Sharmila Oncogene

Oncogene (2006) 25, 3335–3345
& 2006 Nature Publishing Group All rights reserved 0950-9232/06 $30.00
www.nature.com/onc

ORIGINAL ARTICLE

MKP1/CL100 controls tumor growth and sensitivity to cisplatin
in non-small-cell lung cancer

S Chattopadhyay1,4, R Machado-Pinilla1,4, C Manguan-Garcı´ a1, C Belda-Iniesta2, C Moratilla1,
P Cejas2, JA Fresno-Vara2, J de Castro-Carpeno2, E Casado2, M Nistal3, M Gonzalez-Baron2
˜ ´
and R Perona1
1
´dicas, C/Arturo Duperier, Madrid, Spain; 2Servicio de
Translational Oncology Unit CSIC/UAM, Instituto de Investigaciones Biome
´
Oncologıa Me 3
´dica, Hospital La Paz, Madrid, Spain and Servicio de Anatomıa Patolo
´ ´gica Hospital La Paz, Madrid, Spain

Non-small-cell lung cancer (NSCLC) represents the most Keywords: non-small-cell lung cancer; MKP1; CL100;
frequent and therapy-refractive sub-class of lung cancer. siRNA; Jun kinase; cisplatin
Improving apoptosis induction in NSCLC represents a
logical way forward in treating this tumor. Cisplatin, a
commonly used therapeutic agent in NSCLC, induces
activation of N-terminal-c-Jun kinase (JNK) that, in turn, Introduction
mediates induction of apoptosis. In analysing surgical
tissue samples of NSCLC, we found that expression of Lung cancer is the leading cause of cancer-related death
MKP1/CL100, a negative regulator of JNK, showed a in Western countries. This invariably results, seconda-
strong nuclear staining for tumor cells, whereas, in normal rily, from being diagnosed at a highly disseminated stage
bronchial epithelia, MKP1 was localized in the cytoplasm when therapeutic options are rarely curative. Currently,
as well as in nuclei. In the NSCLC-derived cell lines H- most therapies for disseminated lung cancer are centered
460 and H-23, we found that MKP1 was constitutively on chemotherapy with drugs such as cisplatin, docetax-
expressed. Expressing a small-interfering RNA (siRNA) el, vinorelbine and gemcitabine. Drug resistance and the
vector for MKP1 in H-460 cells resulted in a more patient’s poor performance status (apart from the late
efficient activation by cisplatin of JNK and p38 than in the stage of diagnosis) are jointly responsible for a median
parental cells, and this correlated with a 10-fold increase survival time of approximately 1 year from the time of
in sensitivity to cisplatin. A similar response was also diagnosis. This low survival rate reflects, in part, the
observed in H-460 and H-23 cells when treated with the ability of tumor cells to maintain survival pathways
MKP1 expression inhibitor RO-31-8220. Moreover, during malignancy progression (Vivanco and Sawyers,
expression of a siRNA-MKP2, an MKP1-related phos- ´ ´
2002; Perona and Sanchez-Perez, 2004) despite the
phatase, had no effect on H-460 cell viability response to therapy.
cisplatin. Tumors induced by H-460 cells expressing Conversely, chemotherapy agents activate signaling
MKP1 siRNA grew slower in nuÀ/nuÀ mice and showed pathways that can be induced in response to mitogen
more susceptibility to cisplatin than parental cells, and stimulation and which, frequently, are deregulated in the
resulted in an impaired growth of the tumor in mice. On tumor as part of the progression–transformation in the
the other hand, overexpression of MKP1 in the H-1299 cell. As a consequence of these processes, the effective-
NSCLC-derived cell line resulted in further resistance to ness of chemotherapy is a balance between pro- and
cisplatin. Overall, the results showed that inhibition of antiapoptotic signals elicited by the treatment and those
MKP1 expression contributes to a slow down in cell that are upregulated in the tumor. As mentioned above,
growth in mice and an increase of cisplatin-induced cell chemotherapy administered to non-small-cell lung
death in NSCLC. As such, MKP1 can be an attractive cancer (NSCLC) patients with non-resectable tumor at
target in sensitizing cells to cisplatin to increase the stages III and IV is, usually, a platinum-based combina-
effectiveness of the drug in treating NSCLC. tion regimen, in which the platinum compound is
Oncogene (2006) 25, 3335–3345. doi:10.1038/sj.onc.1209364; cisplatin. The drug (cis-diaminedichloroplatinum;
published online 6 February 2006 c-DDP, cisplatin) is a DNA-reactive agent that induces
lesions in DNA that include intra-strand 1,2-d(GpG),
1,2-d(ApG) and 1,3-d(GpNpG) and inter-strand cross-
linking (Eastman, 1985). Like many anticancer agents,
cisplatin induces a sustained activation of the mitogen-
´
Correspondence: Dr R Perona, Instituto de Investigaciones Biomedi- activated protein (MAP) kinases, N-terminal-c-Jun
cas, C/Arturo Duperier, 4, Madrid 28029, Spain. ´ ´
kinase (JNK) and p38 (Sanchez-Perez et al., 2000).
E-mail: RPerona@iib.uam.es
4
These authors contributed equally to this work. Each MAP kinase sub-family is activated by a specific
Received 11 April 2005; revised 29 November 2005; accepted 29 upstream MAP kinase kinase, which phosphorylates
November 2005; published online 6 February 2006 threonine as well as tyrosine residues within a conserved

