![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 14, Issue 5, 1757-1768, May 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Cell Biology, University of Alberta, Edmonton, Alberta, Canada T6G 2H7
Submitted August 12, 2002;
Revised November 27, 2002;
Accepted January 30, 2003
Monitoring Editor: Mary Beckerle
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Coordinated Assembly of Tight Junction and Adherens Junction
A growing body of evidence indicates that the individual junctions within
the junctional complex are jointly regulated in assembly and function. These
data indicate that the first step in junctional complex formation requires
E-cadherin–mediated cell adhesion
(Gumbiner et al.,
1988
), followed by transient ZO-1 localization to the adherens
junction before recruitment to the newly forming tight junction
(Rajasekaran et al.,
1996). However, there is also data to suggest that this hierarchy
of junctional complex assembly is not absolute. Balda et al.
(1993) observed that treating
Madin-Darby canine kidney (MDCK) cells with a diacylglycerol analog in low
Ca2+ media induced redistribution of ZO-1, but not
E-cadherin, to the junctional membrane. More recently, Troxell et al.
(2000) have shown that tight
junction assembly is extensive in MDCK cells expressing a mutant E-cadherin
protein lacking the extracellular domain required for cell-cell adhesion.
Recent data from our laboratory substantiated the notion of cross-talk
between the tight junction and adherens junction
(Wittchen et al.,
2000). We studied the effects of exogenously expressing the
amino-terminal half of the tight junction protein ZO-3 (NZO-3) on tight
junction physiology and junctional complex assembly. Expression of this
construct in MDCK cells caused a significant delay in transepithelial
electrical resistance (TER) recovery in a calcium switch assay, indicating
that assembly of tight junction barrier properties was perturbed. Expression
of NZO-3 also disrupted the normal recruitment of tight junction proteins and
the adherens junction proteins E-cadherin and 
-catenin to the plasma
membrane during the early stages of junctional complex formation. While many
investigators have provided evidence that disruption of adherens junction
components negatively regulates the tight junction, these results show that
the opposite is also the case: expression of a mutant tight junction protein
which disrupts tight junction assembly and physiology can also impede adherens
junction assembly.
Actin Cytoskeleton during Junctional Assembly and Barrier
Function
We also looked at the effects of NZO-3 expression on the actin cytoskeleton
and observed that recruitment of perijunctional F-actin during the junctional
assembly process was similarly delayed
(Wittchen et al.,
2000). It is well established that there is reciprocity between
actin filament integrity and junctional physiology. For example,
pharmacological perturbation of the F-actin cytoskeleton by cytochalasin D
causes a drop in TER across an MDCK cell monolayer
(Stevenson and Begg, 1994).
Likewise, disassembly of the junctional complex by incubation in calcium-free
media disrupts the junction-associated peripheral actin ring
(Wittchen et al.,
2000).
Dynamic regulation of the actin cytoskeleton is controlled in part by the
Rho family of GTPases, members of the Ras superfamily of small GTPases.
Remodeling the actin cytoskeleton is critical to cell morphology changes,
cytokinesis, substrate adhesion, cell spreading, and migration
(Kaibuchi et al.,
1999). A key feature of all GTPases is their cyclical activation
and deactivation by the binding of GTP followed by the intrinsic GTPase
conversion to GDP, thus allowing them to act as molecular switches.
The Rho GTPase family members RhoA, Rac1, and Cdc42 each play a specific
role during cytoskeletal reorganization. RhoA controls the formation of stress
fibers and focal adhesions based on the observation that microinjection of
constitutively active RhoA into fibroblasts induces the formation of stress
fibers and focal adhesions (Ridley and
Hall, 1992). Furthermore, microinjection of the catalytic domain
of Rho-kinase, a downstream effector of RhoA, into fibroblasts also caused the
formation of stress fibers and focal adhesions
(Amano et al., 1997).
In general, the presence of abundant stress fibers and focal adhesions is
associated with a nonmotile phenotype characterized by strong cell substratum
attachment (Herman et al.,
1981; Ridley et al.,
1995; Nobes and Hall,
1999). Rac and Cdc42 activities are linked to the formation of
membrane protrusions at the leading edge of migrating cells. Rac promotes the
formation of lamellipodia, broad actin filament-based extensions of the plasma
membrane (Ridley et al.,
1992). Cdc42 promotes assembly of filopodia
(Nobes and Hall, 1995),
actin-based, spike-like protrusions at the cell periphery. Both lamellipodia
and filopodia are characteristic components of membrane ruffling, the dynamic
movement of plasma membrane at the free edge of cells engaged in cell
spreading and migration.
Rho GTPases and Junctional Complex Assembly and Function
There is a growing body of evidence that supports a role for the Rho
GTPases in assembly and physiological regulation of the junctional complex in
epithelial cells. An early observation that connected tight junction
physiology with RhoA activity showed that treatment of epithelial cells with
C3 transferase, an inhibitor of RhoA, decreased TER and increased paracellular
flux, indicating that tight junction barrier properties were compromised
(Nusrat et al.,
1995). This inhibition of RhoA also disrupted perijunctional actin
formation and caused a redistribution of ZO-1 and occludin away from the cell
surface (Nusrat et al.,
1995). More recently, Jou et al.
(1998) demonstrated that
expression of either dominant negative or constitutively active RhoA and Rac
in MDCK cells reduced TER and perturbed tight junction fence function,
indicated by the unrestricted diffusion of membrane lipids from the apical to
the lateral membrane.
The assembly of adherens junctions also seems to involve the RhoA pathway.
Inhibition of p160ROCK, a downstream effector of RhoA, prevents movement of
E-cadherin, an adherens junction protein, and the tight junction proteins ZO-1
and occludin to the plasma membrane during junctional complex assembly
(Walsh et al., 2001).
