psd1p), a component of the inner mitochondrial membrane Here, we report that deletion of OXA1 encoding an inner mitochondrial membrane protein translocase markedly affects the mitochondrial phosphatidyletha-nolamine level In an oxa1D mutant, cellular and mitochondrial levels of phosphatidylethanolamine were lowered similar to a mutant with PSD1 deleted, and the rate of phosphatidylethanolamine synthesis by decarboxyl-ation of phosphatidylserine in vivo and in vitro was decreased This was due to a lower PSD1 transcription rate in the oxa1D mutant compared with wild-type and compromised assembly of Psd1p into the inner mito-chondrial membrane Lack of Mba1p, another component involved in the assembly of mitochondrial proteins into the inner mitochondrial mem-brane, did not affect the amount of phosphatidylethanolamine or the assembly of Psd1p Deletion of the inner membrane protease Yme1p enhanced Psd1p stability suggesting that Yme1p contributed substantially
to the proteolytic turnover of Psd1p in wild-type In summary, our results demonstrate a link between the mitochondrial protein import machinery, assembly and stability of Psd1p, and phosphatidylethanolamine homeo-stasis in yeast mitochondria
Abbreviations
IMM, inner mitochondrial membrane; PSD, phosphatidylserine decarboxylase; PtdCho, phosphatidylcholine; PtdEtn,
phosphatidylethanolamine; PtdIns, phosphatidylinositol; PtdSer, phosphatidylserine.
Trang 2PtdEtn may also cause the petite phenotype, and⁄ or
(b) petite mutations may generally affect the formation
of mitochondrial PtdEtn Based on this hypothesis, we
screened a yeast petite mutant collection [8] for strains
with abnormal phospholipid patterns Among the
candidate strains identified (R Nebauer, unpublished
results), a mutant with OXA1 deleted exhibited marked
PtdEtn depletion
Oxa1p is a polypeptide involved in the insertion of
mitochondrially encoded proteins into the IMM, but it
also mediates the assembly of nuclear-encoded proteins
into this submitochondrial fraction [9] The import of
proteins synthesized on cytoplasmic ribosomes into
mitochondria starts with translocation across the outer
mitochondrial membrane, mediated by a general
import machinery, the translocase of the outer
mem-brane complex Assembly of polypeptides into the
IMM requires an energized IMM and another
trans-location machinery, the translocase of the inner
membrane complex [9–12] IMM proteins are targeted
to mitochondria by N-terminal targeting signals,
imported into the mitochondrial matrix and sorted to
the IMM via a specific export pathway [9,13–15]
Proteins with their N-termini protruding into the
inter-membrane space attain their inter-membrane orientation by
physical interaction with Oxa1p [16], although the
function of this protein is not limited to proteins that
undergo N-terminal tail export Recently, Mba1p was
identified as a protein that interacts with the Oxa1p
insertion machinery of the IMM [17] Mba1p binds to
the large subunit of mitochondrial ribosomes and
thereby cooperates with the C-terminal
ribosome-bind-ing domain of Oxa1p to ensure proper insertion of
proteins into the IMM
Like the majority of mitochondrial proteins, the
mitochondrial PtdSer decarboxylase Psd1p is encoded
by a nuclear gene, synthesized as a larger precursor on
cytoplasmic ribosomes and imported
post-translation-ally into mitochondria [18] As indicated in the
Uni-Prot knowledge base (http://www.uniprot.org⁄ ), the
yeast Psd1p proenzyme has one potential
mitochon-drial targeting sequence and an a-chain and b-chain
linked by a defined cleavage site [19] According to
von Heijne [20] or applications available at ExPASy
(http://www.expasy.org/) [21], Psd1p localized to the
IMM [1] has at least one transmembrane domain The
N-terminus of Psd1p contains motifs for protein
tar-geting to mitochondria and specifically to the IMM⁄
intermembrane space [18,22]
In this study, we analyzed the roles of Oxa1p,
Mba1p and the IMM protease Yme1p in the
forma-tion of PtdEtn by Psd1p We demonstrate that in an
oxa1D mutant inefficient assembly of Psd1p into the
IMM leads to decreased PtdEtn levels in yeast mito-chondria No such effect could be observed in an mba1D strain Moreover, we show that lack of the IMM protease Yme1p prevents degradation of Psd1p resulting in partial protection of its enzymatic activity Thus, specific components of the mitochondrial bio-synthetic machinery indirectly affect phospholipid homeostasis in this organelle
Results
The oxa1D mutant has an abnormal phospholipid composition
Screening of a set of petite (respiratory-deficient) yeast strains [8,23] for defects in the PtdEtn and phosphati-dylcholine (PtdCho) biosynthetic pathways revealed a number of candidate genes whose deletion caused changes in the amounts of at least one of the major phospholipids PtdCho, PtdEtn, and⁄ or phosphatidyl-inositol (PtdIns) in the cell homogenate and⁄ or mito-chondria (R Nebauer, unpublished data) One of these strains exhibiting decreased cellular PtdEtn levels com-pared with wild-type was the oxa1D mutant (Table 1), which is known to bear a defect in protein transloca-tion from the mitochondrial matrix to the IMM (see above) Fluorescence microscopic inspection employing DAPI staining revealed mitochondrial DNA in wild-type and oxa1D The amount of mitochondrial DNA appeared to be lower in oxa1D than in wild-type Thus, the petite phenotype of the mutant was not caused by a rho-mutation