MKP1 controls sensitivity to cisplatin in NSCLC
S Chattopadhyay et al
3336
´
T–X–Y motif (Hibi et al., 1993; Derijard et al., 1994; differs from normal epithelium. By using the NSCLC-
Kyriakis et al., 1994). These residues are dephosphory- derived cell line H-460, which showed increased levels of
lated by dual-specific protein phosphatases, resulting in MKP1, we inhibited MKP1 expression using small
the inactivation of MAP kinases (Keyse, 1995). The interfering RNA (siRNA) technology. We found that
activation of the MAP kinase family members leads to the resulting cells are 10-fold more sensitive to cisplatin,
the phosphorylation of numerous cellular effectors, and that tumors induced by these cells, as well, were
including protein kinases such as MAPKAPK1, MAP- more sensitive to cisplatin than tumors from control
KAPK2/3 and Mnk1/2 and transcription factors such as cells, indicating that MKP1 plays a critical role in
c-jun, ATF-2, MEF2c and CHOP, which, ultimately, negative regulation of cisplatin-induced cell death in
are responsible for the fate of the cell (Alpert et al., NSCLC.
1999). The activation of JNK by cisplatin takes place via
the MEKK1/SEK1 cascade and is directly related to cell
´ ´
death (Sanchez-Perez and Perona, 1999). Dephosphor-
Results
ylation and inactivation of JNK by the dual-specific
phosphatases CL100 and hVH5 result in protection Expression of MKP1 in NSCLC biopsies
´
against cisplatin-induced apoptosis (Sanchez-Perez ´
To check the frequency of expression of the MKP1
et al., 2000). Further, both phosphatases are able to
phosphatase in NSCLC, we evaluated expression in
inhibit transcriptional activation of c-Jun, the principal
sections of tumor samples of 32 patients of NSCLC
physiological substrate of JNK (Kyriakis et al., 1994).
(Table 1). Of the total, nearly 60% correspond to
´ ´
We have shown previously (Sanchez-Perez and Perona,
squamous cell carcinomas, 25% to adenocarcinomas
1999) that c-jun is necessary for the induction of
and the rest to undifferentiated large-cell carcinomas.
apoptosis in response to cisplatin in a cell line derived
Briefly, half of the samples were grouped into stage IB,
from c-jun knockout mice and which appears to be more
and the rest into stage IIB. We found that in tumor
resistant than normal cells to cisplatin-induced cell
tissues, staining was mainly nuclear in nearly 80% of
death. This effect is specific to c-jun as transfection of c- cases (Figure 1 and Table 2). By contrast, stain was
jun into the c-junÀ/À cell line restores its endogenous
localized both in the cytoplasm and nuclei in normal
activity, and which results in a phenotype similar to that
of parental (wild-type (wt)) cells. The role of c-jun in
apoptosis has been described in other cell lines, such as
PC12 cells, which undergo apoptosis following NGF Table 1 Clinicopathological characteristics of the patients’ tumors
withdrawal in a c-jun-dependent manner (Ham et al., Characteristic N
1995). In agreement with these results, constitutive
Patients 32
expression of c-jun in NIH3T3 cells induces apoptosis
upon serum deprivation (Ham et al., 1995; Bossy et al., Gender
1997). Male 30
MKP1/CL100 is an immediate-early gene whose Female 2
expression is regulated by mitogenic, inflammatory
Age (years)
and DNA-damage stimuli (Beltman et al., 1996; Mean (range) 62 (42–79)
´ ´
Sanchez-Perez et al., 1998; Li et al., 2001). Although
MKP1 levels in normal cells are low, increased levels of Histological sub-type
MKP1 have been found in human ovarian carcinoma, Adenocarcinoma 8
Squamous 19
breast and prostate cancer (Srikanth et al., 1999; Undifferentiated large cell carcinoma 5
Denkert et al., 2002; Wang et al., 2003). In this report,
we have investigated the expression and function of Stage
MKP1 in the chemotherapy response to cisplatin in IA 2
NSCLC. In surgical samples from patients with IB 16
IIB 14
NSCLC, we observed that the expression in the tumor

Figure 1 Immunohistochemistry of patient tumor samples showing MKP1 localization. Sections of 5 mm thickness were cut from
formalin-fixed, paraffin-embedded tissue blocks of surgical tissues from the patients. The primary antibody incubated with the tissue
slides and visualized with peroxidase-based EnVisionTM kit and enzymatic activity revealed (a) moderately differentiated squamous
lung carcinoma, (b) Poorly differentiated squamous lung carcinoma and (c) Bronchial non-tumor epithelium.

Oncogene

3337
Table 2 Sub-cellular localization of MKP1 in tumor tissues a 0 0.5 1 3 6 8 9 10 12 15 22 24 36 48 Time in hours
MKP1 in NSCLC In nucleus In cytosol In cytosol or nucleus pJNK

No./total % No./total % No./total %
pP38
Normal epithelial 10/24 42* 3/24 12* 11/24 46*
Tumor cells 26/32 81* 2/32 6* 4/32 13* MKP1/CL100

pERK
*Statistical significance Po0.01.
ß-tubulin

bronchial epithelium surrounding the tumor, showing b 0 0.5 1 3 6 9 Time in hours

weaker nuclear staining than in tumor tissue (Figure 1). pJNK
All values were statistically significant. The results
showed that this distribution of MKP1/CL100 expres- pP38
sion in the cytoplasm and nuclei was homogeneously
Figure 2 Kinetics of activation of JNK, p38 and ERK in H-460
distributed among the different subgroups of patients. cells treated with cisplatin. (a) H-460 cells were seeded, then
These results are in agreement with those observed by incubated in serum-depleted medium for 16 h followed by the
Vicent et al. (2004), who also detected higher levels of addition of 5 mg/ml cisplatin for the indicated time periods. Protein
MKP1 protein in NSCLC-derived cell lines. Our results extracts of cells at each time point were obtained, and 20 mg of total
protein was subjected to Western blotting and the membrane
suggested that nuclear expression of MKP1 could be an hybridized with antibodies against p-JNK, p-P38, MKP1 and
early event in the disease progression of NSCLC. pERK as indicated in the figure. The loading control of total
protein was checked using an antibody against b-tubulin. (b) H-460
cells were seeded then incubated in serum-depleted medium for
JNK and p38 signaling pathways are activated by 16 h, irradiated with 40 J/m2 with UVC and then lysed at the
indicated time points. Protein extracts of cells at each time point
cisplatin in H-460 cells were obtained and 20 mg of total protein was subjected to Western
In order to explore the physiological role of the high blotting, and the membrane hybridized with antibodies against p-
levels of MKP1 observed in the tumor samples, we JNK and p38. The experiments corresponding to panels a and b
commenced this investigation by analysing its possible were repeated three times with similar results.
role in chemotherapy responses. We decided to study the
signaling pathways induced by cisplatin in H-460 cells,
as we have previously identified MKP1 as an inhibitor Inhibition of MKP1/CL100 expression by siRNA
of the cell death response to cisplatin. Cells were treated sensitizes H-460 cells to cisplatin treatment
with cisplatin, and at different times the kinetic ´ ´
We had reported previously (Sanchez-Perez et al., 2000)
parameters for JNK and p38 activation were studied. that enforced expression of MKP1/CL100 in 293T
JNK was activated in H-460 cells with a late-onset cells protects against cisplatin-induced cell death. With
kinetic, and the activation persisted for at least up to this in mind, we checked whether the inhibition
24 h (Figure 2a), decreasing by 36 h. The kinetic of MKP1 expression could improve cisplatin cytotoxi-
parameters for p38 activation were very similar to that city responses in H-460 cells. We transfected H-460
of JNK. We had previously found, in different cell cells with a siRNA expression vector for MKP1 and
systems, that the kinetic of activation of JNK was pSRMKP1RNAi, and selected a stable expression using
modulated by the dual phosphatase MKP1/CL100 puromycin resistance (H-460MKP1siRNA cells). Con-
´ ´
(Sanchez-Perez et al., 2000). We subsequently investi- trol cells were obtained by transfection with the
gated the expression levels of this protein in H-460 cells pSuperRetro empty vector. We did not isolate indivi-
following the addition of cisplatin. We found that, as dual clones of siRNA-expressing cells so as to pre-empt
expected, H-460 cells constitutively expressed MKP1, clone-related variations in the results. We observed that
and the expression decreased to undetectable levels expression of this construct was able to attenuate MKP1
within 24 h (Figure 2a). This correlated well with the mRNA levels, monitored via reverse transcriptase (RT)–
kinetics of activation of JNK and p38, which increase in PCR (Figure 3a). When we assessed MKP1 protein
the time the MKP1 levels are downmodulated. ERK levels by Western blotting (Figure 3b), the results were
activation upon cisplatin treatment varies from one cell very close to the ones obtained for the MKP1 mRNA
type to another (Persons et al., 1999; Woessmann et al., levels. Finally, we studied MKP1 expression in H-
2002). We found that in H-460 cells ERK1/2 was 460MKP1siRNA cells by immunofluorescence
activated at a very late time (24–36 h), suggesting that (Figure 3c) and we observed a complete inhibition in
this process is not dependent on the decrease in MKP1 MKP1 nuclear expression in a major proportion of
levels. On the other hand, in order to verify that the late H-460MKP1siRNA cells relative to control cells. As
activation of JNK is specific to cisplatin activation, we we had used a pool of stable clones, we were unable
irradiated H-460 cells with 40 J/m2 of UVC and detected to obtain a complete inhibition in MKP1 mRNA
an early and transient activation of JNK and p38 or protein levels, but we were able to find cells
´
(Figure 2b), as described previously (Sanchez-Perez ´ using immunofluorescence that showed complete abro-
et al., 1998). gation of MKP1 expression. We also obtained an
Oncogene

