Báo cáo khoa học: Molecular characterization of Osh6p, an oxysterol binding protein homolog in the yeast Saccharomyces cerevisiae pot

Molecular characterization of Osh6p, an oxysterol binding
protein homolog in the yeast Saccharomyces cerevisiae
Penghua Wang
1
, Wei Duan
1
, Alan L. Munn
2,3
and Hongyuan Yang
1
1 Department of Biochemistry, Faculty of Medicine, National University of Singapore, Republic of Singapore
2 Institute of Molecular and Cell Biology, A*STAR Biomedical Research Institutes, Singapore, Republic of Singapore
3 Institute for Molecular Bioscience, University of Queensland, St Lucia, Queensland, Australia
The oxysterol binding protein (OSBP) and its related
proteins (ORP) constitute a large conserved family of
proteins in eukaryotes [1,2]. OSBP homologs are pre-
sent in many species including humans and the yeast
Saccharomyces cerevisiae (the OSBP homolog in
yeast is OSH). These proteins all share a conserved
 400 amino acid OSBP related domain (ORD),
which contains an ‘OSBP fingerprint’ ‘EQVSHHPP’
[1].
Recent studies on the OSBP homologs of humans
and Saccharomyces cerevisiae have demonstrated the
importance of OSBP proteins in sterol and sphingo-
lipid metabolisms. The canonical OSBP is believed
to play a role in regulating sterol biosynthesis,
Keywords
OSBP; OSH; Osh6p; oxysterol-binding
protein; sterol homeostasis
Correspondence

H. Yang, Department of Biochemistry,
National University of Singapore, Singapore,
119260 Republic of Singapore
Fax: +65 67791453
Tel: +65 68747996
E-mail:
(Received 25 May 2005, revised 22 July
2005, accepted 28 July 2005)
doi:10.1111/j.1742-4658.2005.04886.x
Oxysterol binding protein (OSBP) and its homologs have been shown to
regulate lipid metabolism and vesicular transport. However, the exact
molecular function of individual OSBP homologs remains uncharacterized.
Here we demonstrate that the yeast OSBP homolog, Osh6p, bound phos-
phatidic acid and phosphoinositides via its N-terminal half containing the
conserved OSBP-related domain (ORD). Using a green fluorescent protein
fusion chimera, Osh6p was found to localize to the cytosol and patch-like
or punctate structures in the vicinity of the plasma membrane. Further
examination by domain mapping demonstrated that the N-terminal half
was associated with FM4-64 positive membrane compartments; however,
the C-terminal half containing a putative coiled-coil was localized to the
nucleoplasm. Functional analysis showed that the deletion of OSH6 led
to a significant increase in total cellular ergosterols, whereas OSH6 over-
expression caused both a significant decrease in ergosterol levels and resist-
ance to nystatin. Oleate incorporation into sterol esters was affected in
OSH6 overexpressing cells. However, Lucifer yellow internalization, and
FM4-64 uptake and transport were unaffected in both OSH6 deletion and
overexpressing cells. Furthermore, osh6D exhibited no defect in carboxy-
peptidase Y transport and maturation. Lastly, we demonstrated that both
the conserved ORD and the putative coiled-coil motif were indispensable
for the in vivo function of Osh6p. These data suggest that Osh6p plays

a role primarily in regulating cellular sterol metabolism, possibly stero
transport.
Abbreviations
ACAT, acyl CoA:cholesterol acyl transferase; CPY, carboxypeptidase Y; DAPI, 4¢-6-diamidino-2-phenylindole; GFP, green fluorescent protein;
GST, glutathione-S-transferase; LY, Lucifer yellow; MVB, multivesicular body; ORD, OSBP related domain; ORP, oxysterol binding protein
related protein; OSBP, oxysterol binding protein; OSH, yeast gene encoding oxysterol binding protein; Osh6p, yeast OSBP homolog; PA,
phosphatidic acid; PtdCho, phosphatidylcholine; PtdEtn, phosphatidylethanolamine; PH, pleckstrin homology; PtdIns, phosphatidylinositol;
PtdInsP, phosphatidylinositol phosphate; SD, synthetic dropout; SE, sterol ester; TAG, triacylglycerol.
FEBS Journal 272 (2005) 4703–4715 ª 2005 FEBS 4703
esterification and sterol-regulated gene transcription.
Lagace et al. [3] reported that overexpression of OSBP
in several cell lines reduced cholesterol ester synthesis
by 50% and showed a 40–60% decrease in acyl-
CoA:cholesterol acyltransferase (ACAT) activity and
mRNA. In contrast, overexpression of OSBP up-regu-
lated the transcription of sterol-regulated genes and
increased the rate of cholesterol biosynthesis. More-
over, OSBP localization is inherently linked to sterol
homeostasis [4,5] and sphingolipid metabolisms [6]. A
recent report suggested that OSBP may be involved in
vesicle-dependent ceramide transport from the endo-
plasmic reticulum to the Golgi [7]. Overexpression
studies with other human ORPs have also provided
further evidence. Stable overexpression of ORP2 led to
a significant reduction of ACAT activity and choles-
terol esters [8]. Overexpression of a splice variant of
ORP4 (ORP4-S) caused a 40% reduction in esterifica-
tion of low density lipoprotein-derived cholesterol [9].
Most interestingly, the role of OSBP in sterol homeo-
stasis seems to be conserved from mammals to the uni-

cellular yeast. The yeast OSHs collectively are essential
for cell viability and loss of all OSH function results in
a drastic increase in total cellular sterol levels and
accumulation of free sterols in the cytoplasm [10,11].
More recently, in a landmark paper Wang et al. [12]
showed that in response to cholesterol binding OSBP
interacts with two phosphatases: HePTP and PP2A.
These phosphatases are brought by OSBP to the vicin-
ity of phosphorylated extracellular signal-regulated
kinase 1 ⁄ 2 (ERK1 ⁄ 2) within the cell to promote its
dephosphorylation.
The functions of OSBP homologs are not limited to
sterol and sphingolipid metabolisms, they are also
implicated in the metabolism of other lipids and in
membrane trafficking. ORP1 and ORP2 appear to play
a role in the Golgi secretory function [13]. Deletion of
OSH4 ⁄ KES1 is able to bypass the essential require-
ment for Sec14p, which is a major phosphatidylinositol
(PtdIns) ⁄ phosphatidylcholine (PtdCho) transfer protein
[14]. A recent study proposed that Osh4p ⁄ Kes1p inter-
faces lipid metabolism and vesicle biogenesis possibly
via its regulation on the adenosine diphosphate-ribosy-
lation factor cycle or Pik1p activity, a Golgi associated
phosphatidylinositol 4 kinase [15]. Lastly, it appears
that OSBP and its homologs may also participate in
other cellular activities such as cell cycle [16], meiosis
and mating [17], mitosis and tumor metastasis [18].
The yeast genome encodes seven OSBP homologs,
Osh1p–Osh7p. Based on sequence homology, the Osh
proteins can be further divided into four subfamilies:

