Interestingly, activation of PPARb enhanced the mRNA expression of CerK and CerK activity.. Results Skin barrier disruption in hairless mice induces mRNA expressions for both PPARb and C
Trang 1activated receptor beta-induced cell survival of
mouse keratinocytes
Kiyomi Tsuji1, Susumu Mitsutake2, Urara Yokose2, Masako Sugiura3, Takafumi Kohama4and Yasuyuki Igarashi1,2
1 Laboratory of Biomembrane and Biofunctional Chemistry, Faculty of Advanced Life Sciences, Hokkaido University, Sapporo, Japan
2 Laboratory of Biomembrane and Biofunctional Chemistry, Faculty of Pharmaceutical Sciences, Hokkaido University, Sapporo, Japan
3 Biological Research Laboratories II, Daiichi-Sankyo Co Ltd., Tokyo, Japan
4 Exploratory Research Laboratories I, Daiichi-Sankyo Co Ltd., Tokyo, Japan
Ceramide (Cer) has been implicated in various cellular
processes including proliferation, apoptosis and cell
signaling [1] Intracellular Cer levels are strictly
regu-lated by several enzymes, including ceramide kinase
(CerK), which converts Cer to ceramide 1-phosphate
(C1P) [2] Previous studies have suggested that CerK
and C1P are involved in many cell functions, including
membrane fusion, phagocytosis and degranulation in mast cells, among others [3] Recently, several studies have established a function for CerK in cell growth and apoptosis For example, in Arabidopsis plants, mutation of CerK was associated with an accumula-tion of Cer and enhanced symptoms during pathogen attack [4] In addition, in mammalian cells such as
Keywords
cell survival; ceramide; ceramide
1-phosphate; CerK; PPARb
Correspondence
Y Igarashi, Laboratory of Biomembrane and
Biofunctional Chemistry, Faculty of
Pharmaceutical Sciences and Faculty of
Advanced Life Sciences, Hokkaido
University, Nishi 6, Kita 12, Kita-ku, Sapporo
060-0812, Japan
Fax: +81 11 706 4986
Tel: +81 11 706 3970
E-mail: yigarash@pharm.hokudai.ac.jp
Website: http://biomem.pharm.hokudai.ac.jp/
english/index.html
(Received 5 April 2008, revised 26 May
2008, accepted 29 May 2008)
doi:10.1111/j.1742-4658.2008.06527.x
Ceramide (Cer) is known to be a lipid mediator in apoptosis and to have
an important role in cell fate, via control of intracellular Cer levels Recently, ceramide kinase (CerK) was identified as an enzyme that converts Cer to ceramide 1-phosphate (C1P) We examined potential functions of CerK in the regulation of keratinocyte survival, and the possible involve-ment of peroxisome proliferator-activated receptor beta (PPARb) PPARb
is known to be a nuclear receptor acting as a ligand-inducible transcription factor and has been implicated in the control of keratinocyte survival In the mouse keratinocyte cell line SP1, serum starvation induced cell death and the accumulation of intracellular Cer, an apoptotic event However, apoptosis was inhibited by activation of PPARb Interestingly, activation
of PPARb enhanced the mRNA expression of CerK and CerK activity Furthermore, the cell survival effect of PPARb was greatly diminished in keratinocytes isolated from CerK-null mice Chromatin immunoprecipita-tion revealed that, in vivo, PPARb binds to the CerK gene via a sequence located in the first intron Electrophoretic mobility-shift assays confirmed that PPARb associates with this sequence in vitro These findings indicated that CerK gene expression was directly regulated by PPARb In conclu-sion, our results demonstrate that PPARb-mediated upregulation of CerK gene expression is necessary for keratinocyte survival against serum starva-tion-induced apoptosis
Abbreviations
ABC, ATP-binding cassette; C1P, ceramide 1-phosphate; Cer, ceramide; CerK, ceramide kinase; ChIP, chromatin immunoprecipitation; EMSA, electrophoretic mobility shift assays; LD, L-165,041; PI, propidium iodide; PPARb, peroxisome proliferator-activated receptor beta; PPRE, PPAR response element; RXR, retinoid X receptor; TEWL, transepidermal water loss.
