In a previous report, we demonstrated the involvement of the calcium-dependent activation of calcineurin PP2B leading to dephosphorylation of the pro-apoptotic protein BAD in 7-ketochole
Trang 1Involvement of several pro-apoptotic but also anti-apoptotic
calcium-dependent transduction pathways
Arnaud Berthier, Ste´phanie Lemaire-Ewing, Ce´line Prunet, Thomas Montange, Anne Vejux,
Jean Paul Pais de Barros, Serge Monier, Philippe Gambert, Ge´rard Lizard and Dominique Ne´el INSERM U498 – Me´tabolisme des lipoprote´ines et interactions vasculaires, Dijon Cedex, France
Oxysterols are probably the components of oxidized
low-density lipoproteins which have the strongest
involvement in the genesis and development of
athero-sclerosis [1–3] Oxysterols mediate the early events of
atherosclerosis observed during the development of the
disease, such as the production of proinflammatory
cytokines, expression of adhesion molecules, and
cyto-toxicity to the cells of the vascular wall and
mono-cytes⁄ macrophages [4–6] This cytotoxicity appears to
be mainly related to the induction of apoptosis [6]
Among cytotoxic oxysterols found in atheromatous
lesions, 7-ketocholesterol is one of the most abundant
and one of the most studied [7] In a previous
report, we demonstrated the involvement of the
calcium-dependent activation of calcineurin (PP2B) leading to dephosphorylation of the pro-apoptotic protein BAD in 7-ketocholesterol-induced apoptosis of THP-1 cells The rise in free-Ca2+ activating calcineu-rin is induced by the translocation of Trpc-1, a com-ponent of the store-operated Ca2+entry channel, into lipid raft domains, which are microdomains of the plasma membrane formed by the lateral packing of glycosphingolipids and cholesterol [8] However, in this study, we show that BAD was dephosphorylated at serine 99 prior to dephosphorylation at serine 75, and the use of calcineurin inhibitors does not completely inhibit BAD dephosphorylation, suggesting the activa-tion of other pro- or anti-apoptotic pathways
Keywords
apoptosis; calcium; 7-ketocholesterol; signal
transduction; THP-1 cells
Correspondence
D Ne´el, INSERM U498 – Laboratoire de
Biochimie Me´dicale, CHU ⁄ Hoˆpital du
Bocage, 2 Bd Mare´chal de Lattre de
Tassigny, BP 77908, 21079 Dijon Cedex,
France
Fax: +33 3 80 29 36 61
Tel: +33 3 80 29 50 03
E-mail: dominique.neel@chu-dijon.fr
(Received 28 January 2005, revised 11 April
2005, accepted 18 April 2005)
doi:10.1111/j.1742-4658.2005.04723.x
Oxysterols, and particularly 7-ketocholesterol, appear to be strongly involved in the physiopathology of atherosclerosis These molecules are suspected to be cytotoxic to the cells of the vascular wall and mono-cytes⁄ macrophages, particularly by inducing apoptosis Previous studies have demonstrated that 7-ketocholesterol-induced apoptosis is triggered by
a sustained increase of cytosolic-free Ca2+, which elicits the mitochondrial pathway of apoptosis by activation of the calcium-dependent phosphatase calcineurin, leading to dephosphorylation of the ‘BH3 only’ protein BAD However, thorough study of the results suggests that other pathways are implicated in 7-ketocholesterol-induced cytotoxicity In this study, we dem-onstrate the involvement of two other calcium-dependent pathways during 7-ketocholesterol-induced apoptosis The activation of the MEKfi ERK pathway by the calcium-dependent tyrosine kinase PYK 2, a survival path-way which delays apoptosis as shown by the use of the MEK inhibitor U0126, and a pathway involving another pro-apoptotic BH3 only protein, Bim Indeed, 7-ketocholesterol treatment of human monocytic THP-1 cells induces the release of Bim-LC8 from the microtubule-associated dynein motor complex, and its association with Bcl-2 Therefore, it appears that 7-ketocholesterol-induced apoptosis is a complex phenomenon resulting from calcium-dependent activation of several pro-apoptotic pathways and also one survival pathway
Abbreviations
MEK 1 ⁄ 2, MAPK-Erk kinase-1 and )2; MSB, microtubule-stabilizing buffer; PTK, protein tyrosine kinases.
