36 E-mail: biohrdc@fmed.uba.ar Received 27 July 2006, revised 23 August 2006, accepted 12 September 2006 doi:10.1111/j.1742-4658.2006.05496.x We have investigated the direct effect of ar
Trang 1through the induction of arachidonic acid release inside this organelle in Leydig cells
Ana Fernanda Castillo, Fabiana Cornejo Maciel, Rocı´o Castilla, Alejandra Duarte, Paula Maloberti, Cristina Paz and Ernesto J Podesta´
Department of Biochemistry, School of Medicine, University of Buenos Aires, Argentina
Arachidonic acid (AA) is a fatty acid with 20 carbons
and four cis double bonds that are the source of its
flexibility and its reactivity with molecular oxygen The
oxidation can happen nonenzymatically or through the
action of three types of oxygenases: cyclooxygenase,
lipoxygenase and cytochrome P450 Most of the effects
of AA are attributable to its conversion by those
enzymes to prostaglandins, leukotrienes and other
bio-active products [1] AA itself also has biological
activ-ity; however, the number of its described actions is
reduced compared to the effects described for the AA
metabolites Moreover, it is not very well documented
whether nonmetabolized AA is released and elicits spe-cial functions in a specific cellular compartment [2] Transport of long-chain fatty acids in cells definitely occurs when they are tightly linked to CoA by esterifi-cation catalyzed by acyl-CoA synthetases [3] In mam-malian and yeast cells [4] it appears that the acyl-CoA synthetases merely enhance uptake indirectly Thus, formation of the polar CoA-ester effectively traps the fatty acid in the cell and functions as part of a facilita-ted distribution in different cellular compartments The mechanisms involved in the compartmentaliza-tion of long-chain acyl-CoA esters and free fatty acids
Keywords
acyl-CoA synthetase; acyl-CoA thioesterase;
arachidonic acid compartmentalization;
Leydig cells; steroidogenesis
Correspondence
E J Podesta´, Department of Biochemistry,
School of Medicine, University of Buenos
Aires, Paraguay 2155–5th, C1121ABG,
Buenos Aires, Argentina
Fax: +54 11 45083672 ext 31
Tel: +54 11 45083672 ext 36
E-mail: biohrdc@fmed.uba.ar
(Received 27 July 2006, revised 23 August
2006, accepted 12 September 2006)
doi:10.1111/j.1742-4658.2006.05496.x
We have investigated the direct effect of arachidonic acid on cholesterol transport in intact cells or isolated mitochondria from steroidogenic cells and the effect of cyclic-AMP on the specific release of this fatty acid inside the mitochondria We show for the first time that cyclic-AMP can regulate the release of arachidonic acid in a specialized compartment of MA-10 Leydig cells, e.g the mitochondria, and that the fatty acid induces choles-terol transport through a mechanism different from the classical pathway Arachidonic acid and arachidonoyl-CoA can stimulate cholesterol trans-port in isolated mitochondria from nonstimulated cells The effect of arach-idonoyl-CoA is inhibited by the reduction in the expression or in the activity of a mitochondrial thioesterase that uses arachidonoyl-CoA as a substrate to release arachidonic acid cAMP-induced arachidonic acid accu-mulation into the mitochondria is also reduced when the mitochondrial thioesterase activity or expression is blocked This new feature in the regu-lation of cholesterol transport by arachidonic acid and the release of arachidonic acid in specialized compartment of the cells could offer novel means for understanding the regulation of steroid synthesis but also would
be important in other situations such as neuropathological disorders or oncology disorders, where cholesterol transport plays an important role
Abbreviations
AA, arachidonic acid; AA-CoA, arachidonoyl-CoA; Acot2, mitochondrial acyl-CoA thioesterase; ACS4, acyl-CoA synthetase 4; BPB,
4-bromophenacyl bromide; 8Br-cAMP, 8-bromo-cAMP; CHX, cycloheximide; CPT1, carnitine-palmitoyl transferase 1; DBI, diazepam-binding inhibitor; NDGA, nordihydroguaiaretic acid; P450scc, cholesterol side-chain cleavage cytochrome P-450 enzyme; PBR, peripheral
benzodiazepan receptor; StAR, steroidogenic acute regulatory protein.