Osh1p and Osh2p; Osh3p; Osh4p ⁄ Kes1p and Osh5p;
Osh6p and Osh7p [10]. Among all individual OSH
gene deletions, Osh6D and osh5D exhibited most eleva-
ted sterol levels, highlighting the importance of Osh6p
and Osh5p in maintaining sterol homeostasis [10].
Here, we characterize Osh6p in greater detail and
show that deletion or overexpression of OSH6 causes
sterol-related defects but does not affect endocytosis
and endocytic trafficking of a marker, multivesicular
body sorting (MVB) or carboxypeptidase Y (CPY)
transport to the vacuole.
Results
Osh6p binds phospholipids
As a short Osh protein, Osh6p consists of an ORD for
lipid binding and a putative coiled-coil motif for pro-
tein–protein interaction. In this study, Osh6p was
demonstrated to bind a pool of phosphatidylinositol
phosphates (PtdInsP) including PtdIns(4)P, PtdIns(5)P,
PtdIns(3,4)P
2
and PtdIns(3,5)P
2
with the strongest
binding to PtdIns(5)P (Fig. 1). The ORD domain
showed a very similar lipid binding pattern as the full-
length protein. While the coiled-coil half and glutathi-
one S-transferase (GST) alone (control) failed to bind
any lipids. As a positive control GST-EEA1-a FYVE
protein preferably bound PtdIns(3)P [19], indicating
the specificity and reliability of this assay. Although

Osh6p showed a higher affinity for PtdIns(5)P than
for PtdIns(4)P in vitro, the physiological level of
PtdIns(4)P is much higher than that of PtdIns(5)P [20],
implying that PtdIns(4)P may be the native ligand of
Osh6p under physiological conditions. Interestingly,
when more lipids and protein were loaded, both Osh6p
and the ORD domain of Osh6p bound phosphatidic
acid (PA) (Fig. 2). The other lipids except for
PtdIns(4)P, PtdIns(5)P, PtdIns(3, 4)P, and PtdIns(3,5)P
in Fig. 1 were also tested under this condition, but
no binding was observed (data not shown). Osh6pORD
showed an eightfold higher affinity for PA than Osh6p
did; indicating the N-terminal half containing the
conserved ORD domain mediates lipid binding. A
weak binding to PtdIns(3)P by all proteins was also
observed, which might be nonspecific.
Characterization of the cellular location of Osh6p
The Saccharomyces cerevisiae yeast encodes seven
OSBP homologs that share an essential overlapping
function; however, each OSH performs distinct roles
probably at different cellular locations [10]. We hereby
examined the cellular location of Osh6p using a
C-terminal tagged green fluorescent protein (GFP) con-
struct and examined its distribution by fluorescence
Yeast OSBP and sterol homeostasis P. Wang et al.
4704 FEBS Journal 272 (2005) 4703–4715 ª 2005 FEBS
microscopy. As shown in Fig. 3A, Osh6p-GFP was
predominantly cytosolic, with a minor pool associated
with patch-like structures in the vicinity of plasma
membrane. However, the N-terminal half Osh6p(1–

254)-GFP showed strikingly different localization. As
seen in Fig. 3A, Osh6p(1–254)-GFP was associated
with one or two punctate structures without discernible
peripheral staining. These punctate structures were fur-
ther found to colocalize with the FM4-64 staining
endosomes (Fig. 3B). Most interestingly, the C-ter-
minal half Osh6p(255–448)-GFP was localized to the
nucleoplasm (Fig. 3A), which was further confirmed
by 4¢-6-diamidino-2-phenylindole (DAPI) staining. As
Fig. 3C shows, GFP staining structures were well colo-
calized with DAPI staining nucleus.
Osh6p is required for sterol homeostasis
Osh proteins collectively play a crucial role in main-
taining sterol homeostasis. In addition, deletion of
each individual OSH was also shown to affect sterol
levels to some extent [10]. Here we examined the role
of Osh6p in sterol homeostasis by gene deletion and
overexpression approaches. First we checked the
OSH6 deletion and overexpression strains by Western
blotting using anti-Osh6p serum. Figure 4A shows that
an  50 kDa band was recognized by anti-Osh6p
serum from cell extracts of wildtype cells, which was
absent from extracts of OSH6 knockout cells. On the
other hand, the amount of Osh6p protein overexpro-
duced using the ADH1 promoter from a 2l plasmid
(pADNSOSH6) was increased by more than 10-fold
compared to vector only (pADNS) (Fig. 4B). Next we
analyzed the steady-state ergosterol levels in osh6D and
overexpression cells. Consistent with a previous report
[10], deletion of OSH6 caused an increase in total

ergosterol levels by  80% (Fig. 4C) in a strain from
the Euroscarf collection (Institute of Microbiology,
Johann Wolfgang Goethe-University, Frankfurt,
Germany). Introduction of OSH6-GFP on a CEN
plasmid into osh6D corrected the ergosterol levels back
to that of wildtype. However, overexpression of OSH6
reduced the ergosterol levels (Fig. 4C) and associated
with nystatin-resistance (10-fold greater than vector
control) (Fig. 4D), indicating a reduction of free
sterols in the plasma membrane. To test the sensitivity
and reliability of GC-MS analysis, we used arv1 and
erg3 (ergosterol biosynthesis) mutants as controls.
Fig. 1. Osh6p binds phosphoinositides.
One hundred picomoles of desired lipids
were spotted on Hybond-C membranes
(Amersham Biosciences), dried and blocked.
The blot was incubated with GST fusion
proteins or GST at a concentration of 30 n
M
or 60 nM, respectively, washed and then
incubated with mouse anti-GST IgG and
anti-mouse secondary antibody sequentially.
Protein-bound lipids were detected
using ECL.
P. Wang et al. Yeast OSBP and sterol homeostasis
FEBS Journal 272 (2005) 4703–4715 ª 2005 FEBS 4705
Deletion of ARV1 results in  50% increase in free
sterols and  75% increase in sterol esters [32] and
erg3D mutants cannot synthesize ergosterol [33]. Con-
sistent with these reports, we observed a 38% increase