The decrease in cellular PtdEtn in oxa1D was com-pensated by increased amounts of PtdIns, and also of lysophospholipids, phosphatidic acid and, to a lesser extent, PtdCho (Table 1) The decrease and compensa-tion in oxa1D were similar to psd1D, which lacks the major enzyme of cellular PtdEtn formation, mitochon-drial Psd1p In oxa1D mitochondria, the effect of PtdEtn depletion was even more pronounced than in total cell extracts Depletion of mitochondrial PtdEtn
in oxa1D was mainly compensated by an increase in PtdIns and, to a lesser extent, PtdCho Although the decrease in PtdEtn in oxa1D mitochondria was compa-rable with that in psd1D, there was a difference in the amount of mitochondrial PtdSer in these two strains
In psd1D, PtdSer imported into mitochondria from the endoplasmic reticulum [3] was not further converted to PtdEtn and accumulated in this organelle to some extent, whereas no such accumulation was observed with oxa1D Lack of such an accumulation appears to
be due to residual Psd1p activity in oxa1D mitochon-dria, as shown below
Trang 3During our studies of oxa1D-dependent PtdEtn depletion in yeast mitochondria we also investigated the effect on PtdEtn homeostasis of other yeast gene products that are related or linked to the Oxa1p-dependent protein translocation machinery These strains were mutants of MBA1, which encodes a com-ponent involved in the Oxa1p-dependent export of mitochondrially encoded proteins into the IMM [17], and YME1, which encodes an intermembrane space-located ATP-dependent AAA protease (ATPase associ-ated with various cellular activities) [24] The mba1D deletion strain exhibited only a slight decrease in total cellular PtdEtn and essentially the same mitochondrial phospholipid pattern as wild-type (Table 1) The yme1D and yme1D oxa1D mutants contained cellular and mitochondrial amounts of PtdEtn (Table 1) that exceeded wild-type levels
Deletion of OXA1 affects the rate of PtdEtn synthesis by Psd1p in vivo
Because mitochondrial Psd1p is the major producer of cellular PtdEtn, we hypothesized that the decrease in total cellular and mitochondrial PtdEtn levels in the oxa1D mutant were due to reduced activity of this enzyme To test this hypothesis, we performed in vivo experiments labeling PtdSer with [3H]serine and fol-lowed its conversion to PtdEtn and PtdCho in a time-dependent manner (see Experimental procedures) All strains tested showed a linear increase in the formation
of the three aminoglycerophospholipids within the selected timeframe, which enabled us to determine the rate of formation, i.e the incorporation of radiolabel per period, for each phospholipid The formation rates for PtdSer, PtdEtn and PtdCho in wild-type cells were set at 100%, and the corresponding rates for mutant strains were calculated accordingly As can be seen from Fig 1, deletion of OXA1 decreased the rate of formation of all aminoglycerophospholipids The rate
of PtdSer synthesis decreased to 80%, the rate of PtdEtn formation to 70% and that of PtdCho synthesis
to 60% of wild-type Because Oxa1p was assumed to compromise only the mitochondrial PtdEtn-synthesiz-ing Psd1p, leavPtdEtn-synthesiz-ing the Golgi-located Psd2p unaffected, the decrease in the rate of PtdEtn synthesis in oxa1D confirmed a defect in Psd1p-dependent PtdEtn forma-tion Under these circumstances, the decreased rate of PtdCho formation seemed to be due to the lowered rate
of PtdEtn formation, whereas reduced PtdSer forma-tion might reflect a response to a feedback regulatory mechanism It should be noted that the steady-state lev-els of individual phospholipids do not necessarily reflect the rates of synthesis of the components
Trang 4In contrast to psd1D, however, the oxa1D mutation
led to a smaller reduction of PtdEtn synthesis That
the psd1D strain and the oxa1D psd1D double mutant
had comparable rates of PtdEtn formation suggests
that Oxa1p acted upstream of Psd1p Not
unexpect-edly, rates of PtdEtn formation in the oxa1D psd2D
mutant were lower than in the psd2D mutant indicating
an additive effect of these two mutations acting on
two different pathways Taken together, deletion of
OXA1 affected synthesis of PtdEtn by Psd1p, but did
not completely abolish the activity of this enzyme
Activity of Psd1p in vitro is impaired in oxa1D
PtdEtn depletion in mitochondria and the decreased
rate of Psd1p-dependent PtdEtn formation in oxa1D
suggested a functional impairment of mitochondrial
Psd1p To address the question of Psd1p enzyme
activ-ity we subjected subcellular fractions of an oxa1D
mutant to enzymatic analyses As can be seen from
Fig 2, the in vitro activity of Psd1p with oxa1D
chondria was only 60% that of wild-type In
mito-chondria from the psd1D mutant there was no
measurable Psd1p activity (data not shown) Psd1p
activity in mitochondria from mba1D was not
decreased, in line with the unchanged mitochondrial
level of PtdEtn in this strain (Table 1)
Studies on the stability of subunits of the
mitochon-drial membrane complexes Cox and ATPase revealed
that these proteins are degraded in the absence of
Oxa1p [25] When functional Oxa1p is missing the
membrane subunits of these complexes cannot be
assembled and are cleaved by the intermembrane space
(i)-AAA protease Yme1p and⁄ or by the matrix
(m)-AAA protease Afg3p⁄ Yta12p To test whether Psd1p
stability was also affected by the presence or absence