Trang 2NIH 3T3 fibroblasts and A549 lung cancer cells,
treat-ment with exogenous C1P at low concentrations
enhanced cell survival, whereas high concentrations of
C1P reduced cell survival and enhanced apoptosis
induced by serum starvation [5] Although these
previ-ous reports suggest that CerK and C1P are involved in
the regulation of cell survival or cell proliferation, the
molecular mechanisms involved remain largely
unknown
Peroxisome proliferator-activated receptors (PPARs)
are members of the nuclear hormone receptor
super-family PPARs form heterodimers with the retinoid
X receptor (RXR) in a ligand-dependent manner
Together, these heterodimers regulate the expression of
target genes by binding to their PPAR response
ele-ments (PPREs) [6,7] Three subtypes of PPARs have
been identified, PPARa⁄ NR1C1, PPARc ⁄ NR1C3 and
PPARb⁄ PPARd ⁄ NR1C2, each with reportedly unique
tissue distribution and distinct cellular functions in
lipid metabolism, diabetes, inflammation and tumor
progression [8] PPARb is ubiquitously expressed and
has been implicated by recent studies in
tumor-promo-tion in various cell types [9] PPARb has also been
suggested to have an important role in skin wound
healing After a skin injury, PPARb expression is
rap-idly elevated in the epidermis at the wound edges;
dele-tion of a single PPARb allele results in delayed wound
healing [10] Furthermore, inflammation-induced
apo-ptosis was found to be enhanced in keratinocytes
isolated from PPARb-null mice [11] These studies
sug-gest a role for PPARb in the regulation of keratinocyte
survival and apoptosis Interestingly, the promoter
activity of PPARb was reported to be increased by the
treatment of exogenous Cer [11] However, any
inter-actions between PPAR and sphingolipids remain to be
further defined
Skin epidermis contains abundant lipids, including
cholesterol, fatty acids and Cer, each of which plays a
critical role in water retention and epidermal
perme-ability barrier functions [12] Recently, an abnormal
sphingolipid distribution pattern was found in the
keratinocytes of patients suffering from harlequin
ichthyosis, a genetic disorder in which the
keratino-cytes carry abnormal lamellar granules [13] In
harle-quin ichthyosis, mutations have been identified in the
ABCA12 gene, which encodes a member of the lipid
transporter ATP-binding cassette (ABC) family that is
also part of the ABCA subfamily ABCA12 is thought
to function in keratinocytes as a glucosylceramide
transporter to lamellar bodies [13] Interestingly,
expression of ABCA12 mRNA has been shown to be
induced by activation of PPARb or PPARc in cultured
human keratinocytes [14]
In this study, we examined the interaction between CerK and PPARb, and its role in regulating keratino-cyte survival We report that in a mouse keratinokeratino-cyte cell line upregulation of CERK expression by activa-tion of PPARb results in a decrease in intracellular Cer levels and enhancement of cell survival
Results
Skin barrier disruption in hairless mice induces mRNA expressions for both PPARb and CerK PPARb plays an important role in keratinocyte sur-vival during skin wound healing, and PPARb expres-sion is elevated at injury sites [10,15] We investigated whether skin barrier disruption by tape-stripping would induce PPARb expression To quantify the skin barrier disruption, transepidermal water loss (TEWL),
an indicator of skin barrier function [16], was mea-sured in the dorsal skin of hairless mice treated with
or without tape-stripping With normal skin barrier function, low levels of water loss from epidermal tissue and low TEWL rates were evident However, tape-stripping the stratum corneum significantly increased the TEWL rate and resulted in skin barrier disruption (Fig 1A) PPARb mRNA expression was determined
in keratinocytes isolated from the epidermal layer of these skins, and was found to be significantly increased
in skin barrier disruption following tape-stripping (Fig 1Ba,b) Furthermore, CERK mRNA expression was also elevated following tape-stripping (Fig 1Ba,c) This is the first evidence that CERK expression increases in response to skin barrier disruption by tape-stripping, and that it is accompanied by increases
in PPARb expression
Stress-induced cell death is inhibited by the treatment of specific PPARb ligand:
L-165,041 (LD) Cer has emerged as an apoptotic lipid mediator, and accumulation of Cer and apoptosis are known to be induced under various stimuli such as tumor necrosis factor-alpha, serum deprivation and c-radiation [17] Previous studies have provided evidence that control
of intracellular Cer levels, through their modulation by enzymes active in sphingolipid metabolism, is impor-tant in drug resistance and cancer cell survival [18,19] Liu et al [18] reported that the activity of glucosyl-ceramide synthase, the enzyme that converts Cer to glucosylceramide, and glucosylceramide levels were increased in adriamycin-resistant breast cancer cells Uchida et al [19] reported that in HL-60⁄ ADR cells
Trang 3transcriptional upregulation of glucosylceramide
syn-thase via doxorubicin-induced activation of Sp1 results
in a decreased Cer level and obtained drug resistance
Moreover, we previously reported that exogenous Cer
is incorporated, hydrolyzed to sphingosine and then
recycled into intracellular Cer itself, and that
accumu-lation of Cer contributes to the induction of apoptosis
[20] However, the signaling mechanism involved in
Cer-mediated apoptosis remains to be further defined
CerK is an enzyme that converts Cer to C1P
Recently, CERK was cloned [2] C1P has been
reported to be involved in the regulation of
pro-gramed cell death in plants and in cell survival in
mammalian cells [4] Arabidopsis, carrying a CerK
mutation, exhibits a spontaneous cell-death
pheno-type and accumulates Cer late in development [4] In
A549 human lung adenocarcinoma cells transfected
with CERK siRNA, downregulation of CerK
reduced cellular proliferation [5] We thought,
there-fore, that CerK could be involved in cell survival
promoted by PPARb, and that a decrease in Cer
content following activation of CerK would cause
suppression of cell death In order to investigate this possibility, we examined whether CerK could be affected by activation of PPARb, using the mouse keratinocyte cell line SP1 Enhanced cell survival has previously been reported in the human keratinocyte cell line HaCat following treatment with the specific PPARb ligand L-165,041 (LD) [21]
We confirmed that SP1 cell survival could be enhanced by PPARb As shown in Fig 2A, serum starvation stress induced limited cell growth and death
in SP1 cells (Fig 2Ac), yet treatment with 1 lm LD inhibited this cell death (Fig 2Ad) Furthermore, as determined using a cell proliferation assay (Fig 2B), the rate of cell survival in SP1 cells treated with serum starvation stress was reduced in a time-dependent manner However, treatment with 1 lm LD inhibited the reduction of cell survival induced by serum starvation stress Flow cytometry analysis further demonstrated that treatment with LD (1 or 10 lm) suppressed the number of annexin V⁄ propidium iodide (PI)-positive cells, representing late apoptotic or necro-tic cells, which had increased upon serum starvation
P< 0.0001
Non-stripped strippedTape
-100
80
60
40
20
0
A
B
8
6
4
2
0
P < 0.0001 8
6
4
2
0
P < 0.0001
CERK Mrpl27
Non-stripped strippedTape - strippedNon- strippedTape
- Non-stripped stripped
Fig 1 Skin barrier disruption in hairless mice induces mRNA expression for both PPARb and CerK (A) Quantification of TEWL indicates skin barrier disruption The dorsal skin of each mouse was tape-stripped five to eight times to perturb the skin barrier function TEWL was mea-sured in tape-stripped mice skin (tape-stripped) or control mice skin (not stripped) as described in Materials and methods Each column rep-resents the mean ± SD of three animals for each group (B) Expression levels of PPARb and CERK mRNA in skin barrier disruption Total RNA was extracted from the epidermis of tape-stripped mice skin (tape-stripped) or control mice skin (not stripped), and the expression lev-els of PPARb, CERK and Mrpl27 mRNA were determined by RT-PCR as described in Materials and methods (a) Agarose gel electrophoresis
of the products of PCR using specific primers for PPARb, CERK or Mrpl27 mRNA (b, c) Density of the mRNA expression of PPARb (b) or CERK (c) The results shown are normalized to the mRNA level of Mrpl27, and are relative to control mice skin (not stripped), and the mean ± SD of three animals for each group.
