Articles in PresS. Am J Physiol Renal Physiol (September 7, 2011). doi:10.1152/ajprenal.00419.2011
1
2
WT1 INTERACTING PROTEIN (WTIP) REGULATES PODOCYTE PHENOTYPE BY CELLCELL AND CELL-MATRIX CONTACT REORGANIZATION
3
4
5
6
7
8
9
10
11
Jane, H. Kim *1, Amitava Mukherjee2, Sethu Madhavan2, Martha Konieczkowski2 and John R.
Sedor, MD1,2,.
1
Departments of Physiology and Biophysics and 2Medicine and the Rammelkamp Center for
Education and Research, MetroHealth System Campus, Case Western Reserve University,
Cleveland, OH, United States
12
13
14
Running Title: WTIP regulates podocyte cell-cell and cell-matrix junctions
15
16
Key words: G-proteins, cytoskeleton, RhoGTPase
17
18
Corresponding author:
19
John R Sedor, MD
2500 MetroHealth Drive R415
Cleveland, OH 44109-1998
E-mail: john.sedor@case.edu
Phone: 216-778-4993
Fax: 216-778-8248
20
21
22
23
24
25
26
1
Copyright © 2011 by the American Physiological Society.
27
ABSTRACT
28
Podocytes respond to environmental cues by remodeling their slit diaphragms and cell-matrix
29
adhesive junctions. WT1 Interacting Protein (Wtip), an Ajuba family LIM domain scaffold protein
30
expressed in the podocyte, coordinates cell adhesion changes and transcriptional responses to
31
regulate podocyte phenotypic plasticity. We evaluated effects of Wtip on podocyte cell-cell and
32
cell-matrix contact organization using gain-of- and loss-of-function methods. Endogenous Wtip
33
targeted to focal adhesions in adherent but isolated podocytes and then shifted to adherens
34
junctions after cells made stable, homotypic contacts. Podocytes with Wtip knockdown (shWtip)
35
adhered but failed to spread normally. Non-contacted shWtip podocytes did not assemble actin
36
stress fibers and their focal adhesions failed to mature. As shWtip podocytes established cell-
37
cell contacts, stable adherens junctions failed to form and F-actin structures were disordered. In
38
shWtip cells, cadherin and β-catenin clustered in irregularly distributed spots that failed to
39
laterally expand. Cell surface biotinylation showed diminished plasma membrane cadherin, β-
40
catenin, and α-catenin in shWtip podocytes, although protein expression was similar in shWtip
41
and control cells. Since normal actin dynamics are required for organization of adherens
42
junctions and focal adhesions, we determined if Wtip regulates F-actin assembly.
43
Undifferentiated podocytes did not elaborate F-actin stress fibers, but when induced to
44
overexpress WTIP, formed abundant stress fibers, a process blocked by the RhoA inhibitor, C3
45
toxin, and a RhoA kinase inhibitor. WTIP directly interacted with Rho guanine nucleotide
46
exchange factor (GEF) 12 (Arhgef12), a RhoA-specific GEF enriched in the glomerulus. In
47
conclusion, stable assembly of podocyte adherens junctions and cell-matrix requires Wtip, a
48
process that may be mediated by spatio-temporal regulation of RhoA activity through
49
appropriate targeting of Arhgef12.
50
2
51
INTRODUCTION
52
Podocytes are highly differentiated glomerular epithelial cells, characterized by numerous
53
interdigitating foot processes (FP), which elaborate highly specialized cell-cell contacts known
54
as slit diaphragms (SD). FPs are defined by three membrane domains: the apical membrane
55
domain, the SD protein complex, and the basal membrane domain (19). The submembranous
56
region of all three compartments is connected to the FP actin cytoskeleton (16, 25). Therefore,
57
the actin cytoskeleton plays a crucial role in determining and maintaining the overall structure of
58
the SD. Changes in the actin cytoskeleton from coordinated stress fibers into a dense actin mat
59
is synonymous with podocyte FP effacement and SD disruption (19, 24, 35). Identification of
60
proteins that regulate or stabilize the actin cytoskeleton is important for understanding the
61
maintenance and function of the glomerular filtration barrier (25, 37). Recently, an increasing
62
number of actin-associated proteins in the podocyte have been identified, thus highlighting the
63
significance of dynamic actin cytoskeleton regulation in the maintenance of the glomerular
64
filtration barrier.
65
The actin cytoskeleton not only provides mechanical support for the cell, but also determines
66
cell shape, enables cell movement, and is required for assembly of normal cell-matrix and cell-
67
cell adhesive contacts. Members of the Rho family of small guanosine triphosphatases
68
(RhoGTPases), RhoA, Rac1 and Cdcd42, have emerged as key regulators of the actin
69
cytoskeleton. The activation of RhoA, Rac1 and Cdcd42 leads to assembly of actin stress fibers,
70
protrusive actin-rich lamellipodia, and protrusive actin-rich filopodia, respectively (7, 13, 14).
71
Furthermore, through their interactions with multiple target proteins, they can coordinate other
72
cellular activities, such as gene transcription, with changes in cellular adhesion. Recent
73
observations also suggest that subcellular pools of RhoGTPases operate to regulate specific
74
morphogenic events in different cellular domains (29).
3
75
Focal adhesions are specialized regions of cell adhesions to the extracellular matrix, where
76
integrin receptors associate with a number of structural and signaling proteins to form a link with
77
the actin cytoskeleton (4). These integrin-associated proteins include focal adhesion kinase
78
(FAK), Src family kinases and scaffolding proteins, such as paxillin and vinculin (15, 27). The
79
importance of these molecules has been underscored by the results of gene knockout and
80
knockdown experiments (12, 22, 42). The activated signaling molecules play crucial roles in
81
regulating multiple events, including cell adhesion and cell migration. Like cell-matrix adhesions,
82
cell-cell adhesions are intimately associated both physically and functionally with the actin
83
cytoskeleton. Cadherins, which are cell-cell adhesion receptors, associate with the actin
84
cytoskeleton with β-catenin and α-catenin. Following cadherin engagement, RhoGTPases are
85
recruited, activated and regulate the actin cytoskeleton necessary for the formation of stable
86
cell-cell adhesions (44). Formation of cell-cell adhesions requires coordination between
87
assembly and disassembly of cell-matrix adhesions (10, 43).
88
Wtip belongs to a subset of LIM-domain containing proteins that include the prototypic members
89
zyxin, Ajuba, and lipoma-preferred partner (LPP) (18, 30, 38, 45), all of which localize to focal
90
adhesions and cell-cell adhesions. All members of this family contain two distinct regions: a C-
91
terminal domain containing three LIM protein-protein interaction motifs and a proline-rich N-
92
terminal domain containing one or two nuclear export sequences (5). Although we had localized
93
WTIP to cell-cell contacts and regions of active actin dynamics in podocytes, the functional
94
impact of WTIP on the assembly of cell-matrix and cell-cell contacts has been unclear. Using
95
gain of function and RNA interference (RNAi)-mediated gene knockdown in mouse podocytes,
96
we examined the roles of Wtip in actin dynamics, RhoA GTPase activity and podocyte cell
97
contact formation. We found Wtip targets to focal adhesions in adherent but isolated podocytes
98
and then shifts to podocyte adherens junctions after cells make homotypic contacts, a process
99
dependent on RhoA-regulated F-actin dynamics.
4
100
EXPERIMENTAL PROCEDURES
101
Antibodies and plasmids. An affinity purified anti-WTIP antibody was generated and
102
characterized as described previously (20). Other antibodies used in this study included: anti-α-
103
tubulin, anti-vinculin, and anti-pan-cadherin (Sigma Aldrich, St. Louis, MO); anti-paxillin, anti-α-
104
catenin, anti-β-catenin (BD Transduction, San Diego, CA); anti-phosphotyrosine (PY99), anti-
105
myc (Upstate Biotechnology); anti-FAK (Biosource, Camarillo, CA); anti-Rho-GDI (Santa Cruz
106
Biotechnology, Santa Cruz, CA); anti-GFP (monoclonal and polyclonal; Clontech, Mountain
107
View, CA); and anti-ZO-1 (Chemicon International, Temecula, CA). For visualization of F-actin,
108
rhodamine-phalloidin was purchased from Molecular Probes (Carlsbad, CA). The expression
109
vectors, pcDNA3-EGFP-RhoAQ63L (Addgene plasmid 12968) and pcDNA3-EGFP-RhoAT19N
110
(Addgene plasmid 12967), were purchased from Addgene (Cambridge, MA). The plasmid, pT-
111
Adv Rho guanine nucleotide exchange factor (GEF) 12 (Arhgef12) (mLARG) was kindly
112
provided by Dr. Alexander Belyavsky (Engelhardt Institute of Molecular Biology, Russian
113
Academy of Sciences, Moscow, Russia). The N-terminal coding sequence (aa 1-415, containing
114
the PDZ domain) was cloned into pEGFPN2 (Clontech) to produce the expression construct,
115
NT-Arhgef12-GFP. The expression plasmid, myc-WTIP has been previously described (38)
116
Cell culture and cell lines. A conditionally immortalized podocyte cell line MPC was a generous
117
gift of Dr. Peter Mundel (Massachusetts General Hospital, Boston, MA) (23). Cells were
118
maintained in RPMI 1640 medium (Cambrex, Walkersville, MD) supplemented with 10% fetal
119
bovine serum (FBS), 100 U/ml penicillin, and 100 mg/ml streptomycin. Under permissive
120
conditions, podocytes were maintained in 5% CO2 and at 33°C in culture medium supplemented
121
with 10 U/ml mouse recombinant -interferon (Sigma Chemical, St. Louis, MO) to enhance the
122
expression of the SV40 Large T-antigen. To induce differentiation, we plated podocytes on type
123
I collagen in 5% CO2 and 37°C without -interferon (non-permissive conditions) for at least 10
124
days. Podocytes (between passages 10 and 25) used in these studies expressed the
5
125
transcription factor, WT1. Before experiments, expression of the podocyte differentiation marker,
126
synaptopodin, was confirmed by immunofluorescence analysis in a parallel well of podocytes.