3338
a b

o
NA

NA
o

etr
etr

siR

siR
erR
erR

P1

P1
up
up

MK

MK
pS
pS
MKP1 MKP1

GAPDH ß-tubulin

c

H460SuperRetro

H460MKP1siRNA

Figure 3 Levels of mRNA and protein in H-460 control and H-460 cells transfected with siRNA of MKP1. (a) Comparison of the
levels of mRNA by RT–PCR in H-460 control (pSuperRetro) and H-460 cells transfected with siRNA against MKP1 (H-
460MKP1siRNA). Total RNA from both H-460pSuperRetro and H-460MKP1siRNA cells was extracted and cDNA prepared and
PCR amplified as indicated in Materials and methods. The quantity of cDNA in each preparation was estimated by PCR amplification
of the GAPDH gene. (b) Comparison of the expression of protein in H-460 pSuperRetro and H-460MKP1siRNA cells. Total protein
(20 mg) was separated by Western blotting and the resulting membrane probed with specific antibody against MKP1. Antibody against
b-tubulin was used to control for the amount of protein loaded. (c) Immunofluorescence studies in H-460pSuperRetro and H-
460MKP1siRNA cells. Cells were grown on coverslips and subsequently fixed and incubated with anti-antibody and with a secondary
antibody conjugated to FITC. The immunofluorescence was visualized with a fluorescence microscope at Â 100 magnification.
Different examples of H-460pSuperRetro and H-460MKP1siRNA cells are shown.

H-460-derived cell line expressing a GFPsiRNA, and no H460 pSuperRetro H460 MKP1siRNA
difference in MKP1 expression was found when 0 30 1 3 6 9 12 14 20 Time (hours) 0 30 1 3 6 9 12 14 20
compared to wtH-460 cells (data not shown). From all
these findings, we concluded that the siRNA construct PARP
designed for MKP1 was able to knock-down the MKP1 p-JNK
expression successfully.
We proceeded, subsequently, to investigate the p-p38
activation of JNK as well as p38 kinase and PARP ß-tubulin
degradation (as a test for apoptosis) in H-460MKP1siR-
NA cells. In H-460SuperRetro cells, caspase activation Figure 4 Changes in activation of JNK, p38 and caspase-3
following cisplatin addition in the H-460MKP1siRNA cells. H-
takes place around 20 h after cisplatin treatment, 460pSuperRetro and H-460MKP1siRNA cells were seeded then
whereas in H-460MKP1siRNA, activation of caspase incubated in serum-depleted medium for 16 h followed by the
occurs already at 14 h (Figure 4). Also, in these cells, addition of 5 mg/ml cisplatin for the indicated time periods. Protein
activation of JNK starts at earlier times than in control extracts of cells at each time point were obtained and 20 mg of total
cells, in fact, basal levels of JNK activation could be protein was subjected to Western blotting and the membrane
hybridized with antibodies against p-JNK, p-P38 and PARP as
observed in H-460MKP1siRNA untreated cells. Similar indicated in the figure. The loading control of total protein was
results were observed for p38 activation. These results checked using an antibody against b-tubulin. For PARP, the full-
are consistent with those obtained in 293T cells length protein corresponds to the upper band and the proteolytic
expressing a dominant-negative form of MKP1, which fragment (85 kD, lower band) is indicated with an arrow. The
experiment was repeated three times with similar results.
were more sensitive for SAPK activation by cisplatin
´
than the corresponding control cells (Sanchez-Perez ´
et al., 2000). When we compared cell viability in
response to cisplatin of H-460MKP1siRNA with their by a clonogenic assay, of surviving cells to cisplatin in
respective control cells, we found that H-460MKP1siR- H-460MKP1siRNA and control cells, as well
NA cells were one order of magnitude more sensitive to (Figure 5b). We obtained a decrease of 40% in cell
cisplatin than their control H-460 cells (Figure 5a). This survival in H-460MKP1siRNA cells treated with cispla-
observation indicated that the manipulation of MKP1 tin, relative to the controls. This was in agreement with
protein levels by RNA interference could, indeed, the cell viability estimations, obtained above.
improve cisplatin responses in cells that overexpressed As we observed such a difference in sensitivity to
this protein, in agreement with the results obtained for cisplatin of H-460MKP1siRNA cells, we checked if these
PARP activation (Figure 4). We monitored cell viability, differences were due to alteration in the growth kinetics of
Oncogene

3339
a H460pSR b H460pSR c H460pSR
H460MKP1siRNA H460MKP1siRNA H460MKP1siRNA
125
125 4500
4000

