![]()
|
|
||||||||
(Received for publication, October 26, 1994; and in revised form, January 20, 1995) From the
The effect of disruption of the Golgi apparatus on
25-hydroxycholesterol-mediated transcriptional suppression and
activation of acyl-CoA:cholesterol acyltransferase was examined. In
Chinese hamster ovary (CHO) cells, brefeldin A (BFA) caused
dose-dependent inhibition of 25-hydroxycholesterol-mediated suppression
of mRNAs for four sterol-regulated genes: 3-hydroxy-3-methylglutaryl
(HMG)-CoA reductase, HMG-CoA synthase, farnesyl-diphosphate synthase,
and the low density lipoprotein receptor. BFA prevented suppression
whether added prior to or following a 4-h pretreatment with
25-hydroxycholesterol. In the presence of BFA (1 µg/ml),
25-hydroxycholesterol-mediated suppression of mRNAs for HMG-CoA
reductase, the low density lipoprotein receptor, and
farnesyl-diphosphate synthase was almost completely blocked. HMG-CoA
synthase mRNA was 80-90% suppressed by 25-hydroxycholesterol
compared with 50-60% suppression in the presence of BFA. These
effects of BFA were not due to alterations in mRNA stability.
Disruption of the Golgi apparatus, as assessed by staining with a
fluorescent lectin, correlated with concentrations of BFA that reversed
mRNA suppression. Monensin was also found to block the effects of
25-hydroxycholesterol on suppression of HMG-CoA reductase. However,
this ionophore decreased the other three sterol-regulated mRNAs to a
similar degree as 25-hydroxycholesterol. In contrast to CHO cells,
BFA-resistant PtK1 cells displayed normal down-regulation of HMG-CoA
reductase and an intact Golgi apparatus in the presence of BFA and
25-hydroxycholesterol. Cholesterol esterification in CHO cells was
stimulated to a similar extent by BFA (1 µg/ml) and
25-hydroxycholesterol, and simultaneous treatment of CHO cells with
both compounds was 60-70% additive. These results suggest that an
intact Golgi apparatus is required for 25-hydroxycholesterol-mediated
suppression of mRNA. The intracellular concentration of cholesterol is maintained
within a narrow range by several feedback mechanisms that halt
intracellular sterol synthesis and uptake via the LDL ( The cellular metabolite responsible for feedback
repression of cholesterol synthesis has not been identified.
Presumably, when sterol concentrations reach a critical level,
cholesterol or a related metabolite triggers a regulatory cascade that
culminates in transcriptional suppression and the other
sterol-regulated events described above. A class of potential
regulatory molecules is oxygenated sterols. These cholesterol
derivatives contain, in addition to the 3-hydroxyl group, a hydroxyl,
keto, or epoxide moiety(4, 5) . Oxysterols, such as
25-hydroxycholesterol, produce regulatory responses in cultured cells
similar to those of LDL, and the most potent are effective at nanomolar
concentrations. The mechanism of oxysterol suppression of cholesterol
synthesis is unclear. A high affinity binding protein for oxysterols,
oxysterol-binding protein (OSBP), has been identified (6, 8, 9) that displays affinity for various
oxysterols proportional to suppression of HMG-CoA reductase and
cholesterol synthesis in cultured cells(4, 7) . A clue
to OSBP function came from immunofluorescence studies on CHO cells
overexpressing the rabbit OSBP cDNA. OSBP was found to undergo
translocation from a cytoplasmic/vesicular compartment to the Golgi
apparatus in the presence of ligand(10) . This suggested that
the Golgi apparatus could be a target for oxysterol action, perhaps by
modifying sterol trafficking between cholesterol-rich plasma membrane
and cholesterol-poor ER(11) . Since localization of OSBP to
the Golgi apparatus was disrupted by BFA (10) , we further
tested whether BFA would disrupt the regulatory actions of OSBP or
other oxysterol signaling pathways requiring an intact Golgi apparatus.
Here, BFA and monensin was used to assess the involvement of the Golgi
apparatus in mediating the effects of 25-hydroxycholesterol on
suppression of mRNAs for several sterol-regulated genes in CHO and
BFA-resistant PtK1 cells. The effect of BFA on stimulation of
cholesterol esterification by 25-hydroxycholesterol was also
investigated.
Figure 1:
25-Hydroxycholesterol-mediated
suppression of mRNA in BFA-treated CHO cells. CHO cells were cultured
in medium B for 18-24 h prior to the start of experiments. Cells
then received 6 ml of medium B containing the indicated concentrations
of BFA (0-2 µg/ml). Following a 15-min incubation at 37
°C, new medium was added containing 25-hydroxycholesterol (2.5
µg/ml) and the indicated concentrations of BFA. Cells were
incubated for 4 h and harvested, and mRNA was quantitated. The results
are expressed relative to values for cells grown in medium B without
additions and are the average of two separate experiments normalized to
glyceraldehyde-3-phosphate dehydrogenase
mRNA.
