Tài liệu Báo cáo khoa học: Rac upregulates tissue inhibitor of metalloproteinase-1 expression by redox-dependent activation of extracellular signal-regulated kinase signaling pptx

Rac upregulates tissue inhibitor of metalloproteinase-1
expression by redox-dependent activation of extracellular
signal-regulated kinase signaling
Rainer Engers
1
, Erik Springer
1
, Verena Kehren
1
, Tatjana Simic
1
, David A. Young
2
, Juliane Beier
3
,
Lars-O. Klotz
3
, Ian M. Clark
2
, Helmut Sies
3
and Helmut E. Gabbert
1
1 Institute of Pathology, Heinrich-Heine-University, Duesseldorf, Germany
2 School of Biological Sciences, University of East Anglia, Norwich, UK
3 Institute of Biochemistry and Molecular Biology I, Heinrich-Heine-University, Duesseldorf, Germany
The Rho-like GTPase Rac mediates distinct actin cyto-
skeleton changes required for cell adhesion, migration
and invasion [1–4]. In addition, Rac has been implica-
ted in G1 cell cycle progression [5], apoptosis [6], secre-

tion in mast cells [7], malignant transformation [8],
and gene expression [9–15]. In the search for the bio-
chemical mechanisms underlying these various biologi-
cal activities of Rac, at least 15 different Rac effector
proteins have been identified so far [16–18]. Moreover,
Rac was shown to regulate c-Jun N-terminal kinase
Keywords
ERK; invasion; Rac; reactive oxygen
species; Tiam1; TIMP
Correspondence
R. Engers, Institute of Pathology,
Moorenstr. 5, D-40225 Duesseldorf,
Germany
Fax: +49 211 8118353
Tel: +49 211 8118341
E-mail:
(Received 21 May 2006, revised 16 August
2006, accepted 21 August 2006)
doi:10.1111/j.1742-4658.2006.05476.x
The Rho-like GTPase Rac regulates distinct actin cytoskeleton changes
required for adhesion, migration and invasion of cells. Tiam1 specifically
activates Rac, and Rac has been shown to affect several signaling pathways
in a partly cell-type-specific manner. Recently, we demonstrated that Rac
activation inhibits Matrigel invasion of human carcinoma cells by tran-
scriptional upregulation of tissue inhibitor of metalloproteinase-1. The
purpose of the present study was to identify key mediators of Tiam1 ⁄
Rac-induced tissue inhibitor of metalloproteinase-1 expression. Mutational
analysis of the human tissue inhibitor of metalloproteinase-1 promoter
revealed a major role for a distinct activating protein-1 site at )92 ⁄ )86 and
a minor role for an adjacent polyoma enhancer A3 site. Moreover, Rac

activation induced the generation of reactive oxygen species and subsequent
reactive oxygen species-dependent activation of extracellular signal-regula-
ted kinase 1,2. In contrast, c-Jun N-terminal kinase and p38 mitogen-
activated protein kinase activities were not affected. In line with this,
Tiam1 ⁄ Rac-induced tissue inhibitor of metalloproteinase-1 expression as
well as Tiam1 ⁄ Rac-induced binding of nuclear extracts to the activating
protein-1 site at )92 ⁄ )86 were inhibited by catalase and by specific inhibi-
tors of the extracellular signal-related kinase-1,2 activators, mitogen-activa-
ted protein kinase kinase-1 and mitogen-activated protein kinase kinase-2
(PD098059, U0126). In conclusion, Rac-induced transcriptional upregula-
tion of tissue inhibitor of metalloproteinase-1 is mediated by reactive oxy-
gen species-dependent activation of extracellular signal-related kinase-1,2
and by transcription factors of the activating protein-1 family.
Abbreviations
AP-1, activating protein-1; cPLA2, cytosolic phospholipase A2; EMSA, electrophoretic mobility shift assay; ERK, extracellular signal-regulated
kinase; FL, full-length; HRE, hypoxia-response element; JNK, c-Jun N-terminal kinase; MEK, mitogen-activated protein kinase kinase;
p38 MAPK, p38 mitogen-activated protein kinase; PEA-3 ⁄ ETS-1, polyoma enhancer A3; RCC, renal cell carcinomas; ROS, reactive oxygen
species; SDS, sodium dodecylsulfate; STAT, signal transducer and activator of transcription; TIMP, tissue inhibitor of metalloproteinase;
UTE-1, upstream TIMP-1 element-1.
4754 FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS
(JNK) [9], p38 mitogen-activated protein kinase (p38
MAPK) [10], cytosolic phospholipase A
2
(cPLA2) [15],
serum response factor [11], nuclear factor jB [13], sig-
nal transducer and activator of transcription (STAT)-3
[14], and the formation of reactive oxygen species
(ROS) in both phagocytic [19] and nonphagocytic cells
[15,20]. Recent studies also suggest a role for Rac in
the activation of extracellular signal-regulated kinase

(ERK) [21–23]. However, these Rac-induced signaling
events seem to be at least in part cell type-dependent.
For instance, Rac has been shown to activate JNK
and p38 MAPK but not the ERKs in Cos and HeLa
cells [9,24]. In contrast, in both human kidney 293T
cells and HepG2 cells, Rac failed to activate JNK
[25,26], and in cardiac myocytes, NIH3T3 cells, human
kidney fibroblasts and Rat-2 fibroblasts, an active
mutant of Rac (V12-Rac1) either cooperated with
c-Raf in ERK activation [21,27,28] or activated ERK
dose-dependently [22].
Although the exact biochemical, functional and cell
type-dependent interplay between these molecules and
signaling pathways downstream of Rac is still far from
being elucidated, it is well established that all of these
signaling events may affect gene expression by modula-
ting different transcription factors [16,29–32]. So far,
however, only little is known about which genes are
indeed regulated by Rac signaling. Recently, we
have demonstrated that sustained activation of Rac
by overexpression of the Rac-specific activator, Tiam1,
or overexpression of V12-Rac1, strongly induces
transcriptional upregulation of tissue inhibitor of
metalloproteinase (TIMP)-1 and post-transcriptional
upregulation of TIMP-2 in both human colon carcino-
mas and renal cell carcinomas (RCCs) in vitro, and
consequently inhibits Matrigel invasion of these cells
[33]. TIMP-1 and TIMP-2 belong to a family of four
different isoforms. Of these, the prototype member of
the TIMP family, TIMP-1, is of particular interest,

because in addition to its classic role as a broad speci-
fic inhibitor of active matrix metalloproteinases and
hence of invasion, metastasis and angiogenesis, it also
exhibits growth factor-like activities or suppresses
apoptosis [34–36]. As Rac has also been reported to
stimulate cell proliferation [5] and to inhibit apoptosis
[6], one might speculate that these effects of Rac are at
least in part also mediated by TIMP-1.
TIMP-1 expression can be stimulated by a wide vari-
ety of agents, such as serum, growth factors, phorbol
esters, cytokines, hormones and viruses, and this
occurs primarily at the transcriptional level [37–43].
The structure of the TIMP-1 promoter has been eluci-
dated in mouse [44], rat [45], and humans [46]. The
most prominent feature shared by all three species
is the proximal ‘noncanonical’ 22 base pair serum
response element, comprising an activator protein-1
(AP-1) site at )92 ⁄ )86 (numbers refer to the human
TIMP-1 promoter [46]), followed by a STAT site and
a polyoma enhancer A3 (PEA-3 ⁄ ETS-1) site. The AP-1
site was shown to be necessary for basal and, in most
cases, also for induced expression of the TIMP-1 gene,
whereas the adjacent PEA-3 ⁄ ETS-1 site seems to play
an additional, albeit minor, role [39,41,46–49]. In addi-
tion, a hypoxia-response element (HRE) has been
found at )27 ⁄ )23 in human kidney fibroblasts [50],
and an additional regulatory element, designated as
upstream TIMP-1 element-1 (UTE-1), has been identi-
fied at )63 ⁄ )53 [51]. UTE-1 is bound by Runx1a,
Runx1b and Runx2 [52], and was reported to be essen-

tial for TIMP-1 transcription in Jurkat cells [51].
In the present study, we have investigated the signa-
ling events mediating Rac-induced transcriptional
upregulation of TIMP-1. We show that activation of
Rac induces ROS formation, followed by ERK1,2 acti-
vation and AP-1-mediated TIMP-1 promoter activa-
tion. In contrast, other signaling pathways such as p38
MAPK, JNK and cPLA2 are not likely to be involved.
Given the well-established roles of (a) AP-1 in regula-
ting the transcription of numerous genes, and (b) of
TIMP-1 as an inhibitor of tumor invasion and meta-
stasis as well as a regulator of cell proliferation and
apoptosis, this newly identified signaling pathway
might play a major role in different Rac-dependent cell
biological properties.
Results
Tiam1
⁄
Rac-induced activation of the human
TIMP-1 promoter is mainly mediated by a distinct
AP-1-binding site
Recently, we have shown by transient cotransfection
that C1199-Tiam1 activates the human TIMP-1 pro-
moter in HepG2 cells and that this effect is mediated
by Rac [33]. To identify the regulatory elements within
the TIMP-1 promoter that mediate this effect, we first
tested different length variants of the human TIMP-1
promoter. To this end, C1199-Tiam1 (in pcDNA1)
was cotransfected with the respective TIMP-1 promo-
ter contructs in HepG2 cells (Fig. 1). All constructs

proved inducible by C1199-Tiam1, but the strongest
induction was seen with the basic promoter construct
()102 ⁄ +95). When smaller TIMP-1 promoter con-
structs such as )80 ⁄ +95 and )73 ⁄ +95 were used,
basal activities, which were determined by cotransfec-
tion with empty vector, dropped dramatically to levels
barely above background, and no obvious inducibility
R. Engers et al. Rac signaling towards TIMP-1
FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS 4755
by C1199-Tiam1 was seen (data not shown). In
addition, the shortest TIMP-1 promoter construct
()42 ⁄ +95) was completely inactive (data not shown).
As similar results were also obtained when another
expression vector for C1199-Tiam1 (pEGFP) was used
(data not shown), these results suggest that the regula-
tory elements mediating Rac-induced TIMP-1 promo-
ter activation are located downstream of position
)102. Furthermore, these results suggest that silencer
elements exist between )738 and )102.
Previous studies have shown that downstream of
)102, the human TIMP-1 promoter harbors binding
sites for transcription factors, including AP-1, STAT,
PEA-3 ⁄ ETS-1, UTE-1, and the promoter-specific
protein factor SP-1. Of these, the AP-1 site has been
reported to be critical for basal and mostly also for
induced transcription [39,41,46–49]. Moreover, Logan
et al. [47] have reported that the AP-1 and PEA-
3 ⁄ ETS-1 sites and the proteins binding to them interact
to enhance transcription from the whole element.
Therefore, we analyzed the role of these two sites in