Trang 2are important unresolved issues [5] The simple
struc-ture of AA and the natural occurrence of so many
close chemical analogues are, not surprisingly,
associ-ated with a lack of specificity The selective actions of
free AA may be explained simply by its specific release
under physiological conditions and by the absence of
such mechanisms for releasing other long-chain fatty
acids, compounds which might otherwise share its
bio-chemical effects [2] Thus, the accessibility of AA to a
specific cellular compartment and the specificity of its
action are certainly linked
The enzymes involved in the release of AA have
been well characterized, with the phospholipase A2
being the most important [6] However, it remains
unclear as to how exactly AA is released in a specific
compartment of the cells under physiological
condi-tions [2] Recently, using steroidogenic cells as an
experimental system, we described an alternative
releasing mechanism for AA as a mediator of hormone
action with the participation of an acyl-CoA
synthe-tase (ACS4) and a mitochondrial acyl-CoA
thioest-erase (Acot2) [7,8] ACS4 has been described as an
AA-preferring acyl-CoA synthetase [9], while Acot2 is
a member of a new thioesterase family with long-chain
acyl-CoA thioesterase activity and it is associated with
the inner mitochondrial membrane [10–13]
In the steroidogenic cells, the step that determines
the rate of steroid synthesis (the rate-limiting step) is
the transport of cholesterol to the inner mitochondrial
membrane [14], a process in which ACS4 and Acot2
play a key role In this mechanism, it has been
sugges-ted that ACS4 and Acot2 may constitute a system to
deliver AA into a specific intracellular compartment,
e.g the mitochondria [8]
AA plays a crucial role in the steroidogenic cells,
mediating the induction of the steroidogenic acute
reg-ulatory (StAR) protein, one of the proteins involved in
cholesterol transport [15–19] Although it is clear from
previous publications that AA plays its role in the
pro-cess through the conversion to its lipoxygenated
metabolites [15,17,20], a direct action of this fatty acid
on cholesterol transport into the mitochondria cannot
be ruled out
In this context, a positive action of nonesterified
fatty acids on cholesterol metabolism has been
des-cribed in the mitochondrial membrane [21] It was also
demonstrated that cholesterol binding to the enzyme
that transforms it into pregnenolone (P450scc) in lipid
vesicles is greatly potentiated when the local membrane
is rendered more fluid by the addition of nonesterified
fatty acids [22]
All the evidence described above led us to propose
the hypothesis that AA might have a direct action on
cholesterol transport into the mitochondria via a
speci-fic release in this organelle This knowledge would be important for the understanding of cholesterol trans-port in the classical steroidogenic as well as in neuro-logical systems, since changes in cholesterol transport
in the central nervous system are part of the phenotype seen in the neuropathology and neurological disorders such as Alzheimer’s, Parkinson’s and Huntington’s dis-eases, and brain injury and inflammation, as well as in animal models of epilepsy [23] This is also valid for cholesterol transport and metabolism in tumors such
as glioma and mammary tumor cells [24,25]
For these reasons, the objective of the present work was to study the release of AA into the mitochondria and a possible direct role of fatty acids on cholesterol transport in this organelle
Results
It is known that the acute response of steroidogenesis
to hormonal stimulation has an absolute requirement:
de novo protein synthesis [26,27] This conclusion is based on the fact that hormone stimulated steroid syn-thesis is totally inhibited by cycloheximide (CHX), a protein synthesis inhibitor The two proteins required
in this step are ACS4 [28] and StAR [29] ACS4 works
in the release of AA, which, in turn, acts on StAR pro-tein induction
Because exogenous AA stimulates steroidogenesis in cells, the first experiment was carried out to study the direct effect of AA on cholesterol transport in the absence of newly synthesized StAR protein For this purpose, MA-10 cells were incubated with exogenous
AA either with or without submaximal concentration
of 8-bromo, 5¢-cAMP (8Br-cAMP) in the presence or absence of CHX Progesterone production as measure-ment of cholesterol transport was evaluated in the culture media after 1 h of incubation, as described
in Experimental procedures (Fig 1) Exogenous AA alone stimulated progesterone production, reaching 50–60% of the maximal value obtained with 8Br-cAMP (Fig 1A) Submaximal doses of 8Br-8Br-cAMP in combination with AA stimulated steroid production at the same level as the stimulation produced by satur-ating doses of 8Br-cAMP 8Br-cAMP-stimulated pro-gesterone production was completely abolished in the presence of CHX However, AA-induced steroid pro-duction was only partially blocked by the protein syn-thesis inhibitor Again, CHX did not totally reduce the synergistic effect of AA on steroid production, either (Fig 1A) Protein synthesis inhibition did not affect progesterone production supported by the water-soluble derivative of cholesterol, 22(R)-OH-cholesterol,
Trang 3which travels freely across the membranes to reach the
inner mitochondrial cholesterol side-chain cleavage
cytochrome P-450 enzyme (P450scc) (Fig 1B)
The widely known fact that cAMP cannot stimulate
steroidogenesis in the absence of protein synthesis is
due to the absence of two crucial proteins, ACS4 and
StAR ACS4 is induced by hormones and it is
neces-sary for AA release [28], which participates in StAR
protein induction Therefore, in the absence of ACS4,
no AA or StAR protein induction occurs, as
previ-ously described [28] This is the reason why CHX
com-pletely abolished cAMP-stimulated steroidogenesis
When exogenous AA is used in the presence of CHX,
the fatty acid bypasses the absence of ACS4 but not
the absence of StAR Then, the partial inhibition
pro-duced by CHX on AA stimulated steroidogenesis was
unexpected The stimulatory effect of AA on steroid
synthesis in the absence of protein synthesis suggests
that AA can per se enhance the cholesterol transport
and steroidogenesis in mitochondria of steroidogenic
cells without de novo protein synthesis In order to test
this hypothesis, firstly, we tested whether AA
exogen-ously added to intact cells could reach the
mitochon-dria Second, we studied the effect of exogenous AA
on cholesterol transport in isolated mitochondria from nonstimulated MA-10 steroidogenic cells
For the first approach, MA-10 cells were labeled with [1-14C] AA during 5 h After this period, the cells were incubated in the presence or in the absence of 8Br-cAMP After this incubation, free AA was measured in the mitochondria, as described in the Experimental pro-cedures Figure 2 shows the uptake of AA into the mito-chondria in basal and stimulated conditions As can
- CHX
+ CHX
AA Control 8Br-cAMP
0.2 m M + AA
8Br-cAMP 0.2 m M
8Br-cAMP
0.5 m M
A
b
b
a
a
a
a
b,c b,c
0
2
4
6
8
10
12
14
B
22(R)OH- cholesterol
0
10
20
30
40
50
60
Fig 1 Effect of cAMP, AA and CHX on progesterone production
by MA-10 cells MA-10 cells were incubated in the presence or
absence of 10 lgÆmL)1CHX for 30 min and then stimulated for 1 h
with 8Br-cAMP (0.2 m M or 0.5 m M ) and ⁄ or 300 l M AA (A), or 5 l M
22(R)-OH-cholesterol (B) in serum-free culture medium containing
0.1% fatty acid-free bovine serum albumin Progesterone
concen-trations were measured by RIA and data are shown as
progester-one production (ngÆmL)1) in the incubation medium Results are
expressed as the mean ± SD from five independent experiments.
(a) P < 0.001 versus control cells without CHX treatment; (b)
P < 0.001 versus respective treated cells in absence of CHX; and
(c) P < 0.01 versus control cells treated with CHX.