RhoA, Rac1, and Cdc42 have been implicated in promoting cell-cell adhesion
(Kuroda et al., 1997;
Takaishi et al.,
1997), and adherens junction formation seems to require RhoA and
Rac1 (Braga et al.,
1997; Sander et al.,
1999). Furthermore, Rac1 is colocalized with E-cadherin at sites
of cell-cell contact during adherens junction assembly and translocates to the
cytoplasm if junctions are disrupted by removing calcium
(Nakagawa et al.,
2001). This group and others also demonstrated that E-cadherin
mediated cell-cell adhesion stimulates Rac1 activation
(Nakagawa et al.,
2001; Noren et al.,
2001).
Two proteins that are found at the junctional complex and are linked to
signaling pathways involving Ras superfamily GTPases are p120 catenin and
AF-6. P120 catenin is known to play a role in cytoskeletal changes linked to
both cell junctions and cell motility
(Braga, 2000). P120 catenin has
dual functions within the cell that directly correlate to its localization.
When bound to E-cadherin at the adherens junction, p120 catenin promotes cell
adhesion, perhaps by cadherin clustering
(Yap et al., 1998).
However, increased cytoplasmic p120 catenin causes reduced stress fiber
formation and increased cell motility via modulation of the activity of the
Rho GTPases (Anastasiadis et al.,
2000; Noren et al.,
2000). AF-6 is a downstream target of Ras, which also binds to the
tight junction protein ZO-1 (Yamamoto
et al., 1997).
Because of the close association of actin with the junctional complex of
epithelial cells, it is becoming apparent how Rho family members might impact
the physiological regulation of fully formed junctions via the actin
cytoskeleton. For example, tight junction permeability is thought to be
regulated by contraction of the perijunctional actin ring through a
contractile force that subtly increases the tension generated between opposing
cell surfaces and results in opening of the tight junction
(Madara and Pappenheimer,
1987; Hecht et al.,
1996). One interesting hypothesis relating to this phenomenon is
that rapid cycling of RhoA and Rac between active and inactive forms fine
tunes the tension subjected on the peripheral actin ring, thus regulating the
degree of tight junction permeability (Jou
et al., 1998).
Our previous finding that exogenous expression of a mutant tight junction
protein affected actin remodeling during junctional assembly
(Wittchen et al.,
2000) led to further investigation of the actin cytoskeleton in
NZO-3–expressing MDCK cells. The results presented herein identify a
more global alteration of actin dynamics in these cells. We show that
expression of NZO-3 decreases the number of stress fibers and focal adhesions
and increases the rate of cell migration in a wound healing assay. We have
identified a potential molecular mechanism underlying these changes and
provide evidence for the involvement of RhoA and p120 catenin. The information
obtained from these studies sheds light on the mechanisms of cytoskeleton
organization by the Rho family of GTPases during junctional assembly and wound
healing, and how tight junction elements might influence these processes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
; Wittchen et al.,
1999,
2000). Immunofluorescence
experiments, wound healing assays, and RhoA activity assays were repeated with
two independent NZO-3/MDCK cell lines (clones D5 and B6), and two independent
CZO-3/MDCK cell lines (data from one line are shown).
Immunohistochemistry
For staining of subconfluent monolayers, MDCK cell lines were plated on
collagen-coated coverslips and allowed to grow until they reached a density of
50%. Cells were fixed and permeabilized using 2.5% paraformaldehyde on
ice for 30 min followed by incubation with 0.2% Triton X-100/Tris-buffered
saline (TBS)+ for 5 min at room temperature. The coverslips were blocked for
15 min at room temperature in TBS+/BLOTTO (5% skim milk powder in TBS +
Ca2+ and Mg2+), and costained with
anti-vinculin antibody (Zymed Laboratories, South San Francisco, CA) and
fluorescein isothiocyanate (FITC)-phalloidin (Sigma-Aldrich, St. Louis, MO) to
detect actin filaments. Vinculin localization was visualized with rhodamine
anti-mouse secondary antibodies (Jackson Immunoresearch Laboratories, West
Grove, PA). After washing 3 x 5 min with TBS +
Ca2+ and Mg2+, coverslips were
mounted on glass slides by using an antibleaching mounting media (VectaShield;
Vector Laboratories, Burlingame, CA), and viewed with an Axioskop fluorescence
microscope (Carl Zeiss, Thornwood, NY). Wounded monolayers were fixed 3 h
postwounding as described above and stained with rhodamine-phalloidin
(Sigma-Aldrich) to detect actin filaments. Images of wounded monolayers were
collected and processed identically; representative images shown.
Wound Healing Assays
In vitro wound healing assays were used to assess cell migration. Cells
were plated at varying densities, and dishes that had just reached confluence
were used. Two linear wounds were scratched in each dish of cells with a p200
pipet tip. Using a phase contrast microscope attached to an MD2 microscope
digitizer (Accustage, Minneapolis, MN) to measure stage position, the width of
the wound at two different points was measured over a 6-h time period. The
average migration rate was calculated by taking the total distance migrated
(in micrometers) divided by the total time (in hours). Data were plotted as
relative migration rate with the value for parental MDCK cells arbitrarily set
to 1 for comparison between experiments. For each cell line, two different
wounds were made per dish, and width was measured at two points per wound (n =
4). Data shown are representative of three to six independent experiments.
RhoA Activity Assays
We performed Rho activity assays as described previously
(Ren et al., 1999;
Noren et al., 2000)
with minor modifications. The glutathione S-transferase
(GST)-Rho-binding domain (RBD) construct (amino acids 7–89 from
rhotekin) was kindly provided by Dr. Keith Burridge (University of North
Carolina, Chapel Hill, NC). Cells were serum-starved overnight before
performing RhoA activity assays because serum activation of RhoA potentially
masks small but physiologically relevant differences in activity
(Noren et al., 2001).
An 80% confluent 10-cm dish of cells was washed in ice-cold
HEPES-buffered saline and lysed by scraping in 300 µl of RIPA buffer (50 mM
Tris, pH 7.2, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 500 mM
NaCl, 10 mM MgCl2, and protease inhibitors [1 µg/ml aprotinin, 1
µg/ml chymostatin, 1 µg/ml leu-peptin, 1 µg/ml pepstatin, and 1 mM
Pefabloc SC; Roche Diagnostics, Indianapolis, IN]). Lysates were clarified by
centrifugation at 12,000 x g at 4°C for 5 min, and equal
volumes of lysates were incubated with 30 µg of GST-RBD beads at 4°C
with rotation for 30 min. An aliquot of lysate was reserved for analysis of
total RhoA. Beads were washed four times with 1 ml of buffer B (TBS + 1%
Triton X-100, 150 mM NaCl, 10 mM MgCl2, and protease inhibitors).