127
Unless otherwise indicated, all experiments were carried out in differentiated podocytes. Mouse
128
podocytes were transfected with the indicated expression vectors using Fugene 6 reagent
129
(Roche Diagnostics Corporation, Chicago, IL) following the manufacturer’s protocol. Podocyte
130
clones transfected with pEGFP-WTIP (see below) were selected in G418 (200 μg/ml) and GFP-
131
WTIP expression was documented by western blotting and fluorescence microscopy.
132
Conditionally immortalized mouse podocytes, stably expressing either a control vector (shEMP)
133
or mouse shWtip were generated as previously described (20). pLKO.1-TRCWTIP1 targets
134
mouse Wtip [NM_207212] nt 1471-1491 (TRCN0000095769): 5’-
135
CCGGCCCGCAACAAAGAAGC GATTTCTCGAGAAATCGCTTCTTTGTTGCGGGTTTTTG-3’;
136
or pLKO.1-TRC WTIP2 targets mouse Wtip [NM_207212] nt 766-786 (TRCN0000095770):5’-
137
CCGGCGCGAGACTACT TTGGCATTTCTCGAGAAATGCCAAAGTAGTCTCGCGTTTTTG-3’.
138
Human podocytes stably expressing tetracycline (TCN)-inducible human WTIP epitope-tagged
139
with V5 (GEC-WTIP-V5) were generated using a ViraPower T-Rex lentiviral expression system
140
and Gateway Technology vectors as previously described (20), according to the manufacturer's
141
protocols (Invitrogen).
142
Transient cell transfection. Mouse podocytes were transiently transfected with pcDNA3-EGFP-
143
RhoAQ63L or pcDNA3-EGFP-RhoAT19N using Fugene 6 reagent (Roche) using the
144
manufacturer’s protocol. After 48 to 72 hr, cells expressing the constructs were identified using
145
fluorescence microscopy. To assess the interaction between Rho guanine nucleotide exchange
146
factor 12 (ARHGEF12 or LARG) and WTIP, COS cells (American Type Culture Collection
147
[ATCC], Manassas, VA) or 293 FT cells (Invitrogen) were transfected with the indicated
148
constructs and processed for immunoprecipitation or immunofluorescence microscopy 48 to 72
149
hr later.
6
150
RNA Isolation and RT-PCR. Total RNA was prepared from undifferentiated control and knock
151
down cell lines, as indicated, using the RNeasy Mini Kit (Qiagen, Valencia, CA). On-column
152
DNAse digestion was performed according to the manufacturer’s protocol. Briefly, 1 μg of total
153
RNA was reverse transcribed using random hexamers and SuperScript III First-Strand
154
Synthesis Super Mix (Invitrogen, Carlsbad, CA). RNA was incubated with annealing buffer and
155
random hexamers at 65oC for 5 min, chilled and incubated with First-Strand Reaction Mix and
156
SuperScript III/ RNaseOUT Enzyme Mix for 10 min at room temperature, followed by 50 min at
157
50oC. The reaction was terminated by heating at 85oC for 5 min. Sample cDNA (2 μl) was used
158
to perform semi-quantitative PCR for Wtip using the GC-RICH PCR System (Roche) under the
159
following conditions: 95oC for 3 min; 95oC for 30 sec, 59oC for 30 sec, 72oC for 1 min (23
160
cycles); 72oC for 5 min, 4oC hold. PCR products were visualized in 1% TBE agarose gels.
161
Primer pairs for Wtip were: Forward 5’-GAGCCTGCCCAGTTCCCTTCC-3’ and Reverse 5’-
162
AGCAGCGGAAGCAGCCTGGGTGGTAG-3’. To amplify Limd1 mRNA, sample cDNA (2.0 μl)
163
was annealed at 60oC with the following primer pairs: Forward 5’-
164
ACCCCACCCAGCATTGAAGAACAT-3’ and Reverse 5’–GGCCAAAGGATCCCAACAGAAGG-
165
3’. To amplify Gapdh mRNA (as a loading control), sample cDNA (1.2 μl) was annealed at 57oC
166
with the following primer pairs: Forward 5’-GGAGCCAAACGGGTCATC-3’ and Reverse 5’–
167
TGTTGCTGTAGCCGTATTCAT-3’. To assess podocyte Arhgef12 message expression, cDNA
168
(2 μl) was used for PCR with HotStarTaq (Qiagen) as follows: 95oC for 15min, 95oC for 30 sec,
169
touchdown annealing (30 sec) from 72oC to 57.5oC and extension at 72oC for 1 min (24 cycles),
170
followed by annealing at 58oC and amplification at 72oC for 1 min (28 cycles). Primer pairs for
171
Arhgef12 PDZ N-terminal were : Forward 5’ TCAAAGAAGATGGAGCAGCCATGC-3’ and
172
Reverse 5’-TCTTTGGGTAGCCGTTCGGTTGTA-3’ Primer pairs for the internal Arhgef12
173
coding sequence were: Forward 5’-AACCAACCTTTCGCCCTGGAAATC-3’ and Reverse 5’-
174
TTGAGATTGGAGGTGTCAAGGCGA-3’.
7
175
Recombinant Adenovirus Generation and Infection. Using PCR, we constructed a GFP-WTIP
176
expression plasmid by cloning the human WTIP coding domain cDNA into pEGFP-C2 (BD
177
Biosciences, Palo Alto, CA). pEGFP-WTIP sequence fidelity and reading frame was confirmed
178
by sequencing. The GFP-WTIP fusion gene was amplified by PCR from pEGFP-WTIP and
179
subcloned into pShuttle-CMV, an AdEasy transfer plasmid for recombinant adenovirus
180
construction. Recombinant transfer vector was linearized and cotransformed with pAdEasy-1
181
DNA into BJ5183 according to the manufacturer's instructions. Bacteria were selected on LB
182
plates containing kanamycin. Plasmids were amplified, purified (Qiagen, Valencia, CA),
183
linearized, and transfected into 293 cells (ATCC) for viral particle generation. Recombinant viral
184
particles were then amplified and purified using the Adeno-X virus purification kit (BD
185
Biosciences) and titered. Infecting podocytes with 200–300 plaque-forming units/cell was
186
sufficient to achieve uniform WTIP expression.
187
Immunofluorescence microscopy and quantification. Cells, cultured on sterile glass coverslips
188
(collagen type I-coated for podocytes), were washed in Dulbecco's PBS, fixed in
189
paraformaldehyde (4%, 10 min at room temperature) and permeabilized with 0.2% Triton X-100
190
in Dulbecco's PBS for 5 min on ice. After blocking in 10% goat serum with 2% BSA and 0.2%
191
fish gelatin, cells were incubated with primary antibodies in PBS either at 37oC for an hour in
192
humidified chamber or at 4oC overnight. Subsequently, coverslips were washed and incubated
193
with secondary antibody at dilutions ranging from 1:200 to 1:300 for 1.5 hr at room temperature.
194
Secondary antibodies included fluorescein isothiocyanate-conjugated horse anti-mouse or
195
Texas Red-conjugated goat anti-rabbit antibodies (Vector Laboratories, Burlingame, CA).
196
Coverslips were mounted in anti-fade, aqueous medium containing 4',6 diamidino-2-
197
phenylindole (Vectashield with DAPI; Vector Laboratories) on standard glass slides. For
198
visualization of F-actin, rhodamine-phalloidin was used following manufacturer’s protocol
199
(Molecular Probes). Antibody staining was visualized using a Nikon epifluorescence E600
200
microscope, and photographs were taken with a SPOT Digital System camera model 2.3.0.
8
201
Confocal images were obtained with a Leica TCS SP2 Confocal system. Digital images were
202
processed and grouped using Adobe Photoshop v6.0 (Adobe Systems Inc., San Jose, CA).