Trang 2Of the various pathways that are involved in cell
survival, the Ras⁄ Raf ⁄ MEK ⁄ Erk pathway plays a
crit-ical role Indeed, Ras-activated Raf operates by
phos-phorylating and activating MAPK-Erk kinase-1 and
-2 (MEK 1⁄ 2) [9,10] MEKs have very narrow
sub-strate specificity, restricted to p44Erk1 and p42Erk2
Phosphorylation of these kinases by MEKs results in
phosphorylation of further downstream targets
inclu-ding p90Rsk, which could phosphorylate and inactivate
BAD at serine 75 [11] On the other hand, it has been
shown that protein tyrosine kinases, such as
proline-rich tyrosine kinase-2 (PYK 2), transduce key
extracel-lular signals through the activation of the MEK⁄ Erk
pathway [12] Moreover, studies have shown that
PYK 2 is activated by an increase in intracellular
Ca2+ concentration, confirming the well-known
pro-cess of Ca2+-induced ERK activation [13]
Bim, like BAD, is a ‘BH3-only’ protein of the Bcl-2
family and an important mediator of apoptosis in
response to loss of survival signal [14,15] There are
three major splice variants of Bim: short (BimS), long
(BimL) and extra-long (BimEL) [16] In most cells,
including THP-1 cells, BimEL is the major species
expressed [17] This isoform of Bim induces apoptosis
by antagonizing the activity of the anti-apoptotic Bcl-2
family members [14] In lymphocytes, Bim has been
shown to be a major transducer of several apoptotic
signals including microtubule destabilization [18]
In this study, we show that 7-ketocholesterol induced,
via the calcium-sensitive tyrosine kinase PYK 2⁄
CAKb⁄ RAFTK ⁄ CADTK, a calcium-dependent
activa-tion of the MEK 1⁄ 2 fi ERK 1 ⁄ 2 survival pathway
delaying several apoptotic mechanisms initiated by the
oxysterol We also demonstrate the involvement of Bim
in 7-ketocholesterol-induced cytotoxicity Indeed, we
show that Ca2+influx leads to the translocation of the
protein from the microtubule dynein motor complex to
mitochondria, and thus its interaction with Bcl-2 [19]
Therefore, 7-ketocholesterol-induced apoptosis appears
to be a complex phenomenon involving several
calcium-dependent transduction pathways
Results
ERK 1⁄ 2 is activated during the first steps of
7-ketocholesterol-induced apoptosis
The effects of 7-ketocholesterol were examined in
rela-tion to the expression of various signalling proteins
in THP-1 cells Exposure of cells to 7-ketocholesterol
resulted in a rapid phosphorylation of ERK 1⁄ 2 within
1 h, as monitored through the use of phospho-specific
antibodies (Fig 1A) ERK phosphorylation peaked at
2–3 h, where phosphoERK reached 30% of total ERK
vs 13% in control, and then declined back towards basal levels at 12 h after exposure to 7-ketocholesterol (Fig 1A,B), whereas apoptosis increased significantly (Fig 1D) No activation of p38 MAPK or JNK was observed (data not shown) and dephosphorylation of PKB at threonine 308 was noted (Fig 1C) This PKB dephosphorylation, leading to its inactivation, appeared as soon as 3 h after 7-ketocholesterol treat-ment However, no phosphorylation of PKB was observed at serine 473 in either control cells or treated
7-keto A
B
C
D
P Erk1/2 Erk1/2
7-keto
Ctrl 1h 2h 3h 6h 12h 18h
Ctrl 1h 2h 3h 6h 12h 18h
300 200 100 150 250
50 0 Ctrl
1h 2h 3h 6h 12h 18h
*
*
*
*
*
Time of 7-keto treatment
P PKB (Thr 308)
% of Apoptotic Cells
Ctrl 7k20
80 70 60 50 40 30 20 10 0
0 6 12 18 24 30 36 42 48
Time (h)
PKB
Fig 1 7-Ketocholesterol induces ERK activation and Akt ⁄ PKB inac-tivation THP-1 cells were treated with 7-ketocholesterol (7-keto,
40 lgÆmL)1) for various incubation times Cell extracts were collec-ted, subjected to SDS ⁄ PAGE and immunoblotted with ERK 1 ⁄ 2, phospho-ERK 1 ⁄ 2, Akt and phospho-Akt thr308 antibodies Repre-sentative western blots of ERK 1 ⁄ 2–phospho-ERK 1 ⁄ 2 and Akt– phospho-Akt are shown (A and C, respectively); (B) phospho-ERK blot densitometry analysis Values are means ± SD (n ¼ 3).
*P < 0.05 vs control group (D) Microscopic quantification of cells with fragmented and ⁄ or condensed nuclei was performed using Hoechst 33342 and the percentage of apoptotic cells was deter-mined Data are means ± SD (n > 5).