Trang 4stress (Fig 2C) These findings demonstrate that SP1
cells are sensitive to ligand activation of PPARb, which
enhanced cell survival or inhibited cell death induced
by serum starvation
Next, we investigated the participation of Cer in
cell survival promoted by PPARb, by examining
intracellular Cer levels Total cellular lipids were
extracted from serum-starved or non-starved SP1
cells cultured for 24 h in the presence or absence of
1 lm LD, and the levels of endogenous Cer were
determined by diacylglycerol kinase assay [22] Cer
levels were significantly increased in cells stressed by
serum starvation, compared with control cells
(Fig 2D) However, similar to results observed in
the survival studies, treatment with LD inhibited the
increase in Cer levels induced by serum starvation
stress These findings suggest that regulation of Cer
levels is involved in enhanced cell survival resulting
from activation of PPARb
Activation of PPARb by LD induces CERK mRNA expression and increases CerK activity in SP1 cells
To investigate the role of CerK and its regulation of Cer levels in the effect of PPARb activation on cell survival, the expression of CERK mRNA in SP1 cells treated with LD was determined using real-time PCR CERK mRNA expression was increased by LD treatment in a dose-dependent (Fig 3Aa) and time-dependent (Fig 3Ab) manner, indicating that activa-tion of PPARb is involved in the gene transcripactiva-tion
of CERK Furthermore, an in vitro kinase assay determined that CerK activities in whole-cell lysates were significantly increased in SP1 cells treated for
24 h with 1 lm LD, compared with untreated cells (Fig 3B) Although serum starvation stress did not affect CerK activity, the rate of increase in CerK activity in cells treated with LD was larger in cells
P < 0.001
250
200
150
100
50
0
D
Stress
LD
A
140
120
100
80
60
40
20
0
0 6 12 18 24 (h)
LD (+ )
LD (−)
B
Stress (+ )
Stress (− ) Stress (+ )
400
300
200
100
0
C
Stress (−)
Stress (+ )
LD (−) LD ( + )
P < 0.0001 P < 0.001
*
*
*
Apoptotic cells (% control)
a b
d
c
Fig 2 Activation of PPARb enhances cell survival and inhibits cell death induced by serum starvation stress (A) SP1 cells under serum star-vation stress treated with LD SP1 cells were cultured for an additional 24 h in the presence [LD (+)] or absence [LD ( ))] of the specific PPARb ligand LD (1 l M ), in medium lacking serum (serum-starvation conditions) or in medium containing fetal bovine serum (control condi-tions) [Stress (+) or Stress ( )), respectively] for 24 h SP1 cells were observed under a phase-contrast microscope (B) Effect of LD on the cell survival rate of mouse keratinocyte cells SP1 cells were cultured in the presence [LD (+)] or absence [LD ( ))] of LD (1 l M ), in serum starvation conditions [Stress (+)] After treatment for the indicated times, the cell survival rate was determined by Cell Counting Kit-8 as described in Materials and methods Significant difference from the corresponding LD ( )) time point (*P < 0.05) (C) Inhibition by LD of serum starvation stress-induced cell death SP1 cells were cultured in serum starvation conditions [Stress (+)] or in control conditions [Stress ( ))], in the presence of LD (0, 1 or 10 l M ) for 24 h Cells were then stained with annexin V ⁄ fluorescein isothiocyanate and PI, and analyzed
by flow cytometry The results shown are relative to untreated, unstressed cells the mean ± SD of three wells, and each experiment was repeated three times (D) Activation of PPARb inhibits the generation of cellular Cer induced by serum starvation stress SP1 cells were left untreated or treated with 1 l M LD under control conditions [Stress ( ))] or serum starvation conditions [Stress (+)] for 24 h Total cellular lipids were extracted using the standard Bligh–Dyer protocol Total cellular Cer levels were measured by diacylglycerol kinase assay as described under Materials and methods and quantified using an Image Analyzer BAS2000 The results shown are relative to untreated, unstressed cells [Stress ( )) ⁄ LD ())], and are the mean ± SD of three experiments.