B 17
16
15
4
3
2
1
0
A
AA
Nuclei Mitochondria
***
a
i
r
d
o
c
o
t
m
i
e
l
c
u
a
i
r
d
o
c
o
t
m
i
e
l
c
u
Fig 2 Effect of cAMP on mitochondrial and nuclear AA content MA-10 cells were labeled for 5 h at 37 C with [1- 14 C] AA (1 lCiÆmL)1per 2 · 10 6 cells) in serum-free media containing 0.5% fatty acid-free bovine serum albumin Then, cells were incubated in either the presence or absence of 1 m M 8Br-cAMP for 30 min After washing the cells, they were scraped and nuclear and mitoch-ondrial fractions were obtained as described in the Experimental procedures The fractions were sonicated and lipids were extracted with ethyl acetate The organic phase was collected and dried under nitrogen The dried extracts were dissolved in chloro-form:methanol (9 : 1, v ⁄ v) and analyzed by thin-layer chromato-graphy on silica gel plates (A) Representative autoradiochromato-graphy showing AA spots in nuclear and mitochondrial fractions (B) Auto-radiography spots quantification by densitometry The autoradio-graphies were quantified by densitometry and the data were normalized against the intensity of the signal of unlabeled AA stained with iodine Bars denote levels (in arbitrary units) of AA in mitochondria and nuclei Results are expressed as the mean ± SD from three independent experiments *** P < 0.001 versus control mitochondria.
Trang 4clearly be seen, 8Br-cAMP increased the uptake of AA
3 times compared with nonstimulated conditions,
with-out changing the content in the nuclear fraction
For the second approach, mitochondria from
non-stimulated MA-10 cells were isolated and incubated in
the presence of AA Figure 3A shows that AA elicited
a stimulatory effect on cholesterol transport measured
as progesterone synthesis As expected, this effect was
not affected by the addition of CHX previous to the
addition of AA Nordihydroguaiaretic acid (NDGA),
an inhibitor of AA metabolism, had no effect on AA
action Although NDGA is commonly used as
lip-oxygenase inhibitor, it is also able to inhibit Acot2
[11] However, in this case, we only evaluated its action
on AA metabolism since the stimulation was
per-formed with AA Other fatty acids, such as oleic and
arachidic acids produced a small, but not significant,
effect The total steroidogenic capacity of the isolated
mitochondria was determined by the incubation with
the water-soluble derivative of cholesterol,
22(R)-OH-cholesterol Neither CHX nor NDGA affected this
total steroidogenic capacity
As we have proposed [8] that the conversion of
AA-CoA to AA within the mitochondria may constitute a
mechanism to deliver AA into specific compartment of
the cells, the next experiment was carried out to
deter-mine the effect of AA-CoA, the substrate of Acot2, on
steroid synthesis in isolated mitochondria Figure 3B
shows that AA-CoA not only stimulated cholesterol
transport in isolated mitochondria but also had a
higher effect than free AA action When another
acyl-CoA, such as oleoyl-acyl-CoA, was tested, it was also
proved capable of increasing progesterone production
in mitochondria to a similar extent
To determine if the effect of AA-CoA on cholesterol
transport was due to its conversion to AA by the
action of Acot2, we studied the effect of the blockage
of Acot2 expression or activity on AA-CoA-stimulated
steroid synthesis Our next experiment was conducted
as described in Fig 3B in the presence and absence of
BPB or NDGA, both inhibitors of Acot2 activity [11]
Figure 3C shows that blockage of Acot2 activity
pro-duces a significant inhibition of progesterone synthesis
stimulated by AA-CoA
To silence the expression of Acot2, we transiently
transfected MA-10 cells with pRc⁄ CMVi plasmid
con-taining an antisense Acot2 cDNA (accession number
Y09333) The effect of antisense plasmid transfection
on Acot2 protein concentrations was studied by
west-ern blot, by means of a specific antibody against the
Acot2 and b-tubulin as control As expected [8],
anti-sense-transfected cells showed a strong reduction in
Acot2 protein levels compared with cells transfected
with vector alone (Fig 4A,B) The stimulatory effect
of AA-CoA on steroid synthesis in mitochondria isola-ted from non stimulaisola-ted MA-10 cells where Acot2 was
a
b
0.15
0.05 0.00
0.25 0.20 0.15 0.10 0.05 0.00
B
a
a,b
a
a
b
a 0.25 0.20 0.15 0.10 0.05 0.00
Malonyl-CoA
- AA-CoA + AA-CoA
Control
C
0.70 0.20
0.10
0.00
0.85
Arachidonic acid
Arachidic acid Oleic acid
A
Control
+ NDGA + CHXNone
22(R)OH-cholesterol
Fig 3 Effect of fatty acids, CoA derivatives of fatty acids, BPB, NDGA and malonyl-CoA on progesterone production in isolated mitochondria Mitochondria were isolated from MA-10 cells and preincubated for 5 min at 37 C in the absence or in the presence
of 10 lgÆmL)1CHX or 100 l M NDGA (A) or 0.1 m M BPB, 100 l M
NDGA or 100 l M malonyl-CoA (C) Mitochondria were then incuba-ted for 20 min with 200 l M AA, oleic acid or arachidic acid or with
5 l M 22(R)-OH-cholesterol (A); 200 l M AA, AA-CoA or oleoyl-CoA (B); or 200 l M AA-CoA (C) Mitochondria were pelleted by centri-fugation and progesterone concentrations were measured in the supernatants by RIA Data are shown as progesterone production (ngÆmg)1mitochondrial protein) in the incubation media (A) Results are expressed as the mean ± SD from six independent experi-ments *** P < 0.001 versus control (B) Results are expressed as the mean ± SD from three independent experiments (a) P < 0.001 versus control mitochondria; (b) P < 0.01 versus AA treated mito-chondria (C) Results are expressed as the mean ± SD from three independent experiments (a) P < 0.001 versus control mitochon-dria; and (b) P < 0.01 versus AA-CoA treated mitochondria.