Trang 3cells (data not shown) As ERK 1⁄ 2 activation is
known to be involved in survival pathways [20], we
then focused our attention on the role of the ERK
pathway, and looked for the possibility of a
relation-ship between ERK 1⁄ 2 activation and apoptosis
induced by 7-ketocholesterol
Inhibition of the MEKfi ERK pathway accelerates
7-ketocholesterol-induced THP-1 apoptosis
To investigate the exact role of ERK in
7-ketocholes-terol-induced apoptosis, we used the MEK inhibitor
U0126 Treatment of THP-1 cells with U0126 resulted
in the inactivation of ERK 1⁄ 2, almost disappearance
of the phospho-ERK spot, in agreement with the
inhi-bition of MEK, on control cells and at both 3 and 6 h
after addition of 7-ketocholesterol (Fig 2A,B) In light
of these results, we investigated the effect of the inhibi-tion of the MEK ERK 1⁄ 2 pathway on THP-1 cell viability after 7-ketocholesterol treatment Whereas U0126 alone has no effect on cell viability, we found that the addition of U0126 accelerates 7-ketocholesterol-induced apoptosis, as shown by the number of apop-totic cells (Fig 2C) Indeed, apoptosis of THP-1 cells
is observed 6 h earlier when cotreated with U0126 and 7-ketocholesterol than when treated with 7-keto-cholesterol alone This difference of 6 h correlates with the transient activation of ERK 1⁄ 2 Thus 7-ketocho-lesterol, which induces THP-1 apoptosis, also activates
an anti-apoptotic pathway
Calcium-dependent activation of ERK 1⁄ 2
As calcium signals could be involved in MAPK activa-tion and as we and others have previously described a calcium influx in oxysterol-induced apoptosis [8,21,22],
we tested whether or not calcium could mediate ERK 1⁄ 2 phosphorylation The role of calcium in MAPK activation was investigated by the cotreatment
of THP-1 with 7-ketocholesterol and verapamil, a cal-cium channel blocker which has been described as
a potential inhibitor of oxysterol-induced apoptosis Under these conditions, verapamil inhibited ERK acti-vation early as 3 h after treatment with 7-ketocholes-terol (Fig 3A,B) These results were strengthened by the use of the intracellular calcium chelator BAPTA which also completly inhibited ERK activation (data not shown)
Next, we wondered how calcium could activate the ERK 1⁄ 2 pathway As the cytosolic calcium-dependent PYK 2, a Src kinase activator, could mediate the acti-vation of the Ras-Raf-Mek-Erk pathway [12,23,24], activation by phosphorylation of tyrosines 579⁄ 580 in PYK 2 was investigated using a phosphorylation site-specific antibody Figure 3C,D shows that PYK 2 was phosphorylated at tyrosines 579⁄ 580 as early as 1 h after the THP-1 cells were treated with 7-ketochole-sterol This phosphorylation peaked at 2 h and then declined back toward basal level at 12 h after the addition of 7-ketocholesterol Phosphorylated PYK 2 peaked around 56% of total PYK 2 at 2 h vs 17% in control However, only weak phosphorylation (15% of PYK phosphorylated as in controls) was detected on these tyrosine residues when cells were cotreated with 7-ketocholesterol and verapamil (Fig 3E,F) suggesting that calcium is responsible for the activation of PYK 2 as previously demonstrated by Lev et al [13] Taken together, these results suggest that calcium uptake activates ERK 1⁄ 2 via the activation of PYK 2
7-keto A
B
C
P Erk1/2
% of Apoptotic Cells
Erk1/2
250
150
200
100
50
0
Ctrl
Ctrl
U
U
3h 3h+U 6h
7k20 7k20 + U
*
*
*
*
*
*
45
35
25
15
5
0
0
Time (h)
40
30
20
10
6h+U Time of 7-keto treatment
Fig 2 The MEK blocker U0126 inhibits 7-ketocholesterol-induced
ERK activation and accelerates apoptosis THP-1 cells were either
untreated (Ctrl) or incubated with U0126 (U, 10 lmolÆL)1) alone or
in association with 7-ketocholesterol (7-keto) for the indicated
times (A) Cell lysates were subjected to SDS ⁄ PAGE, and
immuno-blot analysis with antibodies against ERK 1 ⁄ 2 or phospho-ERK 1 ⁄ 2
was performed (B) Phospho-ERK blot densitometry analysis
Val-ues are means ± SD (n ¼ 3) (C) Microscopic quantification of cells
with fragmented and ⁄ or condensed nuclei was performed using
Hoechst 33342 and the percentage of apoptotic cells was
deter-mined Data are means ± SD (n ¼ 4) (*P < 0.05).
Trang 4ERK 1/2 inhibits 7-ketocholesterol-induced
apoptosis by phosphorylation of BAD
Having established the role of ERK 1⁄ 2 in the
7-keto-cholesterol survival pathway, we wondered how ERK
inhibited cell death During apoptosis involving
mito-chondria, Bcl-2 family members play a critical role
We have previously described that 7-ketocholesterol
induced BAD dephosphorylation at serines 75 and 99,
but BAD dephosphorylation at serine 75 was incom-plete [8] Moreover, Scheid et al [11] showed that ERK could phosphorylate BAD at serine 75, via p90RSK We therefore tested the hypothesis that ERK could reduce apoptosis by phosphorylating BAD To characterize the importance of the ERK pathway in the activa-tion⁄ inhibition of BAD during the early steps of 7-ketocholesterol-induced apoptosis, the status of BAD phosphorylation at serine 75 was investigated following treatment with the MEK inhibitor U0126 (Fig 4A,B) Western blot analysis revealed that BAD was only slightly dephosphorylated at serine 75 at 12 h of treat-ment with 7-ketocholesterol alone (a 15–20% decrease