Tiam1 ⁄ Rac-induced TIMP-1 promoter activation. To
this end, different luciferase–TIMP-1 promoter con-
structs (in the context of )105 ⁄ +95) with mutated (i.e.
inactive) binding sites for AP-1 and ⁄ or PEA-3 ⁄ ETS-1
transcription factors as well as the corresponding
wild-type )102 ⁄ +95 promoter construct (control) were
used for cotransfection experiments with empty vector
and C1199-Tiam1, respectively (Fig. 2A). Thus, select-
ive inactivation of the AP-1 or PEA-3 ⁄ ETS-1 site or
inactivation of both sites simultaneously resulted in a
marked reduction of basal transcription, as evidenced
by cotransfection with empty vector (mock). In com-
parison with the respective basal promoter activities,
C1199-Tiam1-induced TIMP-1 promoter activation
was strongly inhibited by selective inactivation of the
AP-1 site and completely inhibited by simultaneous
inactivation of the AP-1 and PEA-3 ⁄ ETS-1 sites. Select-
ive inactivation of the PEA-3 ⁄ ETS-1 site had an addi-
tional but weaker inhibitory effect. Moreover, deleting
32 bases around the UTE-1 region (within the
)102 ⁄ +95 CAT construct) did no affect C1199-Tiam1-
induced TIMP-1 promoter activation (data not shown),
excluding a role for the UTE-1 site. These results sug-
gest that the AP-1 site at )92 ⁄ )86 plays a major role in
Tiam1 ⁄ Rac-induced TIMP-1 promoter activation, and
that the adjacent PEA-3 ⁄ ETS-1 site is of additional,
albeit minor, importance.
The role of AP-1 transcription factors in Tiam1 ⁄
Rac-induced TIMP-1 promoter activation was further
supported in human RCC cells by analyzing binding

of nuclear extracts of mock-transfected and C1199-
Tiam1-transfected RCC cells to the AP-1 site at
)92 ⁄ )86 of the human TIMP-1 promoter (Fig. 2B) in
electrophoretic mobility shift assays (EMSAs). Thus,
nuclear extracts of C1199-Tiam1-transfected RCC cells
bound markedly stronger to the wild-type AP-1 site
than nuclear extracts of mock-transfected control cells.
Binding to the wild-type AP-1 site was specific, as it
could be entirely inhibited by competition with the
unlabeled wild-type AP-1 oligonucleotide, but hardly
inhibited by competition with the unlabeled mutated,
and hence inactivated, *AP-1 oligonucleotide. More-
over, in contrast to moderate and strong binding of
nuclear extracts of mock-transfected and C1199-
Tiam1-transfected RCC cells, respectively, to the wild-
type AP-1 site, only very weak if any binding to the
mutated *AP-1 site could be observed.
Effects of sustained Rac activation on MAPK
activities, ROS formation, and activities of Rho
and Cdc42
After having shown that Tiam1 ⁄ Rac-induced upregula-
tion of TIMP-1 is mainly mediated by AP-1 transcrip-
tion factors, we investigated through which signaling
pathway Rac induces AP-1-mediated upregulation of
TIMP-1 and whether this pathway is cell density-
dependent. To address the latter question, TIMP-1
Fig. 1. Activation of different length variants of the human tissue
inhibitor of metalloproteinase-1 (TIMP-1) promoter by C1199-Tiam1.
In HepG2 cells, different length variants of the human TIMP-1 pro-
moter (in pBLCAT3; numbers refer to the 5¢-end and 3¢-end of the

DNA sequence in relation to the transcriptional start point, which
was defined as +1) were transiently cotransfected with C1199-
Tiam1 or empty vector (pcDNA1), respectively. C1199-Tiam1-
induced effects are presented as n-fold activation of the respective
TIMP-1 promoter constructs in comparison to the effects of empty
vector (control). All length variants of the human TIMP-1 promoter
proved to be inducible by C1199-Tiam1, but the strongest induction
was seen with the basic promoter construct ()102 ⁄ +95). The
observed effects did not result from differences in protein expres-
sion upon transfection, as verified by immunoblotting. Data shown
are means ± standard deviations of an experiment performed in
triplicate, and are representative of three independent experiments.
Rac signaling towards TIMP-1 R. Engers et al.
4756 FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS
secretion was determined by immunoblotting in sup-
ernatants of stably transfected RCC cells at different
levels of confluency. As shown in Fig. 3A, stable over-
expression of C1199-Tiam1 and V12-Rac1 strongly
induced TIMP-1 secretion at both 50% and 100%
confluency of cells, suggesting that the pathway medi-
ating Tiam1 ⁄ Rac-induced upregulation of TIMP-1 is
independent of the cell density. So far, Rac has been
reported to induce numerous signaling events ⁄ path-
ways through partly different mechanisms [9–23,27,28].
Of these, the JNK, p38 MAPK and ERK1,2 signaling
pathways as well as ROS are known to affect gene
A
B
Fig. 2. Identification of regulatory elements within the human tissue inhibitor of metalloproteinase-1 (TIMP-1) promoter, mediating its activa-
tion by Tiam1 ⁄ Rac signaling. (A) Effects of C1199-Tiam1 on different )102 ⁄ +95 TIMP-1 promoter mutants, as determined by luciferase

reporter gene assays. In HepG2 cells, C1199-Tiam1 and empty vector (pcDNA1) were transiently cotransfected with different )102 ⁄ +95
TIMP-1 promoter constructs [WT, wild type )102 ⁄ +95; *AP-1, )102 ⁄ +95 with selectively inactivated activating protein-1 (AP-1) site; *PEA-3,
)102 ⁄ +95 with selectively inactivated polyoma enhancer A3 (PEA-3 ⁄ ETS-1) site; *AP-1 ⁄ *PEA-3, )102 ⁄ +95 with AP-1 and PEA-3 ⁄ ETS-1
sites both inactivated] and luciferase activity was subsequently determined (left, absolute values (light units); right, relative values showing
C1199-Tiam1-induced effects as fold-induction over mock). Selective inactivation of the AP-1 or PEA-3 ⁄ ETS-1 sites or inactivation of both
sites simultaneously resulted in a marked reduction of basal transcription, as evidenced by cotransfection with empty vector (mock). In com-
parison to the respective basal promoter activities, C1199-Tiam1-induced TIMP-1 promoter activation was strongly inhibited by selective inac-
tivation of the AP-1 site and completely inhibited by simultaneous inactivation of the AP-1 and PEA-3 ⁄ ETS-1 sites. Selective inactivation of
the PEA-3 ⁄ ETS-1 site had an additional but weaker inhibitory effect. The observed effects did not result from differences in protein expres-
sion upon transfection, as verified by immunoblotting. Results are means ± standard deviations of an experiment performed four times and
are representative of three independent experiments. (B) Binding of nuclear extracts of mock-transfected and C1199-Tiam1-transfected cells
to the wild-type activating protein-1 (AP-1) and mutated (*AP-1), and hence inactivated, AP-1 site ()92 ⁄ )86) of the human TIMP-1 promoter,
as determined by electrophoretic mobility shift assay (EMSA). After serum starvation for 24 h, nuclear extracts of mock-transfected (M) and
C1199-Tiam1-transfected (T) clearCa-28 renal cell carcinoma (RCC) cells were incubated with a labeled 21 base pair double-stranded oligonu-
cleotide, containing the AP-1 or *AP-1 site ()92 ⁄ )86) of the human TIMP-1 promoter, for 30 min, run on a polyacrylamide gel, and analyzed
by autoradiography. Binding specificity was verified by competition with an excess (10·) of unlabeled AP-1 and *AP-1 oligonucleotides,
respectively. In a control reaction, labeled oligonucleotides were incubated without nuclear extracts. C3H10T1 ⁄ 2 mouse fibroblast cells (10)
were used as a positive control. Stable overexpression of C1199-Tiam1 resulted in markedly increased AP-1 binding, which could be blocked
by competition with the unlabeled AP-1 but not with the unlabeled *AP-1 oligonucleotide. Data are from a single experiment ⁄ autoradiograph.
The order of the lanes has been changed for clarity of presentation. Results are representative of three independent experiments.
R. Engers et al. Rac signaling towards TIMP-1
FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS 4757
transcription at least in part through AP-1 transcrip-
tion factors. Since Tiam1 ⁄ Rac-induced signaling events
are at least partly cell type-dependent, we first ana-
lyzed which signaling pathways known to affect the
transactivation activities of AP-1 transcription factors
and known to be potentially activated by Rac were
indeed activated upon stable overexpression of C1199-
Tiam1 or V12-Rac1 in human RCC cells. By immuno-