The bound fraction (active RhoA) was analyzed by resuspending the beads in
2x gel sample buffer, boiling 5 min, and running on SDS-PAGE. Active
RhoA (bound fraction) and total RhoA were analyzed by Western blotting with an
anti-RhoA antibody (monoclonal antibody 26C4; Santa Cruz Biotechnology, Santa
Cruz, CA). The results were quantified by densitometry of multiple Western
blots from four independent experiments. RhoA activity was determined by
determining the ratio of the amount of RhoA sedimented by the GST-RBD beads to
the total amount of RhoA in the whole cell lysate (active/total) to compare
activity of RhoA from different samples.
GST Pull-Down Assays
GST fusion proteins were expressed and purified as described previously
(Haskins et al.,
1998; Wittchen et
al., 1999). Equivalent amounts (estimated by Coomassie Blue
staining) of GST-FLZO-3, GST-NZO-3, GST-CZO3, or GST alone were bound to
glutathione-Sepharose beads. MDCK cells were lysed in RIPA buffer, and an
equal amount of lysate was added to each batch of affinity resin. Batch
binding was carried out overnight at 4°C with rotation. Beads were washed
4x with buffer B plus protease inhibitors, resuspended in 2x gel
sample buffer, boiled 5 min, and loaded for SDS-PAGE. Blots were incubated
with the following antibodies: anti-RhoA (monoclonal antibody 26C4; Santa Cruz
Biotechnology), anti-cdc42 (catalog no. sc-87; Santa Cruz Biotechnology),
anti-Rac1 (catalog no. R56220
[GenBank]
; Transduction Laboratories, Lexington, KY),
anti-p120catenin (catalog no. P17920
[GenBank]
; Transduction Laboratories), or anti-AF-6
(catalog no. A60520; Transduction Laboratories).
Direct Binding Experiments
Human p120-3ABC in GST expression vector pGEX5–1 was kindly provided
by Dr. Frans van Roy and Dr. J van Hengel (University of Ghent, Ghent,
Belgium) (van Hengel et al.,
1999). Equivalent amounts of GST and GST-p120 were immobilized on
glutathione-Sepharose beads. Histidine-tagged CZO-3 was expressed in Sf9
insect cells by using a baculovirus eukaryotic expression system (Invitrogen,
Carlsbad, CA) and purified as described previously
(Haskins et al.,
1998; Wittchen et
al., 1999). Eluted CZO-3 was diluted 1:20 in RIPA buffer,
added to GST and GST-p120 beads, and allowed to bind 1 h at 4°C with
rotation. Beads were washed 4x with buffer B plus protease inhibitors,
resuspended in 2x GSB, and boiled for 5 min before loading for SDS-PAGE.
CZO-3 retained by GST-p120 beads was detected by blotting with a polyclonal
antibody raised against amino acids 754–898 in the C terminus of ZO-3
(rab5F3 anti-FP2).
For binding of NZO-3 to CZO-3, equivalent amounts (estimated by Coomassie Blue staining) of GST and GST-NZO-3 were immobilized on glutathione-Sepharose beads followed by the addition of histidine-tagged CZO-3. Binding, processing of beads, and detection of bound CZO-3 were carried out as described above.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
) prompted us to look at other aspects of actin dynamics in
NZO-3–expressing cells. We first noted that NZO-3/MDCK cells contained
fewer actin stress fibers than parental untransfected cells or
CZO-3–expressing cells. Figure
1A shows NZO-3/MDCK cells and parental MDCK cells plated at
subconfluent density stained with rhodamine-phalloidin to visualize F-actin.
Thick actin stress fiber bundles are abundant in parental cells, whereas
NZO-3/MDCK cells present an almost complete absence of basally located stress
fibers. In addition, the limited stress fibers present in NZO-3/MDCK cells are
thinner. The perijunctional apical actin ring of NZO-3/MDCK cells is normal
compared with parental untransfected cells. Costaining the same cells for
vinculin as a marker for focal adhesions revealed that NZO-3/MDCK cells also
had fewer and smaller focal adhesions
(Figure 1A). Cells expressing
CZO-3 contained abundant stress fibers and seemed identical to parental
untransfected cells (Figure
1B).
|
Based on this reduced stress fiber and focal adhesion phenotype, we
hypothesized that NZO-3/MDCK cells would migrate faster than parental cells
due to weaker attachment to the substratum. Many studies have demonstrated an
inverse correlation between the presence of actin stress fibers and increased
motility (Herman et al.,
1981; Ridley et al.,
1995; Nobes and Hall,
1999). We performed in vitro wound healing assays to test our
hypothesis. Cell monolayers that had just reached confluence were scratched
with a pipet tip to create a linear wound, and the width of the wound at
specific locations was measured over a 6-h period. The relative migration rate
for two independent NZO-3/MDCK cell lines (B6 and D5) was compared with
parental untransfected cells, and we found that both NZO-3–expressing
cell lines migrated approximately fourfold faster than parental MDCK cells
(Figure 2A). In separate
experiments, migration rate of CZO-3/MDCK cells was not significantly
different from parental MDCK cells (Figure
2B). When wounded monolayers were stained with phalloidin to
visualize actin, not only are the number of stress fibers reduced, as observed
in Figure 1, but also there is
less F-actin condensed at the wound-facing leading edge of NZO-3/MDCK cells
compared with parental cells (Figure
3). CZO-3/MDCK cells showed a similar amount of F-actin at the
leading edge to parental MDCK cells (Figure
3). These results identify a global effect on the actin
cytoskeleton caused by NZO-3 expression in MDCK cells. Pursuing a possible
mechanism through which actin cytoskeleton dynamics is affected, we
hypothesized that the Rho family of GTPases might be involved. Given that
strong evidence exists that RhoA activity is responsible for the formation of
stress fibers and focal adhesions (Ridley
and Hall, 1992), we asked whether RhoA activity was decreased in
NZO-3–expressing MDCK cells. When loaded for equivalent amount of total
protein, we observed no difference in the total amount of RhoA protein from
whole cell lysates of NZO-3/MDCK cells compared with parental and CZO-3/MDCK
cells (Figure 4A). To determine
the level of Rho activity, we used an affinity precipitation technique that
specifically pulls out active (GTP-bound) RhoA from the total RhoA pool. A
representative Western blot of the active RhoA fraction versus the total RhoA
from the same lysate is shown in Figure
4B. These results were quantified by densitometry, and the
relative activity of RhoA was plotted as a ratio of active RhoA relative to
the total amount of RhoA in each sample
(Figure 4C). Both NZO-3/MDCK
cell lines had a significantly lower amount of active RhoA compared with
parental cells (p < 0.05). CZO-3/MDCK cells had significantly higher RhoA
activity compared with parental cells; multiple exposures of the Western blot
were analyzed to confirm this observation.