of phosphorylated BAD), whereas cotreatment of THP-1 with the oxysterol and U0126 induced the same rate of dephosphorylation of BAD as early as 3 h
We next asked whether the acceleration of BAD dephosphorylation induced by U0126 could have an impact on mitochondria integrity Cells were pretreated with U0126, followed by various 7-ketocholesterol treatments, and the transmembrane mitochondrial potential was measured (Fig 4C) In the absence of a MEK inhibitor, mitochondria depolarization first appeared no later than 12 h after the beginning of treatment, whereas in the presence of U0126, the trans-membrane mitochondrial potential declined as early as
6 h We next wondered whether U0126 could also accelerate cytochrome c and Smac⁄ Diablo leakage from the mitochondria Western blot analysis was per-formed and we showed that U0126 increased the release of cytochome c and Smac⁄ Diablo induced by 7-ketocholesterol treatment (Fig 4D) Taken together, these results indicate that BAD is phosphorylated at ser75 after ERK activation by 7-ketocholesterol and this phosphorylation delays apoptosis by preventing mitochondrial damage
Secondary decreases of PYK 2 and ERK 1⁄ 2 acti-vities do not appear to be related to a decrease in cytoplasmic calcium
As we previously described that 7-ketocholesterol treatment of THP-1 cells induced an increase of cyto-plasmic free calcium [8], a time course of intracellular calcium fluctuations was performed between addition
of 7-ketocholesterol and appearance of a significant number of apoptotic cells to investigate the relation-ship between changes in intracellular calcium, activa-tion of PYK 2-ERK 1⁄ 2 pathway and apoptosis As shown in Fig 5 calcium concentration increased until
12 h when, whereas PYK 2 and ERK 1⁄ 2 activities peaked at 2–3 h as shown above So the secondary decrease of PYK 2 and ERK 1⁄ 2 activities did not appear to be related to a decrease in cytoplasmic free calcium
7-keto A
B
C
D
E
F
7-keto
P Erk1/2
Erk1/2
P PYK2
PYK2
P PYK2
PYK2
200
Ctrl
7-keto
Vera 3h 3h+Vera 6h 6h+Vera
Ctrl Vera 3h 3h+Vera 6h 6h+Vera
Ctrl Vera 3h 3h+Vera 6h 6h+Vera
Ctrl Vera 3h 6h 6h+Vera Time of 7-ketotreatment
Time of 7-keto treatment
Time of 7-keto treatment
*
3h+Vera
100
150
50
0
200
120
160
80
40
0
200
100
150
50
0
Fig 3 7-Ketocholesterol-induced calcium influx activates ERK
phos-phorylation through PYK 2 THP-1 cells were either untreated (Ctrl)
or incubated with L -type calcium channel blocker, verapamil (Vera,
100 lmolÆL)1) alone or in association with 7-ketocholesterol for
the indicated times Cell extracts were collected, subjected to
SDS ⁄ PAGE and immunoblotted with ERK 1 ⁄ 2 and
phospho-ERK 1 ⁄ 2 (A), or PYK 2 and phospho-PYK 2 (C, E) antibodies (B, D, F)
Respective group densitometry results Values are means ± SD
(n ¼ 3) *P < 0.05 vs control group.
Trang 5Bim activation depends on calcium uptake
Besides BAD, Bim functions also as a sensor toward
apoptotic stimuli by heterodimerization and
inactiva-tion of anti-apoptotic multi-BH domain proteins
Moreover, it has been shown that Bim could be under
control of the MEKfi ERK pathway [25] Indeed,
ERK 1⁄ 2 is known to exert its anti-apoptotic activity
in promoting phosphorylation and consequently the proteasome-dependent degradation of Bim [26] Inter-estingly, in THP-1 cells, BimEL is expressed in control cells and the level of BimEL, as well as Bcl-2, is not apparently changed by 7-ketocholesterol treatment (Fig 6A) Moreover, the use of the ERK inhibitor, U0126, does not affect BimELor Bcl-2 expression, sug-gesting that the transient activation of ERK 1⁄ 2 by
Calcium Influx
Ctrl Vera 7-keto 7-keto + Vera
Time of 7-keto treatment (Hours)
70
60
50
40
30
20
10
0
Fig 5 Time course of intracellular calcium after 7-ketocholesterol treatment of THP-1 cells THP-1 cells were incubated in the pres-ence of 7-ketocholesterol (40 lgÆmL)1, 7-keto), with or without verapamil (100 lmolÆL)1, Vera) or in the absence of oxysterol (Ctrl) After different incubation times, cells were loaded with fluo-3 ⁄ AM and the dye fluorescence was measured by flow cytometry Each fluorescent data point is normalized to the maximal fluo-3 fluores-cence induced in cells treated with ionomycin (2 lmolÆL)1) Data are the means ± SD.
P Bad (S75)
Bad
Hsc 70
Ctrl
800
600
400
200
0
7-keto A
B
C
D
% of Cells with Depolarized Mitochondria
Smac / Diablo
S100 Fraction Cytochrome c
Hsc 70
Time (h)
6h+U
Ctrl U 18h 18h+U 24h 24h+U 30h 30h+U
7-keto
Ctrl U
7K20 7K20 + U
70
60
50
40
30
20
10
0
Time of 7-keto treatment
†
*
*
*
†
†
Fig 4 7-Ketocholesterol-induced ERK phosphorylation inhibits BAD
dephosphorylation, mitochondria depolarization and cytochrome
c–Smac ⁄ Diablo release (A) Western blot analysis of BAD and
phospho-BAD was performed during 7-ketocholesterol (7-keto) and
U0126 (U) treatment of THP-1 cells (B) Phospho-BAD blot
densi-tometry analysis Values are means ± SD (n ¼ 3) *P < 0.05 vs.