Results of experiments in which cells were pretreated with
25-hydroxycholesterol (2.5 µg/ml) for 4 h prior to the addition of
BFA and oxysterol are shown in Fig. 2. BFA increased HMG-CoA
reductase and LDL receptor mRNA levels in a dose-dependent fashion by
30-40% relative to control, but did not return mRNA to fully
induced levels. HMG-CoA synthase mRNA was suppressed more completely
than the other three mRNAs, but the degree of reversal by BFA (30%
increase relative to control) was similar when compared with the LDL
receptor and HMG-CoA reductase. FPP synthase mRNA was also measured in
these experiments and was found not to be suppressed as efficiently by
25-hydroxycholesterol compared with the other three mRNAs or induced by
BFA to the same degree.
Figure 2:
Reversal of mRNA suppression by BFA in CHO
cells pretreated with 25-hydroxycholesterol. Cells were cultured for
18-24 h in medium B prior to the addition of 6 ml of medium B
containing 25-hydroxycholesterol (2.5 µg/ml). After a 4-h
incubation at 37 °C, cells received fresh medium B containing
25-hydroxycholesterol (2.5 µg/ml) and the indicated concentrations
of BFA. After 3 h, cells were harvested, total RNA was isolated, and
mRNA was quantitated as described under ``Experimental
Procedures.'' Results are expressed relative to control cells
grown in medium B without oxysterol or BFA and are the average of two
separate experiments normalized to expression of
glyceraldehyde-3-phosphate dehydrogenase.
The morphology of the Golgi apparatus in
cells treated with increasing concentrations of BFA was monitored using
FITC-labeled lentil lectin (Fig. 3). FITC-labeled lentil lectin
staining of untreated CHO cells revealed brightly fluorescent
structures clustered at one pole of the nucleus, indicative of the
Golgi apparatus (Fig. 3A). Treatment with 0.1 µg/ml
BFA (Fig. 3B) for 4 h did not alter this pattern
appreciably; however, treatment with 0.2 µg/ml BFA (Fig. 3C) produced some fragmenta-tion of the Golgi
apparatus and increased diffuse staining. This latter BFA concentration
produced a mixed population of cells with intact, partially disrupted,
or absent Golgi staining. Higher concentrations of BFA (0.5 and 1
µg/ml; Fig. 3, D and E, respectively)
resulted in uniform disruption of the Golgi apparatus and diffuse
staining around the nucleus, indicating absorption of lentil
lectin-binding oligosaccharides into the ER. BFA concentrations that
produced partial or complete disruption of the Golgi apparatus
correlated with reversal of 25-hydroxycholesterol suppression of mRNAs
as shown in Fig. 1and 2. FITC-labeled lentil lectin staining of
the Golgi apparatus in the presence of 25-hydroxycholesterol (2.5
µg/ml) and increasing BFA concentrations showed patterns of
staining indistinguishable from results shown in Fig. 3(data
not shown).
Figure 3:
Fluorescence localization of the Golgi
apparatus in BFA-treated CHO cells. Cells were cultured in medium B for
18 h prior to the addition of fresh medium B containing 0 (A),
0.1 (B), 0.2 (C), 0.5 (D), or 1.0 (E) µg/ml BFA. After treatment with BFA for 4 h, cells
were fixed and stained with FITC-labeled lentil lectin as described
under ``Experimental Procedures.'' Control cells received
ethanol. Bar, 10 µm.
The concentration of BFA required to disrupt the Golgi
apparatus in these studies was found to be higher then previously
reported: 0.1 µg/ml BFA is usually sufficient to collapse the Golgi
apparatus into the ER(24) . This can be explained by the
relatively long incubation times required to achieve suppression of
sterol-regulated transcription and active degradation of BFA in CHO
cells(25) . However, extended incubations with relatively high
concentrations of BFA did not adversely affect cell viability.
Incubation of CHO cells for 4 h with 0.1 and 1.0 µg/ml BFA
inhibited protein synthesis (as measured by
[
Figure 4:
Suppression of mRNA by
25-hydroxycholesterol in control and BFA-treated CHO cells. A,
CHO cells were cultured in medium B for 18-24 h prior to the
start of experiments. Cells were pretreated with 6 ml of medium B
containing either 1 µg/ml BFA (
Figure 5:
BFA does not reverse suppression of
HMG-CoA reductase mRNA by 25-hydroxycholesterol in PtK1 cells. CHO and
PtK1 cells were grown in medium B (supplemented with 1 mM pyruvate for PtK1 cells) for 24-36 h prior to treatment for
6 h with medium B containing ethanol solvent (no additions (NA)), 2.5 µg/ml 25-hydroxycholesterol (25-OH), 1
µg/ml BFA, or both (BFA/25-OH). Cells received BFA 15 min
prior to oxysterol addition. RNA was harvested, and HMG-CoA reductase
and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA
levels were quantitated as described under ``Experimental
Procedures.'' Results are expressed relative to cells that
received no additions and are the means ± S.D. of five separate
experiments.