Trang 5knocked down was significantly reduced compared
with mitochondria isolated from mock-transfected cells
(Fig 4C) Acot2 knockdown did not produce any
effect on progesterone synthesis in mitochondria
trea-ted with 22(R)-OH-cholesterol (Fig 4D) This provides
evidence that the reduction in Acot2 expression does
not affect mitochondrial integrity
The results described above indicate the necessity of
Acot2 in AA-CoA-stimulated steroidogenesis,
indica-ting also that the effect of AA-CoA is due to its
con-version to AA into the mitochondria If this is the
case, inhibition of AA-CoA uptake into the
mitochon-dria should inhibit steroid synthesis Indeed, we
inhib-ited the carnitine-dependent acyl-CoA transport with
malonyl-CoA and the stimulatory effect of AA-CoA
on mitochondrial steroid synthesis was significantly
reduced (Fig 3C)
The requirement of Acot2 for the action of
AA-CoA on steroid synthesis suggests the participation of
this enzyme in the mitochondrial cAMP-induced AA
accumulation Then, we next tested the effect of Acot2
on mitochondrial [1-14C]-AA accumulation induced by
8Br-cAMP, using the same strategy described in
Fig 3C and Fig 4: inhibition of Acot2 activity and
expression, respectively As shown in Fig 5, BPB
inhibited cAMP induced accumulation of labeled AA
into the mitochondria (Fig 5A,B) In accordance with
this effect produced by BPB on AA mitochondrial
content, there is an increase in AA-CoA retained in
the postmitochondrial fraction of cells treated with this compound (Fig 5C) The same effect was observed in cells where Acot2 expression was blunted (Fig 6)
Discussion
In the present paper, we show for the first time that cAMP can regulate the release of AA in a specialized compartment of the cells, e.g the mitochondria, and that the fatty acid induces steroid synthesis through a mechanism different from the classical pathway invol-ving the stimulation of StAR protein expression This biological effect can be seen by the addition of exogen-ous AA to intact MA-10 Leydig cells (300 lm) or isolated mitochondria (200 lm) In determining the
Acot2 β-tubulin
***
D
c R p /
i V M
R p /
-i V M
t o c A
22(R)OH-cholesterol Control
pRc/CMVi-Acot2 antisense
C
a
b,c
0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00
8 4 0
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0
Acot2 antisense
***
***
AA-CoA Control
c R p /
i V M
R p /
-i V M
t o c A
/
Fig 4 Effect of AA-CoA on progesterone production in isolated
mitochondria from Acot2 knockdown MA-10 cells MA-10 cells
were transfected with pRc⁄ CMVi or pRc ⁄ CMVi-Acot2 antisense
cDNA plasmids After 72 h, mitochondria were isolated from
MA-10 transfected cells (A) Representative western blot of
mitochon-dria from MA-10 transfected cells The membrane was blotted
sequentially with anti-Acot2 and anti-b-tubulin sera (B) Western
blot quantification by densitometry Bars denote relative levels
of Acot2 expression in arbitrary units *** P < 0.001 versus
pRc ⁄ CMVi transfected cells (C) Mitochondria from MA-10
trans-fected cells were incubated for 20 min at 37 C in the absence or
in the presence of 200 l M AA-CoA (D) Mitochondria from MA-10
transfected cells were incubated for 20 min at 37 C in the
absence or in the presence of 5 l M 22(R)OH-cholesterol In (C) and
(D), mitochondria were pelleted by centrifugation and progesterone
concentrations were measured in the supernatants by RIA Data
are shown as progesterone production (ng⁄ mg mitochondrial
pro-tein) in the incubation media Results are expressed as the mean ±
SD from three independent experiments In (C), (a) P < 0.001
versus control mitochondria from pRc ⁄ CMVi-transfected cells;
(b) P < 0.01 versus control mitochondria from pRc ⁄ CMVi-Acot2
antisense-transfected cells; and (c) P < 0.01 versus AA-CoA
treated mitochondria from pRc ⁄ CMVi-transfected cells In (D),
*** P < 0.001 versus mitochondria from respective-transfected
cells without AA-CoA treatment.
Trang 6effective concentration of AA, we also assess AA’s
bio-activity Although in some cases concentrations
between 1 and 10 lm AA are sufficient to detect a
bio-logical effect [2], other effects require 100–300 lm [2]
However, at this high concentration, AA may not even
be in solution Previous studies performed with
con-centrations of AA of the same magnitude as ours
(200 lm) observed either none [30] or a small but
signi-ficant [31,32] effect of exogenously added AA on basal
steroid synthesis in MA-10 and rat Leydig cells,
respectively In the present paper, we show the
stimu-lation of steroidogenesis using 300 lm AA in the pres-ence of albumin These different results can be explained by the fact that protein binding can increase the overall concentration of fatty acids present in an aqueous environment by effectively decreasing the insoluble fraction [2] Albumin, in particular, binds specifically to fatty acids [33] In our experiments, we added albumin in the preparation of the AA solution, which allowed us to detect a significant stimulatory effect of exogenous AA on steroid production of non-stimulated steroidogenic cells (50–60% compared with cAMP) The stimulation of steroid synthesis by exog-enously added AA was lower than the stimulation reached by cAMP or 22(R)-OH-cholesterol These results are similar to those obtained by us in rat Ley-dig cells and by other authors in the same or other steroidogenic tissues [20,32,34]
When exogenous AA is added together with sub-maximal doses of 8Br-cAMP, there is a synergistic
a
b
Acot2 antisense
B
AA
8Br-cAMP Control
pRc/CMVi-Acot2 antisense
A
a
b
0 1 2 3 4 5 6 7 control 8Br-cAMP control 8Br-cAMP
Fig 6 Effect of Acot2 knockdown on AA accumulation into mito-chondria MA-10 cells were transfected with pRc ⁄ CMVi or pRc ⁄ CMVi-Acot2 antisense cDNA plasmids as described in Fig 5 After 72 h, MA-10 transfected cells were labeled and stimulated as described in Fig 2 (A) Representative autoradiography showing AA spots in mitochondrial fractions (B) Autoradiography spots quantifi-cation by densitometry Bars denote levels (in arbitrary units) of AA
in mitochondria from MA-10 transfected cells treated with or with-out 8Br-cAMP Results are expressed as the mean ± SD from three independent experiments (a) P < 0.01 versus mitochondria from control pRc ⁄ CMVi-transfected cells; and (b) P < 0.001 versus mitochondria from 8Br-cAMP treated pRc ⁄ CMVi antisense trans-fected cells.