Ctrl or U group; P < 0.05 vs Ctrl or oxysterol-treated cells (C)
Transmembrane mitochondrial potential was measured by flow
cytometry using DiOC 6 (3) dye After the incubation period,
fluores-cence associated with DiOC6(3) was measured by flow cytometry
and 10 000 cells were analysed for each assay Results represent
the means ± SD (n ¼ 4) (*P < 0.05) (D) THP-1 cells were either
untreated (Ctrl), treated with U0126 (U) or with 7-ketocholesterol
(7-keto) for 18, 24 or 30 h alone or in association with U0126, and
cytosol fractions (S100 fraction) were collected Subcellular
frac-tions were subject to SDS ⁄ PAGE and immunoblot analyses were
performed with antibodies against cytochrome c or Smac ⁄ Diablo.
Hsc70 was used as internal control loading The blots are
represen-tative of two independent experiments.
7-keto A
Ctrl
Ctrl
Bcl-2
Bcl-2
Fig 6 7-Ketocholesterol and ERK activation do not regulate Bim expression THP-1 cells were either untreated (Ctrl), treated with U0126 (U) or with 7-ketocholesterol (7-keto) alone or in association with U0126 for the indicated period, and cell extracts were collec-ted Lysates were analysed by western blotting using Bim and Bcl-2 antibodies (A, B).
Trang 67-ketocholesterol does not induce the degradation of
BimELin THP-1 cells (Fig 6B)
In light of these results, we wondered if BimEL was
involved in 7-ketocholesterol-induced apoptosis, because
BAD does not seem to be the only pro-apoptotic
molecule involved [27] Indeed, Puthalakath et al
[19] described that the pro-apoptotic activity of Bim
can also be regulated by interaction with the dynein
motor complex and the microtubules As we previously
described that 7-ketocholesterol induced a calcium
uptake and as it is known that calcium uptake could
destabilize microtubules [28] dissociating Bim from
microtubule dynein motor complex, we wondered if
7-ketocholesterol-induced calcium influx could modify microtubule structure and BimEL localization Thus,
we examined the effects of 7-ketocholesterol, alone or
in association with verapamil, on microtubule organ-ization using classical fluorescence microscopy Both control cells and verapamil treated cells displayed a typical randomly oriented, intact microtubular network (Fig 7A,B) In 7-ketocholesterol-treated cells, a pro-gressive disorganization⁄ reorganization of microtubules occurred These modifications were time-dependent and affected more than 90% of the cells (Fig 7C,E and data not shown) This reorganization⁄ disorganization was partially prevented in the presence of verapamil
Ctrl
A
B
C
D
E
F
Vera
7-keto
6h
7-keto
6h +
7-keto
12h
7-keto
12h +
Vera
Vera
Fig 7 Analysis of calcium involvement in 7-ketocholesterol-induced THP-1-microtubule-network disruption Cells were analysed by indirect immunofluorescence microscopy using anti-(a-tubulin) Igs and Hoechst 33342
as described in Experimental procedures showing microtubules and nuclear structure
of (A) untreated cells (Ctrl) (B) verapamil-treated cells (Vera), 7-ketocholesterol-trea-ted cells alone (C, E) or in association with verapamil (D, F).
Trang 7(Fig 7D,F), suggesting that calcium uptake could be
the event that initiates microtubule destabilization
Thus, we wondered if inactivated Bim is targeted
to microtubules and if 7-ketocholesterol treatment of
THP-1 cells could change its localization
Co-immuno-precipitation experiments revealed that in untreated
cells BimEL is complexed with microtubules and not
with Bcl-2, whereas as 7-ketocholesterol treatment
time progresses, BimEL colocalized with Bcl-2 and
not with microtubules, the part of Bcl2 in anti-Bim
immunoprecipitate (IP Bim) growing from 18% in
control to 50% after 24 h of treatment, a 2.7-fold
increase (Fig 8A,B) Moreover, cotreatment of the
cells with 7-ketocholesterol and verapamil appears to
abolish the dissociation of BimEL from the
microtu-bules and consequently its translocation to Bcl-2, the
part of Bcl2 in IP Bim is only of 24% after 24 h of
cotreatment with verapamil We next investigated the
localization of BimEL at the mitochondrial level under
7-ketocholesterol treatment alone or in association
with verapamil The results from this investigation
confirmed the increase of BimEL levels in the
mito-chondria and that this increase is inhibited by the
presence of verapamil (Fig 8C) So, BimEL
relocaliza-tion at the mitochondrial level is dependent on
cal-cium influx induced by 7-ketocholesterol
Discussion
In a prior report [8], we demonstrated that a major step
in the apoptotic response to 7-ketocholesterol of human monocytic cell line THP-1 is the sustained influx of extracellular Ca2+ leading to the dephospho-rylation of the pro-apoptotic ‘BH3-only’ protein BAD Moreover we demonstrated that this dephosphoryla-tion was mediated by calcineurin (PP2B) However, this dephosphorylation was incomplete and occurred more quickly at serine 99 than at serine 75, suggesting the existence of alternative mechanisms leading to apopto-sis Moreover, this idea was confirmed by the work of Panini et al describing that calcium-dependent activa-tion of cPLA2 led to the stimulaactiva-tion of arachidonate release and to apoptosis [29] However, the lack of a complete inhibition of apoptosis in cPLA2 (–⁄ –) macro-phages also led this group to point out the existence of other pathways regulating oxysterol-induced cell death Therefore, oxysterol-induced apoptosis appears to be a complex phenomenon with multiple initiation pathways and several apoptotic mechanisms