in total ergosterol in arv1D compared to wildtype and
no ergosterol was detected in erg3D by GC-MS
(Fig. S1). To examine whether osh6 mutants affected
sterol ester levels, cells were stained with Nile Red,
a dye that specifically stains neutral lipids. As shown
in Fig. 4E, an average of six lipid droplets per cell
(n ¼ 100) was observed in wildtype cells; while there
were about four only in the OSH6 overexpressing cells.
Exposure times were equal for both strains and the
brightness of lipid bodies was almost the same. Dele-
tion of OSH6 resulted in a significant increase of total
sterol levels (Fig. 4C); however, no change in the num-
ber of lipid droplets was observed (data not shown).
In order to elucidate how Osh6p affected sterol lev-
els, we tested sterol esterification and the rate of sterol
biosynthesis in both osh6D and OSH6 overexpressing
strains. Sterol biosynthesis was slightly accelerated in
osh6D but not affected in OSH6 overexpressing cells
(data not shown). Sterol esterification in OSH6 dele-
tion and overexpressing cells was decreased, but tri-
acylglycerol biosynthesis was not significantly affected
(Fig. 5), indicating that fatty acid uptake and transport
was normal. As positive controls, deletion of the major
sterol esterification gene ARE2 reduced
3
H-labeled
sterol esters by  75% [27], and the mutation of VPS4
(vacuolar protein sorting 4) caused a 40% decrease
[34] (Fig. S2).
Osh6p is not essential for fluid-phase endocytosis

or endocytic traffic of membrane markers from
the cell surface to the vacuole
We next investigated the role of Osh6p in certain mem-
brane trafficking pathways. Because endocytosis is
severely impaired in cells with seven OSHs mutated, we
first examined Lucifer yellow uptake in the osh6D and
OSH6 overexpressing mutants. Lucifer yellow (LY) is
taken up and delivered to the lumen of the vacuole in
wildtype cells. As shown in Fig. 6A, osh6D exhibited no
deficiency in uptake or transport of LY to the vacuole.
In addition, even when OSH6 was highly overexpressed
from the ADH1 promoter on a 2l plasmid, no signifi-
cant defect was observed (data not shown). We then
examined whether Osh6p was important for FM4-64
uptake and transport. FM4-64 is a lipophilic dye that
intercalates into cell membranes and is delivered to the
vacuolar limiting membrane via the endocytic pathway
[21]. The osh6D cells were labeled with FM4-64 at
15 °C and warmed to 30 °C (chase), then assessed for
distribution of FM4-64 at different time points. At
early time points of chase FM4-64 labels small punctate
structures representing early endosomes, and at later
time points it accumulates in large late endosomal ⁄
prevacuolar structures adjacent to the vacuole. Finally
FM4-64 reaches the vacuole limiting membrane in wild-
type cells. As shown in Fig. 6B, several small punctate
structures representing early endosomes were seen at
PA
PC
PE

PS
PI(3)P
10
3
500 250 125 62 31 pmol
GST
GST-Osh6p
GST-Osh6pCC
GST-Osh6pORD
PA
PA
PA
PI(3)P
PI(3)P
PI(3)P
Fig. 2. Osh binds PA as described in Fig. 1 except for the concen-
trations used for GST fusions and GST, which were 60 n
M and
120 n
M, respectively. The amount of lipid spotted is indicated on
the top of blots.
Yeast OSBP and sterol homeostasis P. Wang et al.
4706 FEBS Journal 272 (2005) 4703–4715 ª 2005 FEBS
0 min of chase. At 5 min of chase, one or two large
dots were observed around the vacuole, which repre-
sent late endosomal ⁄ prevacuolar structures. At 10 min,
punctate staining diminished and vacuolar staining in-
creased, and finally punctate staining was almost lost
with predominant vacuolar staining at 30 min. Com-
pared with wildtype, no delay of FM4-64 transport was

observed in osh6D. In addition, FM4-64 transport was
not affected in OSH6 overexpressing cells (data not
shown).
Osh6p is not essential for carboxypeptidase Y
maturation or multivesicular body sorting
Although it has been recently demonstrated that loss of
all OSH function did not affect carboxypeptidase Y
(CPY) maturation [11], the role of individual Osh pro-
teins in CPY sorting has not been investigated. It is
likely that members of the Osh protein family may have
antagonistic effects on each other [10]; therefore, we
tested whether Osh6p play a role in biosynthetic pro-
tein trafficking of the soluble CPY to the vacuole. CPY
is modified by core glycosylation in the endoplasmic
reticulum (p1 form with molecular mass of 67 kDa),
and transported to the Golgi for further glycosylation
(p2 form with molecular mass of 69 kDa). The p2 form
of CPY traverses the Golgi and is packaged into vesi-
cles destined for the vacuolar protein sorting pathway
which delivers it to the vacuole. Upon delivery to the
vacuole it is processed to its mature form (M form with
molecular mass of 61 kDa). Deficiency in this process
leads to accumulation of and secretion of p2 CPY. As
shown in Fig. 7, at 0 min of chase, p1 CPY was the
predominant form. The p2 form was seen at the 5 min
time point. At a later time (10 min), mature CPY was
observed together with some remaining p2 and p1 pre-
cursors. At the 30 min time point, most CPY was in
the mature 61 kDa form. Compared with wildtype,
GFP FM4-64 Merge

GFP
Merge
DAPI
DIC
GFP
Osh6p (1-254)
-GFP
GFP DIC
Osh6p -GFP
Osh6p (255-448)
-GFP
AC
B
Fig. 3. Localization of Osh6p. (A) Exponentially growing cells (Y10000) expressing GFP fusions from YEplac181 vector were mounted on a
glass slide and visualized using a Leica fluorescence microscope. (B) Colocalization of Osh6p(1–254)-GFP with FM4-64 positive compart-
ments. Cells (Y10000) were labeled with FM4-64 in ice-water for 30 min and then shifted to 15 °C for 20 min allowing FM4-64 to be inter-
nalized. Cells were immediately put back on ice and washed thoroughly to remove excess dye. GFP and FM4-64 images were acquired via
a GFP filter and Texas Red filter, respectively. Arrows indicate some colocalization. (C) Colocalization of Osh6p(255–448)-GFP with DAPI
stained nucleus. Scale bar: 5 lm. DIC, differential interference contrast.
P. Wang et al. Yeast OSBP and sterol homeostasis
FEBS Journal 272 (2005) 4703–4715 ª 2005 FEBS 4707
osh6D showed no delay in CPY maturation. Although
even wildtype cells secret a minute fraction of CPY pre-
cursors, we could not detect CPY in the extracellular
extracts from either wildtype or osh6D cells in this
study. This discrepancy may be due to insufficient pro-
teins loaded for detection. We next checked the role
of Osh6p in the multivesicular body (MVB) pathway
using GFP fusions of the surface receptor-Ste3p and
vacuolar hydrolase-carboxypeptidase S. MVB is a pro-