blotting with antibodies specific for phosphorylated
forms of the respective kinases, we demonstrated that
sustained Rac activation in human RCC cells resulted
in activation of ERK1,2, but had no effect on p38
MAPK activity (Fig. 3B). Moreover, in contrast to
anisomycin, which served as a positive control, sus-
tained activation of Rac also failed to induce JNK
activation in these cells (data not shown). However,
Rac activation led to elevated levels of ROS (Fig. 3C),
as reflected by increased 2¢,7¢-dichlorofluorescein fluor-
escence, resulting from oxidation of an H
2
O
2
-sensitive
probe, 2¢,7¢-dichlorodihydrofluorescein. ROS formation
induced by C1199-Tiam1 and V12-Rac1, respectively,
could be blocked by catalase, supporting a role for
H
2
O
2
. The effect of V12-Rac1 on ROS formation in
human RCC cells was less pronounced than the effect
of C1199-Tiam1, which is in accordance with previous
observations in epithelial cells [53].
Because in NIH3T3 and HeLa cells, respectively,
some effects of Rac were found to require Rac-induced
downregulation of Rho activity [54,55], we also ana-
lyzed the effects of C1199-Tiam1 and V12-Rac1 on Rho

activity in human RCC cells by biochemical pull-down
assays. Thus, stable overexpression of C1199-Tiam1
strongly activated Rac, but neither C1199-Tiam1 nor
V12-Rac1 affected Rho activity (Fig. 3D). This suggests
that Rac-induced upregulation of TIMP-1 is neither
mediated by Rho nor requires downregulation of Rho.
Moreover, C1199-Tiam1 and V12-Rac1 had no effect
on Cdc42 activity (Fig. 3D). Similar results as found in
human RCC cells were obtained when C1199-Tiam-
transfected and V12-Rac1-transfected human colon
cancer cells (DusCol-1B) were used (data not shown).
Together, these results suggested that Rac-induced up-
regulation of TIMP-1 might be mediated by ERK1,2
rather than by JNK or p38 MAPK signaling, and imply
a possible role for ROS. Moreover, downregulation of
Rho is not likely to be involved.
Tiam1
⁄
Rac-induced upregulation of TIMP-1 is
reversed by specific inhibitors of ROS and ERK1,2
signaling pathways
To analyze the roles of ERK1,2 and ROS in Rac-
induced upregulation of TIMP-1, we next treated
C1199-Tiam1-transfected RCC cells (clearCa-28) with
specific inhibitors of these signaling pathways and
investigated their effects on TIMP-1 secretion by
quantitative TIMP-1 immunoassays. Thus, two differ-
ent specific inhibitors of the direct upstream kinases
of ERK1,2 ) mitogen-activated protein kinase kinase
(MEK1,2) (PD098059 and U0126) ) as well as cata-

lase attenuated C1199-Tiam1-induced upregulation
of TIMP-1 in a dose-dependent manner (Fig. 4A),
suggesting a role for ERK1,2 and H
2
O
2
. In contrast,
a specific inhibitor of p38 MAPK (SB203580) had no
effect (Fig. 4A). The effects of both MEK1,2 inhibi-
tors and catalase cannot simply be explained by toxic
effects or effects on growth and apoptosis, as superna-
tants were calibrated for the respective cell numbers
and cells appeared phenotypically normal by phase
contrast microscopy. Furthermore, dimethylsulfoxide,
Fig. 3. Effects of sustained Rac activation on tissue inhibitor of metalloproteinase-1 (TIMP-1) secretion, reactive oxygen species (ROS) for-
mation and the activities of extracellular signal-related kinase (ERK) 1,2, p38 mitogen-activated protein kinase (p38 MAPK), Rho, and Cdc42
in stably transfected renal cell carconoma (RCC) cells. (A) Effects of sustained Rac activation on TIMP-1 secretion at different levels of cell
confluency. Secretion levels of TIMP-1 were determined by immunoblotting from concentrated serum-free supernatants as previously des-
cribed [33]. Expression of C1199-Tiam1 and V12-Rac1 was verified by immunoblotting with Tiam1-specific and myc-epitope-specific antibod-
ies, respectively. Results are representative of two independent experiments. (B) Effects of sustained Rac activation on the activities of
ERK1,2 and p38 MAPK. Cells were serum-starved for 24 h and lysed, and kinase activities were determined by immunoblotting and com-
pared with the expression levels of respective total proteins. As a positive control for p38 MAPK activation, mock-transfected cells were
treated for 1 h with anisomycin (1 lgÆmL
)1
). Results are representative of at least three independent experiments. (C) Effects of sustained
Rac activation on ROS formation as determined by measuring oxidation of an H
2
O
2
-sensitive probe, 2¢,7¢-dichlorodihydrofluorescein, in

C1199-Tiam1-transfected and V12-Rac1-transfected cells, respectively, relative to mock-transfected control cells after 24 h of serum starva-
tion. Rac activation led to elevated levels of ROS, and this effect could be blocked by catalase, supporting a role for H
2
O
2
. Results are
means ± standard deviation of three independent experiments. (D) Effects of sustained Rac activation on Rho and Cdc42 activities, as deter-
mined by biochemical pull-down assays. C1199-Tiam1 activates Rac, but neither C1199-Tiam1 nor V12-Rac1 affect the activities of Rho or
Cdc42. Active Rac and active Cdc42 were precipitated with a biotin–CRIB peptide, and active Rho was precipitated with GST-C21. Precipi-
tates were probed with antibodies against Rac1, Cdc42, and Rho, respectively. Aliquots of the respective lysates served as controls for ana-
lyzing total amounts of Rac, Cdc42, and Rho, respectively. Results are representative of three independent experiments.
Rac signaling towards TIMP-1 R. Engers et al.
4758 FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS
which was used as a vehicle for both MEK1,2 and
p38 MAPK inhibitors, had no effect on TIMP-1
secretion, as verified in separate control experiments
(data not shown) and as indicated by the fact that
SB203580 (dissolved in dimethylsulfoxide) failed to
reverse the effect of C1199-Tiam1. In addition, the
effects of MEK1,2 inhibitors and catalase on C1199-
Tiam1-induced TIMP-1 secretion in quantitative
TIMP-1 immunoassays were verified by immunoblot-
ting (data not shown). Similar results as observed
for C1199-Tiam1-transfected RCC cells were also
obtained with C1199-Tiam1-transfected human colon
carcinoma cells (Fig. 4A), indicating that the observed
effects are not restricted to a single cell line, but
rather are of more general importance.
The role of JNK in Tiam1 ⁄ Rac-induced upregula-
tion of TIMP-1 was analyzed by CAT reporter gene

assays rather than by means of the commonly used
JNK inhibitor SP600125, because recent investigations
have suggested a lack of specificity for this compound
[56,57]. Thus, C1199-Tiam1-induced TIMP-1 promoter
activation in HepG2 cells could not be inhibited by
cotransfection with a dominant negative cDNA of
JNK (Fig. 4B), although high expression levels were
obtained. In line with this, overexpression of constitu-
tively active JNK failed to activate the human TIMP-1
promoter (Fig. 4B). From these results, we conclude
that Tiam1 ⁄ Rac-induced upregulation of TIMP-1 is
mediated by ROS and ERK1,2, whereas JNK and p38
MAPK are not likely to be involved.
A
B
DC
R. Engers et al. Rac signaling towards TIMP-1
FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS 4759
Tiam1
⁄
Rac-induced binding of nuclear extracts to
the AP-1 site at
)
92
⁄ )
86 is reversed by catalase
and by specific inhibition of ERK1,2 signaling
To further substantiate the role of ROS and of ERK1,2
signaling in Tiam1 ⁄ Rac-induced and AP-1-mediated
upregulation of TIMP-1, we investigated whether Tiam1 ⁄