In this study, we examined two paradoxical effects induced by 7-ketocholesterol in THP-1 cells Indeed, our results demonstrate a calcium-dependent activation
of one survival pathway, the PYK 2fi MEK 1 ⁄ 2 fi
IP Bim A
B
C
Ctrl Vera 6h 6h
+Vera
Tubulin α
tubulin 500
400 300 200 100 0
Ctrl Ve ra 6h
6h+V era
12h+V era
18h+V era
24h+V
era
Bcl-2
Time of 7-keto treatment
7-keto Ctrl 6h 12h 18h
WB : Bim
- Vera + Vera
Mitochondria
*
*
†
†
†
†
Bcl-2 Bim EL
12h 12h 18h 18h 24h
Ctrl 6h 12h 18h 24h 24h
Total Extract
Fig 8 7-Ketocholesterol-induced calcium
influx activates the dissociation of Bim from
microtubules and consequently its
transloca-tion to Bcl-2 at the mitochondrial level (A)
THP-1 cells were untreated (Ctrl), treated
with verapamil (Vera) or with
7-ketocholes-terol alone (7-keto) or in association with
verapamil for 6, 12, 18 or 24 h After
treatment, anti-Bim immunoprecipitates
were collected, subjected to SDS ⁄ PAGE,
and immunoblotted with antibodies specific
for a-tubulin, Bcl-2 or Bim Western blots
were also performed on total extract to
check the levels of a-tubulin, Bcl-2 or Bim.
(B) Tubulin and Bcl-2 blot densitometry
analysis Values are means ± SD, (n ¼ 3).
*P < 0.05 vs Ctrl or Vera group; P < 0.05
vs Ctrl or oxysterol-treated cells (C) THP-1
cells were untreated (Ctrl), treated with
verapamil (Vera) or with 7-ketocholesterol
alone (7-keto) or in association with
verapa-mil for 6, 12 or 18 h Mitochondrial fractions
were collected and subjected to western
blot analysis with antibodies against Bim.
The blots are representative of three
independent experiments.
Trang 8ERK 1⁄ 2 pathway allowing BAD phosphorylation on
serine 75, as well as one additional apoptotic pathway
inducing the translocation of Bim from microtubules
to Bcl-2 at the mitochondrial level, which leads to
mit-ochondrial damage and apoptosis In fact, the use of
U0126, a MEK 1⁄ 2 inhibitor which accelerates
7-keto-cholesterol-induced THP-1 apoptosis, clearly suggests
that this signalling pathway acts as a survival pathway
In most, but not all systems studied, activation of
MEK 1⁄ 2 and ERK 1 ⁄ 2 is associated with the
inhibi-tion of cell death [30,31] The mechanism by which this
occurs is not completely understood, but could be
rela-ted to the phosphorylation of BAD, and thus, to its
inactivation [11] Moreover, direct anti-apoptotic
actions of ERK 1⁄ 2 related to the
phosphoryla-tion⁄ inactivation of caspase-9 [32] and to the
phos-phorylation⁄ degradation of Bim [26] have been
recently described It is therefore surprising that
expo-sure of leukaemic cells to 7-ketocholesterol at
concen-trations that trigger apoptosis was associated with the
clear activation of ERK 1⁄ 2, even if it is a transient
phenomenon Moreover, it seems that other MAPKs
(JNK or p38 MAPK) were not activated under
7-keto-cholesterol treatment of THP-1 cells, suggesting that
ERK activation is not under the activation of other
MAPKs as described by Numazawa et al [33]
Con-trary to Rusin˜ol et al [27], who described Akt⁄ PKB
degradation under oxysterol treatment of macrophage
cells, we never noticed a degradation of Akt⁄ PKB in
THP-1 cells during the course of 7-ketocholesterol
treatment but we did show that Akt⁄ PKB was
inacti-vated by dephosphorylation at threonine 308
As we previously described a calcium influx in
7-ketocholesterol-induced THP-1 apoptosis, we
exam-ined the role of calcium in ERK 1⁄ 2 activation
Hence, the L-Type calcium channel blocker verapamil
and intracellular calcium chelator BAPTA completely
inhibited 7-ketocholesterol-induced ERK activation
Interestingly, calcium-dependent activation of the
MEK 1⁄ 2 fi ERK 1 ⁄ 2 pathway has been previously
described in 7b-hydroxycholesterol-treated aortic
smooth muscle cells [21] Nevertheless, the authors
did not describe the effects or the mechanism of
ERK activation Protein tyrosine kinases (PTKs)
transduce key extracellular signals that trigger various
biological events, such as cytoskeletal rearrangement
and mitogenesis Among the PTKs, PYK 2 (also
known as CAKb, RAFTK or CADTK), which exists
mainly in the cytoplasm [34], is abundantly expressed
in haemopoeitic cells and in the brain Moreover,
PYK 2 is activated by stimuli that increase the
concentrations of intracellular Ca2+ Indeed elevation
of intracellular calcium triggers activation of PYK 2
as described by Lev et al [13] and a maximal cata-lytic activity was observed after phosphorylation of PYK 2 at tyrosines 579 and 580 in the kinase domain activation loop [35] Thus, in untreated THP-1 cells,