203
Leica Quantify software was used to measure the lengths of the focal adhesions. For each
204
experimental condition, the lengths (in pixels) of focal adhesions from 150 cells were measured.
205
The average length was determined and then used to generate a normalized value of mean
206
pixel intensity for focal adhesions. Data are means ± S.E. Unpaired t-tests were conducted to
207
determine significance (asterisk; P < 0.05).
208
To quantify adherens junction assembly, we used published definitions of sequential stages in
209
cell–cell adhesion, which characterize the transitions in cadherin and actin localization as
210
adherens junctions mature (1). Image J software was used to measure the mean pixel intensity
211
of cadherin clusters [stage 1 junctions (1)] in forming junctions, and colocalization of cadherin at
212
actin tips [stage 2 junctions (1)] was quantified as yellow pixel intensity from cadherin [FTIC] and
213
rhodamine actin overlays. For each stage of cell-cell adhesion, a fixed region of interest was
214
used to measure the mean pixel intensity from 100 different cells. The area of the region of
215
interest was determined and then used to generate a normalized value of mean pixel intensity
216
for cadherin clustering and colocalization with actin. Data are means ± S.E. Unpaired t-tests
217
were conducted to determine significance (asterisk; P < 0.05).
218
Time-lapse imaging of live cells and F-actin quantification. For visualizing the EGFP-actin in
219
living cells, 24 hr after the control and Wtip knockdown cells were plated, pEGFP-actin was
220
transiently transfected using Fugene6 (Roche). After incubation for 48 hr, cells were observed
221
under the indicated conditions using live cell imaging parameters on the Leica TCS SP2
222
Confocal system. ImageJ software was used to create a ZProjection stack of the time-lapse
223
images for GFP-actin dynamics. Mean actin intensity was determined using the line scan
224
measure function across the nucleus of 20 individual cells in 4 separate experiments. Data are
225
means ± S.E. Unpaired t-tests were conducted to determine significance (asterisk; P < 0.05).
9
226
For F-actin quantification from static images, the ImageJ line scan function was also used as
227
described.
228
Coprecipitation and immunoblotting. Proteins were extracted from transfected and stable cells
229
using immunoprecipitation (IP) lysis buffer containing 1% Triton-X-100, 150mM NaCl, 10mM
230
Tris HCl, 0.5% deoxycholate, 1mM sodium orthovanadate, along with protease inhibitors and
231
analyzed by immunoprecipitation and immunoblotting as we have previously described (31, 38).
232
Following centrifugation to remove debris, supernatants were matched for protein, precleared
233
with protein G-Sepharose GammaBind beads, and incubated overnight at 4°C with primary
234
antibody. The following day, 50 µl of GammaBind beads were added on the following day and
235
incubated for 1 hr. Control experiments were done in parallel at the same time using appropriate
236
non immune IgG. For GFP-tagged protein only, GFP was expressed and precipitated to confirm
237
specificity. Beads were collected by low speed centrifugation in microcentrifuge tubes for 2-3
238
minutes and washed three times using IP lysis buffer. Bound proteins were released by boiling
239
in 2X SDS sample buffer for 5 min. Eluted proteins were separated by 4-20% SDS-PAGE,
240
transferred to Immobilon (Millipore Corp., Billerica, MA) membranes, and analyzed by
241
immunoblotting. Bound antibody was detected by chemiluminescence (Western Lightning;
242
PerkinElmer Life Sciences).
243
RhoA activity assay. RhoA activity was determined using a configuration-specific monoclonal
244
antibody based RhoA Activation Assay Kit (New East Biosciences, Malvern, PA) following the
245
manufacturer’s protocols. Briefly, GEC-WTIP-V5 cells were grown to approximately 80-90%
246
confluence and then stimulated in the presence or absence TCN. Culture media was aspirated
247
and washed twice with ice-cold PBS. One ml of ice-cold 1X Assay/lysis buffer was added to the
248
cells and placed on ice for 10-20 min. Cells were scraped and collected and lysates were
249
cleared by centrifugation for 10 min (12,000xg at 4oC) and placed on ice. Aliquots of 0.5-1.0 ml
250
of cell lysates were adjusted to 1ml with 1X Assay/lysis buffer. One μl of anti-active RhoA
10
251
monoclonal antibody was added to each sample, followed by 20 μl of resuspended protein A/G
252
agarose bead slurry. Samples were incubated at 4oC for 1 hr with gentle agitation. Beads were
253
pelleted by centrifugation for 1 min at 5,000xg. Supernatant was discarded and the beads were
254
washed three times with 0.5 ml of 1 X assay/lysis buffer, centrifuging and aspirating the
255
supernatant each time. After the last wash, the beads were pelleted and supernatant was
256
removed. Beads were resuspended in 20 μl of 2X reducing SDS-PAGE sample buffer. Each
257
sample was boiled for 5 min and centrifuged for 10 sec at 5,000xg.
258
Cell surface biotinylation assay. Cell surface proteins were biotinylated by incubating the cells
259
with 1.5mg/ml sulfo-NHS-SS-biotin (Thermo Scientific, Rockford, IL)) for 1 hr at 4oC and free
260
biotin was quenched with a blocking solution (50nM NH4Cl in PBS containing 1mM MgCl2 and
261
0.1mM CaCl2). Cells were then either directly extracted in a RIPA buffer, or stripped to remove
262
the extracellular bound biotin with 50 mM glutathione, 75 mM MaCl2, 75 mM NaOH, and 2%
263
bovine serum albumin, at 4oC, and RIPA extracted. Cell extracts were centrifuged and
264
incubated with streptavidin magnetic beads (Dynal, Olson, Norway) to collect biotinylated
265
proteins. Extracted proteins were then separated by SDS-PAGE and analyzed by
266
immunoblotting using antibodies against pan-cadherin, β-catenin, and α-catenin. Total cell
267
lysates were analyzed by immunoblotting to determine the expression levels of these proteins in
268
podocytes expressing the control and Wtip shRNA vectors.
269
Statistical Analysis. The data from all of the experimental groups were expressed as the means
270
± S.E. An unpaired Student's t-test was used to compare differences between control and
271
experimental groups. Statistical significance was defined as p < 0.05.
272
11
273
RESULTS
274
Endogenous Wtip localized to focal adhesions and cell-cell adhesions. We had previously
275
shown that ectopically expressed WTIP localizes with cell-cell and cell matrix contacts. To
276
examine localization of endogenous Wtip in mouse podocytes, we used an affinity-purified
277
antibody, which we have characterized (20). As shown in Fig 1A, Wtip colocalized with vinculin
278
and paxillin at focal adhesions in cultured, differentiated podocytes prior to the establishment of
279
cell-cell contacts. After cell contacts were established, Wtip colocalized with both β-catenin and
280
cadherin, detected with a pan-cadherin antibody, at cell-cell contact sites (Fig 1B). In addition, a
281
pool of endogenous Wtip localized to the nucleus in some cells, which is consistent with our
282
previous studies that demonstrated, via cellular fractionation, that ectopically expressed human
283
WTIP was dynamically regulated and shifted between plasma membrane, cytosol and the
284
nucleus (20) in response to environmental cues. These results indicate that Wtip localized
285
specifically at cell-matrix and cell-cell adhesion sites and suggested dynamic remodeling of cell-
286
cell junctions regulated the intracellular localization of Wtip, consistent with a role for Wtip in
287
plasticity of podocyte phenotype. In addition, intracellular localization of ectopically expressed
288
WTIP recapitulated patterns of the endogenous protein.
289
Redistribution of WTIP to focal adhesions upon disruption of adherens junction in low
290
calcium media. Since Wtip localization was regulated by the state of cell-cell contact, we next
291
studied the effect of dynamic remodeling of adherens junctions on Wtip localization using a
292
calcium switch assay in podocytes expressing GFP-WTIP delivered by recombinant adenovirus.
293
WTIP was localized to cell-cell contacts in presence of normal calcium (NC, 1.8 mM; Fig 1C, left
294
panel). After the cells were incubated in low calcium (LC, 5 μM) media, WTIP localized at focal
295
adhesion-like structures as early as 2 hr (not shown) and persisted there for 6 hr (Fig 1C, center
296
panel). Furthermore, when podocytes were re-incubated in NC medium, WTIP again targeted to
297
cell-cell contacts (Fig 1C, right panel). Previous studies have shown that the vinculin
12
298
redistributes similarly in a calcium switch assay (28). Therefore to confirm that WTIP localized to
299
focal adhesions in the absence of stable adherens junctions, we stained podocytes in LC
300
medium with antibodies that recognize the focal adhesion markers, phosphotyrosine (P-
301
tyrosine) and vinculin. GFP-WTIP was colocalized with both P-tyrosine and vinculin at the focal
302
adhesions (Fig 1D).