cess whereby the limiting membrane of late endosomes
invaginates and buds into the lumen of the organelle,
and is responsible for the biosynthetic delivery of lyso-
somal hydrolases and the down-regulation of numerous
activated cell surface receptors. However, no defect was
observed in either osh6D or OSH6 overexpressing cells
(data not shown).
Characterization of the functional domains
of Osh6p
As a short Osh protein, Osh6p consists of a conserved
ORD domain that is believed to mediate lipid binding
and a putative coiled-coil motif for protein interaction.
We were also interested to know whether these
domains were important for the function of Osh6p.
We expressed GST fusion proteins in JRY6326 strain
(pMET2-OSH2, osh1-osh7D). In the presence of methio-
nine, OSH2 expression is suppressed, leading to cell
0
4
8
12
16
Ergosterol (µ g/mg dry weight)
WT osh6∆ osh6∆ WT WT
vector vector
OSH6-GFP pADNS pADNSOSH6
osh6
WT
Osh6p
Dpm1p

pADNS pADNSOSH6
VATPase60
Osh6p
Nile Red DIC
pADNS
pADNS
OSH6
pADNS
pADNS
OSH6
pADNS
OSH6
pADNS
1µg/ml nystatin
No nystatin
Fig. 4. Osh6p is required for maintaining sterol homeostasis. (A,B) Western blotting. Osh6p was detected using rabbit anti-Osh6p serum at
1 : 300 dilution. Dpm1p and VATPase60 were used as internal loading controls. pADNS represents the Y10000 strain harboring vector (pADNS);
pADNSOSH6 shows Y10000 strain transformed with pADNSOSH6 overexpressing OSH6 from the ADH1 promoter. (C) Total sterols (free and
esterified) were extracted with hexane and blow-dried under a stream of nitrogen. Ergosterol was identified and quantified by GC-MS. Results
were obtained from two independent experiments (n ¼ 3). The x-axis denotes strain genotype. Wildtype (WT) vector: Y10000 cells transformed
with YCplac111-scGFP; osh6D vector: Y15074 cells transformed with YCplac111-scGFP; osh6D OSH6-GFP: Y15074 cells carrying YCp OSH6-
GFP; WT pADNS ⁄ WT pADNSOSH6: Y10000 cells harbouring pADNS ⁄ pADNSOSH6. (D) Nystatin assay. Cells at mid log phase were serially
diluted by 10-fold and grown on SD solid medium lacking leucine with (upper) or without (lower) 1 lgÆmL
)1
nystatin at 30 °C for 48 h. (E) Visual-
ization of lipid droplets. Cells at stationary phase were stained with Nile Red and visualized via a Texas Red filter using a Leica fluorescence
microscope. Scale bar: 5 lm. DIC, differential interference contrast.
Yeast OSBP and sterol homeostasis P. Wang et al.
4708 FEBS Journal 272 (2005) 4703–4715 ª 2005 FEBS
growth arrest. Reintroduction of any OSH gene can

restore cell growth except in OSH1 for which over-
expression is required [10]. When the coiled-coil
domain, ORD domain and full-length Osh6p were
overexpressed in JRY6326 (Fig. 8), only full-length
Osh6p protein could rescue JRY6326 in the presence
of methionine, which suggested the ORD and coiled-
coil domains were both essential for Osh6p function.
Discussion
Osh proteins collectively play a crucial role in main-
taining sterol homeostasis. Recent studies have largely
focused on the function of all seven OSH gene prod-
ucts [10,11]. Those studies provided invaluable in-
sights into understanding the function of this family
of proteins in yeast and offered guidance to future
research. On the other hand, although the entire Osh
protein family shares at least one essential function,
each individual Osh protein possesses unique roles.
One prominent example is KES1, the mutation of
which, but none of the other OSH genes, bypasses
the essential requirement for SEC14 [14]. Therefore, it
is necessary to analyze individual OSH genes to gain
further insights into the function of this family of
proteins.
The lipid binding properties have long been estab-
lished for OSBP and its homologs. OSBP binds oxy-
sterols via its C-terminal half containing the conserved
ORD domain and phosphoinositides via its pleckstrin
homology (PH) domain [22–24]. Here we show that
the non-PH containing Osh6p can also bind phospho-
inositides and PA (Figs 1 and 2) with different affinity.

In addition, bacterially expressed Osh6p shows no
affinity for ergosterol, diacylglycerol (DAG) or cera-
mide in our assay system (data not shown). Our results
represent a novel and exciting finding given the fact
that the bypass sec14 mutants require a basal PA level
to exert their suppression effects [25,26]. Consistent
with this, the three OSBP homologs: Kes1p, ORP1and
ORP2 which are potentially involved in the Golgi
secretory function are able to bind PA with varying
affinity [13–15]. Further, our results demonstrate that
the N-terminal half (amino acids 1–300) containing the
conserved OSBP domain is sufficient and indispensable
for Osh6p binding to phospholipids. Unlike the long
Osh proteins, Osh6p contains no canonical PH domain
that mediates phospholipid binding [23,24]. However,
it is possible that Osh6p may bind phospholipids
through a domain other than PH. In support of this,
the short Kes1p containing no PH domain was also
shown to bind PtdIns (4,5)P [15]. Although the role of
the conserved ORD domain is unclear, our results
imply that it could recognize specific lipid ligand(s).
As an effort to understand the role of Osh6p in
the vesicular transport, we examined the cellular
location of Osh6p. Although the cellular location of
the full-length protein could not be pinpointed in this
study, the patch-like structures might possibly repre-
sent endoplasmic reticulum. Interestingly, the N ter-
minus (1–254) is localized to endosomes and the
C terminus (255–448) to the nucleoplasm. To our
knowledge, Osh6p may be the first of the OSBP

homologs to be shown to localize to the nucleoplasm
and the physiological relevance is worthy of further
investigation. In addition, our data suggest that like
long Osh proteins, Osh6p may be able to associate
with multiple membranes through different functional
domains. Consistent with the result from lipid bind-
ing analysis (Fig. 1), the ORD domain may bind
phospholipids on the endosomal membranes thus tar-
geting Osh6p to membranes; whereas the C terminus
containing the coiled-coil motif could interact with a
protein or protein complex in the other compartment
0
500
1000
1500
2000
2500
0
2000
4000
6000
8000
WT osh6 ∆ pADNS pADNSOSH6
WT
osh6 ∆
pADNS pADNSOSH6
TAG (cpm/mg dry weight)
SE (cpm/mg dry weight)
Fig. 5. Oleate incorporation. Sterol esterification (A) or triacylglycerol
(TAG) synthesis (B) was determined by incorporation of [