of these mitochondrial hydrolases, we analyzed Psd1p activity in the respective single mutants or in double mutants in combination with oxa1D Deletion of YME1encoding the i-AAA protease led to a consider-able increase in Psd1p activity (Fig 2), which is in line with the increased PtdEtn level in a deletion mutant compared with wild-type (Table 1) This observation was surprising because overexpression of the PtdEtn biosynthetic pathway enzymes phosphatidylserine syn-thase 1 (Pss1p) and⁄ or Psd1p did not change the PtdEtn level (R Birner-Gruenberger, unpublished results) The yme1D oxa1D double mutant showed an intermediate value for the Psd1p activities from the single mutants Yme1p appears to contribute markedly
Fig 1 Deletion of OXA1 causes a
decreased rate of PtdEtn synthesis in vivo.
Wild-type and mutant strains were labeled
Incorporation of label into PtdSer, PtdEtn
and PtdCho was determined by
liquid-scintil-lation counting after separation of
phospho-lipids by TLC (see Experimental procedures).
The formation rate of PtdSer, PtdEtn and
PtdCho of wild-type (black bars) was set at
100% Values are means from three
inde-pendent experiments with mean deviations
as indicated by the error bars.
Fig 2 The oxa1D mutation affects Psd1p activity in vitro Enzy-matic assays were performed with isolated mitochondrial fractions from wild-type BY4742, mba1D, oxa1D, yme1D and yme1D oxa1D Values are expressed relative to wild-type which was set at 100% and are means from three independent experiments with mean deviations as indicated by the error bars.
Trang 5to the proteolytic turnover of Psd1p Deletions of
either subunit of the m-AAA protease yta12D and
afg3D, respectively, seemed to have no effects on
Psd1p turnover (data not shown) and were not
investi-gated further
A decreased transcription rate for PSD1 and a
defect in Psd1p maturation are the molecular
basis of the decreased rate of PtdSer
decarboxylation in oxa1D mitochondria
One obvious explanation for the decreased amount of
Psd1p in oxa1D mitochondria was a possible reduction
in the PSD1 transcription rate in the mutant To
address this question we performed RT-PCR analyses
of PSD1 mRNA with wild-type and oxa1D (see
Exper-imental procedures) These analyses revealed a
reduc-tion in PSD1 mRNA in the mutant The transcripreduc-tion
rate for PSD1 was repressed in oxa1D to 50% that
of wild-type Thus, downregulation of PSD1
expres-sion at the transcriptional level appears to be one
reason for the decreased Psd1p activity in oxa1D
Because Oxa1p had been shown to facilitate
mem-brane assembly in several mitochondrial proteins (see
above), it was tempting to speculate that it was also
necessary for correct insertion of Psd1p into the IMM
To test this hypothesis, we performed import
experi-ments of radioactively labeled Psd1p into isolated
mitochondria These in vitro assays (see Experimental
procedures) allowed analysis of protein assembly into
mitochondrial membranes independent of the
tran-scriptional level of a respective gene The full-length
precursor form of Psd1p was synthesized by a coupled
transcription⁄ translation reaction and incubated with
wild-type and oxa1D mitochondria Complete
process-ing of Psd1p occurred in three proteolytic steps
(Fig 3A) The primary translation product of 57 kDa
was cleaved to a first intermediate of 52 kDa, most
likely during or immediately after the import process
This cleavage step is in agreement with the finding that
a positively charged amino acid stretch at the
N-termi-nus of Psd1p serves as a mitochondrial targeting
sequence Processing of Psd1p was continued by
cleav-age of a 2 kDa fragment representing the
inter-membrane space sorting signal, yielding the second
intermediate of 50 kDa Assembly of Psd1p into the
IMM was completed by (autocatalytic) cleavage of the
50 kDa intermediate to one a-chain and one b-chain
(4 and 46 kDa mature forms) The 4 kDa a-subunit
was not detected in electrophoretic analysis
In oxa1D (Fig 3B), import and processing of Psd1p
occurred more slowly than in wild-type resulting in
a lower ratio of mature form to precursors Thus,
deletion of OXA1 decreased both the transcription rate
of PSD1 and the Psd1p assembly rate into the IMM Both effects appear to result in a reduced amount of enzymatically active Psd1p in the IMM and thus in a decreased capacity to form PtdEtn
Discussion
The biosynthetic scheme shown in Fig 4 summarizes the possible ways in which the mitochondrial level of PtdEtn can be affected First, is the supply of PtdSer
to the mitochondria as a precursor for PtdEtn forma-tion by Psd1p This process includes synthesis of Ptd-Ser in the endoplasmic reticulum by the PtdPtd-Ser synthase Pss1p and translocation of PtdSer to the site
of Psd1p-catalyzed decarboxylation in the IMM Sec-ond, mitochondrial factors may, directly or indirectly,
A
B
Fig 3 Proteolytic processing of Psd1p Maturation of Psd1p was measured in wild-type BY4742 (A) and oxa1D (B) The primary translation product of 57 kDa (not shown in the diagram) was cleaved to a 52 kDa intermediate (d), which was further processed
to yield a 50 kDa polypeptide (h) The final processing step leads
to the formation of the mature 46 kDa b-subunit of Psd1p (*) For each time point, the amount of every single processing intermedi-ate was expressed as percent of the sum of all intermediintermedi-ates Values are means from three independent experiments with mean deviations as indicated by the error bars.