Growth Percentage
Percentage of survival

100
100 3500

% clone formation
3000
75 2500
75
2000
1500
50 50 1000
500
25 25 0
1 3 5 7 9 11
Number of days
0 0
0 0.1 1 10
Dose in g/ml cell number: 10000
Figure 5 Effect of expression of siRNA on survival of H-460 cells. (a) Survival curves of H-460pSuperRetro (pRS) and H-
460MKP1siRNA cells. Both cell lines were seeded in 24-well plates and cisplatin was added the following day at concentrations ranging
from 0 to 10 mg/ml. After 48 h, the cells were fixed with glutaraldehyde, stained with crystal violet and the percentage staining
calculated according to standard procedures. Data represent the means of two experiments performed in quadruplicate. (b) Clonogenic
assays of H-460pSuperRetro (pRS) and H-460MKP1siRNA cells. Both cell lines were seeded in triplicate and treated for 48 h with
5 mg/ml cisplatin. Cells were then allowed to grow in cisplatin-free medium until they formed colonies. Relative colony formation of
cells treated with cisplatin compared to untreated cells was determined by crystal violet staining. Data represent the means of two
experiments performed in triplicate. (c) Growth curves of H-460pSuperRetro (pRS) and H-460MKP1siRNA cells. One thousand cells
of each cell line were seeded in triplicate and allowed to grow in 12-well plates. Growing cells were fixed on alternate days, stained with
crystal violet and the percentage staining calculated according to standard procedures. Data represent the means of two experiments
performed in triplicate.

these cells due to the RNA interference vector expression. We have previously published that enforced expression
Both, control- and siRNA-transfected H-460 cells grew at of MKP1 was able to induce survival capacity to cisplatin
comparable rates when cultured in vitro (Figure 5c). ´ ´
treatment (Sanchez-Perez et al., 2000). To prove this in
In order to check if the sensitization of H- NSCLC, we constitutively expressed MKP1 in the H1299
460MKP1siRNA cells was an event linked to this cell cell line (that does not express either MKP1 or MKP2)
line and also if an alternative method for manipulating either alone or in combination with the pSRMKP1RNAi
MKP1 levels would have the same effect, we used both vector. As expected, H1299MKP1 cells expressed the
H-460 and H-23 cell lines, the latter also expressing high myc-tagged MKP1 protein and the levels were strongly
basal levels of MKP1 (Vicent et al., 2004). Treatment of reduced in H1299MKP1 cells also transfected with the
both cell lines with the inhibitor of MKP1 expression pSRMKP1RNAi vector (H1299MKP1–MKP1siRNA)
RO-31-8220 (Beltman et al., 1996) resulted, with a (Figure 8a). We then studied the survival capacity to
different timing, in a reduction in MKP1 expression to cisplatin treatment and found that H1299 cells expressing
almost undetectable levels (Figure 6a). H-460 cells MKP1 were more resistant to cisplatin than parental cells
treated for 1 h with RO-31-8220 showed an earlier (Figure 8b). Furthermore, H1299 cells expressing both
(from 9 to 3 h) and more intense activation of JNK and MKP1 and MKP1siRNA behave like the parental cells,
also p38 than untreated cells (Figure 6b). In H-23, cells indicating again that the expression of MKP1 protects
treated for 6 h with RO-31-8220 also showed an earlier NSCLC cells against cisplatin treatment.
activation of both JNK and p38 kinases (Figure 6c). As
R031-8220 was also described as a PKC inhibitor, we Knock down of MKP2 expression did not influence cell
assayed the effect of another PKC inhibitor, GO6976, viability response to cisplatin
on JNK activation by cisplatin and observed no effect MKP1/CL100 is a member of a dual phosphatase family
on JNK activation, in agreement with the results of of proteins. Among the family members, MKP2 showed
other groups (data not shown) (Standaert et al., 1999). a 60% sequence homology to MKP1, and also similar
The treatment of both H-460 and H-23 cells with RO- substrate specificity (Hirsch and Stork, 1997). To verify
31-8220 and cisplatin was able to sensitize these cells to that the results obtained in H-460MKP1siRNA cells
cisplatin when compared to control cells (Figure 7), were specific for the downregulation of MKP1 expres-
indicating that inhibition of MKP1 expression, by either sion, we generated an H1299 cell line overexpressing
RNAi or by chemical inhibitors, was a valid strategy to MKP2. H1299MKP2 cells expressed the myc-tagged
sensitize NSCLC cells to cisplatin. These results are also MKP2 protein (Figure 9a), and when we studied the
supported by the observation that MKP1-deficient survival capacity to cisplatin, we observed almost no
MEFs are more sensitive to anysomicin-induced apop- differences with the control cells (Figure 9b). We also
tosis (Wu and Bennett, 2005), further emphasizing the obtained a siRNA expression vector for MKP2 and
importance of MKP1 in regulating stress-induced cell stably transfected it in H-460 cells to monitor the
death. decrease in MKP2 protein levels using Western blotting
Oncogene

3340
a H23 RO 10 M H460
0 1 3 6 9 12 24 h 0 1 3 6 9 12 24 h
MKP1 MKP1

ß-Tubulin ß-Tubulin

b H460 H460 Ro 10 M 1 hour
0 30 1 3 6 9 12 24 h 0 30 1 3 6 9 12 24 h
p-JNK p-JNK

p-P38 p-P38
JNK JNK

c H23 H23 Ro 10 M 6 hours
0 30 1 3 6 9 12 h 0 30 1 3 6 9 12 h
p-JNK p-JNK

p-P38 p-P38
JNK JNK
Figure 6 Changes in activation of JNK, p38 and MKP1 expression following treatment of H-23 and H-460 cells with R0-31-8220. (a)
Changes in MKP1 protein levels after treatment with RO-31-8220. Both H-23 and H-460 cells were treated with RO-31–8220 at 10 mM
during the indicated times. Total protein (20 mg) was used for Western blot analysis of MKP1 expression. b-Tubulin was used for
loading control of c-Jun. (b) Changes in JNK and p38 activities in H-460 cells treated with R0-331-8220 and cisplatin. H-460 cells were
treated with cisplatin 5 mg/ml and when indicated, pretreated during 1 h with 10 mM RO-31-8220. Total protein (20 mg) was used for
Western blot analysis of pJNK and pP38. JNK1 was used as a loading control. (c) Changes in JNK and p38 activities in H-23 cells
treated with R0-331-8220 and cisplatin. H-460 cells were treated with cisplatin 5 mg/ml and, when indicated, pretreated during 6 h with
10 mM RO-31-8220. Total protein (20 mg) was used for Western blot analysis of pJNK and pP38. JNK1 was used as a loading control.