3
H]oleic acid
into sterol esters (SE) or TAG and expressed as
3
H-labeled SE or TAG
per mg of dry cells. Results were obtained from two independent
experiments (n ¼ 4). The x-axis denotes strain genotype. cpm,
counts per minute; WT, Y10000; osh6D, Y15074; pADNS (vector
control) and pADNS OSH6 indicate Y10000 cells transformed with
pADNS ⁄ pADNSOSH6.
P. Wang et al. Yeast OSBP and sterol homeostasis
FEBS Journal 272 (2005) 4703–4715 ª 2005 FEBS 4709
WT
DICFITC
Fig. 6. Osh6p is not essential for Lucifer yellow (LY) uptake and FM4-64 transport. (A) Wildtype (Y10000) or osh6D (Y15074) cells at early
log phase were allowed to internalize LY for 1 h and images were captured using a Leica fluorescence microscope. DIC, differential interfer-
ence contrast; FITC, fluorescent image. (B) Cells at early log phase were labeled with FM4-64 at 15 °C for 20 min. After removal of excess
FM4-64, cells were chased for 0, 5, 10, and 30 min at 30 °C. FM4-64 staining was visualized using a Leica fluorescence microscope
equipped with a Texas Red light filter. Upper panels: DIC and FM4-64 images of Y10000 (wildtype). Lower panels: DIC and FM4-64 images
of Y15074 (osh6D). Time of chase is indicated at the bottom. Scale bar: 5 lm.
Yeast OSBP and sterol homeostasis P. Wang et al.
4710 FEBS Journal 272 (2005) 4703–4715 ª 2005 FEBS
(for example the nucleoplasm or endoplasmic reticu-
lum membrane).
Excess intracellular sterols are normally converted to
sterol esters, a process that is catalyzed by Are1p and
Are2p in yeast [27]. Sterol esterification is primarily
regulated by the availability of sterol substrates. How-
ever, excess ergosterol observed in osh6D does not
result in enhanced sterol esterification (Fig. 5A) or

increased lipid droplets as assessed by Nile Red stain-
ing (not shown). It is therefore possible that osh6D
cells accumulate free sterols in intracellular compart-
ments other than the ER, due to defects in lipid traf-
ficking. In support of this, it was recently reported
that filipin staining sterols accumulated in the internal
membranes of OSH null mutants [11]. In addition,
although some OSH mutants including osh6D have
increased total ergosterol levels, they exhibit no nysta-
tin sensitivity [10], indicating a possible disruption of
sterol transport and accumulation of free sterols in
membranes other than the plasma membrane. In sup-
port of this idea, two recent studies suggested OSBP
and its homologs are potentially involved in vesicle-
independent intracellular lipid transport [9,28]. More
recently, Wang et al. [12] showed that upon binding to
cholesterol OSBP served as a scaffold and recruited
two phosphatases, thus regulating the cholesterol-medi-
ated ERK1 ⁄ 2 signaling.
Although deletion of all OSH genes impaired endo-
cytosis and the vacuole morphology [11], the results
presented here suggest that Osh6p is not essential
for some membrane trafficking pathways. Neither
endocytic (Fig. 6) nor biosynthetic transport to the
vacuole (Fig. 7) is affected in the absence of Osh6p or
the presence of a high level of Osh6p. The discrepancy
may be due to the fact that a more drastic disruption
of sterol homeostasis is incurred upon loss of all Osh
function. For instance, total cellular ergosterol increase
by 3.5-fold in OSH mutant cells compared with an

80% increase in osh6D cells.
In summary, this study provides a detailed charac-
terization of Osh6p, one of the seven OSBP homo-
logs in yeast. We show, for the first time, a direct
binding of PA by the ORD domain and the cellular
location of Osh6p. We further demonstrate that dele-
tion or overexpression of OSH6 affects sterol homeo-
stasis but not endocytosis or CPY secretion. Our
results suggest that the primary molecular function
of Osh6p is likely to be in sterol metabolism, prob-
ably sterol transport.
Experimental procedures
Materials
Mouse anti-CPY, Dpm1p, VATPase60 and rabbit anti-
GST IgGs were obtained from Molecular Probes (Eugene,
OR, USA). Mouse anti-GST IgG was purchased from
Santa Cruz Biotechnology (Santa Cruz, CA, USA). YPD
medium contained 2% (w ⁄ v) dextrose, 2% (w ⁄ v) peptone
and 1% (w ⁄ v) yeast extract (Gibco-BRL ⁄ Life Technologies,
Paisley, UK). Synthetic dropout medium (SD) was com-
prised of 0.67% (w ⁄ v) yeast nitrogen base, 2% (w ⁄ v)
P2
P1
M
P2
P1
M
WT
osh6∆
0 5 10 30

0 5 10 30 min
Intracellular Extracellular
Fig. 7. Osh6p is not essential for CPY maturation. Cells were radio-
labeled with [
35
S]methionine ⁄ cysteine and chased for various times
with cold methionine ⁄ cysteine. Cells were lysed and CPY was
immunoprecipitated from cell lysate (intracellular) and extracellular
growth medium ⁄ periplasm (extracellular). Proteins were resolved
by SDS ⁄ PAGE and detected by radiophotography. p1, core glycos-
ylated endoplasmic reticulum form; p2, fully glycosylated Golgi
form; M, mature vacuolar form; wildtype, SEY6210; osh6D,
SEY6201. Times of chase are indicated under the blot.
OSH6
OSH6CC
OSH6ORD
Vector
OSH6
OSH6CC
OSH6ORD
Vector
Fig. 8. Functional domains of Osh6p. pYEX4T-1, pYEX OSH6,
pYEXOSH6CC, and pYEXOSH6ORD were introduced into JRY6326
cells and grown on YM solid medium containing 2 m
M methionine
for 72 h at 30 °C.
P. Wang et al. Yeast OSBP and sterol homeostasis
FEBS Journal 272 (2005) 4703–4715 ª 2005 FEBS 4711
dextrose plus dropout powder. YM (yeast minimal) med-
ium was made of 2% (w ⁄ v) dextrose and 0.67% (w ⁄ v) yeast

nitrogen base. LB broth contained 1% (w ⁄ v) tryptone,
0.5% (w ⁄ v) yeast extract, and 1% (w ⁄ v) sodium chloride.
Construction of plasmids and strains
Construction of expression plasmids were performed
according to the universal protocols [29]. For plasmids and
strains used in this study, see Tables 1 and 2.
Biochemical assays
Expression and purification of GST fusion protein was per-
formed essentially according to a method of Dowler et al.
[30].
Fluorescence microscopy
FM4-64 staining was performed according to a previously
described method [21]. Cells at early log-phase (D
600
¼ 0.2)
were stained with FM4-64 at a final concentration of
1.2 lgÆmL
)1
at 15 °C for 20 min for preliminary labeling
and the excess dye was removed by washing with ice-cold
labeling medium. Cells were then resuspended in fresh labe-
ling medium and incubated at 30 °C. An aliquot was
removed at 0, 5, 10 and 30 min time points and endocytosis
was stopped by addition of ice-cold NaF ⁄ NaN
3
to a final
concentration of 12 mm. All samples were thoroughly
washed with ice-cold wash buffer (1· NaCl ⁄ P
i
,10mm