303
Depletion of Wtip abrogated the assembly of actin stress fibers and maturation of focal
304
adhesions. We next examined the effect of Wtip depletion on focal adhesion complex formation
305
in mouse podocytes with stable knockdown of Wtip (shWtip) and cells expressing a control
306
construct (shEMP), which have been previously characterized (20). Staining with rhodamine-
307
phalloidin revealed that majority of Wtip knockdown cells failed to elaborate actin stress fibers
308
(Fig 2A). Because focal adhesions are required for the formation of actin stress fibers and
309
because Wtip was localized at FAs prior to generation of stable adherens junctions, we next
310
examined whether maturation of FAs was altered after Wtip knockdown. Normally focal
311
complexes containing paxillin and FAK are formed at the leading edge of an extended
312
lamellipodial protrusion and cell periphery. These nascent focal complexes then grow to mature
313
focal adhesions as other focal adhesion proteins, such as vinculin, are recruited. To assess the
314
state of focal complex maturation, shEMP and Wtip knockdown cells were fixed and stained for
315
vinculin, paxillin, and tyrosine phosphorylation. Individual vinculin signals in Wtip knockdown
316
cells were significantly smaller at the cell periphery compared to shEMP podocytes, as
317
assessed by quantification of relative mean intensity (Fig 2, panels B and C, P<0.05).
318
Immunostaining for paxillin revealed no significant difference in signal of relative mean intensity
319
between shEMP and Wtip knockdown. However, the localization of paxillin appeared to be more
320
disorganized in Wtip knockdown cells (Fig 2D). Immunostaining for tyrosine phosphorylation
321
(anti-pY99) also revealed a decrease in overall tyrosine phosphorylation in Wtip knockdown
322
compared to shEMP cells, consistent with failure of focal complex maturation into focal
13
323
adhesions (Fig 2E). The knockdown of Wtip was specific as assessed by RT-PCR. ShWtip
324
podocytes did not express Wtip transcripts but mRNA abundance of Limd1, a closely related
325
LIM domain protein family member and Gapdh was similar in shEMP and Wtip knockdown cells
326
by RT-PCR (Fig 2F). In addition, focal contact maturation failure was not caused by Wtip shRNA
327
off-target effects. Expression of the focal adhesion proteins, paxillin, vinculin, and FAK, was
328
equivalent between control and Wtip knockdown cells (Fig 2G). These data suggested
329
recruitment and accumulation of Wtip was not essential for assembly of nascent focal
330
complexes but was necessary for maturation of focal complexes into focal adhesions and actin
331
stress fiber formation in podocytes.
332
Wtip is involved in the formation of cell-cell adhesions. As shown in Fig1B, Wtip also
333
localized to cell-cell adhesion sites after stable cell-cell contacts were formed, suggesting Wtip
334
depletion would cause defects in the formation of adherens junctions. Compared to shEMP
335
cells, Wtip knockdown podocytes displayed a marked inability to form cadherin-based,
336
homotypic cell-cell adhesions (Fig 3A, double arrowheads) and stress fibers (Fig 3A, histogram,
337
P<0.05). Instead of forming adhesions with neighboring attached cells, Wtip knockdown
338
podocytes appeared to extend lamellipodia and filopodia, which extended under or over
339
neighboring cells. Further analysis of adherens junction formation revealed that parallel lines of
340
actin filaments were assembled continuously between neighboring cells in shEMP cells,
341
indicative of proper adherens junction formation (Fig 3B, top). By contrast, actin filaments
342
between neighboring Wtip knockdown cells were disordered and tangled (Fig 3B, bottom).
343
These results suggested that perturbed actin dynamics at cell-cell contact sites in Wtip
344
knockdown podocytes impaired normal adherens junction assembly. To quantitate differences
345
in cell-cell adhesion between Wtip and control shRNA podocytes, we used a previously
346
published method that characterized changes in cadherin and actin cytoskeleton organization
347
(1). Initial junction formation was categorized into two stages: clustering of cadherin puncta
14
348
along the length of the forming contact (stage 1) and maturation of the cadherin puncta into
349
plaques at the edges of the contact at actin tips (stage 2). Based on relative fluorescence
350
intensity, a statistically significant difference between shWtip and shEMP podocytes was
351
identified in formation of both stage 1 (cadherin puncta, Fig 3B, asterisks) and stage 2
352
(colocalization of cadherin at actin tips, Fig 3B, carrots) junction formation in Wtip knockdown
353
cells compared to control cells (Fig 3C, P<0.05). These results suggest that Wtip depletion
354
perturbs actin filaments organization at cell-cell contact sites, preventing normal assembly of
355
cadherin-based adherens junctions between neighboring podocytes.
356
To further examine the effects of Wtip depletion on cell-cell contact formation, we induced
357
formation of cell adhesions in control and Wtip-depleted cells by changing the Ca2+
358
concentration in the culture medium from low to normal calcium. Cells were fixed and stained
359
with rhodamine-phalloidin and anti-pan-cadherin antibody at 20 minutes and 5 hours after
360
calcium addition, when cell-cell adhesions were initially reforming and were completely
361
assembled, respectively. In shEMP cells, actin filaments longitudinally ramified from cell-cell
362
contacts (Fig 3D and E, top panels). In contrast, cadherin-based cell-cell contacts were
363
infrequent in Wtip-depleted cells (Fig 3D and E, bottom panels) and the tips of F-actin filaments
364
were not anchored by cadherin puncta. These results show that Wtip plays a role in the control
365
of actin filament organization upon cell-cell contact and, therefore, in the formation of cadherin
366
mediated cell-cell adhesions.
367
In shEMP podocytes, cadherin was detected primarily at the cell-cell contacts (Fig 3B, top
368
inlay). In contrast, in Wtip depleted cells, we observed retention of cadherin in intracellular
369
compartments and only weak staining at the plasma membrane (Fig 3B, bottom inlay) so we
370
next analyzed both total and cell surface-biotinylated cadherin. Whole cell lysate protein was
371
precipitated with streptavidin beads and precipitates were immunoblotted with pan-cadherin, α-
372
catenin, and β-catenin antibodies. Cadherin, α-catenin, β-catenin, and RhoGDI abundance in
15
373
the whole cell lysates was similar in both shEMP and Wtip knockdown cells (Fig 3F, WCL).
374
However, levels of cadherin, α-catenin, and β-catenin were decreased the streptavidin-
375
precipitated fraction in Wtip knockdown cells compared to shEMP podocytes (Fig 3F). Wtip
376
appeared necessary for recruitment, assembly or retention of the adherens junction complex in
377
the plasma membrane suggesting a mechanism by which Wtip depletion interferes with
378
formation of stable cell-cell contacts.
379
WTIP coprecipitated with the adherens junction proteins, cadherin, β-catenin, and α-
380
catenin. Using podocytes stably expressing GFP-WTIP, we next tested if WTIP interacts with
381
junctional proteins or simply co-distributes with them. Anti-GFP immunoprecipitates from GFP-
382
WTIP-expressing podocytes were immunoblotted with anti-pan-cadherin antibody (recognizes
383
N-, E-, P-, and R-cadherins) or antibodies against Wtip, ZO-1, α-catenin or β- catenin. Cadherin,
384
α-catenin and β-catenin (Fig 3G, left) specifically coprecipitated with GFP-WTIP, but we were
385
unable to detect substantial amounts of ZO-1 in these immunoprecipitates. No adherens
386
junction proteins coprecipitated with GFP (Fig 3G, right).
387
Wtip depleted cells exhibit aberrant actin cytoskeleton dynamics. F-actin structures were
388
disordered in shWtip podocytes. Since normal actin dynamics are necessary for formation of
389
both adherens junctions and focal adhesion complexes, we hypothesized that Wtip may directly
390
or indirectly regulate F-actin dynamics. Although LIM domain proteins can directly interact with
391
actin, we were unable to show that Wtip and actin physically interacted (Kim and Sedor,
392
unpublished observations). Using time lapse microscopy, we examined the dynamics of actin
393
organization in shWtip and control podocytes transfected with an EGFP-actin expression vector.
394
GFP-labeled actin stress fiber formed and persisted in shEMP cells (Fig 4A, arrows), while actin
395
stress fibers only rarely formed in shWtip cells during the observation period (up to 14 min) (Fig
396
4B, arrow heads). Quantification of mean actin intensity in a Z-axis projection of the image stack
397
was statistically less in shWtip versus shEMP podocytes, consistent with failure to assemble F16
398
actin (Fig 4C, P<0.05). In addition, Wtip knockdown cells exhibited reoccurring clusters of
399
polymerized actin (arrow heads), which was not observed in control cells (arrows), suggesting a
400
global perturbation of actin dynamics. Overall, Wtip knockdown cells had more membrane
401
protrusions and filopodia than control cells. These data suggest that Wtip is not only required for
402
the proper formation of actin stress fibers, but that it also plays a broader role in the regulation
403
of the actin cytoskeleton.