Rac-induced binding of nuclear extracts to the AP-1 site
at )92 ⁄ )86 of the human TIMP-1 promoter could be
reversed by specific inhibitors of these pathways. To this
end, C1199-Tiam1-transfected RCC cells were incuba-
ted in the absence or presence of the respective inhibi-
tors, and nuclear extracts of these cells were analyzed
for binding to the AP-1 site at )92 ⁄ )86 by EMSAs. In
line with the results shown above, catalase, as well as
PD098059 and U0126, inhibited Tiam1 ⁄ Rac-induced
binding of nuclear extracts to the AP-1 site (Fig. 5). As
verified in control experiments, dimethylsulfoxide, which
was used as a vehicle, as well as the p38 MAPK inhi-
bitor SB203580, had no effect (data not shown). These
results indicate that Tiam1 ⁄ Rac-induced binding of
nuclear extracts to the AP-1 site at )92 ⁄ )86 and subse-
quent TIMP-1 promoter activation are mediated by
ROS and by MEK1,2 ⁄ ERK1,2.
Tiam1
⁄
Rac-induced ERK1,2 activation is
ROS-dependent
Next, we investigated whether stimulation of ROS for-
mation and ERK1,2 signaling by Rac occurred in a
linear signaling cascade or via independent mecha-
nisms. As shown above, Rac-induced upregulation of
TIMP-1 was inhibited by catalase, suggesting a major
role for H
2
O
2

in mediating this effect. Therefore, we
analyzed whether H
2
O
2
affected ERK1,2 activity in
mock-transfected clearCa-28 control cells. Indeed,
incubation with H
2
O
2
(300 lm, 30 min) strongly
induced ERK1,2 activation, and this effect was inhib-
ited by catalase, whereas heat-inactivated catalase
failed to do so (Fig. 6A). Similar effects were observed
when mock-transfected human colon carcinoma cells
were used (data not shown).
In contrast, inhibition of ERK1,2 signaling by
PD098059 or U0126 could not reverse Tiam1 ⁄ Rac-
induced ROS formation at concentrations that inhib-
ited Tiam1 ⁄ Rac-induced upregulation of TIMP-1
(Fig. 6B). From this, we conclude that activation of
Rac induces both ROS formation and ERK1,2 activa-
tion as part of a linear signaling cascade (Rac fi
ROS fi MEK1,2 fi ERK1,2) rather than via inde-
pendent mechanisms.
Discussion
We have recently shown that in addition to promoting
E-cadherin-mediated cell–cell adhesion, transcriptional
upregulation of TIMP-1 is another mechanism through

which Rac may inhibit invasion of epithelial cells [33].
In the present study, we provide evidence that Rac-
induced transcriptional upregulation of TIMP-1 is
mediated by a signaling cascade leading from Rac to
TIMP-1 transcription via ROS (H
2
O
2
) fi MEK1,2 fi
ERK1,2 fi AP-1.
Several reports have implicated the AP-1 site at
)92 ⁄ )86 as essential for basal as well as for induced
activity of the TIMP-1 promoter [39,41,46–49], but
other inducible elements, including HRE at )27 ⁄ )23
[50] and UTE-1 at )63 ⁄ )53 [51], have also been des-
cribed. By mutational analysis of the human TIMP-1
promoter and CAT ⁄ luciferase reporter gene assays, we
demonstrated that Tiam1 ⁄ Rac-induced promoter acti-
vation is mainly mediated by the AP-1 site at )92 ⁄ )86
and that the adjacent PEA-3 ⁄ ETS-1 site at )79 ⁄ )74
plays an additional, albeit minor, role. This is in line
with the observations of Logan et al. [47], who have
reported that the AP-1 and PEA-3 ⁄ ETS-1 sites and the
Fig. 4. Roles of extracellular signal-related kinase (ERK) 1,2, p38 mitogen-activated protein kinase (p38 MAPK), c-Jun N-terminal kinase (JNK)
and reactive oxygen species (ROS) on Tiam1 ⁄ Rac-induced upregulation of tissue inhibitor of metalloproteinase-1 (TIMP-1). (A) Effects of cat-
alase and specific inhibitors of mitogen-activated protein kinase kinase (MEK) 1,2 (PD098059, U0126) and p38 MAPK (SB203580) on Tiam1 ⁄
Rac-induced TIMP-1 secretion in both human renal cell carcinoma (RCC) (clearCa-28) and human colon carcinoma cells (DusCol-1B), as
determined by means of TIMP-1-specific immunoassays (PD, PD098059; U, U0126; SB, SB203580; Cat, catalase; hia, heat-inactivated).
Results represent mean values and standard deviations of two independent experiments (*P < 0.05 in comparison to C1199-Tiam1).
Dimethylsulfoxide which was used as a vehicle for MEK1,2 and p38 MAPK inhibitors, had no effect, as verified in control experiments (data

not shown) and as indicated by the fact that SB203580 (dissolved in dimethylsulfoxide) failed to attenuate C1199-Tiam1-induced TIMP-1
secretion. (B) No role of JNK in Tiam1 ⁄ Rac-induced TIMP-1 promoter activation as determined by CAT reporter gene assays. In Hep-G2
cells, the human )102 ⁄ +95 TIMP-1 promoter–CAT construct was transiently cotransfected with different expression constructs as indicated
(ca-JNK, constitutively active JNK; dn-JNK, dominant negative JNK) (upper part). Results are presented as n-fold activation of the TIMP-1 pro-
moter construct in comparison to the effects of empty vectors. Ca-JNK failed to activate the TIMP-1 promoter, and dn-JNK failed to inhibit
the C1199-Tiam1-induced TIMP-1 promoter. This did not result from a lack of protein expression, as verified by immunoblotting (lower part).
Data shown are means ± standard deviations of an experiment performed in triplicate, and are representative of three independent experi-
ments.
Rac signaling towards TIMP-1 R. Engers et al.
4760 FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS
B
A
R. Engers et al. Rac signaling towards TIMP-1
FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS 4761
proteins binding to them may interact to enhance
TIMP-1 transcription from the whole element. The
role of AP-1 in Tiam1 ⁄ Rac-induced TIMP-1 expres-
sion was substantiated by EMSAs showing markedly
increased binding of nuclear extracts to the AP-1 site
at )92 ⁄ )86 upon stable overexpression of C1199-
Tiam1. This effect proved specific, as it could be
entirely blocked by an excess of unlabeled wild-type
AP-1 oligonucleotide, whereas an excess of unlabeled
mutated, and hence inactivated, AP-1 (*AP-1) oligonu-
cleotide had hardly any effect. In line with this, nuclear
extracts of C1199-Tiam1-transfected and mock-trans-
fected cells, respectively, exhibited only weak, if any,
binding to *AP-1. In contrast to the AP-1 site at
)92 ⁄ )86, the UTE-1 site at )63 ⁄ )53 is not likely to be
involved. Deleting 32 bases around the UTE-1 region

did no affect C1199-Tiam1 ⁄ Rac-induced TIMP-1 pro-
moter activation (data not shown), excluding a role for
the UTE-1 site.
Rac has been shown to induce numerous signaling
events ⁄ pathways through partly different mechanisms
and in a partly cell type-dependent manner [9–
23,27,28]. Of these mechanisms, JNK, p38 MAPK,
and ERK1,2, as well as the generation of ROS, are
known to affect gene transcription at least in part
through AP-1 transcription factors [30,58]. In human
RCC cells, sustained activation of Rac induced
increased ROS formation as well as activation of
ERK1,2, whereas no effects on JNK and p38 MAPK
activities were observed. Moreover, Rac-induced
upregulation of TIMP-1 could be reversed by specific
inhibitors of MEK1,2 and by catalase, as shown
by quantitative TIMP-1-specific immunoassays. In
contrast, specific inhibition of p38 MAPK and JNK
signaling had no effects. These results were substan-
tiated by EMSAs showing that Tiam1 ⁄ Rac-induced
binding of nuclear proteins to the important regula-
tory AP-1 site ()92 ⁄ )86) of the human TIMP-1 pro-
moter was markedly blocked by specific inhibitors of
the ERK1,2 signaling pathway and by catalase. There-
fore, our results suggest that Rac-induced and AP-1-
mediated upregulation of TIMP-1 requires both ROS
and ERK1,2, whereas JNK and p38 MAPK are not
likely to be involved. Moreover, Rac-induced TIMP-1
expression does not require Rac-dependent downregu-
lation of Rho activity, as reported for some effects of

Rac in NIH3T3 fibroblasts and HeLa cells [54,55], as
neither C1199-Tiam1 nor V12-Rac1 affected Rho
activity in human RCC cells. In addition, C1199-
Tiam1 exclusively activated Rac and exhibited no
effect on Cdc42 activity, supporting its role as a speci-
fic activator of Rac. Similar results as observed for
human RCC cells were also obtained for human colon
carcinoma cells. This excludes a cell type-specific
effect, and rather suggests that Rac-induced upregula-
tion of TIMP-1 via ROS and ERK1,2 may be a more
general mechanism.
Despite the fact that so far only little is known about
the signaling pathways mediating TIMP-1 expression,
our results are in line with recent studies showing that a
specific inhibitor of the ERK pathway reduced both reti-
noic acid-induced and basal TIMP-1 production,
whereas specific p38 MAPK inhibitors rather enhan-
ced TIMP-1 expression [43]. Similarly, specific MEK
inhibitors reversed erythropoietin-induced, oncostatin
M-induced and 12-O-tetradecanoylphrobol 13-acetate
(TPA)-induced TIMP-1 transcriptional activation ⁄
expression [42,59,60], whereas a potent inhibitor of
p38 MAPK failed to do so [60]. In addition, two recent
studies suggested a role for ROS in TIMP-1 expression
[61,62]. However, in neither of these studies was a role
for Rac in TIMP-1 expression investigated.
Rac has been reported to stimulate ROS formation
in phagocytic [19] and nonphagocytic [15,20] cells.
Fig. 5. Role of reactive oxygen species (ROS) and extracellular sig-
nal-related kinase (ERK) 1,2 in Tiam1 ⁄ Rac-induced activating pro-

tein-1 (AP-1) binding of nuclear extracts. Mock-transfected and
C1199-Tiam1-transfected human renal cell carcinoma (RCC) cells
were serum-starved (0.5% fetal bovine serum) for 24 h and subse-
quently incubated in serum-free medium (Optimem) for another
24 h. Catalase (1 mgÆmL
)1
; 2940 unitsÆmg
)1
) and specific inhibitors
of MEK 1,2 (PD098059, 25 l
M; U0126, 10 lM) were applied 90 min
or 24 h, respectively, prior to cell lysis. As a negative control for
catalase, cells were treated with heat-inactivated (hia) catalase.
Binding of nuclear proteins to the regulatory important AP-1-binding
site ()92 ⁄ )86) was determined by electrophoretic mobility shift
assays (EMSAs) as described. Results are representative of at least
three independent experiments. As verified in control experiments,
dimethylsulfoxide, which was used as a vehicle for MEK1,2 inhibi-
tors, had no effect (data not shown).
Rac signaling towards TIMP-1 R. Engers et al.
4762 FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS
Accordingly, Rac stimulated ROS (e.g. H
2
O
2
) forma-
tion in human RCC and colon cancer cells, and in
addition induced upregulation of TIMP-1. Upregula-
tion of TIMP-1 was reversed by catalase, indicating
that this effect is mediated by H