NaF ⁄ NaN
3
) to remove excess dye.
Lucifer yellow and DAPI staining was performed essen-
tially following a method of Yeo et al. [31] and Levine
et al. [35].
Nile Red staining was done following the method of
Yang et al. [27] with minor modification. Cells were grown
to stationary phase at 30 °C and 1 mL of cells was harves-
ted by brief centrifugation. Cells were washed twice with
1· NaCl ⁄ P
i
and resuspended in 1 mL of staining solution
(1· NaCl ⁄ P
i
,1lgÆmL
)1
Nile Red). Nile Red staining pat-
tern was visualized using a Leica DMLB fluorescence
microscope (Chatsworth, CA, USA) via a Texas Red filter.
Table 1. Plasmids. Unless specified, all the constructs were made in this study.
Plasmids Description Ref. or source
pGEXOSH6 pGEX4T-1, GST, OSH6(1–448)
pGEXOSH6ORD pGEX4T-1, GST, OSH6(1–300)
pGEXOSH6CC pGEX4T-1, GST, OSH6(301–448)
pYEXOSH6 pYEX4T-1, GST, OSH6(1–448)
pYEXOSH6CC pYEX4T-1, GST, OSH6(301–448)
pYEXOSH6ORD pYEX4T-1, GST, OSH6(1–300)
pGEX4T-1 Ptac, GST, Amp
R

Amersham
pYEX4T-1 Pcup1, GST, URA3, leu2-d, 2l BD biosciences
pADNSOSH6 pADNS, OSH6(1–448)
pADNS P
ADH1
, LEU2, 2l
YCpOSH6-GFP YCplac111, OSH6 promoter, OSH6(1–448), GFP
YEpOSH6(1–254)-GFP YEplac181, OSH6 promoter, OSH6(1–254),GFP
YEpOSH6(255–448)-GFP YEplac181, OSH1 promoter, OSH6(255–448),GFP
YEpOSH6-GFP YEplac181, OSH6 promoter, OSH6(1–448),GFP
YCplac111-scGFP YCplac111, GFP, CEN, LEU2 [31]
YEplac181-scGFP YEplac181, GFP, 2l, LEU2 [31]
YEp-p
OSH1
-GFP YEplac181, OSH1 promoter, GFP
Table 2. Strains.
Strain name Genotype Ref. or source
Y10000 BY4742, MATa his3D1 leu2D0 lys2D0 ura3D0 Euroscarf
Y15074 BY4742, MATa his3D1 leu2D0 lys2D0 ura3D0 osh6D::KanMX4 Euroscarf
JRY6326 SEY6210, TRP1::PMET3-OSH2 osh1D::kan-MX4 osh2D::kan-MX4
osh3D::LYS2 osh4D::HIS3 osh5D::LEU2 osh6D::LEU2 osh7D::HIS3
[10]
SEY6210 MATa ura3-52 his3D200 lys2-801 leu2-3,112 trp1D901 suc2D9 [10]
SEY6201 SEY6210, osh6D::LEU2 [10]
Y12667 BY4742, MATa his3D1 leu2D0 lys2D0 ura3D0 erg3D::KanMX4 Euroscarf
Y15151 BY4742, MATa his3D1 leu2D0 lys2D0 ura3D0 arv1D::KanMX4 Euroscarf
Y15588 BY4742, MATa his3D1 leu2D0 lys2D0 ura3D0 vps4D::KanMX4 Euroscarf
Y15394 BY4742, MATa his3D1 leu2D0 lys2D0 ura3D0 are2D::KanMX4 Euroscarf
Yeast OSBP and sterol homeostasis P. Wang et al.
4712 FEBS Journal 272 (2005) 4703–4715 ª 2005 FEBS

Pulse-chase radiolabeling and immunoprecipitation
The CPY maturation experiment was carried out according
to a method of Yeo et al. [31] with minor modifications.
Twenty units at D
600
of overnight cell culture in SDYE [SD
medium containing 0.2% (w ⁄ v) yeast extract and minimum
supplements] was harvested and washed with 50 mL of SD
containing necessary supplements. Cells were resuspended
in 2.5 mL labeling mix (SD containing necessary supple-
ments and 2 mgÆmL
)1
BSA) and incubated with 50 lL
of [
35
S]methionine ⁄ cysteine (7.9 mCiÆmL
)1
PerkinElmer,
(Wellesley, MA, USA)) for 10 min. Fifty microliters of
50· chase solution (0.1 m Na
2
SO
4
,1mgÆmL
)1
cysteine,
1mgÆmL
)1
methionine) was added, and 0.5 mL sample was
transferred immediately to a microfuge tube containing

50 lL stop solution (0.2 m NaF and NaN
3
) on ice (labeled
as 0 min). Subsequently, at times of 5, 10 and 30 min,
0.5 mL sample was removed and treated with stop solution
as previously. Cells were converted to spheroplasts and then
lysed. CPY was immunoprecipitated from the growth
medium ⁄ periplasm (extracellular) or spheroplast lysate
(intracellular) using anti-CPY IgG and Protein A sepharose
beads. After stringency wash, bound CPY was eluted and
separated by SDS ⁄ PAGE, and finally detected by radiopho-
tography.
GC-MS analysis of ergosterol
Sterol lipids were extracted using a method of Beh et al.
[10]. For triplicate analysis of a same culture, 150 mL of
exponentially growing yeast (D
600
¼ 0.5–0.8) were split into
three equal volumes and harvested by brief centrifugation
(5 mL was removed for dry weight analysis). Cells were
washed once with 50 mL distilled water. The cell pellets
were resuspended in 1.25 mL of 0.1 m HCl and boiled for
20 min. Cells were washed twice with 5 mL of distilled
water and then the cell pellets were resuspended in 0.5 mL
of 67% (v ⁄ v) methanol. Cells were lyzed with glass beads
by vigorous vortexing for 3 min. For total sterol analysis
(free and esterified), 1.25 mL of methanol and 0.63 mL of
60% KOH were added to the slurry followed by heating at
70 °C for 90 min. Sterols were extracted with 2 mL of
hexane four times. Ergosterol was identified and quantified