|
|
|
We next performed binding experiments from whole cell lysates to attempt to
identify a physical link between the RhoA signaling pathway and the tight
junction protein ZO-3. We first looked at whether NZO-3 interacts with any
members of the Rho family of GTPases. Using NZO-3 expressed as a GST fusion
protein, we performed GST pull-down experiments with MDCK whole cell lysates,
and probed the bound fraction with antibodies to RhoA, Rac1, or Cdc42.
Figure 5A shows that NZO-3 does
not interact with Rho, Rac, or Cdc42 in these assays. However, we did identify
two novel protein interactions by using this GST pull-down technique. P120
catenin and AF-6 are proteins that have been localized to the junctional
complex and linked to signaling pathways implicated in cytoskeleton regulation
and cell motility. Therefore, we determined whether they interacted with any
of the ZO-3 constructs. Equivalent amounts (estimated by Coomassie Blue
staining) of GST alone, FLZO-3, NZO-3, or CZO-3-GST fusion were bound to
glutathione-Sepharose beads and incubated with an equal amount of MDCK cell
lysate. Immunoblotting the bound fraction with an anti-p120 catenin antibody
revealed that p120 catenin binds to both FLZO-3 and CZO-3, with a low level of
binding to NZO-3 (Figure 5B).
We also determined whether any of these ZO-3 constructs could bind to the
Ras-effector protein AF-6 (Yamamoto et
al., 1997). We discovered that the FLZO-3 construct binds to
AF-6 and that the binding region is in the C terminus, because AF-6 was
retained by the CZO-3 beads but not the NZO-3 beads
(Figure 5B). In addition to the
GST pull-down experiments showing binding of CZO-3 to p120 catenin and AF-6,
we confirmed that the interaction between p120 catenin and CZO-3 was direct by
performing direct binding assays by using purified proteins. We found that
purified histidine-tagged CZO-3 was retained specifically on affinity resin
containing p120 catenin, and not on beads with GST alone
(Figure 5C).
|
The question of how exogenous expression of NZO-3 alters the activity of
RhoA despite the fact that it is the C-terminal half of ZO-3 that binds p120
catenin and AF-6 might be answered by the presence of intramolecular
interactions within ZO-3. These intramolecular interactions may subsequently
regulate intermolecular interactions of ZO-3. To address this possibility, we
determined whether the amino-terminal half of ZO-3 can interact with the
C-terminal half. This interaction could theoretically occur via the C-terminal
SH3 domain and a consensus PXXP binding motif
(Ren et al., 1993)
found in the proline-rich region of the N-terminal half of the molecule. Equal
amounts of purified GST and NZO-3/GST were immobilized on
glutathione-Sepharose beads (Figure
6A), followed by the addition of purified histidine-tagged CZO-3.
The bound fraction was immunoblotted with an antibody specific for the C
terminus of ZO-3. Figure 6B
shows that CZO-3 binds directly to NZO-3 substantially above background.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
). Herein, we show more global effects on F-actin dynamics in
NZO-3–expressing cells. NZO-3/MDCK cells have fewer actin stress fibers
and fewer and smaller focal adhesions than untransfected parental cells or
cells transfected with the C-terminal half of ZO-3. NZO-3–expressing
cells also migrate faster than parental MDCK cells in a wound healing assay.
Because Rho GTPases are known to be important regulators of actin cytoskeletal
dynamics, we explored the potential relationships between NZO-3 expression and
Rho GTPase activity and found that NZO-3–expressing cells have lower
RhoA activity compared with parental MDCK cells and CZO-3/MDCK cells. We
further investigated the binding interactions of the ZO-3 protein constructs
and found that although ZO-3 does not seem to interact with RhoA itself, CZO-3
interacts with p120 catenin as well as the Ras target protein AF-6 from MDCK
whole cell lysates.
NZO-3 Expression Affects the Actin Cytoskeleton and Increases Cell
Migration
Our observation that NZO-3/MDCK cells have fewer stress fibers compared
with parental untransfected MDCK cells
(Figure 1) initiated this
investigation. That this was observed in multiple clonal cell lines expressing
NZO-3 and that the stress fibers of CZO-3/MDCK cells are indistinguishable
from parental cells indicates that the observation in NZO-3/MDCK cells is
specific. Because the actin cytoskeleton is tethered to focal adhesions via
actin stress fiber bundles, it is not surprising that the number and size of
focal adhesions are also diminished in NZO-3/MDCK cells
(Figure 1). The meaning of the
result that NZO-3/MDCK cells have less actin staining at the free edge of the
wound compared with parental MDCK cells
(Figure 3) is unclear, because
migrating cells generally have a prominent F-actin condensation in the leading
edge lamellipodia (Waterman-Storer et
al., 1999). It could be that although there is less F-actin
visible at the leading edge in NZO-3/MDCK cells, F-actin that is present may
be more dynamic with a more rapid turnover and less availability for binding
phalloidin. Membrane ruffling activity at the leading edge is controlled by
the activities of Rac1 and Cdc42; however, we were unable to detect consistent
differences in the activity of these GTPases in NZO-3/MDCK cells versus
parental MDCK cells.