2
O
2
. The fact that cata-
lase does not penetrate cell membranes and therefore
exerts its effects extracellularly might suggest that
H
2
O
2
-mediated upregulation of TIMP-1 requires inter-
cellular communication. This, however, is not neces-
sarily the case. Since H
2
O
2
easily penetrates cell
membranes, extracellular metabolization of H
2
O
2
by
catalase will consequently also result in a reduction of
intracellular H
2
O
2
concentrations, and thus catalase
may also affect intracellular signaling by H
2

O
2
. This is
supported by the fact that both C1199-Tiam1-induced
and V12-Rac1-induced intracellular ROS (e.g. H
2
O
2
)
formation, as determined by 2¢,7¢-dichlorofluorescein
fluorescence, was entirely abrogated by catalase
(Fig. 3B).
In addition to its effect on ROS (e.g. H
2
O
2
) forma-
tion, sustained Rac activation also induced activation
of ERK1,2 in both human RCC and colon cancer
cells. In our search for the functional interplay
between ROS and ERK1,2 signaling, we demonstrated
that Rac-induced ERK activation was mediated by
H
2
O
2
(Fig. 6A). This is in line with recent studies
showing that H
2
O

2
may activate ERK [63–65]. Never-
theless, our results are in contrast to other studies, in
which overexpression of constitutively active Rac failed
to activate ERK1,2 [9,24]. In these studies, however,
the effects of Rac on ROS formation were not tested,
and this might be crucial in terms of Rac-dependent
ERK activation. As Rac signaling is at least in part
cell type-dependent, the observed differences in ERK
activation might be explained by cell type-dependent
differences in Rac-induced ROS formation.
In conclusion, our results suggest that Rac-induced
transcriptional upregulation of TIMP-1 takes place via
the following signaling cascade: Rac fi ROS (H
2
O
2
)
formation fi ERK1,2 fi AP-1 fi TIMP-1. Given the
well-established roles of AP-1 as a regulator of gene
transcription and of TIMP-1 as a regulator of cell prolif-
eration, apoptosis, invasion and metastasis, this newly
identified signaling pathway might play a major role in
different Rac-dependent cell biological processes.
Experimental procedures
Cell lines, stable gene transfection, and cell
culture conditions
The human clear cell RCC cell line, clearCa-28, and the
human colon carcinoma cell line, DusCol-1B, stably trans-
fected by retroviral transduction with either empty vector

(pLZRS), an active mutant of Tiam1 (C1199-Tiam1) [53] or
myc-epitope-tagged, constitutively active V12-Rac1, have
been described previously [33]. In the present study, C1199-
Tiam1, comprising the C-terminal 1199 amino acids, was
used instead of full-length (FL)-Tiam1, because this protein
is more stable and active than FL-Tiam1 [53,66,67]. Stably
transfected cell lines were maintained in DMEM supple-
A
B
Fig. 6. Rac-induced extracellular signal-related kinase (ERK) 1,2 acti-
vation is reactive oxygen species (ROS)-dependent. (A) In mock-
transfected renal cell carcinoma (RCC) cells, H
2
O
2
(300 lM, 30 min)
induces activation of ERK1,2, which can be inhibited by active, but
not by heat-inactivated (hia), catalase (1 mgÆmL
)1
each), as shown
by immunoblotting. Catalase and hia-catalase, respectively, were
applied 30 min prior to adding H
2
O
2
. Results are representative of
three independent experiments. (B) No effect of PD098059 (PD)
and U0126 (U) on C1199-Tiam1-induced ROS formation. ROS for-
mation was determined by measuring oxidation of an H
2

O
2
-sensi-
tive probe, 2¢,7¢-dichlorodihydrofluorescein, in mock-transfected
and C1199-Tiam1-transfected cells (a.u., arbitrary units). The data
shown are means ± standard deviations of an experiment per-
formed in triplicate, and are representative of two independent
experiments.
R. Engers et al. Rac signaling towards TIMP-1
FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS 4763
mented with 10% fetal bovine serum (both Sigma, Tauf-
kirchen, Germany) and antibiotics. G418 (Sigma) was used
as a selection marker for the presence of pLZRS at a con-
centration of 500 lgÆmL
)1
. HepG2 cells were obtained from
the German Collection of microorganisms and cell cultures
(DSMZ, Braunschweig, Germany) and were maintained in
RPMI 1640 medium (Sigma) supplemented with 10% fetal
bovine serum and antibiotics. Murine C3H10T1 ⁄ 2 fibro-
blasts were cultured in minimal essential medium with Ear-
le’s salt and l-glutamine (Invitrogen, Karlsruhe, Germany)
containing 10% fetal bovine serum (Invitrogen) and anti-
biotics. Cells were incubated at 37 °C in an atmosphere of
5% CO
2
.
Reagents and antibodies
Specific inhibitors of p38 MAPK (SB203580), as well as
specific inhibitors of the ERK1,2 activators, MEK1 and

MEK2 (PD098059 and U0126), were purchased from
New England Biolabs (Frankfurt am Main, Germany).
Anisomycin and bovine liver catalase were obtained from
Sigma (Taufkirchen, Germany). Phospho-specific antibodies
against p38 MAPK and ERK1,2, as well as antibodies
against total p38 MAPK, ERK1,2, JNK1 and Tiam1, were
obtained from Santa Cruz Biotechnology (Santa Cruz, CA,
USA). The polyclonal TIMP-1-specific antibody was
purchased from Chemicon (Chandler’s Ford, UK). Myc-
epitope-tagged V12-Rac1 was detected using monoclonal
antibody 9E10.
Determination of MAPK activity and
immunoblotting
Protein expression of Tiam1 and myc-epitope-tagged V12-
Rac1 was analyzed by immunoblotting as previously des-
cribed [33]. To analyze expression and ⁄ or phosphorylation
levels of other cellular proteins, cells were maintained for
24 h in medium supplemented with 0.5% fetal bovine
serum and lysed in 200 lL of Laemmli sample buffer. Sam-
ples were then sonicated, and proteins were separated by
SDS ⁄ PAGE. Activation of p38 MAPK and ERK1,2 was
determined by immunoblotting with antibodies specific for
phosphorylated, activated forms of these kinases. To com-
pare phosphorylation levels of proteins with their respective
total expression levels, membranes were stripped by incuba-
tion in a buffer containing 62.5 mm Tris ⁄ HCl, 2% (w ⁄ v)
sodium dodecylsulfate (SDS) and 100 mm b-mercaptoetha-
nol for 30 min at 50 °C, blocked, and reincubated with
antibodies specific for the respective total proteins. For
detection, an enhanced chemiluminescence detection system

(Amersham, Munich, Germany) was used. As a positive
control for p38 MAPK activation, mock-transfected cells
were treated for 1 h with anisomycin (1 lgÆmL
)1
).
To determine JNK activity, cells were serum-starved
(0.5% fetal bovine serum) for 24 h, lysed in a buffer con-
taining 1% Triton X-100, 150 mm NaCl, 20 mm Tris ⁄ HCl
(pH 7.5), 1 mm EDTA, 1 mm EGTA, 1 mm sodium
orthovanadate, 2.5 mm sodium pyrophosphate, 1 mm
b-glycerophosphate, 1 mm phenylmethylsulfonyl fluoride,
and 1 lgÆmL
)1
leupeptin, and subsequently frozen for at
least 2 h at )80 °C. Cell debris was removed by centrifu-
gation (20 800 g, 20 min, Mikro22R, Hettich, rotor type
1151), and protein concentrations were determined by
the Bradford method. JNK was immunoprecipitated from
equal amounts (200 lg each) of total protein by means of a
polyclonal anti-JNK1 antibody (C-17; Santa Cruz Biotech-
nology). Then, cell lysates were incubated for 2 h on ice
with the antibody to JNK1, and immunocomplexes were
captured by overnight incubation with protein A agarose
beads (Biomol, Hamburg, Germany) at 4 °C and sub-
sequent centrifugation (1000 g, 2 min). After removal of
supernatants, JNK activity was determined by incubating
protein A agarose–JNK complexes with 50 lLof10mm
Tris ⁄ HCl buffer (pH 7.5), containing 150 mm NaCl,
10 mm MgCl
2