Trang 5undergoing serum starvation stress than in unstressed
cells [23] These findings suggest that CerK, through
its regulation of Cer levels, plays an important role in
the effect of PPARb activation on cell survival in mouse keratinocytes
PPARb binds to a putative PPRE in CERK and transactivates the CERK gene
The results above suggest that PPARb can regulate expression of the CERK gene PPARs bind to response elements (PPREs) in target genes and regu-late transcription Functional PPREs should reside in the regulatory region of the target gene Although the promoter region in the mouse CERK gene has not been identified, transcription start sequences at the 5¢-end of the gene reside 100 bp upstream of the protein-coding sequence, as identified by the Database
of Transcriptional Start Sites (DBTSS) [24] and our experimental results (data not shown) Analysis of the region near the protein-coding sequence in the mouse CERK gene sequence (GenBank accession number NC_000081) was performed using nubiscan, an
in silico tool for predicting nuclear receptor binding sites [25] This analysis revealed a putative PPRE, which we refer to as putative CERK-PPRE, in intron 1 of the mouse CERK gene (Fig 4A) In order
to determine whether PPARb binds to the mouse CERK gene in vivo, chromatin immunoprecipitation (ChIP) was carried out using SP1 cells, untreated [LD ())] or treated [LD (+)] with 1 lm LD for 24 h Using immunoprecipitated chromatin, the CERK gene sequence containing putative CERK-PPRE was analyzed by PCR In the ChIP DNA obtained with the anti-PPARb IgG, the amount of DNA containing putative CERK-PPRE was significantly greater in the chromatin of cells treated with LD compared with untreated controls (Fig 4B,C) The results were com-parable in the ChIP DNA obtained with acetylated histone H4 antibodies No PCR products with CERK-negative were obtained in ChIP DNA, indicating that the binding of putative CERK-PPRE to PPARb is specific Furthermore, the interaction between PPARb and CERK was confirmed by an EMSA using biotin-labeled putative CERK-PPRE and nuclear extract from SP1 cells As shown in Fig 4D, the binding of biotin-labeled putative CERK-PPRE to nuclear extract from SP1 cells was detected as shift bands The levels
of these shift bands were reduced in the presence of competitors, including unlabeled putative CERK-PPRE and CERK-PPRE-Wild, a known consensus sequence However, no reduction in shift band levels was observed in the presence of PPRE-Mutant, a mutation sequence of PPRE-Wild These results demonstrate specific binding of putative CERK-PPRE to PPARb
in nuclear extracts of SP1 cells, and provide further
300
250
200
150
100
50 0
0 0.1 0.5 1.0 5.0 10
LD (μ M)
A
B
300
250
200
150
100
50 0
0 3 6 12 24 48
Time (h)
a
b
* * *
* *
*
CERK mRNA (% control)
400
300
200
100
0
P < 0.001
P < 0.05
Stress
LD
− − + +
− + − +
CerK activity (% control)
Fig 3 Activation of PPARb induces CERK mRNA expression and
increased CerK activity in SP1 cells (A) CERK mRNA expression is
induced by treatment with LD SP1 cells were treated with LD (0,
0.1, 0.5, 1, 5 or 10 l M ) under serum starvation conditions for 24 h
(a), or with 1 l M LD in serum-free medium for the indicated times
(b) Total RNA was extracted from the cells, and the mRNA
expres-sion levels of CERK and Mrpl27 (a housekeeping gene) were
deter-mined by quantitative real-time PCR as described in Materials and
methods The results shown are the mean ± SD of three wells,
and each experiment was repeated three times *P < 0.05
com-pared with the controls at 0 l M LD (a) or 0 h (b) (B) CerK activity is
enhanced in cells treated with LD SP1 cells were left untreated
[LD ( ))] or treated [LD (+)] with 1 l M LD in serum starvation
condi-tions or control condicondi-tions [Stress (+) or Stress ( )), respectively].
After treatment for 24 h, cell lysates were collected and analyzed
for in vitro CerK activities, as described in Materials and methods
using C18-Ceras a substrate with [32P]ATP[cP] [32P]C1P was
quan-tified using an Image Analyzer BAS2000 (Fuji Film) The results
shown are relative to untreated, unstressed cells, are the
mean ± SD of three experiments.