b a
0
4
8
12
16
20
+ BPB
B
Control 8Br-cAMP 8Br-cAMP
+ BPB
AA
b
a
8Br-cAMP + BPB 0
25
14 C]AA-CoA
–3 /mg protein)
C
Fig 5 Effect of Acot2 activity inhibition on AA accumulation into
mitochondria and AA-CoA accumulation in the postmitochondrial
fraction MA-10 cells were labeled as described in Fig 2 When
indicated, cells were incubated with 0.1 m M BPB for 30 min prior
to the stimulation with 8Br-cAMP (A) Representative
autoradiogra-phy showing AA spots in mitochondrial fractions (B)
Autoradiogra-phy spots quantification by densitometry The autoradiographies
were quantified by densitometry and the data were normalized
against the intensity of the signal of unlabeled AA stained with
iod-ine Bars denote levels (in arbitrary units) of AA in mitochondria.
Results are expressed as the mean ± SD from three independent
experiments (a) P < 0.001 versus mitochondria from control cells;
(b) P < 0.05 versus mitochondria from 8Br-cAMP-treated cells (C)
AA-CoA content in the postmitochondrial fraction Data are shown
as 14 C-AA-CoA in cpm ⁄ mg protein in the postmitochondrial fraction.
Results are expressed as the mean ± SD from three independent
experiments ***P < 0.001 versus control.
Trang 7effect in which steroid synthesis reaches maximal
activation (Fig 1) This result agrees with that
obtained by other authors [35], who suggested that a
critical threshold of cAMP or cAMP-dependent
pro-tein kinase activation is required for the synergistic
effect of AA on cAMP-stimulated StAR protein
expression and steroidogenesis The fact that the cells
treated with exogenous AA can be stimulated with
8Br-cAMP indicates that mitochondrial function
regarding the regulation of cholesterol metabolism
remains intact Moreover, it is known that any
distur-bance or swelling of the mitochondria produces a total
loss of regulation, rendering mitochondria that
pro-duce full steroidogenesis In our case, treatment with
AA also does not cause loss of the regulation of
ster-oid biosynthesis supported by 22(R)-OH-cholesterol
(data not shown) This explanation rules out the
possi-bility that AA or bovine serum albumin would have
effect on mitochondrial integrity This consideration is
also valid when fatty acids or AA-CoA are used in
iso-lated mitochondria, as both treatments can still be
increased by stimulation with 22(R)-OH-cholesterol
(data not shown)
As is already known [26,27], the cAMP-dependent
transport of cholesterol from the mitochondrial outer
to inner membrane can be blocked by a protein
syn-thesis inhibitor such as CHX However, this protein
synthesis inhibitor is not totally able to abolish the
sti-mulation produced by exogenously added AA These
results strongly suggest that AA can exert a role on
cholesterol transport without the induction of StAR
protein
The demonstration that AA and⁄ or AA-CoA
stimu-late cholesterol transport in isostimu-lated mitochondria
sug-gests that the accumulation of AA can occur by direct
uptake of AA itself inside the mitochondria or by the
previous esterification to AA-CoA by ACS4 and
subse-quent action of Acot2 to render free AA in the
mito-chondria The fact that cAMP increases AA uptake
into the mitochondria and that this effect on AA
accu-mulation is reduced when Acot2 activity or expression
are blocked strongly indicates that the operating
mechanism is dependent on the concerted action
of ACS4⁄ Acot2 In this mechanism, cAMP acts to
increase AA-CoA formation in the cytosol The CoA
derivative enters the mitochondria through the
CPT1-dependent pathway The specificity of this mechanism
to release AA inside the mitochondria is shown by the
fact that the content of labeled AA in another
organ-elle such as the nucleus is neither increased by cAMP
nor reduced by the inhibition of Acot2 (Fig 2) This is
the first time that AA incorporation into a specific
subcellular compartment (the mitochondria) has been
shown as a consequence of the action of this second messenger Steroidogenic cells express Acot2 and also
a cytosolic isoform, Acot1, which is 92.5% homolog-ous to the mitochondrial enzyme We have ruled out the possibility that mitochondria would uptake AA produced by the action of Acot1 outside the mitochon-dria, as part of our experiments were performed with isolated mitochondria where Acot1 was not present Moreover, while the overexpression of Acot2 results in
an increase of hormone induced steroid synthesis [8], overexpression of Acot1 does not produce this effect; conversely, it produced a slight inhibition of the pro-cess (data not shown) This last result supports the notion that Acot2 is the thioesterase involved in the release of AA inside the mitochondria
Our model explaining how AA is released into the mitochondria also supports the concept that the select-ive actions of free AA may be explained simply by its specific release under physiological conditions and by the absence of such mechanisms for releasing other long-chain fatty acids, compounds that might otherwise share its biochemical effects This is demonstrated by the fact that when the mitochondria are stimulated with other fatty acids, the response is lower than with AA; however, there is a significant response of steroid-ogenesis to a different Acyl-CoA (oleoyl-CoA, Fig 4) Thus, the specificity of the action is not due to the fatty acid itself but to the acyl-CoA available to the mito-chondrial Acot2 In our case, AA-CoA is formed pref-erentially because of the specificity of ACS4 on AA [9] The mitochondrial inner membrane is not permeable