We also investigated the
effect of forskolin on Golgi disruption by BFA. Forskolin was
previously shown to prevent disruption of the Golgi apparatus by BFA
via a cAMP-independent mechanism(24) . This was not the case
for mRNA suppression, as illustrated by the lack of effect of forskolin
on BFA reversal of oxysterol suppression (Fig. 6). Treatment of
cells with 100 µM forskolin had no effect on base-line
mRNA levels of HMG-CoA reductase or HMG-CoA synthase. Similarly,
forskolin in combination with BFA and 25-hydroxycholesterol did not
suppress mRNA levels to those of 25-hydroxycholesterol-treated cells
during a 4-h incubation period. Forskolin (100 µM)
prevented disruption of the Golgi apparatus by a 30-min treatment with
0.1 µg/ml BFA, as determined by FITC-labeled lentil lectin
fluorescence staining, but was ineffective when cells were treated for
4 or 6 h with 1 µg/ml BFA (data not shown).
Figure 6:
Forskolin does not reverse BFA suppression
of HMG-CoA reductase and HMG-CoA synthase mRNAs by
25-hydroxycholesterol. CHO cells were cultured in medium B for
18-24 h prior to the start of experiments. Cells were pretreated
with forskolin (FK; 100 µM) and BFA (1 µg/ml)
for 60 and 30 min, respectively, prior to the addition of
25-hydroxycholesterol (25-OH; 2.5 µg/ml) for 4 h (cells
treated with BFA/25-hydroxycholesterol/forskolin received forskolin
first). Cells not receiving these additions (no additions (N/A)) were mock-treated with ethanol solvent. Results are
expressed relative to values for cells grown in medium B and are the
average of two separate experiments normalized to expression of
glyceraldehyde-3-phosphate dehydrogenase.
Figure 7:
25-Hydroxycholesterol suppression of mRNA
in CHO cells treated with monensin. CHO cells were grown in medium B
for 18 h prior to the start of experiments. A, fresh medium B
containing 2.5 µg/ml 25-hydroxycholesterol (25-OH); 10
µM monensin (Mon) and 2.5 µg/ml
25-hydroxycholesterol; 100 µM monensin and 2.5 µg/ml
25-hydroxycholesterol; or 100 µM monensin was added to
cells for 4 h. Cells were pretreated with the indicated concentration
of monensin for 30 min prior to oxysterol addition. Results are the
means ± S.D. of four experiments. B, cells received
medium B with 25-hydroxycholesterol (2.5 µg/ml) for 4 h. Fresh
medium was added containing the indicated concentrations of monensin
and 2.5 µg/ml 25-hydroxycholesterol for an additional 3 h. Total
RNA was isolated, and mRNAs for the LDL receptor (
Similar to the results for BFA shown
in Fig. 2, monensin (50-100 µM) reversed the
suppressive effects of 25-hydroxycholesterol on HMG-CoA reductase after
CHO cells had been exposed to oxysterol for 4 h (Fig. 7B). However, the ionophore did not reverse
suppression of mRNA for the LDL receptor, HMG-CoA synthase, or FPP
synthase. The effect of monensin on Golgi apparatus structure was
assessed by fluorescence staining with FITC-labeled lentil lectin (Fig. 8). Monensin caused the Golgi apparatus to lose its
compact perinuclear staining pattern and to assume a more disaggregated
and dilated structure. However, staining did not appear to become
diffuse and associated with other organelles as observed with BFA.
Protein synthesis in CHO cells (as measured by
[
Figure 8:
Fluorescence localization of the Golgi
apparatus in monensin-treated CHO cells. Cells were cultured in medium
B for 18 h prior to the addition of fresh medium B containing 0 (A), 10 (B), or 100 (C) µM monensin. After treatment with monensin for 4 h, cells were fixed
and stained with FITC-labeled lentil lectin as described under
``Experimental Procedures.'' Control cells received ethanol. Bar, 10 µm.
Prevention of oxysterol suppression of
mRNAs by BFA or monensin could be due to message stabilization. To test
this possibility, we treated CHO cells with actinomycin D (10
µg/ml) in the presence of 1 µg/ml BFA, 100 µM monensin, or no additions and determined transcript levels for
HMG-CoA reductase, HMG-CoA synthase, and glyceraldehyde-3-phosphate
dehydrogenase after 2 and 4 h (Fig. 9). The half-life of HMG-CoA
reductase was 4.4 h in control cells, compared with 5.2 h in
BFA-treated cells and 9.2 h in monensin-treated cells.
Glyceraldehyde-3-phosphate dehydrogenase mRNA was considerably more
stable, and half-lives ranged from 20 h in control and BFA-treated
cells to 45 h in monensin-treated cells (glyceraldehyde-3-phosphate
dehydrogenase mRNA levels decreased by <10% after 4 h). HMG-CoA
synthase mRNA stability was not appreciably altered by BFA or monensin
treatment (half-lives ranged from 7 to 5.5 h). Similarly, LDL receptor
mRNA stability was unaltered by BFA or monensin (data not shown).