as described by Yamasaki et al [36], PYK 2 was poorly phosphorylated at tyrosines 579⁄ 580, but after 7-ketocholesterol treatment we saw an increase of its phosphorylation Moreover, our data demonstrate that verapamil inhibits 7-ketocholesterol-mediated-PYK 2 phosphorylation, suggesting the role of this protein tyrosine kinase in ERK activation However the secondary decreases of PYK 2 and ERK1⁄ 2 acti-vites are not related to a decrease of cytoplasmic free calcium suggesting that 7-ketocholesterol treatment of THP-1 induces others transduction pathways leading
to a secondary inactivation of these kinases
We next examined the ability of the MAPK signal-ling pathway to inhibit, via p90RSK, the apoptotic effect of 7-ketocholesterol by phosphorylating BAD Treatment of cells with the combination of the MEK inhibitor U0126 and 7-ketocholesterol caused an ear-lier dephosphophorylation of BAD than with the oxy-sterol alone The induction of cell death was also faster following cotreatment than treatment with 7-ketocholesterol alone, indicating that inhibition of BAD phosphorylation at serine 75 increases apoptosis This process could accelerate the disruption of the mito-chondrial transmembrane potential and Smac⁄ diablo and cytochrome c release into the cytosol Hence, our findings suggest that the MAPK signalling pathway pro-motes cell survival by a mechanism that modulates the cell death machinery directly by phosphorylating and thereby inactivating the pro-apoptotic protein BAD Our previous results also suggest the possibility of the involvement of other Ca2+-initiated cell death pathways Therefore our results showing the release of the pro-apoptotic ‘BH3 only’ protein Bim and its association with Bcl-2 in a calcium-dependent man-ner complement our knowledge on the mechanism
of 7-ketocholesterol-induced apoptosis A calcium-dependent destabilization of microtubules, as previ-ously described by Keith et al [33], probably induces the dissociation of Bim from the microtubule dynein motor complex, allowing inactivation of anti-apopto-tic, multi-BH domain proteins such as Bcl-2 or Bcl-XL
in the mitochondria Previously Palladini et al des-cribed a 7-ketocholesterol induced destabilization of vimentin filament architecture without significant alter-ations of the microtubule network on a bovine aortic endothelial cell line Nevertheless they showed that 7-ketocholesterol induced tubulin aggregates and so they did not exclude a possible role of microtubules in oxysterol-induced endothelial cell apoptosis [37] It has
Trang 9been reported that activation of the ERK 1⁄ 2 signaling
pathway promoted phosphorylation and
proteasome-dependent degradation of Bim, but this result was not
obtained in our study The possible explanation could
be that the length and intensity of activation of
ERK 1⁄ 2 is not sufficient to induce a detectable
change in Bim concentration The short length of
ERK activation and the dephosphorylation of Akt⁄
PKB could be related to the activation of phosphatases
such as protein phosphatase 1a [38] Hence,
7-keto-cholesterol-induced apoptosis appears to be a
com-plex phenomenom implicating several transduction
pathways, two pro-apoptotics and surprisingly one
anti-apoptotic, that to our knowledge have not been
previously described However further studies will be
necessary to try to quantify the part of these different
pathways in 7-ketocholesterol induced cytotoxicity All
of these mechanisms seem to be related to the
sus-tained rise of free cytosolic calcium triggered by the
transfer of Trpc-1 into lipid raft domains that we have
previously described [8] and involve the mitochondrial
pathway of apoptosis with the implication of the
pro-apoptotic BH3 proteins BAD and Bim
As all oxysterols are not able to induce apoptosis and
as cytotoxicity can vary according cell type, it will be
interesting to use other oxysterols and other cell types
and especially other cell lines of the monocytes⁄
macro-phages lineage to investigate the implication of these
different pathways and particularly sustained calcium
ion influx, which appears to be the key event in
oxy-sterol-mediated cytotoxicity In this way the first
experi-ments performed by our group on U937, another
monocytic cell line, with 7-ketocholesterol (A Berthier,
unpublished results) produced results quite similar to
those obtained with THP-1 cells
Experimental procedures
Reagents and antibodies
The THP-1 human monocytic cell line was from the
Ameri-can Tissue and Culture Collection (Manassas, VA, USA)
from Molecular Probes, Inc (Eugene, OR, USA)
dimethylsulfoxide, verapamil, pepstatin A, aprotinin,
paraformaldehyde, Hoechst 33342, NaF,
b-glycerophos-phate and Triton X-100 were from Sigma (Sigma-Aldrich,
CA, USA) The anti-BAD monoclonal antibody and the
BAD phospho Ser112 (human Ser75) polyclonal
anti-body were from Upstate Biotechnology (Lake Placid, NY, USA), and the anti-cytochrome c monoclonal antibody was from Pharmingen (San Diego, CA, USA) The anti-(Hsc