to acyl-CoAs [3]; we wanted to know how AA-CoA reaches the mitochondrial Acot2 The experiment using malonyl-CoA (Fig 5) indicates that AA-CoA follows the usual pathway involving carnitine-palmitoyl transf-erase 1 (CPT1) [3] This enzyme plays a central role in mitochondrial fatty acid oxidation However, in our case, it seems that CPT1 directs AA to another func-tion In this context, it has been proposed that a potential route for long-chain acyl-CoAs to cross the mitochondrial outer membrane could be the voltage-dependent anion selective channel, also called mitoch-ondrial porin and located in the contact sites [36] It is very interesting that a protein obligatory for choles-terol transport in steroidogenic cells, the peripheral benzodiazepine receptor (PBR), is also located in the mitochondrial contact sites and includes the voltage-dependent anion selective channel in its structure together with the adenine nucleotide carrier [37] PBR
is involved in cholesterol transport to the cytochrome P450 side chain cleaving enzyme localized on the outer surface of the mitochondrial inner membrane [37] The endogenous ligand of this receptor is an acyl-CoA
Trang 8binding protein known also as diazepam-binding
inhib-itor (DBI) [37,38] It can be postulated that the role of
DBI is to facilitate the transport of fatty acids through
the mitochondrial outer membrane The homology
between the DBI and an acyl-CoA binding protein
cer-tainly enhances this possibility The specific interaction
between DBI and its endogenous receptor, the PBR
located on the outer⁄ inner mitochondrial membrane
contact sites [37,38], may direct the AA-CoA to this
organelle Contact sites between the mitochondrial outer
and inner membrane could represent the
microenviron-ment for bringing the machinery together to transport
AA-CoA [37] into the mitochondria and facilitate AA
release, which in turn facilitates cholesterol transport
How intramitochondrial AA could stimulate cholesterol
transfer from the outer to the inner mitochondrial
mem-brane can also be explained by the action of AA on the
membrane permeability in the contact sites This
sugges-tion is also in line with experiments showing that AA
induces the specific membrane permeability in heart and
liver mitochondria by opening the mitochondrial
per-meability transition pore [39,40] The pore is a multiple
protein complex located in the mitochondrial contact
sites [41] and, in mitochondria of steroidogenic cells, it
participates in cholesterol transport
In the present paper, we demonstrate that 20–30% of
total steroid production can be elicited without the
necessity for StAR synthesis This is in accordance with
the pathological situation where deletions or mutations
of StAR are detected in humans born with the steroid
deficiency disease, lipoid adrenal congenital hyperplasia
[17,20,42–44] Disruptions of the StAR gene in mice
produce similar phenotypes [18,45] The effect of these
deletions establishes that StAR is necessary for 80–90%
of adrenal cholesterol metabolism [19,46] In other
words, our results may explain the mechanism by
which in these situations there is a remaining 20% of
steroid synthesis, due to the direct effect of AA⁄
AA-CoA produced within the mitochondria by the
action of ACS4⁄ Acot2 together with DBI ⁄ PBR
Thus, it can be postulated that in the acute phase
(early response) of steroid synthesis, the release of AA
into the mitochondria is the first stimulator of
choles-terol transport The sustained phase of the acute
response will then need the induction of StAR We
cannot exclude that an extraordinarily small amount
of intramitochondrial StAR present in resting
condi-tions and not detectable by current techniques can
contribute to the effect of AA on cholesterol transport
in mitochondria
The absence of hormone⁄ cAMP-induced steroid
synthesis when protein synthesis is inhibited can be
explained now by the inhibition in the induction of
ACS4 [28] during the early response and the inhibi-tion of ACS4 and StAR inducinhibi-tions during the sus-tained phase In both phases, the presence of DBI⁄ PBR may be necessary This new feature in the regulation of cholesterol transport by AA and the release of AA in a specialized compartment of the cells could offer novel means for understanding the regulation of steroid synthesis, but would also be important in other situations such as the neurosteroid biosynthesis or oncology disorders, where cholesterol transport, ACS4 and PBR play an important role [23–25,47]
Experimental procedures Materials
Fatty acid-free bovine serum albumin, AA, arachidic and oleic acids, 4-bromophenacyl bromide (BPB), oleoyl-CoA, malonyl-CoA, 8Br-cAMP, 22(R)-OH-cholesterol, cyclohexi-mide (CHX) and Waymouth MB752⁄ 1 cell culture media were purchased from Sigma Chemical Co (St Louis, MO, USA) Nordihydroguaiaretic acid (NDGA) and AA-CoA were from Fluka (Buchs, Switzerland) Sera, antibiotics and trypsin-EDTA were from Gibco-Life Technologies Inc (Gaithersburg, MD, USA) All other reagents were of the highest grade available
Cell culture The MA-10 cell line is a clonal strain of mouse Leydig tumor cells that produce progesterone rather than testoster-one as the major steroid The cells were generously provi-ded by M Ascoli, University of Iowa, College of Medicine (Iowa City, IA, USA) and were handled as originally des-cribed [48]