Trang 6affect the activity of Psd1p, thereby decreasing or
increasing the efficiency of mitochondrial PtdEtn
for-mation Third, import and export of PtdEtn may
con-tribute to a balance in the level of this phospholipid in
mitochondria Finally, although not addressed
specifi-cally in this scheme, transcriptional⁄ translational
regu-lation of PSD1 expression has to be taken into
account
Similar to plants [26], increased levels of yeast
mito-chondrial Psd1p are not necessarily accompanied by
an increase in the amount of mitochondrial PtdEtn In
strains overexpressing Pss1p and⁄ or Psd1p neither the
PtdSer nor the PtdEtn level was markedly changed
compared with wild-type (R Birner-Gruenberger,
unpublished data) These findings imply that the
amount of mitochondrial PtdEtn is tightly controlled
by one of the above-mentioned regulatory mechanisms
Alternatively, the wild-type level of Psd1p may already
represent an excess of activity which cannot be
enhanced further by increasing the amount of protein
A search for components affecting mitochondrial
PtdEtn levels led to the identification of mitochondrial
components interacting directly with
mitochon-drial Psd1p One example of such a component is the
mitochondrial prohibitin, Phb1p⁄ Phb2p Recent studies
in our laboratory demonstrated the synthetic lethality
of a psd1D phb1⁄ 2D double mutant [7] It was
specu-lated that the decreased PtdEtn level in mitochondria
caused by psd1D might be harmful in the phb1⁄ 2D
back-ground, which by itself causes an increase in
mitochon-drial PtdEtn In view of the results of this study, this
hypothesis appears to be wrong, because depletion of
the mitochondrial PtdEtn level by oxa1D to an amount
comparable with that in psd1D did not lead to synthetic
lethality with phb1⁄ 2D (R Nebauer, unpublished
results) Thus, it is the direct interaction of Psd1p
and Phb1⁄ 2p or even a more complex effect through
combination of the two gene products that may be important for mitochondrial function
In this study, we demonstrate another mode of action that affects Psd1p activity in yeast mitochon-dria, namely disturbance of the import and assembly
of this polypeptide into mitochondrial membranes We show that Oxa1p facilitates the import of Psd1p to its proper destination in the IMM Oxa1p has been char-acterized previously as a helper protein for the assembly of a number of other IMM proteins [9]
In wild-type yeast cells, import into mitochondria, pro-cessing and assembly into mitochondrial membranes of Psd1p is accomplished by a three-step mechanism simi-lar to Chinese hamster ovary cells [27] According to Boeckmann et al [28], Psd1p contains all the features
of a typical IMM⁄ intermembrane space protein, namely a positively charged N-terminal sequence fol-lowed by a hydrophobic stretch The three cleavage steps are accomplished by the mitochondrial-process-ing peptidase (MPP), the intermembrane space prote-ase Imp1p and autocatalysis
In oxa1D, the Psd1p processing rate was decreased (Fig 3) This resulted in slower utilization of the pre-cursor polypeptide in the mutant than in wild-type, delayed formation of intermediates and finally a decreased appearance of the mature form Although only one intermediate step in the process, namely translocation of the 50 kDa intermediate to the IMM, appears to be directly affected by Oxa1p, the whole process of Psd1p assembly into the IMM occurs more slowly in the mutant than in the wild-type The resid-ual Psd1p activity in oxa1D appears to be due to alter-native import pathways
In addition to the reduced rate of Psd1p import into mitochondria, the decreased transcription rate of PSD1 in oxa1D seems to play a role in imbalanced PtdEtn formation of the mutant We can only
specu-Fig 4 Factors affecting the PtdEtn level in mitochondria Pss1p (phosphatidylserine synthase 1), Psd1p (phosphatidylserine decarboxylase 1), Psd2p (phosphatidylserine decarboxylase 2), Dpl1p (dihydrosphingosine 1-phosphate lyase 1), import of PtdSer into mitochondria (x), export of PtdEtn from mitochondria (y), import of PtdEtn into mitochondria (z) and factors (F) affecting level and activity of Psd1p in mito-chondria may contribute to the mitomito-chondrial PtdEtn.