by GC-MS QP5000 (Shimadzu Corporation, Kyoto, Japan).
Oleate incorporation
Sterol esterification or triacylglycerol (TAG) synthesis was
measured by incorporation of [
3
H]oleic acid into sterol esters
or TAG following a method of Yang et al. [27]. Ten millilit-
ers of exponentially growing cells (D
600
¼ 0.7–0.8) were labe-
led with 1 lLof[
3
H]oleic acid 5.0 mCiÆmL
)1
, Amersham
Biosciences (Uppsala, Sweden) for 1 h at 30 °C. Cells were
washed twice with 5 mL of ice-cold wash buffer [0.5% (v ⁄ v)
NP-40, 20 mm NaN
3
] and once with distilled water. Cells
were split into three equal aliquots and lyophilized overnight.
The dried cells were lyzed in 50 lL of lysis buffer
[1700 UÆmL
)1
lyticase, 10% (v ⁄ v) glycerol, and 0.02% (w ⁄ v)
sodium azide] at 37 °C for 15 min. The cell suspension was
then subject to one freeze-thaw cycle. As an internal stand-
ard, 0.1 lLof[
14
C]cholesterol (0.1 mCiÆmL

)1
, Amersham) in
200 lL isopropanol was added to the cell suspension. Lipids
were extracted from the cell suspension with 5 mL of hexane
by vigorous vortexing. Phase separation was induced by addi-
tion of 5 mL of KCl ⁄ methanol (2 m KCl ⁄ methanol ¼ 2:1,
v ⁄ v). The hexane extracts were blow-dried under a nitrogen
stream. Sterol esters or TAG were separated by TLC and
quantified using a scintillation counter.
Protein lipid overlay assay
Protein lipid overlay assay was carried out following a
method of Dowler et al. [30]. Phosphatidylinositol phos-
phate (PtdInsP) strips containing indicated lipids were
blocked in blocking solution [3% (w ⁄ v) fatty acid free BSA
in TBST (50 mm Tris ⁄ HCl, pH 7.5, 150 mm NaCl and 0.1%
(v ⁄ v) Tween-20)] at room temperature for 2 h. The PtdInsP
strips were then incubated with GST fusion proteins at a
concentration of 30–60 nm for fusion proteins or 60–120 nm
for GST only for 2 h at room temperature. After washing
with six changes of TBST over 0.5 h, the PtdInsP strips were
incubated with anti-GST monoclonal antibody (1 : 3000)
for 1 h, washed again as previously, and then incubated with
mouse anti-IgG horseradish peroxidase conjugate (1 : 4000)
for 1 h. The PtdInsP strips were finally washed with 12
changes of TBST over 1 h. The protein-bound lipid ligands
were detected by enhanced chemiluminescence (Amersham
Biosciences).
Acknowledgements
This work was supported by a Young Investigator
Award from the National University of Singapore, a

grant from the National Medical Research Council,
Singapore (to H.Y.). We wish to thank Dr C.T. Beh
for yeast strains and Dr N. Lehming for plasmids.
References
1 Lehto M & Olkkonen VM (2003) The OSBP-related
proteins: a novel protein family involved in vesicle
transport, cellular lipid metabolism, and cell signaling.
Biochim Biophys Acta 1631, 1–11.
2 Olkkonen VM & Levine TP (2004) Oxysterol binding
proteins: in more than one place at one time? Biochem
Cell Biol 82, 87–98.
3 Lagace TA, Byers DM, Cook HW & Ridgway ND (1997)
Altered regulation of cholesterol and cholesteryl ester
synthesis in Chinese-hamster ovary cells overexpressing
P. Wang et al. Yeast OSBP and sterol homeostasis
FEBS Journal 272 (2005) 4703–4715 ª 2005 FEBS 4713
the oxysterol-binding protein is dependent on the pleck-
strin homology domain. Biochem J 326, 205–213.
4 Ridgway ND, Dawson PA, Ho YK, Brown MS &
Goldstein JL (1992) Translocation of oxysterol binding
protein to Golgi apparatus triggered by ligand binding.
J Cell Biol 116, 307–319.
5 Ridgway ND, Lagace TA, Cook HW & Byers DM
(1998) Differential effects of sphingomyelin hydrolysis
and cholesterol transport on oxysterol-binding protein
phosphorylation and Golgi localization. J Biol Chem
273, 31621–31628.
6 Storey MK, Byers DM, Cook HW & Ridgway ND
(1998) Cholesterol regulates oxysterol binding protein
(OSBP) phosphorylation and Golgi localization in Chi-

nese hamster ovary cells: correlation with stimulation
of sphingomyelin synthesis by 25-hydroxycholesterol.
Biochem J 336, 247–256.
7 Wyles JP, McMaster CR & Ridgway ND (2002) Vesi-
cle-associated membrane protein-associated protein-A
(VAP-A) interacts with the oxysterol-binding protein to
modify export from the endoplasmic reticulum. J Biol
Chem 277, 29908–29918.
8 Laitinen S, Lehto M, Lehtonen S, Hyvarinen K, Heino
S, Lehtonen E, Ehnholm C, Ikonen E & Olkkonen VM
(2002) ORP2, a homolog of oxysterol binding protein,
regulates cellular cholesterol metabolism. J Lipid Res
43, 245–255.
9 Wang C, JeBailey L & Ridgway ND (2002) Oxysterol-
binding-protein (OSBP)-related protein 4 binds
25-hydroxycholesterol and interacts with vimentin
intermediate filaments. Biochem J 361, 461–472.
10 Beh CT, Cool L, Phillips J & Rine J (2001) Overlapping
functions of the yeast oxysterol-binding protein homo-
logs. Genetics 157, 1117–1140.
11 Beh CT & Rine J (2004) A role for yeast oxysterol-bind-
ing protein homologs in endocytosis and in the mainte-
nance of intracellular sterol-lipid distribution. J Cell Sci
117, 2983–2996.
12 Wang PY, Weng J & Anderson RG (2005) OSBP is a
cholesterol-regulated scaffolding protein in control of
ERK 1 ⁄ 2 activation. Science 307, 1472–1476.
13 Xu Y, Liu Y, Ridgway ND & McMaster CR (2001)
Novel members of the human oxysterol-binding protein
family bind phospholipids and regulate vesicle trans-