Cells were incubated in the presence or absence of
10 lgÆmL)1CHX for 30 min and then stimulated with 8Br-cAMP (0.2 mm or 0.5 mm), 300 lm AA or 5 lm 22(R)-OH-cholesterol in the culture medium containing 0.1% fatty acid-free bovine serum albumin Progesterone production was measured by radioimmunoanalysis (RIA) [7], and data are shown as progesterone production (ngÆmL)1) in the incubation medium
Preparation of mitochondrial fraction Mitochondria were obtained as previously described [17] Briefly, all MA-10 cell cultures were washed with phos-phate-buffered saline, scraped in 10 mm Tris⁄ HCl (pH 7.4),
250 mm sucrose, 0.1 mm EDTA (TSE buffer), homogenized with a Pellet pestle motor homogenizer (Kontes) and centri-fuged at 800 g during 15 min A second centrifugation at
16 000 g during 15 min rendered a mitochondrial pellet and
Trang 9a supernatant (postmitochondrial fraction) The
mitochond-rial pellet was resuspended in TSE buffer
Progesterone production in isolated
mitochondria
Thirty microliters of mitochondrial fraction (200 lg of
protein) were added to 165 lL of medium consisting of
34 mm Tris⁄ HCl (pH 7.4), 20 mm KCl, 4 mm MgCl2 and
108 mm mannitol, containing 0.3% fatty acid-free bovine
serum albumin When indicated, 200 lm AA, 200 lm
AA-CoA or 5 lm 22(R)-OH-cholesterol were added The
mix-ture was completed by adding TSE buffer to complete a
final reaction volume of 500 lL (fatty acid-free bovine
serum albumin final concentration 0.1%) The incubations
were carried out at 37C for 20 min with gently shaking
and were stopped by cooling the tubes in an ice⁄ water bath
As indicated in each figure, inhibitors such as 10 lgÆmL)1
CHX, 100 lm NDGA, 0.1 mm BPB or 100 lm
malonyl-CoA were added to the reaction mixture and preincubated
5 min prior to the addition of AA, AA-CoA or
22(R)-OH-cholesterol
After the incubation time, mitochondria were pelleted by
centrifugation 16 000 g for 15 min and progesterone
con-centrations were measured in the supernatants by RIA
Data are shown as progesterone production per mg of
mitochondrial protein (ngÆmg)1protein)
[1-14C]Arachidonic acid incorporation
in MA-10 cells
MA-10 cells were labeled following a previously described
methodology [17], with minor modifications [14C]-AA
(New England Nuclear, Boston, MA, USA; specific activity
53.0 mCiÆmmol)1) was added to the cultures in a
concentra-tion of 1 lCiÆmL)1 per well (2· 106
cells) in serum-free Waymouth MB752⁄ 1 containing 0.5% fatty acid-free
bovine serum albumin [17] After 5 h of incubation at
37C in a humidified atmosphere containing 5% CO2, the
cells were incubated in the presence or absence of 1 mm
8Br-cAMP for 30 min When indicated, cells were
incuba-ted with 0.1 mm BPB for 30 min prior to the stimulation
with 8Br-cAMP
After these treatments, the cells were washed with
serum-free Waymouth medium containing 0.5% fatty acid-serum-free
bovine serum albumin Nuclear and mitochondrial pellets
were obtained as previously described [17] and resuspended
in 20 mm Hepes⁄ KOH (pH 7.4), 250 mm sucrose, 1 mm
EDTA, 10 mm KCl and 1.5 mm MgCl2 containing 500 ng
of unlabeled AA, and were then sonicated Protein
concen-tration was measured and lipids were extracted from equal
amounts of nuclear or mitochondrial proteins (500 lg in
both cases) from each treatment Lipid extraction was
per-formed twice with ethyl acetate (six volumes per one
vol-ume of nuclear or mitochondrial fraction) The organic
phase was then collected and dried under nitrogen at 25C and analyzed by two successive thin-layer chromatographies
on silica gel Radioactive spots were developed using a Storm Phosphorimager (Amersham Biosciences, Sweden) after 1 week of exposition The postmitochondrial fraction was treated as described for the mitochondria and the [14C]-AA-CoA formation was evaluated by extraction from the aqueous phase according to the literature [49]
Plasmid transfection MA-10 cells were transiently transfected by electroporation
as previously described [8,50] Transfection efficiency varied from 40 to 50% and was estimated by counting fluorescent cells transfected with pRc⁄ CMVi plasmid [51] containing the enhanced form of green fluorescent protein (EGFP) [8] MA-10 cells were transfected either with pRc⁄ CMVi plas-mid containing the Acot2 antisense cDNA [8] or with the empty vector as control Approximately 72 h after transfec-tion, cells were used as described in the respective figures
SDS/PAGE and immunoblot assay Mitochondrial proteins were separated by SDS⁄ PAGE (10% acrylamide gels) and electrophoretically transferred to poly(vinylidene difluoride) membranes (Bio-Rad Laborator-ies Inc., Hercules, CA, USA) as described previously [8] Acot2 protein was detected using anti-Acot2 antibodies [11] and immunoreactive bands were detected by chemilumines-cence using enhanced chemilumineschemilumines-cence reagents (GE Healthcare, Chalfont St Giles, UK)
Protein quantification and statistical analysis Protein was quantitated by the method of Bradford [52] using bovine serum albumin as a standard Statistical ana-lysis was performed by t-test or anova followed by the Student–Newman–Kuels test
Acknowledgements This work was supported in part by National Research Council (CONICET), University of Buenos Aires (UBA) and National Agency of Scientific and Techno-logical Promotion (ANPCyT) Thanks are due to the technical assistance provided by F Meuli