404
WTIP overexpression enhances formation of actin stress fibers. The mammalian Ajuba
405
family LIM proteins, which include Ajuba and Limd1 as well as Wtip, are adaptor proteins within
406
multi-protein complexes that connect cell-cell and cell matrix contact proteins to the
407
cytoskeleton. Given the critical role of actin dynamics in regulating podocyte function and
408
phenotype, we next examined if Wtip influenced F-actin cytoskeleton assembly using rhodamine
409
phalloidin to label actin. Differentiated podocytes express F-actin stress fibers. However,
410
undifferentiated podocytes characteristically lack stress fiber formation. We assayed F-actin
411
stress fiber assembly state in an undifferentiated human podocyte cell line stably transfected
412
with a tetracycline (TCN)-regulated V5-tagged WTIP transgene (GEC-WTIP-V5). Mean actin
413
staining was calculated with the Leica Quantify software analysis program using a single line
414
scan across the diameter of each individual cell (100 cells evaluated in each experiment, n=3
415
separate experiments). Control, undifferentiated podocytes incubated in the presence or
416
absence of TCN demonstrated no difference in actin stress fiber assembly, and mean intensity
417
of actin fluorescence was similar in TCN-treated and untreated cells (Fig 5A). When
418
undifferentiated GEC-WTIP-V5 cells were incubated with TCN to induce WTIP-V5
419
overexpression, F-actin stress fibers robustly assembled (Fig. 5B) compared to untreated GEC-
420
WTIP-V5. The bar graphs depict quantification of the actin stress fiber formation in GEC-WTIP-
421
V5 with or without incubation with TCN (Fig 5B; p<0.01, TCN-treated GEC-WTIP-V5 compared
17
422
to untreated GEC-WTIP-V5 cells). In the absence of TCN, podocytes did not express WTIP-V5
423
and no stress fibers were observed (Fig 5C).
424
WTIP stress fiber formation was regulated through RhoA GTPase dependent pathways.
425
Downstream effectors of the RhoA GTPase include RhoA-associated coiled-coil kinase (ROCK),
426
a kinase recruited to the plasma membrane by active RhoA and required for stress fiber
427
formation (2). To determine whether F-actin stress fiber formation was under the regulation of
428
the traditional RhoA effector pathways, undifferentiated GEC-WTIP-V5 podocytes were pre-
429
incubated with a cell permeable C3 toxin, a known inhibitor of RhoA, and then stimulated with
430
tetracycline to induce WTIP-V5 expression. C3 toxin prevented actin stress fiber formation in
431
podocytes overexpressing WTIP-V5 compared to cells not preincubated with C3 toxin (Fig 6A,
432
middle panel). In addition, the activation of ROCK is known to modulate the organization of the
433
actin-based cytoskeletal systems, including the formation of stress fibers (34). GEC-WTIP-V5
434
treated with a ROCK inhibitor, Y27632, also failed to form actin stress fibers (Fig 6A, right
435
panel). Finally, RhoA activity was assessed by immunoblotting using a configuration-specific
436
monoclonal antibody-based RhoA activity assay in the GEC-WTIP-V5 in the presence or
437
absence of TCN. In GEC-WTIP-V5 stimulated with TCN, RhoA activity was increased compared
438
to control podocytes (Fig 6B). These data link WTIP-V5 induction with the activation of the RhoA
439
pathway and stress fiber assembly, identifying WTIP as a molecular player upstream of the Rho
440
signaling pathway.
441
We next determined if Wtip promoted stress fiber formation by adding monomeric actin to
442
barbed ends of F-actin. GEC-WTIP-G5 were either pre-incubated with low dose cytochalasin D
443
(CD 10 nM) for 1 hr or were untreated, then exposed to TCN for 24 hr and fixed and processed
444
for WTIP-V5 expression and F-actin formation. This concentration of CD inhibits membrane
445
ruffling and prevents F-actin elongation by binding to the barbed end of the filament and
446
preventing monomeric actin addition (8). Low dose CD pre-treatment dramatically decreased F-
18
447
actin stress fiber formation compared to untreated control after WTIP induction (Fig 6C). As a
448
result, we believe that WTIP may promote addition of monomeric actin to the barbed ends of
449
growing F-actin filaments, a pathway regulated through by RhoA.
450
Wtip and RhoA expression in shWtip podocytes reestablished F-actin stress fibers. Wtip
451
knockdown cells showed inefficient formation of focal adhesions and reduced formation of actin
452
stress fibers. To assess if altered actin dynamics in shWtip podocytes was specific to Wtip
453
knockdown or resulted from off-target effects, we next determined if transient overexpression of
454
EGFP-WTIP in shWtip cells rescued stress fiber formation. The human WTIP nucleotide
455
sequence differs at 2 bases from the murine Wtip sequence, making it resistant to a murine
456
Wtip shRNA. Transient EGFP-WTIP transfection into shWtip podocytes restored stress fiber
457
formation comparable to control cells (Fig 7A), suggesting that loss of stress fibers was a result
458
of Wtip knockdown not an shRNA off target effect. Since our previous experiments
459
demonstrated Wtip is an upstream regulator of RhoA signaling and because RhoA activity is
460
required for both focal adhesion and stress fiber formation (3), we next overexpressed
461
constitutively active EGFP-RhoAQ63L and dominant negative EGFP- RhoAT19N in Wtip
462
knockdown podocytes. When EGFP- RhoAQ63L was expressed in Wtip depleted cells, actin
463
stress fibers were formed and perturbations of actin polymerization caused by Wtip depletion
464
were suppressed (Fig 7B, top). These results are consistent with the premise that Wtip
465
regulates directly or indirectly RhoA activity. In contrast, EGFP- RhoAT19N overexpression had
466
no effect on stress fiber formation (Fig 7B, bottom) and demonstrated that RhoA activity was
467
required. Rescue of stress fiber assembly with the overexpression of constitutively active RhoA
468
in Wtip knockdown cells suggested that, even in the absence of Wtip, overexpressed RhoA
469
likely drives the maturation of focal adhesions from focal complexes. Even though Wtip was not
470
required for the assembly of actin stress fibers and/or mature focal adhesions, Wtip may be
471
involved in coordinating proper regional activation of RhoA at focal complexes. Taken together,
19
472
our data suggest that Wtip localized at cell-matrix and cell-cell contact sites and plays a crucial
473
role in Rho-GTPase-mediated actin remodeling, which is essential for proper cell-matrix and
474
cell-cell contact formation.
475
Wtip Interacted with the N-terminal PDZ domain of Arhgef12. Guanine nucleotide exchange
476
factors (GEFs) increase RhoA activity and play a key role in actin cytoskeleton formation. Based
477
on the previous results, we hypothesized that Wtip interacted with a GEF to regulate regional
478
actin dynamics and appropriate assembly of both focal adhesions and podocyte adherens
479
junctions. The Wtip sequence has a C-terminal PDZ-binding domain, VTEL, which is highly
480
similar to the canonical PDZ binding domain in PlexinB1, VTDL. We identified a family of
481
RhoGEFs (PDZ-RhoGEF and leukemia-associated RhoGEF [LARG or ARHGEF12]), which
482
contain PDZ domains and mediate targeted RhoA activation. We focused on Arhgef12 as a
483
Wtip interaction partner, since its human orthologue was recently reported to be enriched in a
484
glomerular expression library prepared from human kidney (21). Both the human and mouse
485
Arhgef12 proteins contain the same functional domains, including the Dlg and ZO-1/2 PDZ
486
domains (47). Immunofluorescence analysis (Fig 8A) and coprecipitation experiments (Fig 8C)
487
using over-expressed myc-WTIP and NT-Arhgef-GFP showed that myc-WTIP interacts with the
488
PDZ containing amino terminal region of Arhgef12. Endogenous Arhgef 12 message was
489
identified by reverse transcription–PCR in mouse podocytes using two different primer sets (Fig
490
8B).
491
20
492
DISCUSSION
493
Structure and dynamics of the actin cytoskeleton are important for regulation of cell-matrix and
494
cell-cell adhesions and for cell migration. In this study, we showed Wtip was localized to focal
495
adhesion and cell-cell contact sites similar to other LIM-domain family members including zyxin,
496
Ajuba, and LPP (11, 17, 40). Interestingly, Wtip localization to the cell-cell contact sites was
497
associated with proper cadherin targeting and actin assembly at the junction (Fig 1). Similarly,
498
Wtip accumulated at cell-matrix adhesions sites where actin was organized into stress fibers as
499
the focal adhesion matures (Fig 1 and 2). These results suggest that Wtip is localized to the
500
cell-matrix and cell-cell contact sites when and where actin cytoskeleton is actively reorganized.
501
In this work, we have focused on the function of Wtip outside the podocyte nucleus. The
502
dynamic distribution of Wtip at cell adhesion sites, the focal adhesions and adherens junctions,
503
in addition to its association with both the focal adhesion and the adherens junction protein
504
complex suggest the importance of Wtip in the stability and/or maintenance of cell adhesions.