, 0.5 mm dithiothreitol, 15 lm ATP, 5 lCi
[c-
32
P]ATP, and 1 mgÆmL
)1
glutathione S-transferase
(GST)–cJun (1–79; Alexis, Gruenberg, Germany) at 37 °C
for 25 min. The reaction was stopped by adding 50 lL
of Laemmli buffer. After denaturation at 90 °C, samples
were centrifuged and separated by 10% SDS ⁄ PAGE. After
electrophoresis, gels were stained, fixed and dried. JNK
activity was determined as phosphorylation of GST–cJun
by means of a phosphoimager. As a positive control, mock-
transfected cells were treated for 1 h with anisomycin
(1 lgÆmL
)1
).
GTPase activity assays
GTPase activity assays were essentially performed as des-
cribed by Sander et al. [54], with the exception that instead
of GST–PAK-CRIB, a biotinylated peptide corresponding
to the CRIB domain of PAK [68], kindly provided by JG
Collard (Netherlands Cancer Institute, Amsterdam, The
Netherlands), was used to precipitate active Rac and active
Cdc42. Briefly, cells were washed and then lysed with a 1%
Nonidet P-40 buffer containing 2 lgÆmL
)1
CRIB peptide.
Cell lysates were then incubated with bacterially produced
GST-C21, containing the Rho-binding domain of the Rho

effector protein Rhotekin [54,69], bound to glutathione-
coupled Sepharose beads. Active Rho–GST-C21 complexes
were precipitated by centrifugation (1000 g, 2 min), and the
remaining supernatants were used to precipitate active
Rac–CRIB and active Cdc42–CRIB complexes with strept-
avidin–agarose in a second step. The beads and proteins
bound to the beads were washed in an excess of lysis
buffer and eluted in SDS sample buffer. Total lysates and
precipitates were analyzed on western blots using anti-
bodies against Rac1, Cdc42, and Rho (all monoclonal anti-
bodies from BD Transduction Laboratories, Heidelberg,
Germany).
Rac signaling towards TIMP-1 R. Engers et al.
4764 FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS
Oxidation of dichlorodihydrofluorescein
Cells were incubated in medium supplemented with 0.5%
fetal bovine serum for 24 h and subsequently treated with
100 lm 2¢,7¢-dichlorodihydrofluorescein diacetate (Sigma)
for 2 h. After being washed twice, cells were collected and
lysed by sonication and one brief freeze–thaw cycle. Lysates
were cleared by brief centrifugation (1000 g, 5 min), and di-
chlorofluorescein fluorescence was determined from the sup-
ernatants on a Perkin-Elmer LS-5 luminometer (excitation
at 498 nm, emission at 522 nm) (Perkin-Elmer, Rodgan-
Ju
¨
gesheim, Germany). Fluorescence data were normalized
against supernatant protein concentrations as determined
by the Bradford method. Catalase (1 mgÆmL
)1

), specific
MEK1,2 inhibitors (25 lm PD098059, 10 lm U0126) or
dimethylsulfoxide (0.38%), the latter of which served as
control for MEK1,2 inhibitors, were applied to serum-
starved cells 1 h prior to adding 2,7-dichlorodihydrofluo-
rescein diacetate.
Quantitative TIMP-1 immunoassays and TIMP-1
immunoblotting
Quantitative TIMP-1 immunoassays were performed as
recently described [33]. Briefly, cells were incubated for 48 h
either in serum-free medium (Optimem) (Gibco, Karlsruhe,
Germany) or serum-free medium supplemented with differ-
ent compounds as indicated. This incubation time was
required for mock-transfected and C1199-Tiam1-transfected
cells to produce measurable amounts of TIMP-1. Superna-
tants were collected and calibrated with the cell numbers.
Subsequently, concentrations of secreted TIMP-1 proteins
were determined by a commercially available TIMP-1-speci-
fic immunoassay kit (Chemicon), according to the manufac-
turer’s instructions. Differences in TIMP-1 secretion were
statistically analyzed by t-test. To verify some of the results
obtained by quantitative immunoassays and to determine
whether the effects of C1199-Tiam1 and V12-Rac1 on
TIMP-1 secretion are cell density-dependent, TIMP-1 secre-
tion was analyzed by immunoblotting as described [33].
Transient transfections, plasmids, and reporter
gene assays
To identify the regulatory elements within the human
TIMP-1 promoter that mediate Tiam1 ⁄ Rac-induced TIMP-
1 expression, CAT reporter gene assays were performed as

described previously [33]. Because of the extremely low
transient transfection efficiencies (< 1%) of clearCa-28 and
DusCol-1B cells, HepG2 cells were used for these experi-
ments [33]. Briefly, HepG2 cells were transiently transfected
using the FUGENE 6 Reagent (Roche, Mannheim,
Germany). Luciferase assays were performed with a high-
sensitivity luciferase assay Kit (Roche) using a Beckmann
scintillation counter (Beckmann, Munich, Germany).
Twenty microliters of (diluted) lysate were mixed with
100 lL of reaction buffer, and all samples were measured
successively.
Tiam1 expression constructs were kindly provided by JG
Collard (Netherlands Cancer Institute, Amsterdam, The
Netherlands), and cloning procedures have been described
previously [66,67,70]. The human TIMP-1 promoter–CAT
constructs )102 ⁄ +95, ) 738 ⁄ +95 and )1718 ⁄ +95 have
been previously described [46]. The numbers of these con-
structs refer to the respective 5¢-ends and 3¢-ends relative to
the transcriptional start point, which was defined as +1. In
order to study the involvement of transcription factor-bind-
ing sites using the highly sensitive luciferase reporter gene
assay (see below), the TIMP-1 promoter–CAT constructs
)102 ⁄ +95, bearing inactivating point mutations of AP-1
and ⁄ or PEA-3 ⁄ ETS-1 sites, *AP-1, *PEA-3 ⁄ ETS-1 and
*AP-1 ⁄ *PEA-3 [46], as well as the intact )102 ⁄ +95 frag-
ment, were subcloned into the pGL3-luciferase vector
(Promega, Mannheim, Germany). A TIMP-1 promoter–
CAT construct, DUTE-1, lacking 32 base pairs around the
UTE-1 site, was generated by PCR (with the upper primer
containing a deletion from )66 to )34) from the )102 ⁄ +95

basal promoter construct. The constitutively active JNK
construct [71], as well as a hemagglutinin (HA)-epitope-
tagged dominant negative JNK construct [72], were kind
gifts of UR Rapp (Wuerzburg, Germany). HA-epitope
tagged dominant negative JNK was subcloned as a SalI ⁄
XhoI fragment into the XhoI site of pMT2SM. Upon tran-
sient transfection of these constructs, protein expression
levels were verified by immunoblotting.
EMSA
Prior to EMSAs, cells were serum-starved (0.5% fetal
bovine serum) for 24 h. To test the role of ERK1,2 and
ROS signaling in C1199-Tiam1-induced AP-1 binding, cells
were subsequently incubated in either serum-free medium
(Optimem) or serum-free medium supplemented with speci-
fic inhibitors of the ERK1,2 pathway (PD98059, 25 lm;
U0126, 10 lm) or with catalase (1 mgÆmL
)1
) as indicated.
For EMSAs, 10 pmol of 21 base pair double-stranded
oligonucleotides, containing either the wild-type (TGGGTG
GATGAGTAATGCATC) (AP-1) or the mutated, and
hence inactivated (TGGGTGGAGGACTAATGCATC),
AP-1 (*AP-1) consensus sites ()92 ⁄ )86) of the human
TIMP-1 promoter were end-labeled with T4 polynucleotide
kinase (New England Biolabs) in the presence of 10 lCi of
[c-
32
P]ATP for 1 h at 37 °C and subsequently purified with
a nucleotide removal kit (Qiagen, Hilden, Germany).
For binding reactions, 5 lg of nuclear extracts of treated

or nontreated mock-transfected and C1199-Tiam1-transfect-
ed RCC cells were incubated with 0.4 pmol of the purified
AP-1 and *AP-1 oligonucleotides, respectively, and 2 lgof
R. Engers et al. Rac signaling towards TIMP-1
FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS 4765
poly-dIdC in EMSA buffer (20 mm Hepes, pH 7.9, 90 mm
KCl, 0.1 mm MgCl
2
, 0.5 lm EDTA, 6.25 lm dithiothreitol,
6.25% glycerol) for 30 min on ice. The specificity of bind-
ing of nuclear extracts to labeled AP-1 oligonucleotide was
verified by competition with nonlabeled wild-type and
mutated AP-1 oligonucleotides, respectively, applied in
10-fold excess. In a control reaction, the labeled AP-1
oligonucleotide was incubated without nuclear extracts.
Samples were mixed with loading buffer (25 mm Tris ⁄ HCl,
0.02% bromophenol blue, 4% glycerol), loaded onto a 6%
nondenaturating polyacrylamide gel, and separated in
0.5 · TBE (25 mm Tris ⁄ borate, 0.5 mm EDTA, pH 7.8) at
200 V for 2.5 h. Gels were dried and exposed to Kodak
(Rochester, NY, USA) X-Omats films at )70 °C.
Acknowledgements
We thank S. Traenkner and S. Khalil for their excellent
technical assistance, and M. Ringler and M. Bellack for
preparation of some of the figures. This work was sup-
ported by the Deutsche Forschungsgemeinschaft (Ga
326 ⁄ 4-1, GA 326 ⁄ 4-3, SFB 503 ⁄ B1) and the Dr Mildred
Scheel Stiftung fu
¨
r Krebsforschung (10-1582-En I).