Trang 7late at present that a negative feedback control caused
by unassembled Psd1p precursor or intermediate
pro-teins might trigger this transcriptional regulation
However, the additive effects of reduced PSD1
tran-scription and Psd1p assembly are sufficient to cause a
limitation of active Psd1p being present in
mitochon-dria of oxa1D
Another component that affects the mitochondrial
level of Psd1p activity is the intermembrane space
protease Yme1p In a yme1D strain, Psd1p activity
exceeded the wild-type level, and in the oxa1D
back-ground, yme1D restored Psd1p activity to a higher
level than wild-type Under the latter conditions,
Oxa1p-independent insertion of Psd1p seems to be
suf-ficient to ensure assembly of a functional enzyme
exhibiting activity higher than wild-type We assume
from these results that Yme1p contributes to Psd1p
degradation and turnover In a yme1D strain, an excess
of Psd1p appears to accumulate in the IMM leading
to the observed effects of enhanced enzyme activity
and increased PtdEtn levels
In summary, our results demonstrate a link between
the mitochondrial machinery of protein assembly and
PtdEtn homeostasis in mitochondria and the whole
cell We have to keep in mind, however, that depletion
of mitochondrial PtdEtn by the various possible effects
described appears to negatively affect proteins involved
in mitochondrial function or membrane properties and
may thus contribute to a petite phenotype (respiratory
defect) By contrast, it should be noted that not all
respiratory defects of mitochondria need to be linked
to lipid defects in mitochondrial membranes as
docu-mented by a recent screening of petite strains in our
laboratory (R Nebauer, unpublished results) Rather
it appears that Psd1p-dependent PtdEtn formation is
affected by a distinct set of mitochondrial proteins,
e.g Oxa1p, which are involved in the correct assembly
of Psd1p into the IMM
Experimental procedures
Strains and culture conditions The yeast strains used in this study are listed in Table 2 Yeast mutants exhibiting a petite phenotype as described by Dimmer et al [8] were obtained from the Euroscarf strain collection (Frankfurt, Germany) S cerevisiae strains were grown under aerobic conditions at 30C on YPD medium containing 1% yeast extract, 2% peptone, and 2% glucose
as the carbon source For large scale cultivation, inocula-tions to a D600 of 0.1 in fresh medium were made by diluting precultures grown to the stationary phase For auxotrophy tests, yeast strains were cultivated on solid syn-thetic medium [29]
Plasmid and strain constructions Primers used in this study are listed in Table 3 The yeast deletion mutants oxa1D::His3MX6 and psd1D::His3MX6 were constructed as described by Longtine et al [30] Prim-ers OXA1-F1 and OXA1-R1 or PSD1-F1 and PSD2-F2, respectively, were used to amplify the His3MX6 disruption cassette The cassette was introduced into the respective strain by lithium acetate transformation [31] Correct inser-tion of the cassette was tested by growing strains on selec-tive media without the respecselec-tive amino acid and by colony PCR with the appropriate primers Double-deletion mutants were constructed by mating the corresponding single-deletion mutants, sporulation of zygotes, and tetrad dissection using standard methods Identity of strains was confirmed by marker-dependent growth and colony PCR
Fluorescence microscopy Visualization of mitochondrial DNA in living cells was per-formed using the fluorescent dye DAPI In brief, cells were grown in YPD medium over night at 30C An inoculation
to a D600of 0.3 in fresh medium was made by diluting of
Table 2 Yeast strains used in this study.