505
Consistent with its localization, depletion of Wtip from podocytes impaired not only stress fiber
506
assembly and focal adhesion formation, but also actin reorganization at cell-cell adhesion sites
507
(Fig 2A). However, total protein expression of focal adhesion proteins FAK, paxillin, and vinculin
508
remained unchanged (Fig 2G). Loss of FAK and paxillin expression in HeLa cells also
509
abrogated efficient N-cadherin mediated cell-cell adhesion (33), which provided evidence for
510
communication between integrin and cadherin systems. However paxillin and FAK localized to
511
focal adhesion structures exclusively, whereas Wtip was localized to both focal adhesion and
512
cell-cell adhesions, suggesting that Wtip is more directly involved in integrin-cadherin
513
intercommunication. At the cell-cell contact sites, Wtip may participate in the cadherin-mediated
514
signaling pathway to regulate actin organization essential for the correct formation of cell-cell
515
adhesions.
21
516
In Wtip depleted cells, formation of the focal complex-an initial step of focal adhesion formation-
517
appeared normal, as shown by the paxillin immunostaining (Fig 2D). Therefore, Wtip is
518
dispensable for formation of the focal complex, although it is indispensible for the maturation of
519
focal complex to focal adhesion (Fig 2C and E). Maturation of focal adhesions is essential for
520
stress fiber formation and stable cell-matrix adhesion (6). Thus, we speculate that at the cell-
521
matrix adhesion sites, Wtip participates in the integrin-mediated signaling pathways.
522
Coordinated sequential activation and inactivation of Rac1 and RhoA are a prerequisite for the
523
formation of stress fibers and mature focal adhesions (36). Our study showed that exogenous
524
expression of constitutively active RhoA in shWtip cells rescues stress fiber formation, which
525
suggests that failure of spatial or temporal regulation of RhoA activity is a cause for impaired
526
actin dynamics in Wtip-depleted cells.
527
In addition, time-lapse observations of actin dynamics revealed loss of stress fibers caused by
528
Wtip depletion was associated with abnormal actin polymerization events even in the absence
529
of specific stimuli. Because sustained Rac activation downregulates Rho activity in fibroblasts
530
(32), and because a Rho inhibitor induces a phenotype suggestive of activation of Rac1 in
531
fibroblasts (26), Wtip might contribute to RhoA activation through downregulation of Rac1
532
activity. Transfection of a constitutively active Rac1 in podocytes induced bursts of actin
533
polymerization that resembled those induced by Wtip depletion (Kim and Sedor, unpublished
534
results). However, it did not perturb stress fibers, suggesting high activity of Rac1 is not
535
sufficient to suppress stress fiber formation in podocytes. By contrast, expression of
536
constitutively active RhoA in Wtip-depleted cells not only restored stress fiber formation, but
537
also repressed abnormal burst of actin polymerization (Fig 7B). RhoA may directly or indirectly
538
inactivate Rac1 in Wtip depleted cells.
539
In conclusion, our results show that Wtip regulates the stable formation of cell adhesions to both
540
extracellular matrix and neighboring cells and suggest that Wtip is necessary for normal
22
541
glomerular filtration barrier function. This process may be mediated by spatio-temporal
542
regulation of RhoA activity through appropriate targeting of Arhgef12 by Wtip. Both animal and
543
cell culture models have demonstrated that regulated Rho family GTPase activity is critical for
544
normal glomerular filtration barrier function and podocyte contact formation. Mice lacking a
545
member of the Rho guanine nucleotide dissociation inhibitor family, RhoGDIα, develop
546
nephrotic syndrome (39). In vitro, small GTPase activity regulated the integrity of cell-cell
547
contacts (9) and the balance of small GTPase activity is critical for normal podocyte process
548
formation, a cell culture phenotype suggestive of foot process retraction (46). Complement-
549
dependent injury upregulated RhoA activity and concomitantly reduced Rac and Cdc42 activity
550
causing foot process retraction (17). Finally, insulin signaling through the insulin receptor rapidly
551
reorganizes the actin cytoskeleton in normal podocytes by regulation of small GTPase activity
552
and loss of podocyte foot processes with actin rearrangement was an early abnormality
553
identified in podocyte-specific, insulin receptor knock-out mice (41). Given the importance of
554
spatial control of Rho family GTPase activity in appropriately regulating cellular function and
555
phenotype (29), scaffolding molecules like Wtip may play a critical role. Further studies of the
556
precise signaling pathway from Wtip to Rho GTPases and, thus actin dynamics, may improve
557
understanding of mechanisms by which Rho family GTPases regulate glomerular filtration
558
barrier function and identify novel targets for therapy.
559
23
560
ACKNOWLEDGEMENTS
561
We thank Dr. Gary Bokoch for the pcDNA3-EGFP-RhoA-Q63L (Addgene plasmid 12968) and
562
pcDNA3-EGFP-RhoA-T19N (Addgene plasmid 12967) constructs. The plasmid, pT-Adv Rho
563
guanine nucleotide exchange factor (GEF) 12 (Arhgef12 or Larg) was kindly provided by Dr.
564
Alexander Belyavsky. Support for this project was provided by National Institute of Diabetes and
565
Digestive and Kidney Diseases Grants DK-07470, P50 DK-054178, DK-064719, and F30
566
DK083897.
567
24
568
REFERENCES
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
1.
Adams CL, Chen Y-T, Smith SJ, and James Nelson W. Mechanisms of Epithelial Cell–Cell
Adhesion and Cell Compaction Revealed by High-resolution Tracking of E-Cadherin– Green Fluorescent
Protein. The Journal of Cell Biology 142: 1105-1119, 1998.
2.
Amano M, Chihara K, Kimura K, Fukata Y, Nakamura N, Matsuura Y, and Kaibuchi K. Formation
of Actin Stress Fibers and Focal Adhesions Enhanced by Rho-Kinase. Science 275: 1308-1311, 1997.
3.
Bhadriraju K, Yang M, Alom Ruiz S, Pirone D, Tan J, and Chen CS. Activation of ROCK by RhoA is
regulated by cell adhesion, shape, and cytoskeletal tension. Experimental Cell Research 313: 3616-3623,
2007.
4.
Craig SW, and Johnson RP. Assembly of focal adhesions: progress, paradigms, and portents.
Current Opinion in Cell Biology 8: 74-85, 1996.
5.
Dawid IB, Breen JJ, and Toyama R. LIM domains: multiple roles as adapters and functional
modifiers in protein interactions. Trends in Genetics 14: 156-162, 1998.
6.
Dold FG, Sanger JM, and Sanger JW. Intact alpha-actinin molecules are needed for both the
assembly of actin into the tails and the locomotion of Listeria monocytogenes inside infected cells. Cell
Motility and the Cytoskeleton 28: 97-107, 1994.
7.
Etienne-Manneville S, and Hall A. Rho GTPases in cell biology. Nature 420: 629-635, 2002.
8.
Franki N, Ding G, Gao Y, and Hays RM. Effect of cytochalasin D on the actin cytoskeleton of the
toad bladder epithelial cell. Am J Physiol Cell Physiol 263: C995-1000, 1992.
9.
Gao S-y, Li C-y, Shimokawa T, Terashita T, Matsuda S, Yaoita E, and Kobayashi N. Rho-family
small GTPases are involved in forskolin-induced cell-cell contact formation of renal glomerular
podocytes in vitro. Cell and Tissue Research 328: 391-400, 2007.
10.
Goodwin M, and Yap A. Classical cadherin adhesion molecules: coordinating cell adhesion,
signaling and the cytoskeleton. Journal of Molecular Histology 35: 839-844, 2004.
11.
Guo B, Sallis RE, Greenall A, Petit MMR, Jansen E, Young L, Van de Ven WJM, and Sharrocks
AD. The LIM Domain Protein LPP Is a Coactivator for the ETS Domain Transcription Factor PEA3. Mol Cell
Biol 26: 4529-4538, 2006.
12.
Hakim ZS, DiMichele LA, Doherty JT, Homeister JW, Beggs HE, Reichardt LF, Schwartz RJ,
Brackhan J, Smithies O, Mack CP, and Taylor JM. Conditional Deletion of Focal Adhesion Kinase Leads to
Defects in Ventricular Septation and Outflow Tract Alignment. Mol Cell Biol 27: 5352-5364, 2007.
13.
Hall A. Rho GTPases and the Actin Cytoskeleton. Science 279: 509-514, 1998.
14.
Hall A, and Nobes CD. Rho GTPases: molecular switches that control the organization and
dynamics of the actin cytoskeleton. Philosophical Transactions of the Royal Society of London Series B:
Biological Sciences 355: 965-970, 2000.
15.
Humphries JD, Wang P, Streuli C, Geiger B, Humphries MJ, and Ballestrem C. Vinculin controls
focal adhesion formation by direct interactions with talin and actin. The Journal of Cell Biology 179:
1043-1057, 2007.