H460 H23 lack of expression of MKP1 on tumorigenicity as well as
response to cisplatin. Both cell lines were inoculated in


120
H460
H460 RO
120 H23
H23 RO
nude mice, and after 14 days when tumor development
100 100 was observed, cisplatin was administered at 5 mg/kg
80 80
60 60
body weight to half of each group of animals every other
40 40 day for a total of five administrations. Tumor measure-
20 20
0 0
ments were performed on the first day of cisplatin
0 0.01 0.1 1 10 0 0.01 0.1 1 10 injection, and the increase in volume size monitored
Dose in g/ml Dose in g/ml from the day of the injection. The results obtained
Figure 7 Effect of RO-31-8220 pretreatment on survival of H-460
showed clearly (Figure 11a and c) that, firstly, tumors
and H-23 cells on viability to cisplatin treatment. Survival curves of induced by H-460MKP1siRNA cells grow more slowly
H-460 and H-23 cells. Both cell lines were seeded in 24-well plates than H-460 control tumors. The expression of MKP1
and, when indicated, pretreated for 1 and 6 h, respectively, with was verified both in control cells-induced tumors and in
RO-31-8220, and cisplatin was added the following day at H-460MKP1siRNA tumors (Figure 11b), indicating the
concentrations ranging from 0 to 10 mg/ml. After 48 h, the cells
were fixed with glutaraldehyde, stained with crystal violet and the functionality of the SiRNA vector in cells injected in the
percentage staining calculated according to standard procedures. mice. In agreement with the results observed in tumor
Data represent the means of two experiments performed in growth, a slower growth of MKP1-deficient MEFs has
quadruplicate. also been reported (Wu and Bennett, 2005). Both cases
are not in complete agreement with our results obtained
(Figure 10a). No differences were observed in viability in vitro, which had shown both cell lines (H-460 and H-
response to cisplatin in H-460MKP2siRNA cells when 460MKP1siRNA) having a similar overall growth
compared with control cells treated with cisplatin kinetic. This suggests that in lung cancer, the tumor
(Figure 10b). These findings indicated that the results environmental conditions are less favorable for H-
obtained in H-460MKP1siRNA cells were specifically 460siRNA cells. Secondly, despite H-460MKP1siRNA
related to the inhibition in MKP1 expression by the tumors growing more slowly than H-460 tumors, they
siRNA vector, and that MKP1/CL100 is the relevant were more sensitive to cisplatin treatment than H-460-
enzyme controlling sensitivity to cisplatin in H-460 cells. induced control tumors, which supports the results
obtained in vitro. The ratio of tumor growth among H-
H-460MKP1siRNA-induced tumors grow slower and are 460 tumors when treated with cisplatin was 1.68,
more sensitive to cisplatin whereas it was 3.77 for H-460siRNA tumors. All these
We inoculated H-460 control and H-460MKP1siRNA data were statistically significant. This implies that these
cells in nuÀ/nuÀ mice and studied the influence of the tumors responded much more efficiently to cisplatin
Oncogene

3341
a a

P2
MK 1/
P1
P

1

MK
UP MK

P-

-
MK

G5

G5
pS SN-
ER

pS

pS
N-
er

N

X

XS
XS
up

pL
pS

pL
pL
MKP2
MKP-1

ß-Tubulin
ß-tubulin

b H1299PLXSN b pSG5
H1299MKP1 120
120 MKP
H1299MKP1-MKP1siRNA

100 100

80 80

% Viabilidad
60
60

40
40

20
20
0
0 0.01 0.1 1 10 100 0
Dose in g/ml
0 0.01 0.1 1 10 100
Figure 8 Effect of expression of MKP1 on survival of H1299 cells. Dosis g/ml
(a) Comparison of the expression of tagged-MKP1 protein in H1299
cells transfected with pLXSNMKP1 or empty vector (pLXSN) and Figure 9 Effect of expression of MKP2 on survival of H-460 cells.
pSuperRetro or pLXSNMKP1 and pSuperMKP1. Total protein (a) Comparison of the expression of tagged-MKP2 protein in H1299
(20 mg) was separated by Western blotting and the resulting membrane cells transfected with pSG5 or pSG5MKP2. Total protein (20 mg)
probed with specific antibody against c-myc. Antibody against b- was separated by Western blotting and the resulting membrane
tubulin was used to control for amount of protein loaded. (b) Survival probed with specific antibody against c-myc. Antibody against b-
curves of H1299 cells transfected with pLXSN, pLXSNMKP1 and tubulin was used to control for amount of protein loaded. (b)
pLXSNMKP1/pSuperMKP1. All the cell lines were seeded in 24-well Survival curves of H1299 cells transfected with pSG5 or
plates, and cisplatin was added the following day at concentrations pSG5MKP2. Both cell lines were seeded in 24-well plates and
ranging from 0 to 10 mg/ml. After 48 h, the cells were fixed with cisplatin was added the following day at concentrations ranging
glutaraldehyde, stained with crystal violet and the percentage staining from 0 to 10 mg/ml. After 48 h, the cells were fixed with
calculated according to standard procedures. Data represent the glutaraldehyde, stained with crystal violet and the percentage
means of two experiments performed in quadruplicate. staining calculated according to standard procedures. Data represent
the means of two experiments performed in quadruplicate.

than parental tumors. Our results support the sugges-
tion that inhibition of expression of MKP1 sensitizes the factor receptors, are also related to an increased survival
tumor to the action of the cisplatin, and that the high of tumor cells, when confronted with chemotherapy (Kim
levels expressed by H-460 cells can induce protection ´ ´
et al., 2003; Perona and Sanchez-Perez, 2004). Apart from
against cisplatin. the signaling pathways activated by oncogenic transfor-
mation in the tumor, it is important to know which other
molecules are involved in the response to the drugs used
Discussion in cancer treatment, and how such responses are modified
by the genetic background of the tumor cells.
Chemotherapy is one of the strategies most widely used in Cisplatin is included in most protocols for the
the treatment of cancer. Often, however, the tumors do treatment of advanced NSCLC, but, to date, the
not succumb to the drugs. This problem is especially patient’s response to this drug is, often, far from
important in the treatment of NSCLC. Lung cancer is a optimum. We considered that sensitizing the cells to
major cause of cancer mortality, and accounts for about the therapeutic agent would be a way of obtaining a
20% of all cancer deaths worldwide. Many genetic better response from the tumor to the drug, and also of
alterations are associated with development of this type combating any resistance to the drug by the tumor.
of tumor (Slebos and Rodenhuis, 1989; Levin et al., 1994; Constitutive expression of MKP1 in NSCLC has
Wiest et al., 1997). Some of them, such as the RAS already been described by other authors (Lim et al.,
mutations or overexpression, or mutations in growth 2003), finding an increase in mRNA levels of MKP1 in
Oncogene