port. J Biol Chem 276, 18407–18414.
14 Fang M, Kearns BG, Gedvilaite A, Kagiwada S,
Kearns M, Fung MK & Bankaitis VA (1996) Kes1p
shares homology with human oxysterol binding protein
and participates in a novel regulatory pathway for yeast
Golgi-derived transport vesicle biogenesis. EMBO J 15,
6447–6459.
15 Li X, Rivas MP, Fang M, Marchena J, Mehrotra B,
Chaudhary A, Feng L, Prestwich GD & Bankaitis VA
(2002) Analysis of oxysterol binding protein homolog
Kes1p function in regulation of Sec14p-dependent
protein transport from the yeast Golgi complex. J Cell
Biol 157, 63–77.
16 Alphey L, Jimenez J & Glover DA (1998) Drosophila
homolog of oxysterol binding protein (OSBP) – implica-
tions for the role of OSBP. Biochim Biophys Acta 1395,
159–164.
17 Park YU, Hwang O & Kim J (2002) Two-hybrid clon-
ing and characterization of OSH3, a yeast oxysterol-
binding protein homolog. Biochem Biophys Res Commun
293, 733–740.
18 Fournier MV, Guimaraes da Costa F, Paschoal ME,
Ronco LV, Carvalho MG & Pardee AB (1999) Identifi-
cation of a gene encoding a human oxysterol-binding
protein-homolog: a potential general molecular marker
for blood dissemination of solid tumors. Cancer Res 59,
3748–3753.
19 Gaullier JM, Ronning E, Gillooly DJ & Stenmark H
(2000) Interaction of the EEA1 FYVE finger with phos-
phatidylinositol 3-phosphate and early endosomes. Role

of conserved residues. J Biol Chem 275, 24595–24600.
20 Rameh LE, Tolias KF, Duckworth BC & Cantley LC
(1997) A new pathway for phosphatidylinositol-4,5-
bisphosphate synthesis. Nature 390, 192–196.
21 Vida TA & Emr SD (1995) A new vital stain for visua-
lizing vacuolar membrane dynamics and endocytosis in
yeast. J Cell Biol 128, 779–792.
22 Dawson PA, Van der Westhuyzen DR, Goldstein JL
& Brown MS (1989) Purification of oxysterol binding
protein from hamster liver cytosol. J Biol Chem 264,
9046–9052.
23 Levine TP & Munro S (1998) The pleckstrin homology
domain of oxysterol-binding protein recognises a deter-
minant specific to Golgi membranes. Curr Biol 8, 729–
739.
24 Levine TP & Munro S (2002) Targeting of Golgi-speci-
fic pleckstrin homology domains involves both PtdIns
4-kinase-dependent and -independent components.
Curr Biol 12, 695–704.
25 Sreenivas A, Patton-Vogt JL, Bruno V, Griac P &
Henry SA (1998) A role for phospholipase D (Pld1p) in
growth, secretion, and regulation of membrane lipid
synthesis in yeast. J Biol Chem 273, 16635–16638.
26 Xie Z, Fang M, Rivas MP, Faulkner AJ, Sternweis PC,
Engebrecht JA & Bankaitis VA (1998) Phospholipase D
activity is required for suppression of yeast phosphati-
dylinositol transfer protein defects. Proc Natl Acad Sci
USA 95, 12346–12351.
27 Yang H, Bard M, Bruner DA, Gleeson A, Deckelbaum
RJ, Aljinovic G, Pohl TM, Rothstein R & Sturley SL

(1996) Sterol esterification in yeast: a two-gene process.
Science 272, 1353–1356.
28 Loewen CJ, Roy A & Levine TP (2003) A conserved
ER targeting motif in three families of lipid binding
proteins and in Opi1p binds VAP. EMBO J 22, 2025–
2035.
Yeast OSBP and sterol homeostasis P. Wang et al.
4714 FEBS Journal 272 (2005) 4703–4715 ª 2005 FEBS
29 Sambrook J, Fritsch EF & Maniatis T (1989) Molecular
Cloning: a Laboratory Manual, Vol. 3, 2nd edn. Cold
Spring Harbor Laboratory Press, Cold Spring Harbor,
New York.
30 Dowler S, Currie RA, Campbell DG, Deak M, Kular G,
Downes CP & Alessi DR (2000) Identification of pleck-
strin-homology-domain-containing proteins with novel
phosphoinositide-binding specificities. Biochem J 351,
19–31.
31 Yeo SC, Xu L, Ren J, Boulton VJ, Wagle MD, Liu C,
Ren G, Wong P, Zahn R, Sasajala P, Yang H, Piper RC
& Munn AL (2003) Vps20p and Vta1p interact with
Vps4p and function in multivesicular body sorting and
endosomal transport in Saccharomyces cerevisiae. J Cell
Sci 116, 3957–3970.
32 Tinkelenberg AH, Liu Y, Alcantara F, Khan S, Guo Z,
Bard M & Sturley SL (2000) Mutations in yeast ARV1
alter intracellular sterol distribution and are complemen-
ted by human ARV1. J Biol Chem 275, 40667–40670.
33 Paltauf F, Kohlwein S & Henry SA (1992) Regulation
and compartmentalization of lipid synthesis in yeast. In
Molecular and Cellular Biology of the Yeast Saccharo-

myces (Jones EW, Pringle JR & Broach JR, eds), pp.
415–500, Cold Spring Harbor Laboratory Press, Cold
Spring Harbor, New York.
34 Wang P, Zhang Y, Li H, Chieu HK, Munn AL &
Yang H (2005) AAA ATPases regulate membrane asso-
ciation of yeast oxysterol binding proteins and sterol
metabolism. EMBO J doi:10.1038.
35 Levine TP & Munro S (2001) Dual targeting of Osh1p,
a yeast homolog of oxysterol-binding protein, to both
the Golgi and the nucleus-vacuole junction. Mol Biol
Cell 12, 1633–1644.
Supplementary material
Fig. S1. Analysis of total ergosterol (free and esteri-
fied) by GC-MS.
Fig. S2. Oleate incorporation. Sterol esterification or
TAG synthesis was determined by incorporation of
[
3
H]oleic acid into sterol esters (SE) or TAG and
expressed as
3
H-labeled sterol esters or TAG per mg
of dry cells.
P. Wang et al. Yeast OSBP and sterol homeostasis
FEBS Journal 272 (2005) 4703–4715 ª 2005 FEBS 4715