Trang 8an overnight culture and cells were harvested in the mid-log
phase DNA was stained with 2.5 lgÆmL)1 of DAPI
dis-solved in NaCl⁄ Piat 30C for 30 min After staining, cells
were rinsed once with NaCl⁄ Pi and then resuspended in
NaCl⁄ Pi Suspensions were placed on a glass slide and
cov-ered with a cover slip Cells were then visualized using a
fluorescence microscope (Axiovert 35, Carl Zeiss, Jena,
Germany) with the appropriate filter set for the
blue-emit-ting fluorochrome DAPI and a 100-fold oil immersion
objective Mitochondrial DNA was visualized as smaller
spots distinct from larger nuclear DNA At least 100 cells
from all strains to be tested were inspected
Labeling of aminoglycerophospholipids in vivo
Labeling of aminoglycerophospholipids in vivo was
deter-mined by following the incorporation of [3H]serine into
PtdSer, PtdEtn and PtdCho as described by Birner et al
[7] For each time point, an equivalent of 10 D600 from
an overnight culture ( 1 mL, corresponding to
1.45· 108cells) was harvested, washed once, suspended in
500 lL YPD and incubated for 30 min at 30C Cells were
labeled with 10 lCi [3H]serine (27 CiÆmmol)1, Perkin–
Elmer, Boston, MA) per time point Samples were taken at
0, 15, 30 and 60 min, put on ice and harvested by
centrifu-gation Chloroform⁄ methanol (2 : 1, v ⁄ v) and glass beads,
3 mL each, were added to the cell pellets For
disintegra-tion of cells samples were shock frozen in liquid nitrogen
and shaken vigorously on an IKA Vibrax VXR for
15 min at 4C Then, lipids were extracted for 30 min by
the method of Folch et al [32] Individual phospholipids
were separated by TLC on Silica gel 60 plates (Merck,
Darmstadt, Germany) with chloroform⁄ methanol ⁄ 25%
ammonia (50 : 25 : 6, v⁄ v ⁄ v) as a developing solvent Spots
on TLC plates were stained with iodine vapor, scraped off
and suspended in 8 mL scintillation cocktail (Packard
Bio-Science, Groningen, the Netherlands) containing 5% water
Radioactivity was determined by liquid scintillation
count-ing using a Packard TriCarb Liquid Scintillation
Analyzer
Preparation of subcellular fractions, protein analysis, and enzymatic analysis
Mitochondria were prepared from spheroplasts by pub-lished procedures [1,33] Relative enrichment of markers and cross-contamination of subcellular fractions were assessed as described by Zinser and Daum [34] Protein was quantified by the method of Lowry et al [35] by using BSA
as a standard SDS–PAGE was carried out as published by Laemmli [36] Western blot analysis of proteins from subcellular fractions prepared as described above was performed as described by Haid and Suissa [37] Immunore-active bands were visualized by enzyme-linked immunosor-bent assay using a peroxidase-linked secondary antibody (Sigma-Aldrich, St Louis, MO) following the manufac-turer’s instructions
PtdSer decarboxylase activity was measured in isolated mitochondria from yeast cells grown in YPD to the loga-rithmic growth phase as reported by Kuchler et al [38] with minor modifications: 100 nmol [3H]PtdSer (specific activity
of 28 900 dpmÆnmol)1) was used as the substrate, and the assay was performed in 0.1 m Tris⁄ HCl, pH 7.2, containing
10 mm EDTA
Import of Psd1p into mitochondria in vitro Import, processing and assembly of Psd1p into mitochon-dria in vitro were assayed following the protocol of Ryan
et al [39] The precursor Psd1p was synthesized in the presence of [35S]methionine (15 mCiÆmL)1; Amersham Bio-sciences, Chalfont, UK) by coupled transcription⁄ translation in a reticulocyte lysate (Promega, Madison, WI) following the manufacturer’s instructions The T7 RNA polymerase system with a PCR-generated DNA fragment
as a template was employed Primers T1 and PSD1-U1 (see Table 2) were used to amplify PSD1 from genomic DNA Yeast mitochondria were isolated as described above and aliquoted at 10 mgÆmL)1 in SEM buffer containing
250 mm sucrose, 1 mm EDTA, 10 mm Mops-KOH, pH 7.2, and stored at )70 C The import assay involved
sequences are homologous to the His3MX6 disruption cassette (OXA1-F1, OXA1-R1, PSD1-F1, PSD1-R1) or to the PSD1 ORF (PSD1-T1) Primer PSD1-U1 is complementary to the region spanning the stop codon of the PSD1 ORF.
Trang 9tion of the radiolabeled Psd1p precursor with isolated
mito-chondria in the presence of NADH (1.8 mm) and ATP
(1.8 mm) in a buffer containing 3% (w⁄ v) fatty acid-free
BSA, 250 mm sucrose, 80 mm KCl, 5 mm MgCl2, 2 mm
KH2PO4, 5 mm methionine, 10 mm Mops-KOH, pH 7.2
[39] After 2, 5, 10, and 15 min samples were withdrawn
and put on ice in the presence of valinomycin (final
concen-tration 0.5 lm) to stop the import reaction Supernatants
were removed by centrifugation at 12 000 g and 4C for
5 min Pellets were washed once in SEM buffer, recovered