The phenotype of fewer actin stress fibers and focal adhesions suggested
that NZO-3/MDCK cells might be more readily motile. There is a
well-established correlation between cell motility and reduced actin stress
fibers and focal adhesions; these F-actin structures are typically considered
characteristics of a nonmotile state
(Herman et al., 1981;
Ridley et al., 1995;
Nobes and Hall, 1999). Wound
healing assays are one way of measuring the migratory behavior of cells.
During wound closure cells migrate to fill the empty region, either in
connection with each other, or in some cases individual cells break off to
migrate individually. The former condition typifies the migration of the MDCK
epithelial cells observed in this study, where there is coordinated migration
of a cell sheet in which cell-cell interactions are maintained. We observed
that NZO-3/MDCK cells migrated faster than control MDCK cells, correlating
well with the reduced amount of actin stress fibers and focal adhesions.
Involvement of the Rho Family of GTPases
Because the Rho GTPases are classical mediators of actin cytoskeleton
remodeling during events such as cell migration, it was of interest to
determine whether there were differences in their activity in the
NZO-3–expressing cells. Specifically, we looked at RhoA activity because
of its well-established role in stress fiber and focal adhesion formation. We
found that RhoA activity was lower in NZO-3/MDCK cells compared with parental
MDCK cells and CZO-3/MDCK cells (Figure
4B). Two independent NZO-3/MDCK cell lines were used to confirm
this observation. It is interesting to note that CZO-3/MDCK cells had
significantly higher RhoA activity compared with parental cells, even though
there was no difference in the number or thickness of stress fibers or the
rate of cell migration during wound closure (Figures
1B and
2B). One explanation is that
there is a maximum threshold level of RhoA activation; activity exceeding this
will not have further additive effects on stress fibers or migration behavior.
Alternatively, negative feedback mechanisms may exist to counteract the
phenotype produced by increased RhoA signaling.
Linking ZO-3 with RhoA Signaling Pathway: Binding Interactions
Identifying novel binding partners for ZO-3 in general will provide
information about its cellular function, and in context of the observations
presented herein, this can provide information on the molecular mechanism
underlying the phenotypes we observe in NZO-3/MDCK cells. One potential
mechanism by which NZO-3 expression could decrease the activity of RhoA is by
interacting with the GTPase and either inhibiting its activity directly, or
indirectly, by preventing it from binding to its guanine nucleotide exchange
factors (GEFs). We tested whether NZO-3 could bind to RhoA from MDCK whole
cell lysate and could not detect an interaction
(Figure 5A). It is also
possible that the NZO-3 interaction with the RhoA GTPases is transient or with
such low affinity that binding is undetectable in the in vitro binding assays
performed. Alternatively, NZO-3 might interact with a Rho GDP-dissociation
inhibitor, keeping RhoA in an inactive (GDP-bound) state; we tested this by
using GST pull-down experiments and found that NZO-3 does not interact with
Rho GDI (our unpublished data).
However, by performing in vitro binding assays, we found that FLZO-3 and
CZO-3, but not NZO-3, interact with p120 catenin from MDCK lysates
(Figure 5B). This is the first
demonstration of a tight junction protein interacting with p120 catenin,
although both ZO-1 and ZO-2 interact with 
-catenin (Itoh et
al., 1997,
1999). The dual roles of p120
catenin in cell-cell adhesion and cell migration make it an ideal candidate to
link tight junction elements and RhoA signaling in the NZO-3/MDCK cell line.
Work by Noren et al.
(2000) has shown that
cytoplasmic p120 catenin binds to Vav-2, a GEF activator of Rac and Cdc42,
suggesting a link between p120 catenin and the Rho GTPase family proteins.
Furthermore, they have shown that increasing the soluble pool of p120 catenin
results in disassembly of focal adhesions and stress fibers. This
overexpression of p120 catenin causes increased cell motility, with
correspondingly decreased RhoA activity and increased Rac and Cdc42 activity
(Noren et al., 2000;
Grosheva et al.,
2001). We compared by immunoblot the relative levels of
cytoplasmic versus membrane-associated p120 in parental MDCK cells and cells
expressing NZO-3 or CZO-3. Using several methods of fractionation, we found
that 90% of total cellular p120 is membrane associated and 10% is
soluble in both cell lines (our unpublished data). It is possible that the
difference in cytoplasmic p120 levels is too small to detect by our methods
yet is still enough to cause a phenotypic effect. Other investigators have
confirmed that there is a direct link between p120 catenin and RhoA by showing
in vitro that p120 catenin can directly inhibit RhoA activation; preventing
GDP dissociation from RhoA by binding to Rho-GDP and acting to sequester it
from its activating GEF (Anastasiadis
et al., 2000).
We also found that FLZO-3 and CZO-3 interact with the Ras target AF-6
(Figure 5B). AF-6 is known to
bind to another tight junction protein (ZO-1), via its Ras-binding domain, and
activated Ras can compete with ZO-1 for binding to AF-6
(Yamamoto et al.,
1997). When epithelial cells are transformed with Ras, they
acquire a fibroblastic, depolarized phenotype
(Schoenenberger et al.,
1991). In conjunction with this, a correlation between Ras
activation and junctional complex disruption has been established
(Mercer, 2000). In
Ras-transformed MDCK cells, TER is low, indicating disrupted barrier
properties, and occludin, claudin-1, and ZO-1 are not localized to the cell
membrane at tight junctions. TER and localization of occludin, claudin-1, and
ZO-1 to tight junctions could be restored by treating Ras-transformed cells
with the mitogen-activated protein kinase kinase inhibitor PD98059
(Chen et al., 2000).