References
1 Sigal E (1991) The molecular biology of mammalian arachidonic acid metabolism Am J Physiol 260, L13– L28
2 Brash AR (2001) Arachidonic acid as a bioactive mole-cule J Clin Invest 107, 1339–1345
Trang 103 Kerner J & Hoppel C (2000) Fatty acid import into
mitochondria Biochim Biophys Acta 1486, 1–17
4 Klingenberg M & Huang SG (1999) Structure and
func-tion of the uncoupling protein from brown adipose
tis-sue Biochim Biophys Acta 1415, 271–296
5 Faergeman NJ & Knudsen J (1997) Role of long-chain
fatty acyl-CoA esters in the regulation of metabolism
and in cell signalling Biochem J 323, 1–12
6 Irvine RF (1982) How is the level of free arachidonic
acid controlled in mammalian cells? Biochem J 204,
3–16
7 Maloberti P, Lozano RC, Mele PG, Cano F, Colonna
C, Mendez CF, Paz C & Podesta EJ (2002) Concerted
regulation of free arachidonic acid and
hormone-induced steroid synthesis by acyl-CoA thioesterases and
acyl-CoA synthetases in adrenal cells Eur J Biochem
269, 5599–5607
8 Maloberti P, Castilla R, Castillo F, Maciel FC, Mendez
CF, Paz C & Podesta EJ (2005) Silencing the expression
of mitochondrial acyl-CoA thioesterase I and acyl-CoA
synthetase 4 inhibits hormone-induced steroidogenesis
FEBS J 272, 1804–1814
9 Kang MJ, Fujino T, Sasano H, Minekura H, Yabuki
N, Nagura H, Iijima H & Yamamoto TT (1997) A
novel arachidonate-preferring acyl-CoA synthetase is
present in steroidogenic cells of the rat adrenal, ovary,
and testis Proc Natl Acad Sci USA 94, 2880–2884
10 Paz C, Dada L, Cornejo Maciel F, Mele P, Cymeryng
C, Neuman I, Mendez C, Finkielstein C, Solano A,
Minkiyu P et al (1994) Purification of a novel 43-kDa
protein (p43) intermediary in the activation of
steroido-genesis from rat adrenal gland Eur J Biochem 224, 709–
716
11 Finkielstein C, Maloberti P, Mendez CF, Paz C,
Cornejo Maciel F, Cymeryng C, Neuman I, Dada L,
Mele PG et al (1998) An adrenocorticotropin-regulated
phosphoprotein intermediary in steroid synthesis is
sim-ilar to an acyl-CoA thioesterase enzyme Eur J Biochem
256, 60–66
12 Svensson LT, Engberg ST, Aoyama T, Usuda N,
Alexson SE & Hashimoto T (1998) Molecular cloning
and characterization of a mitochondrial peroxisome
proliferator-induced acyl-CoA thioesterase from rat
liver Biochem J 329, 601–608
13 Hunt MC, Yamada J, Maltais LJ, Wright MW, Podesta
EJ & Alexson SE (2005) A revised nomenclature for
mammalian acyl-CoA thioesterases⁄ hydrolases J Lipid
Res 46, 2029–2032
14 Crivello JF & Jefcoate CR (1980) Intracellular
move-ment of cholesterol in rat adrenal cells: kinetics and
effects of inhibitors J Biol Chem 255, 8144–8151
15 Walton KM & Dixon JE (1993) Protein tyrosine
phos-phatases Annu Rev Biochem 62, 101–120
16 Solano AR, Dada LA, Luz Sardanons M, Sanchez ML
& Podesta EJ (1987) Leukotrienes as common
inter-mediates in the cyclic AMP dependent and independent pathways in adrenal steroidogenesis J Steroid Biochem
27, 745–751
17 Solano AR, Dada L & Podesta EJ (1988) Lipoxygenase products as common intermediates in cyclic AMP-dependent and -inAMP-dependent adrenal steroidogenesis in rats J Mol Endocrinol 1, 147–154
18 Wang X & Stocco DM (1999) Cyclic AMP and arachi-donic acid: a tale of two pathways Mol Cell Endocrinol
158, 7–12
19 Wang XJ, Dyson MT, Jo Y, Eubank DW & Stocco
DM (2003) Involvement of 5-lipoxygenase metabolites
of arachidonic acid in cyclic AMP-stimulated steroido-genesis and steroidogenic acute regulatory protein gene expression J Steroid Biochem Mol Biol 85, 159–166
20 Mele PG, Dada LA, Paz C, Neuman I, Cymeryng CB, Mendez CF, Finkielstein CV, Cornejo Maciel F & Podesta EJ (1997) Involvement of arachidonic acid and the lipoxygenase pathway in mediating luteinizing hor-mone-induced testosterone synthesis in rat Leydig cells Endocr Res 23, 15–26
21 Jefcoate C (2002) High-flux mitochondrial cholesterol trafficking, a specialized function of the adrenal cortex
J Clin Invest 110, 881–890
22 Dhariwal MS & Jefcoate CR (1989) Cholesterol meta-bolism by purified cytochrome P-450scc is highly stimu-lated by octyl glucoside and stearic acid exclusively in large unilamellar phospholipid vesicles Biochemistry 28, 8397–8402
23 Papadopoulos V, Lecanu L, Brown RC, Han Z & Yao
ZX (2006) Peripheral-type benzodiazepine receptor in neurosteroid biosynthesis, neuropathology and neurolo-gical disorders Neuroscience 138, 749–756
24 Liang YC, Wu CH, Chu JS, Wang CK, Hung LF, Wang YJ, Ho YS, Chang JG & Lin SY (2005) Involve-ment of fatty acid-CoA ligase 4 in hepatocellular carci-noma growth: roles of cyclic AMP and p38 mitogen-activated protein kinase World J Gastroenterol 11, 2557–2563
25 Papadopoulos V, Guarneri P, Kreuger KE, Guidotti A
& Costa E (1992) Pregnenolone biosynthesis in C6–2B glioma cell mitochondria: regulation by a mitochondrial diazepam binding inhibitor receptor Proc Natl Acad Sci USA 89, 5113–5117
26 Garren LD, Gill GN, Masui H & Walton GM (1971)
On the mechanism of action of ACTH Recent Prog Horm Res 27, 433–478
27 Crivello JF & Jefcoate CR (1978) Mechanisms of corti-cotropin action in rat adrenal cells I The effects of inhibitors of protein synthesis and of microfilament for-mation on corticosterone synthesis Biochim Biophys Acta 542, 315–329
28 Cornejo Maciel F, Maloberti P, Neuman I, Cano F, Castilla R, Castillo F, Paz C & Podesta EJ (2005) An arachidonic acid-preferring acyl-CoA synthetase is a