Trang 6evidence that PPARb directly regulates expression of
the CERK gene through PPRE
Next, we assessed the function of putative
CERK-PPRE using the Dual-Luciferase Reporter Assay
System In keeping with the genomic organization of
the CERK gene, the 1807 bp fragment containing
putative CERK-PPRE, or the 1007 bp fragmant not
containing putative CERK-PPRE, in the CERK
intron 1 region was subcloned downstream of the
minimal promoter and the Luc2P reporter gene in pGL4.27[luc2P⁄ minP ⁄ Hygro], respectively named minP-Luc2P-putative CERK-PPRE or minP-Luc2P-D putative CERK-PPRE Reporter gene transfection studies showed that the region containing putative CERK-PPRE has the capacity to significantly influence transcriptional activity By contrast, constructs not containing putative CERK-PPRE were approximately equal to the control luciferase activity from cells
Putative CERK-PPRE
ChIP DNA
CERK-negative
300 400 200 0
P < 0.01
LD (−) LD (+ )
Intron -1
Putative CERK-PPRE: AGGCCAcAGGCCA
IgG anti-PPAR β anti-acH4
B
C A
SP1 cell nuclear extracts Biotin-labeled putative CERK-PPRE
Competitors (non-labeled) putative CERK-PPRE PPRE-Wild PPRE-Mutant
+ + + + +
Free
Shift band
D
Input DNA CERK gene
E
1500
1000
500
0
P < 0.0001
minP-Luc2P-ve ctor
minP-Lu
c2P-Δ putative CERK-PPRE minP-Lu c2P-p utative CERK-PPRE
Fig 4 PPARb binds to putative CERK-PPRE and transactivates the CERK gene (A) Schematic representation of the CERK gene illustrating the position (bar) and sequence of the putative PPRE (putative CERK-PPRE) (B, C) ChIP demonstrates PPARb binding to putative CERK-PPRE
in vivo SP1 cells were untreated or treated with 1 l M LD in serum starvation medium for 24 h Nuclear extracts were collected and subjected
to a ChIP assay, as described in Materials and methods, using antibodies against PPARb or acetylated histone H4, or IgG as a negative control ChIP DNA and aliquot of pre-immunoprecipitation samples of nuclear extracts (Input DNA) were analyzed by PCR with primers for putative CERK-PPRE or CER-negative as PCR negative control with primers for unrelated putative PPRE, as described in Materials and methods PCR products of ChIP DNA and Input DNA were analyzed with 1% agarose gel electrophoresis (B) The bands of PCR products corresponding to the binding of PPARb to putative CERK-PPRE were normalized to the bands of PCR product of Input DNA, and are relative to untreated sample [LD ( ))] (C) The results shown are the means ± SD of three experiments (D) EMSAs indicate that PPARb binds to putative CERK-PPRE in vitro SP1 cells were treated for 24 h with 1 l M LD in serum starvation Nuclear extracts from the cells (5 lg) were incubated with biotin-labeled puta-tive CERK-PPRE oligonucleotide (20 fmol) Competition assays were performed with non-biotinylated oligonucleotides (4 pmol) of putaputa-tive CERK-PPRE, a specific DNA binding consensus sequence for PPARs (PPRE-Wild), or a mutant sequence of PPRE-Wild (PPRE-Mutant) Arrows indicate the labeled putative CERK-PPRE oligonucleotide in specific complex with SP1 nuclear extracts (Shift band) and unbound (Free) (E) Transfection assays indicate that PPARb transactivates the CERK gene SP1 cells were transiently cotransfected with pCMX–mPPARb, pCMX–mRXRa, the luciferase reporter constructs: minP-Luc2P-putative CERK-PPRE or minP-Luc2P-D putative CERK-PPRE, and pRL-SV40 con-trol vector Transfected cells were treated for 24 h in Phenol Red-free Dulbecco’s modified Eagle’s medium containing with 10% charcoal stripped fetal bovine serum and 1 l M LD Results were normalized with Renilla luciferase activity to correct for variability in transfection efficiency Values represent the means ± SD of three wells, and each experiment was repeated twice.
Trang 7transfected with an empty vector (minP-Luc2P-vector).
These findings suggest that putative CERK-PPRE is
the functional PPRE in mouse CERK gene
Enhanced cell survival associated with activation
of PPARb is diminished in keratinocytes from
CerK-null mice
The role of CerK in cell survival associated with PPARb
activation was further examined using primary
kerati-nocytes isolated from CerK-null mice [26] These mice
lack exon 6 of the mouse CERK gene, which encodes a
region of the protein that is known to be essential for
kinase activity and contains a diacylglycerol kinase-like
catalytic domain [2] As shown in Fig 5A, cell death
and limited cell growth following serum starvation
stress for 24 h were apparent in primary keratinocytes
isolated from wild-type mice In wild-type cells treated
with 1 lm LD under serum starvation conditions, cell
death was inhibited, similar to results observed in SP1
cells (Fig 2A) However, in primary keratinocytes
iso-lated from CerK-null mice, the inhibitory effect of LD
on cell death induced by serum starvation stress was
diminished (Fig 5B) To confirm this finding, serum
deprivation-induced cell death was investigated by flow
cytometry analysis with annexin V and PI staining A
significant increase in the percentages of annexin V⁄
PI-positive cells was observed in cultures of wild-type
keratinocytes undergoing serum starvation stress for
24 h, but, as expected, treatment with 1 lm LD
suppressed this number (Fig 5C) In CERK-KO
keratinocyte cultures, however, the inhibitory effect of
LD was once again diminished (Fig 5D) These findings indicate that CerK is necessary for the enhanced cell survival associated with activation of PPARb Taken together, the data demonstrate that upregulation of CERK by PPARb suppresses Cer accumulation induced
by serum starvation stress, which results in cell survival and inhibition of cell death in mouse keratinocytes
Discussion
The study reported here revealed that in mouse kerati-nocytes upregulation of CERK through activation of PPARb results in decreased intracellular Cer levels and increased cell survival with less cell death (Fig 6) There have been previous reports that PPARb activa-tion improved skin wound healing by enhanced kerati-nocyte survival⁄ anti-apoptosis [10,15] In this study, the effects on cell survival of PPARb were diminished
in CERK-KO keratinocytes (Fig 5), suggesting the biological importance of CerK in PPARb functions This study provides the first evidence for the necessity
of CerK in mouse keratinocyte survival associated with activation of PPARb
Cer is known to have an important role in apoptosis and cell-cycle arrest induced by various stressors Intracellular Cer levels are adjusted by several sphingo-lipid production pathways, such as de novo ceramide synthesis by serine palmitoyltransferase and cleavage
of sphingomyelin by sphingomyelinase, as well as by Cer metabolism pathways, including conversion to
CERK-KO keratinocytes
P < 0.005
300
200
100
0
300
200
100
0 Stress
LD
Stress (−)
Stress (+ )
LD (−) LD( +)
Stress (−)
Stress (+ )
LD (−) LD( +)
CERK-KO keratinocytes
P > 0.08
C D
Stress
LD
c d
c
Apoptotic cells (% control)
d
Fig 5 Enhanced cell survival associated
with activation of PPARb is diminished in
keratinocytes from CerK-null mice (A–D)
Mouse primary keratinocytes isolated from
wild-type mice (A, C) or CerK-null mice
(B, D) were grown to 80% confluence then
grown for 24 h under serum starvation
[Stress (+)] or control [Stress ( ))] conditions,
in the presence [LD (+)] or absence [LD ( ))]
of 1 l M LD (A, B) Cells were observed
under a phase-contrast microscope (C, D)
Cells were stained with annexin V ⁄
fluores-cein isothiocyanate and PI, and analyzed by
flow cytometry Results shown are relative
to untreated, unstressed cells, are the
means ± SD of three wells, and each
experiment was repeated three times.