Figure 9:
Effect of BFA and monensin on stability of
mRNAs for HMG-CoA reductase, HMG-CoA synthase, and
glyceraldehyde-3-phosphate dehydrogenase. CHO cells were cultured in
medium B for 18 h prior to the addition of medium B containing 10
µg/ml actinomycin D and either 100 µM monensin
(
In this study, we sought to identify the Golgi apparatus as a
target for the action of oxysterols in suppressing transcription of
sterol-regulated genes. The recent identification of a membrane-bound
transcription factor in the ER that undergoes sterol-regulated
proteolysis to a soluble nuclear form is the first clear evidence that
membrane sterol content regulates
transcription(30, 31) . However, it is still unclear
how ER cholesterol content is regulated and the immediate target for
this pool of regulatory sterol. Results from this study indicate that
disruption of the Golgi apparatus by BFA makes the aforementioned
pathway for sterol-regulated transcription insensitive to oxysterol. Measurement of steady-state mRNA levels of four sterol-regulated
genes revealed that BFA consistently prevented suppression by
25-hydroxycholesterol in CHO cells. BFA is a fungal toxin that
selectively disrupts the cis/medial/trans-elements of the
Golgi apparatus(32) , with limited effects on the trans-Golgi network, endosomal system, and
lysosomes(33, 34) . BFA blocks anterograde transport
from the ER to the Golgi apparatus in the presence of sustained
retrograde movement, causing absorption of Golgi membranes and contents
into the ER(32) . This is achieved by inhibiting GTP exchange
on the ADP-ribosylation factor, thus preventing coatomer assembly and
vesicle formation(35, 36) . The degree to which BFA
prevented oxysterol suppression of transcription varied, as did the
extent of suppression by 25-hydroxycholesterol of the four mRNAs. The
reasons for these differences are unclear, but suggest that the
promoters of the four genes respond differently to oxysterol and
disruption of the Golgi apparatus by BFA. This was most evident for
HMG-CoA synthase, where BFA pretreatment ( Fig. 1and Fig. 4) was only partially effective in preventing suppression
by 25-hydroxycholesterol. This suggests that there is more than one
mechanism for suppression of HMG-CoA synthase mRNA and that BFA is only
effective in blocking one component of transcriptional suppression.
There is some evidence that sterols act via different response elements
and transcription factors depending on the promoter. HMG-CoA reductase
is suppressed by sterols, but suppression is not mediated by the same
sterol regulatory element (37) or sterol regulatory
element-binding proteins (30, 38, 39) responsible for sterol-regulated
transcription of the LDL receptor or HMG-CoA synthase. Similarly,
sterol-mediated suppression of FPP synthase transcription does not seem
to involve sterol regulatory element-like promoter
sequences(23) . However, there must be a common
sterol-dependent mechanism that initiates transcriptional suppression
of these genes that is in turn inhibited by BFA. Several findings
from this study suggest that an intact functional Golgi apparatus is
required for oxysterol-mediated transcriptional suppression. First,
disruption by BFA of the fluorescence staining pattern of a Golgi
apparatus-specific lectin was found to correlate with inhibition by
25-hydroxycholesterol of sterol-regulated mRNA. Second, PtK1 cells were
found to be resistant to BFA effects on 25-hydroxycholesterol-mediated
suppression of HMG-CoA reductase mRNA. This rules out the possibility
that BFA antagonizes 25-hydroxycholesterol by acting on a target
outside the Golgi apparatus. Results with PtK1 cells were also
important since forskolin did not antagonize BFA in our system. In
other studies, forskolin was only partially effective in reversing
stimulation of sphingolipid synthesis (28, 40, 41) or cholesterol esterification (28, 29) by BFA. Third, BFA partially restored mRNA
levels in CHO cells pretreated with 25-hydroxycholesterol for 4 h. This
finding is particularly important since it suggests that an intact
functional Golgi apparatus is necessary for suppression by oxysterol to
be maintained after an initial regulatory signal has been received. Findings similar to those described above for BFA were also seen
with another Golgi apparatus-specific agent, monensin. However, in the
case of monensin, reversal of 25-hydroxycholesterol suppression was
only observed for HMG-CoA reductase mRNA; HMG-CoA reductase mRNA
degradation was reduced; and mRNAs for the other three genes were
decreased. Differences in the mechanism of action of BFA and monensin
could account for some of these results. Unlike BFA, monensin did not
completely fragment the Golgi apparatus, which appeared to vacuolate as
previously reported(27) , and is specific for the trans-elements of the organelle. It should also be noted that
a concentration of monensin 10-fold greater than that required to
elicit effects in other systems (27) was necessary to reverse
HMG-CoA reductase mRNA suppression. BFA is known to stimulate
cholesterol esterification and to modify lipid metabolism in several
cell types(28, 29, 40, 41) , a
result that is confirmed here using CHO cells. In this study, BFA and
25-hydroxycholesterol stimulated cholesterol esterification to a
similar extent, but when added to cells simultaneously, 70% of the
individual activity was reached. There are two possible explanations.
First, acyl-CoA:cholesterol acyltransferase is not maximally activated
by either 25-hydroxycholesterol and BFA, and adding both compounds to
cells is not 100% additive due to saturation of enzyme activity.