Our results provide more evidence for a linkage between the Ras signaling
pathway and junctional complex regulation via the interaction of AF-6 and
ZO-3. That AF-6 and p120 catenin have both been localized to the junctional
complex and have been linked to signaling pathways implicated in cytoskeleton
regulation and cell motility underscores the potential significance of their
binding to ZO-3 and involvement in the phenotype observed in the
NZO-3–expressing cells.
Possible Mechanism to Link NZO-3 Expression to Rho Pathway
A reasonable model that explains how exogenous expression of NZO-3 alters
the activity of RhoA, whereas it is the C-terminal half of ZO-3 that binds
p120 catenin and AF-6, is not immediately intuitive. However, intramolecular
interactions within ZO-3 could explain regulatory interactions involving ZO-3
and p120 catenin. In parental MDCK cells
(Figure 7A), endogenous ZO-3
could exist in two states: a "closed" conformation where the
N-terminal half binds to the C-terminal half, or an "open" state
that allows for interaction with other partners such as p120 catenin. However,
in NZO-3/MDCK cells excess exogenous NZO-3 could inhibit the binding of other
proteins, including p120 catenin, to the C-terminal half of endogenous ZO-3
(Figure 7B). P120 catenin would
then available to inhibit RhoA signaling, either indirectly via Vav-2
activation (Noren et al.,
2000), or directly, by sequestering RhoA from its activating GEFs
(Anastasiadis et al.,
2000). Based on previous results
(Noren et al., 2000;
Grosheva et al.,
2001), the end result of this p120-mediated decrease in RhoA
activity is a decrease in the amount of stress fibers and increased motility,
similar to what we observed in this study. Finally, in the case of CZO-3/MDCK
cells (Figure 7C), p120 catenin
could either bind to the C terminus of endogenous ZO-3 or to the exogenous
CZO-3 fragment. Binding of p120 catenin to the CZO-3 fragment could sequester
p120 catenin and prevent it from interacting with other binding partners
(e.g., Vav2 and Rho-GDP) in the cytoplasm, providing an explanation for the
relative increase in RhoA activity observed in CZO-3/MDCK cells
(Figure 4B). The model in
Figure 7 suggests a mechanism
whereby conformational changes of native ZO-3 may be used in situ to regulate
cytoskeletal dynamics important for normal junctional complex physiology and
cellular events such as migration.
|
This hypothetical mechanism involving regulation of protein function via
intramolecular interactions is supported by previous studies with other
proteins. For example, ezrin (an ERM family protein) exists in two
conformationally distinct states: a "dormant" state, which is a
closed conformation where the N terminus of the protein binds to the C
terminus in a head-to-tail manner; and an "active" state that
opens up the protein and exposes otherwise masked binding sites to allow other
intermolecular binding interactions to take place
(Bretscher et al.,
2000). In the case of ezrin, the active state is able to interact
with the membrane via its N terminus and the cytoskeleton via its C terminus
(Bretscher et al.,
1997). Another example of intramolecular regulatory interactions
is found in vinculin. In this case, the loss of the intramolecular interaction
of the head domain and tail domain accounts for the increased affinity for
talin of the head domain compared with intact vinculin
(Johnson and Craig, 1994).
Disruption of the head–tail interaction also reveals an otherwise hidden
F-actin binding site in the tail region of vinculin
(Johnson and Craig, 1995).
Finally, SAP97 (a membrane-associated guanylate kinase family protein like
ZO-3), has an N-terminal domain that interacts with the SH3 and GUK domain of
the same molecule, thus altering other protein–protein interactions
(Wu et al.,
2000).
In summary, our results indicate a novel link between the tight junction protein ZO-3 and cell signaling pathways regulating the actin cytoskeleton and migration of epithelial cells. Expression of the N-terminal half of ZO-3 caused a noticeable decrease in the amount of stress fibers and fewer and smaller focal adhesions. This phenotype was accompanied by an increased migratory ability of these cells. We correlated this actin phenotype and increased migration with decreased RhoA activity in NZO-3/MDCK cells. We also show that the C-terminal half of ZO-3 binds to both p120 catenin and AF-6, providing a mechanistic link between ZO-3 and the observed phenotype. Finally, we provide direct binding evidence that NZO-3 can interact with the C terminus of ZO-3, and we hypothesize a model where altered binding interactions involving ZO-3, p120 catenin, and possibly AF-6 in NZO-3–expressing cells negatively influence RhoA GTPase activity. This study reveals a potential link between the tight junction protein ZO-3 and Rho GTPase-related signaling events.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
Footnotes |
|---|
* Corresponding author. E-mail address: bstevenson{at}salk.edu.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Anastasiadis, P.Z., Moon, S.Y., Thoreson, M.A., Mariner, D.J., Crawford, H.C., Zheng, Y., and Reynolds, A.B. (2000). Inhibition of RhoA by p120 catenin. Nat. Cell Biol. 2, 637–644.[CrossRef][Medline]
Balda, M.S., Gonzalez-Mariscal, L., Matter, K., Cereijido, M., and
Anderson, J.M. (1993). Assembly of the tight junction: the role
of diacylglycerol. J. Cell Biol.
123,
293–302.
Braga, V. (2000). The crossroads between cell-cell adhesion and motility. Nat. Cell Biol. 2, E182–E184.[CrossRef][Medline]
Braga, VM., Machesky, L.M., Hall, A., and Hotchin, N.A.
(1997). The small GTPases Rho and Rac are required for the
establishment of cadherin-dependent cell-cell contacts. J. Cell
Biol. 137,
1421–1431.
Bretscher, A., Reczek, D., and Berryman, M. (1997). Ezrin: a protein requiring conformational activation to link microfilaments to the plasma membrane in the assembly of cell surface structures. J. Cell Sci. 110, 3011–3018.[Abstract]
Bretscher, A., Chambers, D., Nguyen, R., and Reczek, D. (2000). ERM-Merlin and EBP50 protein families in plasma membrane organization and function. Annu. Rev. Cell Dev. Biol. 16, 113–143.[CrossRef][Medline]
Chen, Y., Lu, Q., Schneeberger, E.E., and Goodenough, D.A.