References
1 Van Aelst L & D’Souza-Schorey C (1997) Rho GTPases
and signaling networks. Genes Dev 11, 2295–2322.
2 Hall A (1998) Rho GTPases and the actin cytoskeleton.
Science 279, 509–514.
3 Evers EE, van der Kammen RA, ten Klooster JP &
Collard JG (2000) Rho-like GTPases in tumor cell inva-
sion. Methods Enzymol 325, 403–415.
4 Evers EE, Zondag GC, Malliri A, Price LS, ten Kloo-
ster JP, van der Kammen RA & Collard JG (2000) Rho
family proteins in cell adhesion and cell migration. Eur
J Cancer 36, 1269–1274.
5 Olson MF, Ashworth A & Hall A (1995) An essential
role for Rho, Rac, and Cdc42 GTPases in cell cycle pro-
gression through G1. Science 269, 1270–1272.
6 Murga C, Zohar M, Teramoto H & Gutkind JS (2002)
Rac1 and RhoG promote cell survival by the activa-
tion of PI3K and Akt, independently of their ability
to stimulate JNK and NF-kappaB. Oncogene 21,
207–216.
7 Norman JC, Price LS, Ridley AJ & Koffer A (1996)
The small GTP-binding proteins, Rac and Rho, regulate
cytoskeletal organization and exocytosis in mast cells by
parallel pathways. Mol Biol Cell 7, 1429–1442.
8 Qiu RG, Chen J, Kirn D, McCormick F & Symons M
(1995) An essential role for Rac in Ras transformation.
Nature 374, 457–459.
9 Coso OA, Chiariello M, Yu JC, Teramoto H, Crespo P,
Xu N, Miki T & Gutkind JS (1995) The small GTP-
binding proteins Rac1 and Cdc42 regulate the activity

of the JNK ⁄ SAPK signaling pathway. Cell 81, 1137–
1146.
10 Bagrodia S, Taylor SJ, Creasy CL, Chernoff J & Cerione
RA (1995) Identification of a mouse p21Cdc42 ⁄ Rac
activated kinase. J Biol Chem 270, 22731–22737 (erratum
appears in J Biol Chem 271, 1250).
11 Hill CS, Wynne J & Treisman R (1995) The Rho family
GTPases RhoA, Rac1, and CDC42Hs regulate tran-
scriptional activation by SRF. Cell 81, 1159–1170.
12 Westwick JK, Lambert QT, Clark GJ, Symons M, Van
Aelst L, Der Pestell RG & Der CJ (1997) Rac regula-
tion of transformation, gene expression, and actin orga-
nization by multiple, PAK-independent pathways. Mol
Cell Biol 17, 1324–1335.
13 Perona R, Montaner S, Saniger L, Sanchez PI, Bravo R
& Lacal JC (1997) Activation of the nuclear factor-kap-
paB by Rho, CDC42, and Rac-1 proteins. Genes Dev
11, 463–475.
14 Simon AR, Vikis HG, Stewart S, Fanburg BL, Cochran
BH & Guan KL (2000) Regulation of STAT3 by direct
binding to the Rac1 GTPase. Science 290, 144–147.
15 Woo CH, Lee ZW, Kim BC, Ha KS & Kim JH (2000)
Involvement of cytosolic phospholipase A2, and the
subsequent release of arachidonic acid, in signalling by
rac for the generation of intracellular reactive oxygen
species in rat-2 fibroblasts. Biochem J 348, 525–530.
16 Aspenstrom P (1999) Effectors for the Rho GTPases.
Curr Opin Cell Biol 11, 95–102.
17 Bishop AL & Hall A (2000) Rho GTPases and their
effector proteins. Biochem J 348, 241–255.

18 Cotteret S & Chernoff J (2002) The evolutionary history
of effectors downstream of Cdc42 and Rac.
Genome
Biol 3, doi:10.1186/gb-2002-3-2-reviews0002.
19 Abo A, Pick E, Hall A, Totty N, Teahan CG & Segal
AW (1991) Activation of the NADPH oxidase involves
the small GTP-binding protein p21rac1. Nature 353,
668–670.
20 Joneson T & Bar SD (1998) A Rac1 effector site con-
trolling mitogenesis through superoxide production.
J Biol Chem 273, 17991–17994.
21 Clerk A, Pham FH, Fuller SJ, Sahai E, Aktories K,
Marais R, Marshall C & Sugden PH (2001) Regulation
of mitogen-activated protein kinases in cardiac myocytes
through the small G protein Rac1. Mol Cell Biol 21,
1173–1184.
22 Woo CH, You HJ, Cho SH, Eom YW, Chun JS, Yoo
YJ & Kim JH (2002) Leukotriene B(4) stimulates Rac-
ERK cascade to generate reactive oxygen species that
mediates chemotaxis. J Biol Chem 277, 8572–8578.
23 Eblen ST, Slack JK, Weber MJ & Catling AD (2002)
Rac-PAK signaling stimulates extracellular signal-regu-
lated kinase (ERK) activation by regulating formation of
MEK1–ERK complexes. Mol Cell Biol 22, 6023–6033.
Rac signaling towards TIMP-1 R. Engers et al.
4766 FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS
24 Minden A, Lin A, Claret FX, Abo A & Karin M (1995)
Selective activation of the JNK signaling cascade and
c-Jun transcriptional activity by the small GTPases Rac
and Cdc42Hs. Cell 81, 1147–1157.

25 Teramoto H, Crespo P, Coso OA, Igishi T, Xu N &
Gutkind JS (1996) The small GTP-binding protein rho
activates c-Jun N-terminal kinases ⁄ stress-activated pro-
tein kinases in human kidney 293T cells. Evidence for a
Pak-independent signaling pathway. J Biol Chem 271,
25731–25734.
26 Schuringa JJ, Dekker LV, Vellenga E & Kruijer W
(2001) Sequential activation of Rac-1, SEK-1 ⁄ MKK-4,
and protein kinase Cdelta is required for interleukin-6-
induced STAT3 Ser-727 phosphorylation and transacti-
vation. J Biol Chem 276, 27709–27715.
27 Frost JA, Xu S, Hutchison MR, Marcus S & Cobb MH
(1996) Actions of Rho family small G proteins and p21-
activated protein kinases on mitogen-activated protein
kinase family members. Mol Cell Biol 16, 3707–3713.
28 Frost JA, Steen H, Shapiro P, Lewis T, Ahn N, Shaw
PE & Cobb MH (1997) Cross-cascade activation of
ERKs and ternary complex factors by Rho family pro-
teins. EMBO J 16, 6426–6438.
29 Gjoerup O, Lukas J, Bartek J & Willumsen BM (1998)
Rac and Cdc42 are potent stimulators of E2F-depen-
dent transcription capable of promoting retinoblastoma
susceptibility gene product hyperphosphorylation. J Biol
Chem 273, 18812–18818.
30 Dalton TP, Shertzer HG & Puga A (1999) Regulation
of gene expression by reactive oxygen. Annu Rev Phar-
macol Toxicol 39, 67–101.
31 Garrington TP & Johnson GL (1999) Organization and
regulation of mitogen-activated protein kinase signaling
pathways. Curr Opin Cell Biol 11, 211–218.

32 Mercurio F & Manning AM (1999) Multiple signals
converging on NF-kappaB. Curr Opin Cell Biol 11,
226–232.
33 Engers R, Springer E, Michiels F, Collard JG & Gab-
bert HE (2001) Rac affects invasion of human renal cell
carcinomas by up-regulating tissue inhibitor of metallo-
proteinases (TIMP)-1 and TIMP-2 expression. J Biol
Chem 276, 41889–41897.
34 Brew K, Dinakarpandian D & Nagase H (2000) Tissue
inhibitors of metalloproteinases: evolution, structure
and function. Biochim Biophys Acta 1477, 267–283.
35 Mannello F & Gazzanelli G (2001) Tissue inhibitors of
metalloproteinases and programmed cell death: conun-
drums, controversies and potential implications. Apopto-
sis 6, 479–482.
36 Henriet P, Blavier L & Declerck YA (1999) Tissue inhi-
bitors of metalloproteinases (TIMP) in invasion and
proliferation. APMIS 107, 111–119.
37 Edwards DR, Rocheleau H, Sharma RR, Wills AJ,
Cowie A, Hassell JA & Heath JK (1992) Involvement
of AP1 and PEA3 binding sites in the regulation of
murine tissue inhibitor of metalloproteinases-1 (TIMP-
1) transcription. Biochim Biophys Acta 1171, 41–55.
38 Ulisse S, Farina AR, Piersanti D, Tiberio A, Cappabi-
anca L, D’Orazi G, Jannini EA, Malykh O,
Stetler-Stevenson WG & D’Armiento M (1994) Follicle-
stimulating hormone increases the expression of tissue
inhibitors of metalloproteinases TIMP-1 and TIMP-2
and induces TIMP-1 AP-1 site binding complex(es) in
prepubertal rat Sertoli cells. Endocrinology 135, 2479–

2487.
39 Uchijima M, Sato H, Fujii M & Seiki M (1994)
Tax proteins of human T-cell leukemia virus type 1
and 2 induce expression of the gene encoding
erythroid-potentiating activity (tissue inhibitor of
metalloproteinases-1, TIMP-1). J Biol Chem 269
,
14946–14950.
40 Silacci P, Dayer JM, Desgeorges A, Peter R, Manueddu
C & Guerne PA (1998) Interleukin (IL)-6 and its soluble
receptor induce TIMP-1 expression in synoviocytes and
chondrocytes, and block IL-1-induced collagenolytic
activity. J Biol Chem 273, 13625–13629.
41 Botelho FM, Edwards DR & Richards CD (1998)
Oncostatin M stimulates c-Fos to bind a transcription-
ally responsive AP-1 element within the tissue inhibitor
of metalloproteinase-1 promoter. J Biol Chem 273,
5211–5218.
42 Kadri Z, Petitfrere E, Boudot C, Freyssinier JM, Fichel-
son S, Mayeux P, Emonard H, Hornebeck W, Haye B
& Billat C (2000) Erythropoietin induction of tissue
inhibitors of metalloproteinase-1 expression and secre-
tion is mediated by mitogen-activated protein kinase
and phosphatidylinositol 3-kinase pathways. Cell
Growth Differ 11, 573–580.
43 Bigg HF, McLeod R, Waters JG, Cawston TE & Clark
IM (2000) Mechanisms of induction of human tissue
inhibitor of metalloproteinases-1 (TIMP-1) gene expres-
sion by all-trans retinoic acid in combination with basic
fibroblast growth factor. Eur J Biochem 267, 4150–4156.