3342
a a

NA
H460 SiRNA + CDDP

o
etr

siR
H460 + CDDP

rR

P2
TUMOR GROWTH H460 SiRNA -CDDP

pe

MK
Su
H460 - CDDP
3.000
αMKP2
Days of CDDP injection
2.500
ß-Tubulin

Tumor Volume (cm3)
2.000
b 125 H460pRScddp
H460Mri/cddp 1.500

100 1.000
Percentage survival

0.500
75

0.000
50 1 4 7 10 13 16 19
Days

25 b MKP1
ß-tubulin

tum er

tum A
0

60 iRN
or
tro sup

or
0 0.01 0.1 1 10

H4 P1s
Re 60P

Dose in g/ml
MK
H4

Figure 10 Levels of MKP2 protein and cell survival in H-460
control and H-460MKP2siRNA cells. (a) Comparison of the c
expression of MKP2 protein in H-460 control and H-460 cells Tumor cell line Relative tumor
transfected with siRNA against MKP2. Total protein (20 mg) of H-
460 control (H-460pSuperRetro) and H-460 cells transfected with growth at 19 days
siRNA against MKP2 (H-460 MKP2siRNA) was separated by
H460 -CDDP 9.8*
Western blotting and the resulting membrane probed with specific
antibody against MKP2. Antibody against b-tubulin was used for H460+CDDP 5.8*
protein loading control. (b) Survival curves of H-460 MKP2siRNA
and H-460pSuperRetro. Both cell lines were seeded in 24-well H460SiRNA-CDDP 6.8*
plates and cisplatin was administered the following day at H460SiRNA+CDDP 1.8*
concentrations ranging from 0 to 100 mg/ml. After 48 h, the cells
were fixed with glutaraldehyde, stained with crystal violet and the * Indicates statistical significance p<0.01
percentage staining calculated according to standard procedures.
Data represent the means of two experiments performed in
quadruplicate. Figure 11 Growth efficiency of H-460pSuperRetro and H-
460MKP1siRNA cells in nude mice. (a) Growth of tumor
xenografts in nude mice. H-460pSuperRetro and H-460MKP1siR-
NA cells were injected on the backs of nude mice. After 14 days
(lower arrow), tumor size was measured and the volume
NSCLC tumors compared to control tissue. A recent determined. Half of the inoculated mice of each type were treated
report (Vicent et al., 2004) also demonstrated elevated with cisplatin, and the rest were left untreated. Cisplatin was
levels of nuclear MKP1 expression in NSCLC biopsies injected 5 times at 5 mg/kg body weight on alternate days when
and derived cell lines. In agreement with this latter indicated (upper arrows). Tumor volume was recorded from the
day of cisplatin injection, and for the next 19 days. (b) MKP1
study, we found differences in sub-cellular distribution expression in tumors. One of the tumors induced by H-
of MKP1: mainly cytoplasmic in normal tissue and 460pSuperRetro and another induced by H-460MKP1siRNA were
nuclear in tumor tissue. As the preferred substrate of lysed. Total protein (20 mg) of H-460pSuperRetro and H-460
MKP1 is JNK and this is translocated to the nuclei MKP1siRNA tumors were separated by Western blotting and the
´ ´
following a stimulus (Sanchez-Perez et al., 1998), it resulting membrane probed with specific antibody against MKP1.
Antibody against b-tubulin was used for protein loading control.
appears reasonable to propose that a preferential (c) Relative growth of tumors induced by H-460pSuperRetro and
distribution of MKP1 to this sub-cellular compartment H-460MKP1siRNA cells. The tumor size for all the animal groups
would be to facilitate its access to the substrate. was expressed as a relation of the 19 days size vs the size at the time
Furthermore, we found that expression of MKP1 was of the first cisplatin injection (day 14). All the data obtained were
statistically significant.
detected from very early stages in tumor development
and maintained until later stages, indicating that it
should be advantageous for tumor progression.
Oncogene

3343
We studied the kinetic activation of the kinases prisingly, tumors induced by H-460MKP1siRNA cells
involved in cisplatin responses in a cell line derived showed a slower kinetic of growth than the ones induced
from NSCLC that constitutively expresses the MKP1 by control cells. This difference between in vitro and in
phosphatase. Although H-460 cells are sensitive to vivo growth of the cell lines can be related to the greater
cisplatin, as shown by the IC50 values, the kinetics for ability of H-460 cells to withstand environmental
JNK and p38 activation are very late onset (around 9 h). conditions than control cells, or perhaps this reflects a
As the kinetic for JNK activation in response to reduction in tumorigenicity of these cells. Supporting
cisplatin depends very much on the dual-specific this possibility is the report (Liao et al., 2003) that
phosphatase MKP1/CL100, we modulated MKP1 levels downregulation of MKP1 expression by antisense
in these cells by inhibiting expression of MKP1 by technology was able to reduce tumorogenicity in
siRNA. Using this approach, we found that the pancreatic cancer cells. Apart from the decrease in
sensitivity of H-460 cells to cisplatin increases by one growth, H-460MKP1siRNA tumors were more respon-
order of magnitude, which indicated that this approach sive to cisplatin than control cells. As a consequence of
was valid for cells that overexpressed MKP1. In such a inhibition of MKP1 expression, a greater than three-fold
case, the activation of the JNK and p38 pathways is reduction in the increase in volume of tumor (non- vs
increased relative to control cells and, therefore, cisplatin-treated transfected cells) is observed; the
cisplatin-induced cell death is also increased. difference in the control cells was only two-fold. Of
MKP2 is a dual-specific phosphatase that has 60% note was the observation that the volume of the
homology at the amino-acid level and similar substrate H-460siRNA tumors increases only two-fold after 20
specificity for MKP1. Also, it is overexpressed in H-460 days of treatment and emphasizes the retardation of
cells. We did not observe any difference with respect to growth in these tumors.
cell survival when MKP2 expression was knocked-down Given the poor response of NSCLC to conventional
by siRNA in H-460 cells, confirming, thereby, that the chemotherapy, the results in the present report are
effect of knocking-down the expression of MKP1 is very highly encouraging, as they support the idea that MKP1
specific. inhibition or inactivation is a good strategy both to
Ectopic expression of MKP1 is able to inhibit inhibit tumor growth and to sensitize cancer cells to
apoptosis induced by other types of stresses, such as conventional chemotherapy, and this results in an
UVC exposure (Liu et al., 1995). MKP1 expression has important decrease in tumor size. Interestingly, MKP1
been reported to be high in different types of tumors. also plays an antiapoptotic role with new therapeutic
For example, DU145, a hormone-refractory prostate agents, such as PS-341. Breast cancer cells overexpres-
cancer cell line, shows high levels of MKP1 (Magi- sing MKP1 are sensitized to PS341, by inhibition of
Galluzzi et al., 1997; Srikanth et al., 1999). Inhibition of MKP1 expression (Small et al., 2004), reinforcing the
MKP1 expression is able to sensitize DU145 cells to Fas idea that inhibition of MKP1 is an important target for
ligand and TNF-a induced apoptosis (Srikanth et al., tumor regression and induction of improvement in
1999), as well as in renal cancer cells (Mizuno et al., chemotherapy-dependant cellular response.
2004). Overexpression of MKP1 in primary human
ovarian carcinoma correlates with shorter progression-
free survival times (Denkert et al., 2002) and, finally, Materials and methods
increases the expression of MKP1 and MKP2 correlated
with low JNK activity in breast cancer (Wang et al., H-460, H-23 and H1299 cell lines were purchased from the
2003). Evidence presented recently demonstrated the American Type Culture Collection (ATCC) and maintained in
MKP1 deficient MEFs are also more sensitive to RPMI supplemented with 10% FBS. Antibodies used were
apoptosis induction by anysomicin than parental cells, anti-MKP1 (M18; Santa Cruz Biotechnologies, Santa Cruz,
CA, USA), anti-pJNK (V7391; Promega, Madison, WI, USA),
owing to a higher activation of JNK and p38 (Wu and anti-c-myc (sc-789; Santa Cruz Biotechnologies), anti-JNK1
Bennett, 2005). All this evidence reported in the (C-17; Santa Cruz Biotechnologies), anti-P38 (C20; Santa Cruz
literature suggested to us that overexpression of Biotechnologies), anti-pP38 (9211S; Cell Signaling, Beverly,
MKP1 in human tumors protects cells from different MA, USA) and anti-b-tubulin (T9026; Sigma, St Louis, MO,
types of stress and, therefore, was conducive to tumor USA), anti-c-jun (H16; Santa Cruz Biotechnologies), anti-
growth. We confirmed this observation in the present MKP2 (2518; Santa Cruz Biotechnologies) and anti-PARP (H-
study in NSCLC. It is not clear to us whether the sub- 250; Santa Cruz Biotechnologies). ECL kit was from
cellular distribution of MKP1 in the nuclei as well as in Amersham (Amersham, UK). Cisplatin and RO-31-8220 were
the cytoplasm in the tumor tissue of NSCLC patients is purchased from Calbiochem. Puromycin and geneticin were
simply due to a concentration effect of overexpression from Sigma-Aldrich.
or whether there are also differences in signaling that
direct MKP1 to different sub-cellular distributions. siRNA constructions and transfections
pSuperRetro was from Oligoengine (Seattle, WA, USA). The
The strategy of inhibition of MKP1 expression in MKP1 and MKP2 pSuperRetro-derived vectors (pSR) were
NSCLC cell lines by RNA interference and the increase constructed by using forward primer 50 -GATCCCCCGCCGG
in chemotherapy effectiveness was not only functional in CCACATCGCCGGCTTCAAGAGAGCCGGCGATGTGG
the cell lines but also had long-term implications, as CCGGCGTTTTTGGAA-30 and reverse primer 50 -AGCTT
observed in the clonogenic survival assay and also in the TTCCAAAAACGCCGGCCACATCGCCGGCTCTCTTGA
tumors induced by the H-460MKP1siRNA cells. Sur- AGCCGGCGATGTGGCCGGCGGGG-30 for MKP1, and
Oncogene