(2000). Restoration of tight junction structure and barrier
function by down-regulation of the mitogen-activated protein kinase pathway in
ras-transformed Madin-Darby canine kidney cells. Mol. Biol.
Cell 11,
849–862.
Grosheva, I, Shtutman, M., Elbaum, M., and Bershadsky, A.D. (2001). p120 catenin affects cell motility via modulation of activity of Rho-family GTPases: a link between cell-cell contact formation and regulation of cell locomotion. J. Cell Sci. 114, 695–707.[Abstract]
Gumbiner, B., Stevenson, B.R., and Grimaldi, A.
(1988). The role of the cell adhesion molecule uvomorulin in the
formation and maintenance of the epithelial junctional complex. J. Cell
Biol. 107,
1575–1587.
Haskins, J., Gu, L., Wittchen, E.S., Hibbard, J., and Stevenson,
B.R. (1998). ZO-3, a novel member of the MAGUK protein family
found at the tight junction, interacts with ZO-1 and occludin. J. Cell
Biol. 141,
199–208.
Hecht, G., Pestic, L., Nikcevic, G., Koutsouris, A., Tripuraneni, J., Lorimer, D.D., Nowak, G., Guerriero, V., Jr., Elson, E.L., and Lanerolle, P.D. (1996). Expression of the catalytic domain of myosin light chain kinase increases paracellular permeability. Am. J. Physiol. 271, C1678–C1684.
Herman, I.M., Crisona, N.J., and Pollard, T.D. (1981).
Relation between cell activity and the distribution of cytoplasmic actin and
myosin. J. Cell Biol. 90,
84–91.
Itoh, M., Nagafuchi, A., Moroi, S., and Tsukita, S.
(1997). Involvement of ZO-1 in cadherin-based cell adhesion
through its direct binding to alpha catenin and actin filaments. J.
Cell Biol. 138,
181–192.
Itoh, M., Morita, K., and Tsukita, S. (1999).
Characterization of ZO-2 as a MAGUK family member associated with tight as
well as adherens junctions with a binding affinity to occludin and alpha
catenin. J. Biol. Chem. 274,
5981–5986.
Johnson, R.P., and Craig, S.W. (1994). An
intramolecular association between the head and tail domains of vinculin
modulates talin binding. J. Biol. Chem.
269,
12611–12619.
Johnson, R.P., and Craig, S.W. (1995). F-actin binding site masked by the intramolecular association of vinculin head and tail domains. Nature 373, 261–264.[CrossRef][Medline]
Jou, T.S., Schneeberger, E.E., and Nelson, W.J.
(1998). Structural and functional regulation of tight junctions
by RhoA and Rac1 small GTPases. J. Cell Biol.
142,
101–115.
Kaibuchi, K., Kuroda, S., and Amano, M. (1999). Regulation of the cytoskeleton and cell adhesion by the Rho family GTPases in mammalian cells. Annu. Rev. Biochem. 68, 459–486.[CrossRef][Medline]
Kuroda, S., Fukata, M., Fujii, K., Nakamura, T., Izawa, I., and Kaibuchi, K. (1997). Regulation of cell-cell adhesion of MDCK cells by Cdc42 and Rac1 small GTPases. Biochem. Biophys. Res. Commun. 240, 430–435.[CrossRef][Medline]
Madara, J.L., and Pappenheimer, J.R. (1987). Structural basis for physiological regulation of paracellular pathways in intestinal epithelia. J. Membr. Biol. 100, 149–164.[CrossRef][Medline]
Mercer, J.A. (2000). Intercellular junctions: downstream and upstream of Ras? Semin. Cell Dev. Biol. 11, 309–314.[CrossRef][Medline]
Nakagawa, M., Fukata, M., Yamaga, M., Itoh, N., and Kaibuchi, K. (2001). Recruitment and activation of Rac1 by the formation of E-cadherin-mediated cell-cell adhesion sites. J. Cell Sci. 114, 1829–1838.[Abstract]
Nobes, C.D., and Hall, A. (1995). Rho, rac, and cdc42 GTPases regulate the assembly of multimolecular focal complexes associated with actin stress fibers, lamellipodia, and filopodia. Cell 81, 53–62.[CrossRef][Medline]
Nobes, C., and Hall, A. (1999). Rho GTPases control
polarity, protrusion, and adhesion during cell movement. J. Cell
Biol. 144,
1235–1244.
Noren, N.K., Liu, B.P., Burridge, K., and Kreft, B.
(2000). p120 catenin regulates the actin cytoskeleton via Rho
family GTPases. J. Cell Biol.
150,
567–580.
Noren, N.K., Niessen, C.M., Gumbiner, B.M., and Burridge, K.
(2001). Cadherin engagement regulates Rho family GTPases.
J. Biol. Chem. 276,
33305–33308.
Nusrat, A., Giry, M., Turner, J.R., Colgan, S.P., Parkos, C.A.,
Carnes, D., Lemichez, E., Boquet, P., and Madara, J.L. (1995).
Rho protein regulates tight junctions and perijunctional actin organization in
polarized epithelia. Proc. Natl. Acad. Sci. USA
92,
10629–10633.
Rajasekaran, A.K., Hojo, M., Huima, T., and Rodriguez-Boulan, E.
(1996). Catenins and zonula occludens-1 form a complex during
early stages in the assembly of tight junctions. J. Cell Biol.
132,
451–463.
Ren, R., Mayer, B.J., Cicchetti, P., and Baltimore, D.
(1993). Identification of a ten-amino acid proline-rich SH3
binding site. Science 259,
1157–1161.