Second, BFA could have a minor inhibitory effect on
25-hydroxycholesterol-stimulated acyl-CoA:cholesterol acyltransferase
activity, but this is masked by the stimulation BFA itself causes. Both
of these scenarios imply that BFA has little or no effect on
25-hydroxycholesterol-stimulated acyl-CoA:cholesterol acyltransferase
activity. BFA has been shown to partially block increased cholesterol
esterification caused by sphingomyelinase treatment of intact cells,
suggesting that it will modify cholesterol influx to the ER and
acyl-CoA:cholesterol acyltransferase activity under some conditions (42) . The observation that BFA and 25-hydroxycholesterol
stimulate acyl-CoA:cholesterol acyltransferase activity suggests that
both compounds cause sequestration of cholesterol within the ER, where
it is available for esterification. This could be accomplished by
stimulating influx of cholesterol from another membrane (i.e. Golgi apparatus) or inhibiting movement of cholesterol out of the
ER. Since BFA does not inhibit cholesterol transport from the ER (43) and based on the mechanism of action of BFA and results
with PtK1 cells, the latter scenario involving redistribution of Golgi
apparatus cholesterol seems more probable. If absorption of Golgi
apparatus cholesterol into the ER is responsible for BFA-mediated
acyl-CoA:cholesterol acyltransferase activation, then the Golgi
apparatus must be a relatively rich source of cholesterol. The
cholesterol content of the Golgi apparatus can be altered by LDL
deprivation or increased cholesterol synthesis, as measured indirectly
by the fluorescence properties of a ceramide analogue that localizes to
the Golgi apparatus (44) . Cisternae of the Golgi apparatus
have a distinct polarity in cholesterol content, with the trans-elements enriched in sterol relative to the cis-elements(45, 46) . Changes in cholesterol
content of the Golgi apparatus might indicate that flow of sterol into
or out of the organelle was perturbed, perhaps as the result of
increased cholesterol synthesis or uptake of lipoproteins. Little is
known of the effects of BFA on other aspects of cholesterol regulation.
BFA did not influence degradation of a HMG-CoA
reductase/ OSBP is a potential candidate for transducing the effects
of oxysterols either by interaction with the Golgi apparatus or by
release from a vesicular or cytoplasmic pool(7) . If OSBP
translocation/attachment to the Golgi apparatus is required for
transducing the effects of oxysterols, then disruption of the Golgi
apparatus would interrupt this process. This is precisely the result
when BFA is used to antagonize 25-hydroxycholesterol suppression of
sterol-regulated mRNAs. There is no direct evidence that OSBP has
signal transduction properties, but recent reports have shown that OSBP
and a number of other proteins with known or suspected roles in signal
transduction have a domain with homology to pleckstrin, a protein
kinase C substrate in platelets(48) . The pleckstrin homology
domain is thought to mediate protein-protein or protein-lipid
interactions(49, 50) . If, as this study suggests,
the Golgi apparatus is involved in transducing the signal for
transcriptional suppression by 25-hydroxycholesterol, then it follows
that lipid or protein sorting and trafficking in the Golgi network are
regulated by oxysterols. Alternatively, absorption of Golgi proteins or
lipids into the ER makes transcription insensitive to
25-hydroxycholesterol by a mechanism that has yet to be identified.
Volume 270,
Number 14,
Issue of April 7, 1995 pp. 8023-8031
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
)receptor(1) . Elevation of cellular cholesterol
results in suppression of transcription of genes for several
cholesterol biosynthetic enzymes and the LDL receptor. Transcription
and translation of HMG-CoA reductase mRNA are reduced, and degradation
of the enzyme is increased by sterol and nonsterol factors(2) .
Excess cholesterol within the cell is rapidly converted to its ester by
the action of acyl-CoA:cholesterol acyltransferase and stored in
cytoplasmic droplets(3) . The net result is decreased
cholesterol synthesis and storage of cholesterol until such time that
it is required for synthesis of membranes, lipoproteins, bile acids, or
hormones.
Materials
BFA was purchased from Calbiochem and
stored as a 1 mg/ml stock solution in ethanol at -20 °C.
[
-
P]dATP and
[1-
H]oleate were from DuPont NEN.
Tran
S-label was from ICN Biomedicals Inc.
25-Hydroxycholesterol was from Steraloids Inc. (Wilton, NH). Silica Gel
G thin-layer chromatography plates were from BDH. S1 nuclease (from Aspergillus oryzae) was purchased from Sigma or Life
Technologies, Inc. FITC-labeled lentil lectin (Lens culinaris)
and the sodium salt of monensin were from Sigma.Cell Culture
CHO-K1 cells (ATCC CCL61) were
cultured in Dulbecco's modified Eagle's medium containing
5% fetal calf serum and 34 µg/ml proline (medium A) in an
atmosphere of 5% CO
at 37 °C. Cells were seeded at a
density of 400,000/100-mm dish in 8 ml of medium A on day 0. On day 3,
medium was replaced with Dulbecco's modified Eagle's medium
containing 5% delipidated fetal calf serum and 34 µg/ml proline
(medium B). Experiments were started 18-24 h after this media
change when cells were
70% confluent. Marsupial kidney PtK1 cells
(ATCC CCL35) were cultured in modified Eagle's medium containing
10% fetal calf serum and 1 mM pyruvate. Delipidated fetal calf
serum was prepared by centrifugation at a density of 1.21 g/ml as
described previously(12) .mRNA Analysis
Total RNA was isolated by
centrifugation through CsCl (13) or by the guanidinium
thiocyanate/phenol/chloroform extraction method(14) . All
probes for quantitative S1 nuclease protection assays were prepared
from M13 templates and purified by 7 M urea, 6% polyacrylamide
gel electrophoresis(15) . HMG-CoA synthase was measured using a
single-stranded antisense probe corresponding to nucleotides
28-416 (HindIII-PstI fragment) of the hamster
cDNA(16) . The S1 probe for FPP synthase was prepared by
cloning a polymerase chain reaction fragment amplified from CHO cDNA
corresponding to nucleotides 520-717 of the rat
cDNA(18) . For the glyceraldehyde-3-phosphate dehydrogenase S1
probe, primers containing 5`-restriction sites were used to amplify a
95-base pair fragment corresponding to nucleotides 467-562 of the
human glyceraldehyde-3-phosphate dehydrogenase cDNA(19) . A
135-base pair S1 probe for PtK1 HMG-CoA reductase was prepared by
polymerase chain reaction amplification of a region corresponding to
amino acids 654-705 of the human enzyme(20) . The
nucleotide sequence of PtK1 reductase was unique and 85% conserved
compared with the human sequence. S17 ribosomal protein mRNA was
quantitated by primer extension (17) and used as an internal
load control in some experiments. S1 analysis of the various mRNAs was
performed on 10-15 µg of total RNA and an excess of
single-stranded DNA probe (0.3 10
cpm/µg).