Trang 8glucosylceramide by glucosylceramide synthase,
degra-dation to sphingosine and conversion to C1P by CerK
In some cancer cells, decreases in Cer levels by
ampli-fied activation of glucosylceramide synthase result in
drug resistance [18,19] Mitra et al [5] reported that
siRNA-induced downregulation of CerK reduced cell
proliferation and promoted apoptosis Previous studies
have suggested that metabolism of eicosanoids is
regu-lated by feedback pathways through PPARs Recently,
Xu et al [27] reported a positive feedback loop
between PPARb and prostaglandin E2 through which
PPARb promotes COX-2 expression and
prostaglan-din E2 synthesis, and subsequent activation of cystolic
phospholipase A2a, which is responsible for
arachi-donic acid release, through mitogen-activated protein
kinase and phosphatidylinositol 3-kinase-mediated
phosphorylation This pathway results in
cholangiocar-cinoma cell growth CerK has also been reported to
act as an upstream modulator of cystolic
phospholi-pase A2a in the inflammatory response [28,29] CerK
may also be implicated in the regulation of PPARb
activation through inflammatory factors such as cystolic
phospholipase A2a
In this study, the skin of CerK-null mice appeared
normal under specific pathogen-free conditions (data
not shown), yet examinations of cultured CERK-KO
keratinocytes showed a diminished effect on cell
sur-vival by activation of PPARb compared with that in
wild-type keratinocytes (Fig 5) PPARb expression has
been reported to be undetectable in the epidermal tissue
of adult mice; however, it is apparently upregulated
in various stress conditions, such as skin wound
healing, that result in enhanced keratinocyte
prolifera-tion [10,15] Reportedly, CerK is also expressed highly
at embryonic day 7 but decreases rapidly thereafter [2]
In the Arabidopsis plant, expression of CERK mRNA
is induced after infection with a bacterial pathogen [4]
Considering all this information, it appears that CerK
function may be upregulated under stress conditions, such as those that induce PPARb expression Future studies will be required into the role of CerK using an animal model
In conclusion, we have shown that PPARb-mediated upregulation of CerK gene expression is necessary for keratinocyte survival against serum starvation-induced apoptosis The interaction between CerK and PPARb may play an important role in regulating epidermal homeostasis in stress environments
Materials and methods
Materials Dispase was obtained from Godo Shusei (Tokyo, Japan) Keratinocyte serum-free medium, epidermal growth factor, bovine pituitary extract, Phenol Red-free Dulbecco’s modi-fied Eagle’s medium, and charcoal stripped fetal bovine serum were obtained from GIBCO BRL (Gaithersburg,
MD, USA) Minimum essential medium without calcium chloride was obtained from Cambrex (Walkersville, MD, USA) Penicillin–streptomycin, 0.125% trypsin–0.01% EDTA and LD were from Sigma (St Louis, MO, USA) Rabbit anti-PPARb IgG was from Santa Cruz Biotechnol-ogy Inc, (Santa Cruz, CA, USA) and the anti-acetyl-Histone H4 serum was from Upstate Biotechnology (Lake Placid, NY, USA) A MEBCYTO Apoptosis Kit was purchased from Medical and Biological Laboratories (Nagoya, Japan), and a Nuclear Extract Kit was from Active Motif (Carlsbad, CA, USA) The ChIP Assay Kit was also a product of Upstate Biotechnology Biotin 3¢-end DNA Labeling Kit and LightShift Chemilumines-cent EMSA Kit were purchased from Pierce (Rockford,
IL, USA) CulturPlate-96, White was purchased from Perkin-Elmer (Boston, MA, USA) and a lipofectamine
2000 reagent was from Invitrogen (Carlsbad, CA, USA) Dual-Luciferase reporter assay system was purchased from Promega (Madison, WI, USA)
Fig 6 Schematic model illustrating the role
of CerK in cell survival mediated by PPARb
in mouse keratinocytes Under injury stress, endogenous ligands activate PPARb Subse-quently, the activated PPARb promotes tran-scription of the CERK gene The conversion
of Cer to C1P by the produced CerK results
in decreased levels of intracellular Cer, and the inhibition of stress-induced cell death The interaction between CerK and PPARb may play an important role in regulating epi-dermal homeostasis in stress environments.