Ren, X.D., Kiosses, W.B., and Schwartz, M.A. (1999). Regulation of the small GTP-binding protein Rho by cell adhesion and the cytoskeleton. EMBO J. 18, 578–585.[CrossRef][Medline]
Ridley, A.J., Comoglio, P.M., and Hall, A. (1995). Regulation of scatter factor/hepatocyte growth factor responses by Ras, Rac, and Rho in MDCK cells. Mol. Cell. Biol. 15, 1110–1122.[Abstract]
Ridley, A.J., and Hall, A. (1992). The small GTP-binding protein Rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors. Cell 70, 389–399.[CrossRef][Medline]
Ridley, A.J., Paterson, H.F., Johnston, C.L., Diekmann, D., and Hall, A. (1992). The small GTP-binding protein rac regulates growth factor-induced membrane ruffling. Cell 70, 401–410.[CrossRef][Medline]
Sander, E.E., ten Klooster, J.P., van Delft, S., van der Kammen,
R.A., and Collard, J.G. (1999). Rac downregulates Rho activity:
reciprocal balance between both GTPases determines cellular morphology and
migratory behavior. J. Cell Biol.
147,
1009–1022.
Schoenenberger, C.A., Zuk, A., Kendall, D., and Matlin, K.S.
(1991). Multilayering and loss of apical polarity in MDCK cells
transformed with viral K-ras. J. Cell Biol.
112,
873–889.
Stevenson, B.R., and Begg, D.A. (1994). Concentration-dependent effects of cytochalasin D on tight junctions and actin filaments in MDCK epithelial cells. J. Cell Sci. 107, 367–375.[Abstract]
Takaishi, K., Sasaki, T., Kotani, H., Nishioka, H., and Takai, Y.
(1997). Regulation of cell-cell adhesion by rac and rho small G
proteins in MDCK cells. J. Cell Biol.
139,
1047–1059.
Troxell, M.L., Gopalakrishnan, S., McCormack, J., Poteat, B.A., Pennington, J., Garringer, S.M., Schneeberger, E.E., Nelson, W.J., and Marrs, J.A. (2000). Inhibiting cadherin function by dominant mutant E-cadherin expression increases the extent of tight junction assembly. J. Cell Sci. 113, 985–996.[Abstract]
van Hengel, J., Vanhoenacker, P., Staes, K., and van Roy, F.
(1999). Nuclear localization of the p120(ctn) Armadillo-like
catenin is counteracted by a nuclear export signal and by E-cadherin
expression. Proc. Natl. Acad. Sci. USA
96,
7980–7985.
Walsh, S.V., Hopkins, A.M., Chen, J., Narumiya, S., Parkos, C.A., and Nusrat, A. (2001). Rho kinase regulates tight junction function and is necessary for tight junction assembly in polarized intestinal epithelia. Gastroenterology 121, 566–579.[CrossRef][Medline]
Waterman-Storer, C.M., Worthylake, R.A., Liu, B.P., Burridge, K., and Salmon, E.D. (1999). Microtubule growth activates Rac1 to promote lamellipodial protrusion in fibroblasts. Nat. Cell Biol. 1, 45–50.
Wittchen, E.S., Haskins, J., and Stevenson, B.R.
(1999). Protein interactions at the tight junction. Actin has
multiple binding partners, and ZO-1 forms independent complexes with ZO-2 and
ZO-3. J. Biol. Chem. 274,
35179–35185.
Wittchen, E.S., Haskins, J., and Stevenson, B.R.
(2000). Exogenous expression of the amino terminal half of the
tight junction protein ZO-3 perturbs junctional complex assembly. J.
Cell Biol. 151,
825–836.
Wu, H., Reissner, C., Kuhlendahl, S., Coblentz, B., Reuver, S., Kindler, S., Gundelfinger, E.D., and Garner, C.C. (2000). Intramolecular interactions regulate SAP97 binding to GKAP. EMBO J. 19, 5740–5751.[CrossRef][Medline]
Yamamoto, T., Harada, N., Kano, K., Taya, S.-I., Canaani, E.,
Matsuura, Y., Mizoguchi, A., Ide, C., and Kaibuchi, K. (1997).
The ras target AF-6 interacts with ZO-1 and serves as a peripheral component
of tight junctions in epithelial cells. J. Cell Biol.
139,
785–795.
Yap, A.S., Niessen, C.M., and Gumbiner, B.M. (1998).
The juxtamembrane region of the cadherin cytoplasmic tail supports lateral
clustering, adhesive strengthening, and interaction with p120ctn. J.
Cell Biol. 141,
779–789.
This article has been cited by other articles:
|
|
 |
|
  R. A. Cardone, A. Bellizzi, G. Busco, E. J. Weinman, M. E. Dell'Aquila, V. Casavola, A. Azzariti, A. Mangia, A. Paradiso, and S. J. Reshkin The NHERF1 PDZ2 Domain Regulates PKA-RhoA-p38-mediated NHE1 Activation and Invasion in Breast Tumor Cells Mol. Biol. Cell, May 1, 2007; 18(5): 1768 - 1780. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Adachi, A. Inoko, M. Hata, K. Furuse, K. Umeda, M. Itoh, and S. Tsukita Normal Establishment of Epithelial Tight Junctions in Mice and Cultured Cells Lacking Expression of ZO-3, a Tight-Junction MAGUK Protein Mol. Cell. Biol., December 1, 2006; 26(23): 9003 - 9015. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Guillemot and S. Citi Cingulin Regulates Claudin-2 Expression and Cell Proliferation through the Small GTPase RhoA Mol. Biol. Cell, August 1, 2006; 17(8): 3569 - 3577. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Cardone, A. Bagorda, A. Bellizzi, G. Busco, L. Guerra, A. Paradiso, V. Casavola, M. Zaccolo, and S. J. Reshkin Protein Kinase A Gating of a Pseudopodial-located RhoA/ROCK/p38/NHE1 Signal Module Regulates Invasion in Breast Cancer Cell Lines Mol. Biol. Cell, July 1, 2005; 16(7): 3117 - 3127. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. G. Laing, B. C. Chou, and T. H. Steinberg ZO-1 alters the plasma membrane localization and function of Cx43 in osteoblastic cells J. Cell Sci., May 15, 2005; 118(10): 2167 - 2176. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Wittchen, R. A. Worthylake, P. Kelly, P. J. Casey, L. A. Quilliam, and K. Burridge Rap1 GTPase Inhibits Leukocyte Transmigratio |
0.05) compared
with parental MDCK cells. CZO-3/MDCK cells had significantly higher Rho
activity compared with parental MDCK cells.