Hybridization and digestion with S1 nuclease for the CHO LDL receptor
and HMG-CoA reductase probes were as described previously(17) .
Hybridization of HMG-CoA reductase (PtK1), HMG-CoA synthase, FPP
synthase, and glyceraldehyde-3-phosphate dehydrogenase S1 probes with
RNA was at 80 °C for 10 min followed by 37 °C for 16 h in 90%
formamide buffer. S1 nuclease-digested products were separated on 7 M urea, 6% polyacrylamide gels. mRNA was quantitated from
autoradiograms by image analysis on a Macintosh Apple OneScanner using
the NIH Image software package (version 1.47).Fluorescence Staining of the Golgi Apparatus
Cells
were fixed in 3% (v/v) formaldehyde, and the Golgi apparatus was
visualized with FITC-labeled lentil lectin as described
previously(9) . Fluorescence microscopy was performed on an
Olympus microscope using a 40 PlanApo objective and an
excitation/emission filter package for FITC fluorescence. Cells were
photographed with Kodak Technical Pan black and white film.
Acyl-CoA:Cholesterol Acyltransferase Assays
The
incorporation of [1-
H]oleate into cholesteryl
[1-
H]oleate in cultured CHO cells was measured as
described previously (12) by incubating cells with 0.1 mM [1-
H]oleate-bovine serum albumin complex
(6000-7000 dpm/nmol) for 1 h at 37 °C.
[1-
H]Oleate-labeled lipid extracts were applied
to thin-layer chromatography plates and separated in a solvent system
of hexane/diethyl ether/acetic acid (90:30:1, v/v). Cholesteryl ester
and triacylglyceride were visualized by brief exposure to iodine or
autoradiography, scraped from the plate, and quantitated by liquid
scintillation counting. Cell protein was measured by the method of
Lowry et al.(21) .Protein Synthesis
CHO cells were preincubated in
methionine-free medium B with BFA or monensin for 2 h, followed by the
addition of [
S]methionine/cysteine (5
µCi/ml) for an additional 2 h. Cell monolayers were washed twice
with cold phosphate-buffered saline and solubilized with 1 ml of 0.5 N NaOH, and aliquots were spotted on glass-fiber filters,
which were then washed in 10% (w/v) trichloroacetic acid followed by
75% (v/v) ethanol. Filters were dried, and radioactivity was measured
and expressed relative to total cell protein. Protein synthesis in BFA-
and monensin-treated cells is expressed as a percentage of control
(4.04 10
dpm/µg of protein/2 h).
Dose-dependent Reversal by Brefeldin A of
25-Hydroxycholesterol Suppression of mRNAs
Cells grown in medium
lacking an exogenous source of cholesterol maximally express mRNAs for
at least three cholesterol biosynthetic enzymes and the LDL receptor.
The addition of LDL or an oxysterol, such as 25-hydroxycholesterol, to
cells produces rapid transcriptional suppression of these
sterol-regulated genes(1, 22, 23) . We found
that treatment of CHO cells with 25-hydroxycholesterol (2.5 µg/ml)
for 4 h was sufficient for 50-80% suppression of mRNAs for the
LDL receptor, HMG-CoA reductase, and HMG-CoA synthase (Fig. 1).
BFA, when added to cells 15 min prior to the addition of
25-hydroxycholesterol, overcame 25-hydroxycholesterol-mediated
suppression of mRNAs for these sterol-regulated genes in a
dose-dependent fashion. The relative levels of HMG-CoA reductase and
LDL receptor mRNAs returned to fully induced values (i.e. cells grown in medium containing delipidated serum) in the
presence of 1 µg/ml BFA, while the level of HMG-CoA synthase mRNA,
which was more fully suppressed then the other two mRNAs, increased
2-fold.