by centrifugation and suspended in SDS⁄ PAGE loading
buffer [36] prior to heating at 95C for 5 min Analysis of
radioactively labeled translation products, intermediates
and mature polypeptides was performed employing
stan-dard methods of SDS–PAGE, autoradiography and
densi-tometric scanning
Phospholipid quantification
For the analysis of total cellular phospholipids yeast cells
harvested from a 500 mL culture grown to the late
logarith-mic phase were disintegrated by shaking with glass beads in
a Merckenschlager homogenizer under CO2 cooling in the
presence of 10 mm Tris⁄ HCl, pH 7.2, and 1 mm
phenyl-methylsulfonyl fluoride (Calbiochem, La Jolla, CA) After
removal of the beads by centrifugation the supernatant
rep-resenting the total cell homogenate was aliquoted and
stored at )70 C Lipids from samples containing 3 mg
protein were extracted by the procedure of Folch et al [32]
using 4 mL chloroform⁄ methanol (2 : 1, v ⁄ v) Isolated
mitochondria (2 mg protein) were subjected to lipid
extrac-tion by the same method
Individual phospholipids were separated by 2D TLC
using chloroform⁄ methanol ⁄ 25% ammonia (70 : 35 : 5,
v⁄ v ⁄ v) as first, and chloroform ⁄ acetone ⁄ methanol ⁄ acetic
acid⁄ water (55 : 20 : 10 : 10 : 5, v⁄ v ⁄ v ⁄ v ⁄ v) as second
developing solvent Phospholipids were visualized on TLC
plates by staining with iodine vapor, scraped off and
quan-tified by the method of Broekhuyse [40]
RNA preparation and real-time PCR
Total RNA was isolated using phenol⁄ chloroform
extrac-tion as described previously [29] and further purified by
RQ1 RNase-free DNase (Promega) treatment according to
the manufacturer’s instructions and subsequent ethanol
pre-cipitation Integrity of RNA was tested by agarose gel
elec-trophoresis and determination of the 260 to 280 nm ratio
of the absorbencies RNA concentration was determined by
measurement of the absorbance at 260 nm
Total RNA was subjected to reverse transcription using
the SuperScriptTMII First Strand Synthesis System
(Invi-trogen, Carlsbad, CA) for real-time PCR (RT-PCR)
Possible traces of contaminating genomic DNA were
removed by DNAse I digestion In detail, 2.5 lg of RNA
with a concentration of 500 ngÆlL)1 were incubated with 10· DNAse I buffer, DNAse I amplification grade and
4 U RNaseOutTM ribonuclease for 15 min at room tem-perature (all reagents from Invitrogen) DNA digestion was stopped by adding 1 lL of EDTA (25 mm) and 2 lL
of H2O, incubating for 5 min at room temperature and further 5 min at 70C The DNase I treated RNA was mixed with 0.5 lg of oligo-dT2-18, 3 lg of random primers, and 4 U of RNaseOutTM ribonuclease, heated for 5 min
at 70C and left at room temperature for another 5 min The RNA sample was mixed with the cDNA synthesis mix, consisting of 5· RT buffer, dithiothreitol (0.1 m), dNTP (10 mm), 4 U RNaseOutTM and 200 U Super-ScriptTMII reverse transcriptase (all reagents from Invitro-gen), and heated to 45C for 1 h The reaction was stopped by heating to 95C for 5 min RT-PCR assays were performed using the Platinum SYBR Green Su-perMix-UDG (Invitrogen) following the manufacturer’s recommendations Primers for RT-PCR were designed using the software tool primer expressTM (ABI) For a
25 lL RT-PCR reaction, 1 lL of primer pair (800 nm), 0.5 lL of diluted cDNA (10 ngÆlL)1) and 12.5 lL of Plati-num SYBR Green SuperMix-UDG were applied In addition, no template controls (NTC) and no RT reaction (No RT) controls were performed The cycling conditions
on an ABI Prism 7000 were set for 2 min at 50C,
10 min at 95C and 40 cycles of 15 s at 95 C and 1 min
at 60C Data were analyzed using the ABI Prism 7000 sds software
Acknowledgements
The authors wish to thank A Hermetter for providing access to the fluorescence microscope (FWF instru-ment) This work was financially supported by the FWF (Fonds zur Fo¨rderung der wissenschaftlichen Forschung in O¨sterreich) projects 14468 and 17321 to GD
References
1 Zinser E, Sperka-Gottlieb CDM, Fasch E-V, Kohlwein
SD, Paltauf F & Daum G (1991) Phospholipid synthesis and lipid composition of subcellular membranes in the unicellular eukaryote Saccharomyces cerevisiae J Bacte-riol 173, 2026–2034
2 Trotter PJ & Voelker DR (1995) Identification of a non-mitochondrial phosphatidylserine decarboxylase activity in the yeast Saccharomyces cerevisiae J Biol Chem 270, 6062–6070
3 Nebauer R, Birner-Gru¨nberger R & Daum G (2003) Biogenesis and cellular dynamics of glycerophospho-lipids in the yeast Saccharomyces cerevisiae Topics Curr Genet 6, 125–168
Trang 104 Birner R & Daum G (2003) Biogenesis and cellular
dynamics of aminoglycerophospholipids Int Rev Cytol
225, 273–323
5 Birner R, Bu¨rgermeister M, Schneiter R & Daum G
(2001) Roles of phosphatidylethanolamine and of its