16.
Ichimura K, Kurihara H, and Sakai T. Actin Filament Organization of Foot Processes in Rat
Podocytes. J Histochem Cytochem 51: 1589-1600, 2003.
17.
James V, Zhang Y, Foxler DE, de Moor CH, Kong YW, Webb TM, Self TJ, Feng Y, Lagos D, Chu CY, Rana TM, Morley SJ, Longmore GD, Bushell M, and Sharp TV. LIM-domain proteins, LIMD1, Ajuba,
and WTIP are required for microRNA-mediated gene silencing. Proceedings of the National Academy of
Sciences 107: 12499-12504, 2010.
18.
Kanungo J, Pratt SJ, Marie H, and Longmore GD. Ajuba, a Cytosolic LIM Protein, Shuttles into
the Nucleus and Affects Embryonal Cell Proliferation and Fate Decisions. Mol Biol Cell 11: 3299-3313,
2000.
25
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
19.
Kerjaschki D. Caught flat-footed: podocyte damage and the molecular bases of focal
glomerulosclerosis. The Journal of Clinical Investigation 108: 1583-1587, 2001.
20.
Kim JH, Konieczkowski M, Mukherjee A, Schechtman S, Khan S, Schelling JR, Ross MD,
Bruggeman LA, and Sedor JR. Podocyte Injury Induces Nuclear Translocation of WTIP via Microtubuledependent Transport. Journal of Biological Chemistry 285: 9995-10004, 2010.
21.
Lindenmeyer MT, Eichinger F, Sen K, Anders H-J, Edenhofer I, Mattinzoli D, Kretzler M, Rastaldi
MP, and Cohen CD. Systematic Analysis of a Novel Human Renal Glomerulus-Enriched Gene Expression
Dataset. PLoS ONE 5: e11545, 2010.
22.
Marg S, Winkler U, Sestu M, Himmel M, Schönherr M, Bär J, Mann A, Moser M, Mierke CT,
Rottner K, Blessing M, Hirrlinger J, and Ziegler WH. The Vinculin-ΔIn20/21 Mouse: Characteristics of a
Constitutive, Actin-Binding Deficient Splice Variant of Vinculin. PLoS ONE 5: e11530, 2010.
23.
Mundel P, Reiser J, Borja AZM, Pavenstädt H, Davidson GR, Kriz W, and Zeller R.
Rearrangements of the Cytoskeleton and Cell Contacts Induce Process Formation during Differentiation
of Conditionally Immortalized Mouse Podocyte Cell Lines. Experimental Cell Research 236: 248-258,
1997.
24.
Oh J, Reiser J, and Mundel P. Dynamic (re)organization of the podocyte actin cytoskeleton in
the nephrotic syndrome. Pediatric Nephrology 19: 130-137, 2004.
25.
Orlando RA, Takeda T, Zak B, Schmieder S, Benoit VM, McQuistan T, Furthmayr H, and
Farquhar MG. The Glomerular Epithelial Cell Anti-Adhesin Podocalyxin Associates with the Actin
Cytoskeleton through Interactions with Ezrin. J Am Soc Nephrol 12: 1589-1598, 2001.
26.
Ota T, Maeda M, Murakami M, Takegami T, Suto S, and Tatsuka M. Activation of Rac1 by Rhoguanine nucleotide dissociation inhibitor-[beta] with defective isoprenyl-binding pocket. Cell Biology
International 31: 92-96, 2007.
27.
Parsons JT. Focal adhesion kinase: the first ten years. J Cell Sci 116: 1409-1416, 2003.
28.
Perez-Moreno M, Avila A, Islas S, Sanchez S, and Gonzalez-Mariscal L. Vinculin but not alphaactinin is a target of PKC phosphorylation during junctional assembly induced by calcium. J Cell Sci 111:
3563-3571, 1998.
29.
Pertz O. Spatio-temporal Rho GTPase signaling – where are we now? Journal of Cell Science 123:
1841-1850, 2010.
30.
Petit MMR, Fradelizi J, Golsteyn RM, Ayoubi TAY, Menichi B, Louvard D, Van de Ven WJM, and
Friederich E. LPP, an Actin Cytoskeleton Protein Related to Zyxin, Harbors a Nuclear Export Signal and
Transcriptional Activation Capacity. Mol Biol Cell 11: 117-129, 2000.
31.
Rico M, Mukherjee A, Konieczkowski M, Bruggeman LA, Miller RT, Khan S, Schelling JR, and
Sedor JR. WT1-interacting protein and ZO-1 translocate into podocyte nuclei after puromycin
aminonucleoside treatment. Am J Physiol Renal Physiol 289: F431-441, 2005.
32.
Sander EE, ten Klooster JP, van Delft S, van der Kammen RA, and Collard JG. Rac
Downregulates Rho Activity. The Journal of Cell Biology 147: 1009-1022, 1999.
33.
Schaller MD. FAK and paxillin. The Journal of Cell Biology 166: 157-159, 2004.
34.
Schliwa M. Action of cytochalasin D on cytoskeletal networks. The Journal of Cell Biology 92: 7991, 1982.
35.
Shankland SJ. The podocyte's response to injury: Role in proteinuria and glomerulosclerosis.
Kidney Int 69: 2131-2147, 2006.
36.
Small JV, Rottner K, Kaverina I, and Anderson KI. Assembling an actin cytoskeleton for cell
attachment and movement. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 1404: 271281, 1998.
37.
Smoyer WE, and Ransom RF. Hsp27 regulates podocyte cytoskeletal changes in an in vitro
model of podocyte process retraction. FASEB J 16: 315-326, 2002.
26
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
38.
Srichai MB, Konieczkowski M, Padiyar A, Konieczkowski DJ, Mukherjee A, Hayden PS, Kamat S,
El-Meanawy MA, Khan S, Mundel P, Lee SB, Bruggeman LA, Schelling JR, and Sedor JR. A WT1 Coregulator Controls Podocyte Phenotype by Shuttling between Adhesion Structures and Nucleus. Journal
of Biological Chemistry 279: 14398-14408, 2004.
39.
Togawa A, Miyoshi J, Ishizaki H, Tanaka M, Takakura A, Nishioka H, Yoshida H, Doi T,
Mizoguchi A, Matsuura N, Niho Y, Nishimune Y, Nishikawa S, and Takai Y. Progressive impairment of
kidneys and reproductive organs in mice lacking Rho GDIalpha. Oncogene 18: 5373-5380, 1999.
40.
Wang Y, and Gilmore TD. Zyxin and paxillin proteins: focal adhesion plaque LIM domain proteins
go nuclear. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 1593: 115-120, 2003.
41.
Welsh GI, Hale LJ, Eremina V, Jeansson M, Maezawa Y, Lennon R, Pons DA, Owen RJ, Satchell
SC, Miles MJ, Caunt CJ, McArdle CA, Pavenstädt H, Tavaré JM, Herzenberg AM, Kahn CR, Mathieson
PW, Quaggin SE, Saleem MA, and Coward RJM. Insulin Signaling to the Glomerular Podocyte Is Critical
for Normal Kidney Function. Cell metabolism 12: 329-340, 2010.
42.
Yano H, Mazaki Y, Kurokawa K, Hanks SK, Matsuda M, and Sabe H. Roles played by a subset of
integrin signaling molecules in cadherin-based cell–cell adhesion. The Journal of Cell Biology 166: 283295, 2004.
43.
Yap AS, Crampton MS, and Hardin J. Making and breaking contacts: the cellular biology of
cadherin regulation. Current Opinion in Cell Biology 19: 508-514, 2007.
44.
Yap AS, and Kovacs EM. Direct cadherin-activated cell signaling. The Journal of Cell Biology 160:
11-16, 2003.
45.
Yi J, Kloeker S, Jensen CC, Bockholt S, Honda H, Hirai H, and Beckerle MC. Members of the
Zyxin Family of LIM Proteins Interact with Members of the p130Cas Family of Signal Transducers. Journal
of Biological Chemistry 277: 9580-9589, 2002.
46.
Zhang H, Cybulsky AV, Aoudjit L, Zhu J, Li H, Lamarche-Vane N, and Takano T. Role of RhoGTPases in complement-mediated glomerular epithelial cell injury. American Journal of Physiology Renal Physiology 293: F148-F156, 2007.
47.
Zinovyeva M, Sveshnikova E, Visser J, and Belyavsky A. Molecular cloning, sequence and
expression pattern analysis of the mouse orthologue of the leukemia-associated guanine nucleotide
exchange factor. Gene 337: 181-188, 2004.