3344
forward primer 50 GATCCCCCGTGCGCTGTAACACCA of drug administration and tumor volume calculated as
TCTTCAAGAGAGATGGTGTTACAGCGCACGTTTTTG l Â w2 Â 0.52, where l is the length and w the width of the
GAAA-30 and reverse primer 50 AGCTTTTCCAAAAACGTG tumors. An average of four mice per treatment was used with
CGCTGTAACACCATCTCTCTTGAAGATGGTGTTACA two tumors on each in order to satisfy statistical constraints.
GCGCACGGGG-30 for MKP2. All mice were housed under similar humane conditions in
The primers were annealed and cloned into pSuperRetro cut accordance with the regulations of the Spanish Government
with HindIII and BglII to generate pSRMKP1RNAi and concerning experimental animal research.
pSRMKP2RNAi. Ampicillin-resistant colonies were selected
and checked by sequencing using pSuperRetro vector sequence
primers 50 ACCTCCTCGTTCGACGG-30 for direct and Immunohistochemistry
50 TGTGAGGGACAGGGGAG-30 for reverse sequencing. Formalin-fixed and paraffin-embedded blocks from 32
PLXSN and pLXSN-CL100 have been described previously surgical biopsies of NSCLC were selected to be analysed by
´ ´
(Sanchez-Perez et al., 2000). pSG5MKP2 plasmid containing a immunohistochemistry. The blocks included, whenever possi-
myc-tagged epitope was kindly donated by Dr Stephen Keyse. ble, invasive carcinoma and surrounding non-tumor epithe-
For stable transfections, 1 mg of the each construct was lium. Institutional approval to conduct the study was
transfected in a 60 mm plate. Transfections were carried out obtained.
using Lipofectamine 2000. Cells were treated with 2.5 mg/ml of Sections of 5 mm were cut from formalin-fixed, paraffin-
puromycin or geneticin during 24–48 h and kept for selection. embedded tissue blocks from the aforementioned patients.
Stable transfection was confirmed by Western blotting. Slides were de-paraffined and endogenous peroxidase activity
was blocked by incubation in 3% H2O2 in methanol for 10 min
at room temperature (RT). Antigens were retrieved by
Cell extracts, Western blots and RT–PCR incubation in EDTA for 45 min at 1551C. The primary
Whole-cell extracts were prepared essentially as described polyclonal antibody (Santa Cruz Biotechnology) was diluted
´ ´
previously (Sanchez-Perez et al., 1998). Western blotting at 1:200 in 1% BSA in TBS. Tissue slides were incubated with
´
was performed using standard methods (Sanchez-Perez ´ the antibody for 1 h at RT. Slides were then rinsed in TBS and
et al., 1998). For RT–PCR, total RNA was isolated from incubated with the peroxidase-based EnVisionTM kit (Dako
cells using Trizol (Life Technologies, Rockville, MD, Corporation, Carpinteria, CA, USA) for 30 min at RT.
USA) according to the manufacturer’s instructions. For Afterwards, specimens were incubated with diaminobenzidine
RT-PCR, single-strand cDNA was synthesized using a reverse chromogenic substrate (Dako Corporation) for 5 min at RT.
MKP1 primer, 50 GCGCTCGTCCAGGAACACCACGGC0 3, Sections were counterstained with hematoxylin, stepwise
and PCR amplification was carried out with the direct dehydrated through graded alcohols and cleared in xylene.
primer, 50 CATGGAAGTGGGCACCCTGGACG0 3 and As a control for nonspecific staining, three tumor tissues slides
the above-mentioned reverse primer. The reaction conditions were stained with, and without, pre-adsorption of the primary
were as follows: 1 Â (951C, 1 min), 25 Â (941C, 45 s, 601C 45 s antibody with a human control peptide (10 mg/ml; Santa Cruz
and 721C, 30 s) and 1 Â (721C, 4 min). The products were Biotechnology). The evaluation of MKP1 immunostaining was
analysed on a 1% agarose gel and stained with ethidium scored by two expert surgical pathologists of the hospital. The
bromide. tumors were scored according to the proportion of nuclei or
cytoplasm stained within the tumor. In all cases, the cutoff
Cell viability, growth kinetic and clonogenic assay point was dictated by the median values and used only for
determinations experimental correlation studies.
Cell viability was determined using a crystal violet Inmunocytochemistry for MKP1 of cultured cells was
staining method followed by colorimetric assay, as described performed in cells fixed with paraformahaldehyde. Cells were
´ ´
previously (Sanchez-Perez et al., 1998). For the growth grown on coverslips and to 60% confluence, fixed and
kinetic studies, cells were plated in triplicate at 1000 cells/well permeabalized with Triton X-100. Cells were stained with the
in a 24-well plate and allowed to grow in normal culture MKP1 antibody and then with a secondary fluorescence-
medium. Cells were fixed with glutaraldehyde at alternate labeled antibody.
days, stained by crystal violet and cell number determined.
The increase in growth was calculated in relation to the cell
number observed on day 1. For the clonogenic assay, Statistical analysis
cells were plated at a density of 10 000 cells/well, in triplicates Statistical analysis was performed using the SPSS data analysis
in six-well plates. Cells were treated for 48 h with cisplatin program v. 10.0. The test used was the w2 with correction
5 mg/ml. Then, the cells were washed and allowed to form for 2 degrees of freedom and for expected sequences lower
colonies. The colony-forming capacity was calculated by than 5.
fixing the cells with glutaraldehyde, and staining with crystal
violet followed by colorimetric assay as mentioned Acknowledgements
above. Controls were untreated cells cultured in parallel
and the percentage colony formation was calculated with ´
We thank I Sanchez-Perez for useful comments, JJ Sanchez
respect to the number of colonies formed by the untreated ´
for advice in statistics and F Nunez and M Marsa for help with
˜
cells. The colony-forming capacity of control cells corresponds the studies in mice. We also thank S Keyse for providing
to 100%. MKP2 plasmid and E Diaz Lopez and V Torres for
expert technical assistance. This work was supported by
Tumorigenicity the following grants: no. 08.1/0048.1/2003 from the
Tumors were induced on nuÀ/nuÀ mice by inoculating a ´
Comunidad Autonoma de Madrid; no. 01/1094, no. 02/0774
million cells of H-460 wt and H-460siRNA cells per mice. ´
and no. 02/1094 from the Fondo de Investigacion Sanitaria;
After 14 days, cisplatin was administered at 5 mg/kg mouse no. RTICC C03/10 from Instituto de Salud Carlos III. The
body weight every alternate day for a total of five adminis- investigators RM-P and CM-G are supported by Grant no.
trations. Tumor measurements were recorded from the 1st day RTICC C03/10.
Oncogene