S]methionine/cysteine incorporation) by 0 and
15%, respectively, relative to untreated controls.Effect of BFA on mRNA Levels in Control and
Oxysterol-treated Cells
The effect of BFA on mRNA suppression
over a 4-h period was investigated to further determine the influence
of this drug on steady-state mRNA levels in control and
oxysterol-treated cells (Fig. 4, A and B). BFA
treatment of CHO cells for 4 h in the absence of 25-hydroxycholesterol
(0-h control plus BFA; Fig. 4, A and B)
produced a variable 0-20% reduction in mRNA levels for the four
sterol-regulated genes relative to control cells grown in delipidated
serum for the same period (0-h control minus BFA; Fig. 4, A and B). Cells treated with 25-hydroxycholesterol and BFA
for 2 and 4 h showed a marked delay or absence of suppression compared
with cells that received oxysterol alone. This delay was most notable
for HMG-CoA reductase and LDL receptor mRNAs, where BFA almost
completely antagonized 25-hydroxycholesterol effects.
25-Hydroxycholesterol suppressed HMG-CoA synthase mRNA by 50% in the
presence of BFA, compared with 80% in cells that received only
oxysterol for 4 h. FPP synthase mRNA was not suppressed as strongly by
25-hydroxycholesterol, consistent with a previous report using HepG2
cells(22) , and BFA prevented oxysterol-mediated suppression.
Expression of the four oxysterol-regulated mRNAs was normalized to mRNA
for S17 ribosomal protein or glyceraldehyde-3-phosphate dehydrogenase,
and the amount of these two mRNAs did not change relative to total RNA
for the treatments employed here. Representative primer extension and
S1 nuclease protection analysis are shown in Fig. 4B.
) or ethanol solvent
(
) for 15 min before the addition of 2.5 µg/ml
25-hydroxycholesterol. Cells were incubated for the indicated times,
RNA was harvested, and mRNAs for sterol-regulated genes were assayed as
described under ``Experimental Procedures.'' Results are the
means ± S.D. of four separate experiments. Values are relative
to expression in cells cultured for 4 h in medium B with no additions. B, shown are representative autoradiograms of S1 nuclease
protection and primer extension assays. The band above the regulated
HMG-CoA synthase S1 nuclease product is residual undigested probe.
Films were developed after 2-12 h of exposure to dried gels at
-70 °C. 25-OH, 25-hydroxycholesterol; GAPDH, glyceraldehyde-3-phosphate
dehydrogenase.
Disruption of the Golgi Apparatus Is Necessary to Reverse
Suppression of mRNA by 25-Hydroxycholesterol
PtK1 cells were
tested for resistance to BFA suppression of the effects of
25-hydroxycholesterol on HMG-CoA reductase mRNA. Marsupial kidney PtK1
cells contain a dominant, nondiffusible factor associated with the
Golgi apparatus that renders the organelle resistant to high
concentrations of BFA(26) . S1 nuclease protection assays were
used to measure HMG-CoA reductase mRNA in PtK1 and CHO cells treated
with BFA and 25-hydroxycholesterol for 6 h (Fig. 5). PtK1
HMG-CoA reductase mRNA was suppressed 40% by 25-hydroxycholesterol
compared with the 70% suppression of CHO reductase mRNA. BFA alone
caused a slight increase in mRNA levels in PtK1 cells and, as shown by
previous results, did not affect CHO HMG-CoA reductase mRNA levels. As
expected, BFA reversed suppression by 25-hydroxycholesterol in CHO
cells, but was ineffective in PtK1 cells. In agreement with a previous
report(26) , the Golgi apparatus of PtK1 cells was intact
following a 6-h treatment with 2 µg/ml BFA (determined by
FITC-labeled lentil lectin staining).
Effect of Monensin on Oxysterol-regulated mRNA
The
effect of monensin, an ionophore known to perturb the trans-cisternae of the Golgi apparatus(27) , on
25-hydroxycholesterol-mediated suppression of mRNA was investigated.
Unlike BFA, monensin alone had significant and varied effects on
sterol-regulated mRNA levels (Fig. 7A). mRNAs for
HMG-CoA synthase, FPP synthase, and the LDL receptor were suppressed to
similar degrees by monensin (100 µM) or
25-hydroxycholesterol (2.5 µg/ml). HMG-CoA reductase mRNA was not
suppressed by 100 µM monensin, and suppression by
25-hydroxycholesterol was prevented by pretreatment with 100 µM monensin. The antagonistic effect of monensin on suppression of
HMG-CoA reductase mRNA is similar to that of BFA shown in Fig. 1and Fig. 4.
), HMG-CoA
reductase (
), HMG-CoA synthase (
), and FPP synthase
(
) were assayed as described under ``Experimental
Procedures.'' Results are the average of two experiments and are
expressed relative to mRNA from cells grown in medium B with no
additions.
S]methionine/cysteine labeling) was inhibited
by 40 and 60% during a 4-h incubation with 10 and 100 µM monensin, respectively.
,
), 1 µg/ml BFA (
, ), or ethanol
solvent alone (
,
). Cells were harvested for RNA immediately
following medium addition (0 h) and at 2 and 4 h. Solidsymbols in the upperpanel indicate
glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA, and opensymbols indicate HMG-CoA reductase mRNA. Results
are representative of two separate
experiments.