Trang 9Hairless mice (HR-1), 4-week-old males, were purchased
from Hoshino Experimental Animal Center (Saitama,
Japan) C57BL⁄ 6J mice were purchased from Clea Japan
(Tokyo, Japan) All animal experiments were performed in
accordance with the Guide for the Care and Use of
Labo-ratory Animals (Hokkaido University Graduate School of
Medicine, Japan) Animals were housed in plastic cages
with metal lids at a temperature of 22 ± 3C, with
50 ± 20% relative humidity, and were exposed daily to
12 h of light and 12 h of darkness
Skin barrier disruption by tape-stripping
In the dorsal skins of hairless mice, skin barrier disruption
was performed by stripping with adhesive tape (P.P.S
Nichiban, Tokyo, Japan: 2.5· 3.0 cm) repeatedly, five to
eight times An Evaporimeter AS-TW1 (Asahi Biomed, Co
Ltd, Yokohama, Japan) was used to measure TEWL, in
accordance with the ventilated chamber method [30]
Mea-surements were carried out at a temperature of 22 ± 3C,
with 50 ± 20% humidity, and were performed in triplicate
at each treatment skin spot
Harvest and culture of keratinocytes
Mouse keratinocytes were isolated from the epidermis of
hairless mice or newborn of C57BL⁄ 6J mice Briefly, the
epidermis was separated from the dermis following an
over-night incubation at 4C in 2.5 UÆmL)1 Dispase
Keratino-cytes isolated from the epidermis were harvested after
treatment with 0.125% trypsin and 0.01% EDTA at 37C
for 5 min Keratinocytes isolated from the epidermis of
new-born mice were incubated in a humidified atmosphere of 5%
CO2 in air at 37C with keratinocyte serum-free medium
containing 0.02 mm Ca2+, 5 ngÆmL)1 epidermal growth
factor and 50 lgÆmL)1bovine pituitary extract The isolated
keratinocytes were used at a subconfluent state (80%
conflu-ency) The mouse keratinocyte cell line SP1 cells, a kind gift
from S.H Yuspa (National Cancer Institute, Bethesda, MD,
USA) [31], was cultured in minimum essential medium
with-out calcium chloride supplemented with 0.02 mm Ca2+, 8%
chelexed fetal bovine serum [32] and antibiotics (100
unitsÆmL)1 penicillin and 0.1 mgÆmL)1 streptomycin) The
cultures were maintained in a humidified atmosphere of
5% CO2in air at 37C
Total RNA isolation
The total RNA from each sample was isolated using an
RNeasy Mini Kit (Qiagen, Chatsworth, CA, USA),
accord-ing to protocols provided by the manufacturer To remove
contaminating genomic DNA, the RNA samples were
treated with RNase-free DNase I (Qiagen) at room temper-ature for 30 min
RT-PCR RT-PCR of each mRNA was performed with Omniscript
RT Kit (Qiagen) following the manufacturer’s instructions using oligo(dT) primers and Taq DNA polymerase (Qiagen) with specific primers Sequences of the specific primers included, for PPARb forward 5¢-GCAGCCTCTTCCTCA ATGAC-3¢, for reverse 5¢-GTACTGGCTGTCAGGGTG GT-3¢; CERK forward 5¢-TCTGCAAGGACAGACCCT CT-3, reverse 5¢-CAAGTGCCATTTGCTGAGAA-3¢; and mitochondrial ribosomal protein L27 (Mrpl27) forward 5¢-GGGATAGTCCGCTACACGAA-3¢, reverse 5¢-ACCA TGTGGTTGTTGGGAA-3¢ The PCR condition of PPARb was as follows: 95C for 30 s, 60 C for 30 s and
72C for 60 s, and 35 cycles were used The PCR condition
of CERK was as follows: 95C for 30 s, 55 C for 30 s and 72C for 60 s, and 35 cycles were used The PCR condition of Mrpl27 was as follows: 95C for 30 s, 60 C for 30 s and 72C for 60 s, and 28 cycles were used PCR products were separated by electrophoresis on 2% agarose gels and visualized by ethidium bromide staining The relative intensity of the gel bands was measured using nih image software, and results were normalized to the mRNA level of Mrpl27, a housekeeping enzyme [33] We performed these experiments using samples from three animal preparations
Assessment of cell survival and cell death The rate of cell survival was determined using Cell Count-ing Kit-8 (Dojindo Laboratories, Kumamoto, Japan), fol-lowing the manufacturer’s instructions Briefly, the cells were seeded onto a six-well plate (1.0· 105cellsÆmL)1), incubated at 37C for 3 days, then, in the presence or absence of serum, cells were treated with or without LD (1 lm) After treatment, a solution of Cell Counting Kit-8,
in 1⁄ 10 volume of the culture medium, was added to each well, and the culture continued at 37C for 4 h Each well was then assessed at D450using an automatic enzyme-linked immunosorbent assay plate reader
Cell death was quantified by flow cytometry analysis Cells were seeded onto six-well plates, grown to 80% con-fluence, then treated with or without varying concentra-tions of LD in the presence or absence of serum After a predetermined time, the cells were trypsinized and washed twice with NaCl⁄ Pi The dead cells were stained with annexin V⁄ and PI using a MEBCYTO Apoptosis kit according to the manufacturer’s instructions flow cyto-metry analysis was carried out on a FACSort cell sorter (Becton Dickinson, Mountain View, CA, USA) using cell quest software
Trang 10Measurement of total intracellular Cer levels
Total cellular Cer levels were measured by the
diacylglyc-erol kinase method [22] Briefly, total cellular lipids were