several biosynthetic pathways in Saccharomyces
cerevisi-ae Mol Biol Cell 12, 997–1007
6 Storey MK, Clay KL, Kutateladze T, Murphy RC,
Overduin M & Voelker DR (2001)
Phosphatidylethanol-amine has an essential role in Saccharomyces cerevisiae
that is independent of its ability to form hexagonal
phase structures J Biol Chem 276, 48539–48548
7 Birner R, Nebauer R, Schneiter R & Daum G (2003)
Synthetic lethal interaction of the mitochondrial
phos-phatidylethanolamine biosynthetic machinery with the
prohibitin complex of Saccharomyces cerevisiae Mol
Biol Cell 14, 370–383
8 Dimmer KS, Fritz S, Fuchs F, Messerschmitt M,
Weinbach N, Neupert W & Westermann B (2002)
Genetic basis of mitochondrial function and
morphol-ogy in Saccharomyces cerevisiae Mol Biol Cell 13,
847–853
9 Stuart RA (2002) Insertion of proteins into the inner
membrane of mitochondria: the role of the Oxa1
com-plex Biochim Biophys Acta 1592, 79–87
10 Neupert W (1997) Protein import into mitochondria
Annu Rev Biochem 66, 863–917
11 Pfanner N & Geissler A (2001) Versatility of the
mito-chondrial protein import machinery Nat Rev Mol Cell
Biol 2, 339–349
12 Rehling P, Pfanner N & Meisinger C (2003) Insertion of
hydrophobic membrane proteins into the inner
mitochon-drial membrane – a guided tour J Mol Biol 326, 639–657
13 Herrmann JM & Neupert W (2003) Protein insertion
into the inner membrane of mitochondria IUBMB Life
55, 219–225
14 Stuart RA & Neupert W (1996) Topogenesis of inner
membrane proteins of mitochondria Trends Biochem
Sci 21, 261–267
15 Herrmann JM, Neupert W & Stuart RA (1997)
Inser-tion into the mitochondrial inner membrane of a
poly-topic protein, the nuclear-encoded Oxa1p EMBO J 16,
2217–2226
16 Hell K, Herrmann JM, Pratje E, Neupert W & Stuart
RA (1998) Oxa1p, an essential component of the N-tail
protein export machinery in mitochondria Proc Natl
Acad Sci USA 95, 2250–2255
17 Ott M, Prestele M, Bauerschmitt H, Funes S, Bonnefoy
N & Herrmann J (2006) Mba1, a membrane-associated
ribosome receptor in mitochondria EMBO J 25, 1603–
1610
18 Voelker DR (1997) Phosphatidylserine decarboxylase
Biochimica Biophysica Acta 1348, 236–244
19 The Uniprot Consortium (2007) The Universal Protein
Resource (UniProt) Nucleic Acids Res 35, 193–197
20 von Heijne G (1992) Membrane protein structure pre-diction Hydrophobicity analysis and the positive-inside rule J Mol Biol 20, 487–494
21 Gasteiger E, Gattiker A, Hoogland C, Ivanyi I, Appel
RD & Bairoch A (2003) ExPASy: the proteomics server for in-depth protein knowledge and analysis Nucleic Acids Res 31, 3784–3788
22 Schatz G (1996) The protein import system of mito-chondria J Biol Chem 271, 31763–31766
23 Atkinson KD, Jensen B, Kolat AI, Storm EM, Henry
SA & Fogel S (1980) Yeast mutants auxotrophic for choline and ethanolamine J Bacteriol 141, 558–564
24 Leonhard K, Herrmann JM, Stuart RA, Mannhaupt G, Neupert W & Langer T (1996) AAA proteases with cat-alytic sites on opposite membrane surfaces comprise a proteolytic system for the ATP-dependent degradation
of inner membrane proteins in mitochondria EMBO J
15, 4218–4229
25 Lemaire C, Hamel P, Velours J & Dujardin G (2000) Absence of the mitochondrial AAA protease Yme1p restores F0-ATPase subunit accumulation in an oxa1 deletion mutant of Saccharomyces cerevisiae J Biol Chem 275, 23471–23475
26 Rontein D, Wu W-I, Voelker DR & Hanson AD (2003) Mitochondrial phosphatidylserine decarboxylase from higher plants Functional complementation in yeast, localization in plants, and overexpression in Arabidopsis Plant Physiol 132, 1678–1687
27 Kuge O, Saito K, Kojima M, Akamatsu Y & Nishijima
M (1996) Post-translational processing of the phosphati-dylserine decarboxylase gene product in Chinese ham-ster ovary cells Biochem J 319, 33–38
28 Boeckmann B, Bairoch A, Apweiler R, Blatter MC, Estreicher A, Gasteiger E, Martin MJ, Michoud K, O’Donovan C, Phan I et al (2003) The SWISS-PROT protein knowledgebase and its supplement TrEMBL in
2003 Nucl Acids Res 31, 365–370
29 Burke D, Dawson D & Stearns T (2000) Methods in Yeast Genetics: A Cold Spring Harbor Laboratory Course Manual Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY
30 Longtine MS, McKenzie A III, Demarini DJ, Shah
NG, Wach A, Brachat A, Philippsen P & Pringle JR (1998) Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharo-myces cerevisiae Yeast 14, 953–961
31 Gietz D, St Jean A, Woods RA & Schiestl RH (1992) Improved method for high efficiency transformation of intact yeast cells Nucleic Acids Res 20, 1425
32 Folch J, Lees M & Sloane Stanley GH (1957) A simple method for the isolation and purification of total lipids from animal tissues J Biol Chem 226, 497–509
33 Daum G, Bo¨hni PC & Schatz G (1982) Import of proteins into mitochondria J Biol Chem 257, 13028– 13033