3345
References

Alpert D, Schwenger P, Han J, Vilcek J. (1999). J Biol Chem Magi-Galluzzi C, Mishra R, Fiorentino M, Montironi R, Yao
274: 22176–22183. H, Capodieci P et al. (1997). Lab Invest 76: 37–51.
Beltman J, McCormick F, Cook SJ. (1996). J Biol Chem 271: Mizuno R, Oya M, Shiomi T, Marumo K, Okada Y, Murai
27018–27024. M. (2004). J Urol 172: 723–727.
Bossy W, Bakiri I, Yaniv M. (1997). EMBO J 16: 1695–1709. ´ ´
Perona R, Sanchez-Perez I. (2004). Br J Cancer 90: 573–577.
Denkert C, Schmitt WD, Berger S, Reles A, Pest S, Siegert A Persons DL, Yazlovitskaya EM, Cui W, Pelling JC. (1999).
et al. (2002). Int J Cancer 102: 507–513. Clin Cancer Res 5: 1007–1014.
´
Derijard B, Hibi M, Wu I-H, Barret T, Su B, Deng T et al. ´ ´
Sanchez-Perez I, Martinez-Gomariz M, Willians D, Keyse
(1994). Cell 76: 1025–1037. SM, Perona R. (2000). Oncogene 19: 5142–5152.
Eastman A. (1985). Biochemistry 24: 5027–5032. ´ ´
Sanchez-Perez I, Murguia JR, Perona R. (1998). Oncogene 16:
Ham J, Babij C, Whitﬁeld J, Pfarr CM, Lallemand D, Yaniv 533–541.
M et al. (1995). Neuron 14: 927–939. ´ ´
Sanchez-Perez I, Perona R. (1999). FEBS Lett 453: 151–158.
Hibi M, Lin A, Minden A, Karin M. (1993). Genes Dev 7: Slebos RJ, Rodenhuis S. (1989). Eur Respir J 2: 461–469.
2135–2148. Small GW, Shi YY, Edmund NA, Somasundaram S, Moore
Hirsch DD, Stork PJ. (1997). J Biol Chem 272: 4568–4575. DT, Orlowski RZ. 2004)). Mol Pharmacol 66: 1478–1490.
Keyse SM. (1995). Biochim Biophys Acta 1265: 152–160. Srikanth S, Franklin CC, Duke RC, Kraft RS. (1999). Mol
Kim DH, Kim JS, Park JH, Ji YI, Kwon YM, Shim YM et al. Cell Biochem 199: 169–178.
(2003). Cancer Res 63: 6206–6211. Standaert ML, Bandyopadhyay G, Antwi EK, Farese RV.
Kyriakis J.M, Banerjee P, Nikolakaki E, Dai T, Rubie EA, (1999). Endocrinology 140: 2145–2151.
Ahmad MF et al. (1994). Nature 369: 156–160. Vicent S, Garayoa M, Lopez-Picazo JM, Lozano MD,
Levin WJ, Casey G, Ramos JC, Arboleda MJ, Reissmann PT, Toledo G, Thunnissen FB. (2004). Clin Cancer Res 10:
Slamon DJ. (1994). Chest 106(Suppl 6): 372S–376S. 3639–3649.
Li J, Gorospe M, Hutter D, Barnes J, Keyse SM, Liu Y. Vivanco I, Sawyers SL. (2002). Nat Rev Cancer 2: 489–501.
(2001). Mol Cell Biol 21: 8213–8224. Wang HY, Cheng Z, Malbon CC. (2003). Can Lett 191:
Liao Q, Guo J, Kleeff J, Zimmermann A, Buchler MW, Korc 229–237.
M et al. (2003). Gastroenterology 124: 1830–1845. Wiest JS, Franklin WA, Drabkin H, Gemmill R, Sidransky D,
Lim EH, Aggarwal A, Agasthian T, Wong PS, Tan C, Sim E Anderson MW. (1997). J Cell Biochem Suppl 29: 64–73.
et al. (2003). Clin Cancer Res 9: 5980–5987. Woessmann W, Chen X, Borkhardt A. (2002). Cancer
Liu Y, Gorospe M, Yang C, Holbrook NJ. (1995). J Biol Chemother Pharmacol 50: 397–404.
Chem 270: 8377–8380. Wu JJ, Bennett AM. 2005. J Biol Chem 280: 16461–16466.

Oncogene

Sharmila Oncogene

Recomendados

Recomendados

Más contenido relacionado

La actualidad más candente

La actualidad más candente (20)

Destacado

Destacado (7)

Similar a Sharmila Oncogene

Similar a Sharmila Oncogene (20)

Último

Último (20)

Sharmila Oncogene