BFA- and Oxysterol-activated Cholesterol
Esterification
If BFA was antagonizing cholesterol synthesis or
movement in the cell, it was reasoned that acyl-CoA:cholesterol
acyltransferase activation by 25-hydroxycholesterol would also be
affected. Acyl-CoA:cholesterol acyltransferase activity in CHO and PtK1
cells treated with 25-hydroxycholesterol and/or BFA for 2 and 4 h is
shown in Table 1. Interpretation of these experiments is
complicated by the fact that cholesteryl ester synthesis in CHO cells
was increased by BFA, as reported previously for other cell
lines(28, 29) . Individually, BFA and
25-hydroxycholesterol increased cholesterol esterification in CHO cells
by a similar degree, and when added together, the stimulation of
cholesterol esterification was 79 and 73% of the calculated
esterification rate for the two compounds at 2 and 4 h, respectively.
However, comparisons for the 4-h time points are difficult owing to the
significant error associated with these measurements. In PtK1 cells,
acyl-CoA:cholesterol acyltransferase activity was increased 4- and
7-fold by 25-hydroxycholesterol at 2 and 4 h, respectively. In PtK1
cells, unlike in CHO cells, BFA increased cholesterol esterification by
only 1.3- and 1.7-fold at 2 and 4 h, respectively, and had no effect
when in combination with 25-hydroxycholesterol. BFA had a complex and
variable effect on triacylglyceride synthesis, either increasing or
decreasing synthesis in the two cell lines depending on the treatment
time.
-galactosidase fusion protein in response to mevalonate,
but did increase the protein half-life 2-fold in the absence of this
degradative signal(47) . The lack of effect of BFA on
mevalonate-mediated degradation precludes the Golgi apparatus in
reductase degradation. However, stabilization in the absence of
mevalonate is difficult to reconcile given that increased ER
cholesterol levels and cholesterol esterification result from BFA
treatment.
)
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
![]()
CiteULike
Complore
Connotea
Del.icio.us
Digg
Reddit
Technorati What's this?
This article has been cited by other articles:
![]() |
T. Nishimura, T. Inoue, N. Shibata, A. Sekine, W. Takabe, N. Noguchi, and H. Arai Inhibition of cholesterol biosynthesis by 25-hydroxycholesterol is independent of OSBP Genes Cells, August 1, 2005; 10(8): 793 - 801. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. J. Perry and N. D. Ridgway The role of de novo ceramide synthesis in the mechanism of action of the tricyclic xanthate D609 J. Lipid Res., January 1, 2004; 45(1): 164 - 173. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. P. Wyles, C. R. McMaster, and N. D. Ridgway Vesicle-associated Membrane Protein-associated Protein-A (VAP-A) Interacts with the Oxysterol-binding Protein to Modify Export from the Endoplasmic Reticulum J. Biol. Chem., August 9, 2002; 277(33): 29908 - 29918. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. J. Schroepfer Jr. Oxysterols: Modulators of Cholesterol Metabolism and Other Processes Physiol Rev, January 1, 2000; 80(1): 361 - 554. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Laitinen, V. M. Olkkonen, C. Ehnholm, and E. Ikonen Family of human oxysterol binding protein (OSBP) homologues: a novel member implicated in brain sterol metabolism J. Lipid Res., December 1, 1999; 40(12): 2204 - 2211. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Nohturfft, R. A. DeBose-Boyd, S. Scheek, J. L. Goldstein, and M. S. Brown Sterols regulate cycling of SREBP cleavage-activating protein (SCAP) between endoplasmic reticulum and Golgi PNAS, September 28, 1999; 96(20): 11235 - 11240. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. W. Underwood, N. L. Jacobs, A. Howley, and L. Liscum Evidence for a Cholesterol Transport Pathway from Lysosomes to Endoplasmic Reticulum That Is Independent of the Plasma Membrane J. Biol. Chem., February 13, 1998; 273(7): 4266 - 4274. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Arita, K. Nomura, H. Arai, and K. Inoue alpha -Tocopherol transfer protein stimulates the secretion of alpha -tocopherol from a cultured liver cell line through a brefeldin A-insensitive pathway PNAS, November 11, 1997; 94(23): 12437 - 12441. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. L. Baum, E. J. Reschly, A. K. Gayen, M. E. Groh, and K. Schadick Sterol Carrier Protein-2 Overexpression Enhances Sterol Cycling and Inhibits Cholesterol Ester Synthesis and High Density Lipoprotein Cholesterol Secretion J. Biol. Chem., March 7, 1997; 272(10): 6490 - 6498. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. B. Neufeld, A. M. Cooney, J. Pitha, E. A. Dawidowicz, N. K. Dwyer, P. G. Pentchev, and E. J. Blanchette-Mackie Intracellular Trafficking of Cholesterol Monitored with a Cyclodextrin J. Biol. Chem., August 30, 1996; 271(35): 21604 - 21613. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Xu, Y. Liu, N. D. Ridgway, and C. R. McMaster Novel Members of the Human Oxysterol-binding Protein Family Bind Phospholipids and Regulate Vesicle Transport J. Biol. Chem., May 18, 2001; 276(21): 18407 - 18414. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||