44 Coulombe B, Ponton A, Daigneault L, Williams BR &
Skup D (1988) Presence of transcription regulatory ele-
ments within an intron of the virus-inducible murine
TIMP gene. Mol Cell Biol 8, 3227–3234.
45 Bugno M, Graeve L, Gatsios P, Koj A, Heinrich PC,
Travis J & Kordula T (1995) Identification of the inter-
leukin-6 ⁄ oncostatin M response element in the rat tissue
inhibitor of metalloproteinases-1 (TIMP-1) promoter.
Nucleic Acids Res 23, 5041–5047.
46 Clark IM, Rowan AD, Edwards DR, Bech HT, Mann
DA, Bahr MJ & Cawston TE (1997) Transcriptional
activity of the human tissue inhibitor of metalloprotei-
nases 1 (TIMP-1) gene in fibroblasts involves elements
in the promoter, exon 1 and intron 1. Biochem J 324,
611–617.
47 Logan SK, Garabedian MJ, Campbell CE & Werb Z
(1996) Synergistic transcriptional activation of the tissue
R. Engers et al. Rac signaling towards TIMP-1
FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS 4767
inhibitor of metalloproteinases-1 promoter via func-
tional interaction of AP-1 and Ets-1 transcription fac-
tors. J Biol Chem 271, 774–782.
48 Bahr MJ, Vincent KJ, Arthur MJ, Fowler AV, Smart
DE, Wright MC, Clark IM, Benyon RC, Iredale JP &
Mann DA (1999) Control of the tissue inhibitor of
metalloproteinases-1 promoter in culture-activated rat
hepatic stellate cells: regulation by activator protein-1
DNA binding proteins. Hepatology 29, 839–848.
49 Jaworski J, Biedermann IW, Lapinska J, Szklarczyk A,
Figiel I, Konopka D, Nowicka D, Filipkowski RK, Het-

man M, Kowalczyk A & Kaczmarek L (1999) Neuronal
excitation-driven and AP-1-dependent activation of tissue
inhibitor of metalloproteinases-1 gene expression in
rodent hippocampus. J Biol Chem 274, 28106–28112.
50 Norman JT, Clark IM & Garcia PL (2000) Hypoxia
promotes fibrogenesis in human renal fibroblasts. Kidney
Int 58, 2351–2366.
51 Trim JE, Samra SK, Arthur MJ, Wright MC, McAulay
M, Beri R & Mann DA (2000) Upstream tissue inhibi-
tor of metalloproteinases-1 (TIMP-1) element-1, a novel
and essential regulatory DNA motif in the human
TIMP-1 gene promoter, directly interacts with a 30-kDa
nuclear protein. J Biol Chem 275, 6657–6663.
52 Bertrand-Philippe M, Ruddell RG, Arthur MJ, Thomas
J, Mungalsingh N & Mann DA (2004) Regulation of tis-
sue inhibitor of metalloproteinase 1 gene transcription by
RUNX1 and RUNX2. J Biol Chem 279, 24530–24539.
53 Hordijk PL, ten Klooster JP, van der Kammen RA,
Michiels F, Oomen LC & Collard JG (1997) Inhibition
of invasion of epithelial cells by Tiam1-Rac signaling.
Science 278, 1464–1466.
54 Sander EE, ten Klooster JP, van Delft S, van der Kam-
men RA & Collard JG (1999) Rac downregulates Rho
activity: reciprocal balance between both GTPases
determines cellular morphology and migratory behavior.
J Cell Biol 147, 1009–1022.
55 Nimnual AS, Taylor LJ & Bar-Sagi D (2003) Redox-
dependent downregulation of Rho by Rac. Nat Cell Biol
5, 236–241.
56 Jiang G, Dallas-Yang Q, Liu F, Moller DE & Zhang

BB (2003) Salicylic acid reverses phorbol 12-myristate-
13-acetate (PMA)- and tumor necrosis factor alpha
(TNFalpha)-induced insulin receptor substrate 1 (IRS1)
serine 307 phosphorylation and insulin resistance in
human embryonic kidney 293 (HEK293) cells. J Biol
Chem 278, 180–186.
57 Bertelsen M & Sanfridson A (2005) Inflammatory path-
way analysis using a high content screening platform.
Assay Drug Dev Technol 3, 261–271.
58 Whitmarsh AJ & Davis RJ (1996) Transcription factor
AP-1 regulation by mitogen-activated protein kinase sig-
nal transduction pathways. J Mol Med 74, 589–607.
59 Sohara N, Trojanowska M & Reuben A (2002)
Oncostatin M stimulates tissue inhibitor of metallo-
proteinase-1 via a MEK-sensitive mechanism in human
myofibroblasts. J Hepatol 36, 191–199.
60 Sato T, Koike L, Miyata Y, Hirata M, Mimaki Y,
Sashida Y, Yano M & Ito A (2002) Inhibition of activa-
tor protein-1 binding activity and phosphatidylinositol
3-kinase pathway by nobiletin, a polymethoxy flavo-
noid, results in augmentation of tissue inhibitor of
metalloproteinases-1 production and suppression of pro-
duction of matrix metalloproteinases-1 and -9 in human
fibrosarcoma HT-1080 cells. Cancer Res 62, 1025–1029.
61 Delanian S, Martin M, Bravard A, Luccioni C & Lefaix
J (2001) Cu ⁄ Zn superoxide dismutase modulates pheno-
typic changes in cultured fibroblasts from human skin
with chronic radiotherapy damage. Radiother Oncol 58,
325–331.
62 Yang ZZ & Zou AP (2003) Homocysteine enhances

TIMP-1 expression and cell proliferation associated with
NADH oxidase in rat mesangial cells. Kidney Int 63,
1012–1020.
63 Chung YW, Jeong DW, Won JY, Choi EJ, Choi YH &
Kim IY (2002) H
2
O
2
-induced AP-1 activation and its
effect on p21(WAF1 ⁄ CIP1)-mediated G2 ⁄ M arrest in a
p53-deficient human lung cancer cell. Biochem Biophys
Res Commun 293, 1248–1253.
64 Daou GB & Srivastava AK (2004) Reactive oxygen
species mediate endothelin-1-induced activation of
ERK1 ⁄ 2, PKB, and Pyk2 signaling, as well as protein
synthesis, in vascular smooth muscle cells. Free Radic
Biol Med 37, 208–215.
65 Kim YK, Bae GU, Kang JK, Park JW, Lee EK, Lee
HY, Choi WS, Lee HW & Han JW (2005) Cooperation
of H
2
O
2
-mediated ERK activation with Smad pathway
in TGF-beta1 induction of p21 (WAF1 ⁄ Cip1). Cell
Signal 18, 236–243.
66 Michiels F, Stam JC, Hordijk PL, van der Kammen
RA, Salinas PC, Feltkamp CA & Collard JG (1997)
Regulated membrane localization of Tiam1, mediated
by the NH2-terminal pleckstrin homology domain, is

required for Rac-dependent membrane ruffling and
C-Jun NH2-terminal kinase activation. J Cell Biol 137,
387–398.
67 Michiels F, Habets GG, Stam JC, van der Kammen
RA & Collard JG (1995) A role for Rac in Tiam1-
induced membrane ruffling and invasion. Nature 375,
338–340.
68 Price LS, Langeslag M, ten Klooster JP, Hordijk PL,
Jalink K & Collard JG (2003) Calcium signaling regu-
lates translocation and activation of Rac. J Biol Chem
278, 39413–39421.
69 Reid T, Furuyashiki T, Ishizaki T, Watanabe G,
Watanabe N, Fujisawa K, Morii N, Madaule P &
Narumiya S (1996) Rhotekin, a new putative target for
Rho bearing homology to a serine ⁄ threonine kinase,
PKN, and rhophilin in the rho-binding domain. J Biol
Chem 271, 13556–13560.
Rac signaling towards TIMP-1 R. Engers et al.
4768 FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS
70 Stam JC, Sander EE, Michiels F, van Leeuwen FN,
Kain HE, van der Kammen RA & Collard JG (1997)
Targeting of Tiam1 to the plasma membrane requires
the cooperative function of the N-terminal pleckstrin
homology domain and an adjacent protein interaction
domain. J Biol Chem 272, 28447–28454.
71 Otto IM, Raabe T, Rennefahrt UE, Bork P, Rapp UR
& Kerkhoff E (2000) The p150-Spir protein provides a
link between c-Jun N-terminal kinase function and actin
reorganization. Curr Biol 10, 345–348.
72 Hoffmeyer A, Grosse-Wilde A, Flory E, Neufeld B,

Kunz M, Rapp UR & Ludwig S (1999) Different mito-
gen-activated protein kinase signaling pathways coop-
erate to regulate tumor necrosis factor alpha gene
expression in T lymphocytes. J Biol Chem 274, 4319–
4327.
R. Engers et al. Rac signaling towards TIMP-1
FEBS Journal 273 (2006) 4754–4769 ª 2006 The Authors Journal compilation ª 2006 FEBS 4769