Bim polyclonal antibodies, the anti-(a-tubulin) monoclonal antibody, protein G-Agarose, and total mouse IgG were from Santa Cruz Biotechnologies (Santa Cruz, CA, USA),
anti-PKB and anti-PKB phospho Thr308 and U0126 were from Cell Signaling Technology (Cell Signaling Technology, Hitchin, UK) and the anti-Bcl-2 Ig was from Dako (Dako, Trappes, France)
Cell culture
Human monocytic THP-1 cells were grown in RPMI 1640 with glutamax-I (Gibco, Eragny, France) and antibiotics
Cell treatment
For all experiments, a 7-ketocholesterol stock solution was
described [39] 7-Ketocholesterol was added to the culture
con-centration is in the range of levels measured in human plasma after a meal rich in fat [40] Verapamil, an L-type calcium channel inhibitor, was added to the culture medium at a final
all experiments, verapamil and U0126 were introduced in the culture medium 30 min before 7-ketocholesterol
Characterization of nuclear morphology
by staining with Hoechst 33342
Nuclear morphology of control and treated cells was studied
by fluorescence microscopy after staining with Hoechst 33342
The morphological aspect of cell deposits, applied to glass slides by cytocentrifugation with a cytospin 4 centrifuge (Shandon, UK), was observed with an Axioskop light micro-scope (Zeiss, Jena, Germany) by using UV light excitation Three hundred cells were examined for each sample
Flow cytometric measurement of mitochondrial transmembrane potential (DWm) with the dye DiOC6(3)
Variations of the mitochondrial transmembrane potential
Trang 10iodide (DiOC6(3); kExmax: 484 nm, kEmmax: 501 nm) used
cyto-metric analyses were performed on a Galaxy flow cytometer
(Partec, Munster, Germany) and the green fluorescence was
signals were measured on a logarithmic scale of four
dec-ades of log For each sample, 10 000 cells were acquired
and the data were analysed with flomax software (Partec)
Flow cytometric measurement of cytosolic
calcium with the dye Fluo-3
salt buffer (pH 7.2) with Pluronic F-127 After loading,
cells were suspended in Hepes buffer (pH 7.4) supplemented
with probenecid (5 mm) to prevent leakage of the dye
Fluorescence was measured by flow cytometry with a
filter For each sample, events were acquired for 60 s and
the data were analysed with flomax software (Partec)
Staining for microtubules
Treated or control cells were rinsed twice in
microtubule-stabilizing buffer (MSB; 50 mm Pipes pH 6.9, 10 mm
EGTA, and 10 mm MgSO4) Cells were immediately fixed
para-formaldehyde and 0.2 mm mannitol Cells were then
washed twice with MSB, fixed on polylysine glass and
MSB, cells were treated for 2 h at room temperature with
presence of the primary mouse monoclonal antibody
against a-tubulin [1 : 100 dilution in MSB containing 0.1%
pres-ence of the secondary AlexaFluor 565-conjugated rabbit
anti-mouse antibody (1 : 250 dilution in MSB containing
0.1% MSB) After two washes in MSB, cells were mounted
in Fluoprep At least 50 cells were examined for each
experiment and three independent experiments were
per-formed for each treatment Observations were perper-formed
on an Axioskop light microscope (Zeiss, Jena, Germany)
by using UV light excitation
Immunoprecipitation and western blotting
For the immunoprecipitation of the Bim protein, cells were
protease inhibitors (0.1 mm phenylmethanesulfonyl fluoride,
eliminated by centrifugation for 10 min at 10 000 g The resulting supernatant was precleared by adding 1 lg total mouse IgG and 50 lL protein G-agarose for 30 min After
was collected, adjusted to 500 lL in lysis buffer and
rotating device After the incubation period, 50 lL protein G-agarose were added and the sample was incubated for
2 h before collecting the immunoprecipitates by
times, the immunoprecipitation extract was suspended in
Alternatively, cells were resuspended in Ripa lysis buffer
contain-ing a mixture of protease and phosphatase inhibitors
b-glycerophosphate, 100 mm NaF) After a 30-min
by centrifugation for 20 min at 10 000 g and the supernatant was collected
mito-chondria to the cytosol was investigated by western blot ana-lysis of THP-1 cells incubated for 18, 24 or 30 h with 7-ketocholesterol alone or in association with U0126 as pre-viously described [42] Mitochondria were obtained before the last 1000 g centrifugation generating the cytosol fraction The protein concentrations were measured by using bicinchoninic acid reagent (Pierce, Rockford, IL, USA) according to the method of Smith et al [43] Seventy
diflu-oride membrane (Bio-Rad, Ivry sur Seine, France) After blocking nonspecific binding sites for 2 h at room
antibody diluted in TPBS After three 10-min washes with TPBS, the membranes were incubated with horseradish peroxidase-conjugated secondary antibody at a dilution of
1 : 2500 for 1 h at room temperature and washed three times in TPBS for 10 min Autoradiography of the immu-noblots was performed using an enhanced chemolumines-cence detection kit (Amersham, Les Ulis, France) Western blots were quantified using a JS800 densitometer using