extracted using the Bligh–Dyer protocol as previously
described [34] Extracts were suspended in micelle buffer
containing 7.5% n-b-d-octyl glucopyranoside and
19.4 mgÆmL)1 a-dioleoylphosphatidylglycerol, then mixed
with 0.1 unit of Escherichia coli diacylglycerol kinase and
1 lCi [32P]ATP[cP], and incubated for 1 h at 37C After
the reaction, lipids were separated by a solvent system of
chloroform⁄ methanol ⁄ 15 mm CaCl2 (7.5 : 4.4 : 1, v⁄ v ⁄ v)
on Silica Gel 60 TLC plates (Merck, Darmstadt, Germany)
Bands corresponding to C1P derived from intracellular Cer
were quantified using an Imaging Analyzer BAS2000 (Fuji
Film, Tokyo, Japan)
Quantitative real-time PCR
Quantification of CERK mRNA was performed using an
ABI Prism 7000 sequence detection system (Applied
Biosys-tems, Foster City, CA, USA) RNA samples were reverse
transcribed to synthesize first-strand cDNA using the
Omniscript RT kit (Qiagen), then analyzed by real-time
PCR using specific primers and the double-stranded DNA
dye SYBR Green I (Qiagen), according to protocols
pro-vided by the manufacturer Specific primer sequences used
included for CERK forward 5¢-GAGTGGCAAGTGACA
TGTGG-3¢ and for reverse 5¢-GCACTTCCGGATAAG
GATGA-3¢; those for Mrpl27 were for forward 5¢-CTGC
CCAAGGGTGCTGTGCTC-3¢ and for reverse 5¢-TTGTT
CTCACCAGACCCTTGAC-3¢ All reactions were run with
a hot-start preincubation step of 10 min at 95C, followed
by cycles of 15 s at 95C and 1 min at 60 C The amount
of template was quantified using the comparative cycle
threshold method as outlined in the manufacturer¢s
techni-cal bulletin Quantified CERK mRNA levels were
normal-ized to the Mrpl27 mRNA level for reporting
Quantification of in vitro CerK activity
CerK activity assays were performed as described by
Bajjalieh et al [3] Briefly, cells were washed three times
with ice-cold NaCl⁄ Pi, then lysed in a buffer containing
10 mm Hepes, 2 mm EGTA, 1 mm dithiothreitol, 40 mm
KCl and complete protease inhibitor mixture (Roche,
Basel, Switzerland) The enzyme reactions were performed
for 30 min at 30C in a reaction mixture containing
20 mm Hepes, 80 mm KCl, 3 mm CaCl2, 1 mm
cardio-lipin, 1.5% b-octyl glucoside and 0.2 mm
diethylenetri-aminepentaacetic acid, with 40 mm Cer (C18:0, d18:1) as
a substrate After the reaction, lipids were extracted and
separated on Silica Gel 60 HPTLC plates (Merck,
Darmstadt, Germany) in chloroform⁄ acetone ⁄
metha-nol⁄ acetic acid ⁄ water (10 : 4 : 3 : 2 : 1, v ⁄ v ⁄ v ⁄ v ⁄ v) as the
solvent system Quantification of bands was carried out using the Imaging Analyzer BAS2000
ChIP assays PPAR forms a heterodimer with RXR, and binds to PPRE sequences of the direct repeat-1 (DR-1) type (a repeat sepa-rated by one nucleotide) on DNA Using the nubiscan program [25], putative PPRE elements were identified within the first intron of the mouse CERK gene (putative CERK-PPRE) ChIP was performed using a ChIP Assay Kit (Upstate Biotechnology), according to protocols pro-vided by the manufacturer, with some modifications Briefly, cells were seeded onto six-well plates, grown to 80% confluence, and then treated with or without LD (1 lm) in serum starvation medium for 24 h To cross-link the DNA, cells were fixed with 1% formaldehyde at 37C for 15 min, then sonicated to fragments ranging in size from 200 to 500 bp ChIP was carried out using PPARb antibodies and acetylated histone H4-specific antibodies, with normal rabbit IgG used as a negative control Reverse cross-linking of DNA fragments was achieved at 65C for
6 h After phenol⁄ chloroform treatment of purified DNA, the DNA was amplified by PCR using primers, for putative CERK-PPRE forward 5¢-GTAGGCATGAGAACGGGA AG-3 and for reverse 5¢-GGGGGTAAGAGGAGGAGA AA-3¢ and for CERK-negative forward 5¢-CCGCAAG AGGCTTTATTGTC-3 and reverse 5¢-TATGCCAAGGA CACGGAGAT-3¢, as a negative control PCR primer The condition for PCR amplification was as follows: 95C for
30 s, 60C for 30 s and 72 C for 60 s; 32 cycles were used, depending on the abundance of DNA
Electrophoretic mobility shift assays PPRE studies were performed using biotin-labeled oligonu-cleotides and nuclear extracts from SP1 cells The nucleo-tide sequences of putative CERK-PPRE, including the sense 5¢-CTCTCCAGGCCACAGGCCAGAGCGG-3¢ and anti-sense 5¢-GAGAGGTCCGGTGTCCGGTCTCGCC-3¢ sequences, were biotin-labeled using a Biotin 3¢-end DNA Labeling Kit (Pierce) Nuclear extracts from SP1 cells were prepared using a Nuclear Extract Kit (Active Motif) EMSA was performed using a LightShift Chemiluminescent EMSA Kit (Pierce), following the manufacturer’s protocols, with some modifications Briefly, nuclear extracts (5 lg) from SP1 cells and 20 fmol biotin-labeled putative CERK-PPRE, alone or with 4 pmol unlabeled probe oligonucleotides (putative CERK-PPRE, PPRE-Wild, sense 5¢-CAAAACTAGGTCAAAGGTCA-3¢ and anti-sense 5¢-GTTTTGATCCAGTTTCCAGT-3¢; or PPRE-Mutant sense 5¢-CAAAACTAGCACAAAGCACA -3¢ and anti-sense 5¢-GTTTTGATCGTGTTTCGTGT-3¢) [35], were incubated in a reaction mixture at room temperature for 20 min The mixtures were then separated by electrophoresis