690
691
692
27
693
FIGURE LEGENDS
694
Figure 1: Endogenous Wtip was localized to focal adhesions and cell-cell contacts in
695
podocytes and is dynamically regulated by cell-cell adhesion. (A) Immunostaining with
696
anti-Wtip (FITC), anti-vinculin and anti-paxillin (TRITC). (B) Immunostaining for adherens
697
junction proteins with anti-β-catenin and anti-pan-cadherin (TRITC). Nuclei were labeled with
698
CY5-TOPRO-3. (C) A calcium switch assay demonstrated GFP-WTIP localization at cell-cell
699
junctions in medium with normal calcium (NC) concentration (1.8 mM). With disruption of
700
adhesion junctions in low calcium (LC) medium (5 μM), GFP-WTIP localized in patches
701
resembling focal adhesions (middle panel). GFP-WTIP retargeted to adherens junctions (left
702
panel). (D) In LC medium, GFP-WTIP colocalized with the focal adhesion markers,
703
phosphotyrosine and vinculin. Scale bars: 20μm.
704
Figure 2: Wtip knock down impaired stress fiber formation and dysregulated focal
705
adhesion maturation. (A) Actin stress fibers were visualized by rhodamine-phalloidin staining
706
in shEMP controls (upper panels) and in shWtip (lower panels) and co-stained with anti-WTIP
707
(FITC). (B) Focal adhesions were immunostained with anti-vinculin (TRITC) in shEMP and
708
shWtip. (C) Focal adhesion maturation was assessed by anti-vinculin immunostaining in shEMP
709
and shWtip cells. Quantification of focal adhesion size was determined as described in Materials
710
and Methods. The bar graph shows mean relative intensity units for vinculin (right). (D) shEMP
711
and shWtip cells were immunostained with anti-paxillin, a marker of focal contacts. The bar
712
graph shows mean relative intensity units for paxillin (right). (E) Phosphotyrosine levels of FAs
713
in shEMP and shWtip cells were assessed by immunostaining with PY99 as an indicator of FA
714
maturation. The bar graph shows mean relative intensity units for phosphotyrosine (right). (F)
715
RT-PCR of shEMP and shWtip cells for Wtip, Limd1 and Gapdh transcripts. (G) Immunoblot
716
analysis of shEMP (control) and shWtip (knockdown) podocytes for Wtip, paxillin, vinculin, FAK,
717
and tubulin. Leica Quantify software was used for quantification. Scale bar: 20μm.
28
718
Figure 3: Knockdown of Wtip affects proper cadherin junction assembly and targeting.
719
shWtip and shEMP cells stained with rhodamine-phalloidin and anti-pan-cadherin (A, B). Nuclei
720
were labeled with CY5-TOPRO-3. Images in panel A show robust cadherin-based cell-cell
721
contacts in shEMP podocytes (arrowhead) but not in shWtip cells (double arrowheads). The
722
histograms quantify actin stress fiber abundance as described in the Methods. (C)
723
Quantification of cell-cell adhesion formation in shEMP and Wtip knockdown cells as described
724
in the Methods and the Results. shWtip podocytes 20 min (D) and (E) 5 hr after switching from
725
low to normal calcium. (F) Biotinylation demonstrated decreased plasma membrane localization
726
of adherens junction proteins cadherin, β-catenin, and α-catenin to the membrane in shWtip vs.
727
shEMP podocytes. Immunoblot analysis of whole cell lysates showed equal expression of
728
cadherin, β-catenin, α-catenin and rho-GDI. (G) GFP-WTIP formed a complex with the indicated
729
adherens junction proteins (left panels) but GFP fails to precipitate these proteins (right panels).
730
Scale bar: 10 μm.
731
Figure 4: Wtip knock down affects dynamic actin assembly. (A) Time- lapse images of
732
shEMP and (B) shWtip cells transiently transfected with eGFP-actin (green). Arrows indicate
733
stress fiber formation in shEMP cells, whereas arrow heads indicate actin clusters and the lack
734
of actin stress fiber formation in shWtip cells. (C) Quantification of stress fiber formation in
735
shEMP versus shWtip cells (P<0.05). Scale bars: 20μm.
736
Figure 5: Overexpression of GFP-WTIP enhances formation of actin stress fibers. (A) F-
737
actin assembly compared in control podocytes with and without tetracycline (TCN). Bar graph
738
demonstrates mean actin staining using a single line scan across the nuclei of each (100 cells,
739
n=3 experiments) (B) F-actin assembly in GEC-WTIP-V5 +/- TCN and quantified in the bar
740
graph (p<0.01, GEC-WTIP-V5 with TCN vs. no TCN). (C) GEC-WTIP-V5 +/- TCN assessed F-
741
actin using rhodamine-phalloidin (left panels) and WTIP-V5 expression (FITC, middle panels).
742
Merged images in right panels. Scale bars: 20 μm.
29
743
Figure 6: WTIP stress fiber formation was regulated through RhoA-dependent pathways.
744
(A) GEC-WTIP-V5 were pre-incubated with either a cell-permeable RhoA inhibitor, C3 toxin
745
(upper middle) or a Rho kinase inhibitor, Y27632 (upper right), or were untreated control for 3 hr
746
followed by incubation with TCN for 24 hr (B) RhoA activation assay in GEC-WTIP-V5 cells +/-
747
TCN. (C). Low dose cytochalasin D (CD, 10 nM) pre-treatment of GEC-WTIP-V5 for 20 min
748
followed by addition of TCN. Scale bars: 10 μm.
749
Figure 7: Overexpression of GFP-WTIP or constitutively active RhoA rescued actin
750
phenotype. (A) Transient transfection of an eGFP-WTIP expression vector demonstrated
751
comparable actin stress fiber formation using rhodamine phalloidin in shWtip and shEMP
752
podocytes. (B) Transient transfection of EGFP-RhoAQ63L (RhoA-Q, constitutively active) and
753
EGFP- RhoAT19N (RhoA-T, dominant negative) constructs into shWtip cells demonstrated
754
RhoA activity is required for stress fiber formation. Arrows indicate actin stress fiber tips. Nuclei
755
were labeled with CY5-TOPRO-3.
756
Figure 8: Myc-WTIP interacted with the PDZ domain-containing, amino terminal (NT) of
757
the RhoA GEF, Arhgef12. (A) COS7 cells were co-transfected with Myc WTIP and NT-
758
Arhgef12-GFP constructs. Immunofluorescence labeling was done with rabbit polyclonal anti-
759
Myc antibody followed by Alexa Fluor 568®-conjugated anti-rabbit secondary antibody. Myc-
760
WTIP (red) expression co-localizes with Arhgef12 (green) in the merged image. Magnified view
761
of selected areas from upper panels showed co-localization in cell extension. (B) RT-PCR was
762
done in mouse podocytes using two primer sets amplifying different regions of Arhgef12
763
message. PCR products of expected size (412bp and 331bp) were observed for the N-terminal
764
PDZ domain containing and internal coding region of Arhgef12 respectively. (C)
765
Coimmunoprecipitation was done in 293 cells transfected with Myc-WTIP and Arhgef12-GFP.
766
Cell lysates were immunoprecipitated using anti-Myc or anti-GFP antibodies and the
767
immunoprecipitates were examined by western blot using anti-GFP or anti-Myc antibodies
30
768
respectively. Rabbit IgG was used as a control for immunoprecipitation. Input represented 5% of
769
cell lysates used in the co-immunoprecipitation.
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
31
789
790
INTENTIONALLY BLANK
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
32
810
INTENTIONALLY BLANK
811
812
813
814
815
816
817
818
819
820
33
34
C
D
35
2
36
2
D
37
2
38
Figure 3
Mean intensity of acn staining (RIU)
160
140
120
100
80
60
*
40
20
0
shEMP
39
shW
Figure 3
*
<
<
*
C
*
*
shWtip
shWtip
Stage 1
Stage 2
40
Figure 3
D
E
41
Figure 3
F
P p
M Wti
E
sh sh
G
untransfected
WCL
IB:
IP:
IgG GFP
IB:
GFP -WTIP
GFP
Pan -Cadherin
Pan -Cadherin
-catenin
-catenin
-catenin
ZO-1
42
transfected
Figure 4
A
B
shWtip
C
160
Mean acn intensity (RIU)
140
120
100
80
60
40
*
20
0
shEMP
shW
43
Figure 5
44
Figure 6
45
Figure 7
A
shWtip
B
shWtip
shWtip
46
Figure 8
A
NT-Arhgef12-GFP
Merge
Myc-WTIP
NT-Arhgef12-GFP
Merge
+
+
Input
+
+
+
+
+
k
:G
FP
Bl
an
-R
T
+R
T
PC
R
Myc-WTIP
Mouse Podocyte mRNA
NT-Arhgef12-GFP
WB: Myc
47
+
Rb
Ig
G
WB: GFP
412 bp
331 bp
bI
IP
+
+
IP
:
:R
IP
Mouse Podocyte mRNA
Input
+
+
IP
Bl
R
PC
-R
T
+R
T
Myc-WTIP
NT-Arhgef12-GFP
:M
yc
C
an
k
B
gG
Myc-WTIP