Pseudopodial and -arrestin-interacting proteomes from migrating breast cancer cells upon PAR2 activation.

Pseudopodial and -arrestin-interacting proteomes from migrating breast cancer cells upon PAR2 activation.Nikolaos Parisis1,2,3, Gergana Metodieva1, Metodi V. Metodiev1, *

1Department of Biological Sciences/Proteomics Unit, University of Essex, Wivenhoe

Park, Colchester, Essex CO4 3SQ, United Kingdom;2Institute of Molecular Genetics (IGMM), CNRS, UMR5535, University of Montpellier I

and II, 34293 Montpellier, France; (current address)3Laboratory of Functional Proteomics, INRA, 34060 Montpellier, France (current

address)

*Author for correspondence ([email protected])

Running title: Pseudopodial and -arrestin proteomes

Key words: PAR-2; pseudopodium; beta-arrestin; cell migration; proteomics

123

4

5

6

7

8

9

10

11

12

Summary

Metastatic cancer cells form pseudopodia (PD) to facilitate their migration. The

proteinase-activated receptor-2 (PAR-2) transduces migratory signals from proteases, and

it forms protein complexes with -arrestin and other signalling molecules that are

enriched in pseudopodia. More generally, however, pseudopodial regulation is poorly

understood. Here, we purified the pseudopodial proteomes of breast cancer cells after

activation of the endogenous PAR-2 and we combined gel-based approaches with label-

free high-resolution mass spectrometry to identify proteins that accumulate at the

pseudopodia upon PAR-2-mediated migration. We identified >410 proteins in the cell

body and >380 in the pseudopodia upon PAR2 activation, of which 93 were enriched in

the pseudopodia. One of the pathways strongly enriched in the PD was the clathrin-

mediated endocytosis signalling pathway, highlighting the importance of the scaffolding

function of -arrestin in PAR-2 signalling via its endocytosis. We therefore

immunoprecipitated -arrestins, and with mass spectrometry we identified 418 novel

putative interactors. These data revealed novel -arrestin functions that specifically

control PAR-2-regulated signalling in migrating breast cancer cells but also showed that

some -arrestin functions are universal between GPCRs and cell types. In conclusion, this

study reveals novel proteins and signalling pathways potentially important for migration

of breast cancer cells.

Introduction

Cell migration is important in embryonic development, tissue homeostasis and in the

operation of the immune system. When it malfunctions, it can contribute to several

pathological conditions, including cancer metastasis [1]. Several chemotactic signals are

transduced into directional cell migration (chemotaxis) via numerous membrane

receptors and intracellular signalling pathways. The main characteristic of a chemotactic

cell is the formation of a leading edge, termed pseudopodium (PD), as a result of

spatiotemporally regulated accumulation of proteins towards the gradient [2]. The

identification of proteins important for gradient sensing, pseudopodia extension,

attachment and retraction is of vital importance for the understanding of chemotaxis.

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

Numerous proteins are involved in pseudopodia formation, such as plasma membrane

receptors, integrins, kinases and phosphatases, GTPases, scaffolding proteins and others,

important for actin dynamics, binding of the cytoskeleton to the extra-cellular matrix

(ECM) as well as the control of further signalling events [3-5]. However, the PD-

enriched proteins, which are dependent on the type of the chemoattractant and the

targeted receptor, as well as the cell type, are largely unknown.

In addition to their role in degrading the extracellular matrix, trypsin-like proteases

secreted from cancer cells have been implicated in the promotion of migration via the

proteolytic activation of the protease-activated receptor 2 (PAR-2) [6]. PAR-2 is a

ubiquitously but unevenly expressed GPCR receptor, which is activated by trypsin and

trypsin-like proteases, including tryptase, clotting factors VIIa and Xa, and has important

functions in immune system responses [7, 8]. PAR-2 has also been found overexpressed

in cancer cell lines, human tumours and in cells of the tumour microenvironment in situ

[9]. Ge et al. reported the secretion of biologically active trypsin-like proteases by the

MDA-MB-231 breast cancer cell line that can promote cell migration [10]. PAR-2-

mediated migratory signals have also been reported in pancreatic cells [11], human

melanoma and prostate carcinoma cells [12] and breast cancer cells [13, 14]. Similarly,

increase on the migration and ERK signalling of BT549 cells showed the activation of

PAR-2 by the clotting factors VIIa and Xa [15], suggesting that chronic inflammation

contributes to tumour metastasis. However, little is known about the downstream

signalling pathways regulated by this receptor. Following receptor activation, -arrestins

antagonize with G-proteins, bind to the receptor, uncouple it from G-proteins [16] and

transfer it via clathrin-coated pits to early endosomes [17]. Interestingly, during this -

arrestin-dependent endocytosis of PAR-2, a multiprotein signalling complex is formed

containing the receptor, -arrestin, raf-1 and ERK [18]. In this complex, -arrestin works

as a scaffold and retains ERK1/2 in the cytoplasm, thus ERK activity is predominantly

cytosolic. Notably, this protein complex is enriched in the pseudopodia, where it

promotes actin assembly along with other unidentified proteins [10, 19]. However, the

proteins that interact with -arrestin upon PAR-2 activation remain unknown.

In this study, we used a breast cancer cell line endogenously expressing the PAR-2

receptor and we applied gel-based, label-free proteomics approaches [20, 21] in order to

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

identify proteins and signalling pathways enriched in the pseudopodia. In addition, we

immunoprecipitated -arrestin and identified PAR-2-regulated -arrestin partners in breast

cancer cells (Fig. 1).

Materials and Methods

Materials

MCF-7 and MDA-MB-231 were a kind gift from Dr. Elena Klenova and HEK293S were

from Dr. Phillip Reeves. Vectors encoding FLAG-tagged -arrestin-1 or FLAG-tagged -

arrestin-2 (arr+FLAG) were kindly provided by Dr. Robert Lefkowitz and were

described elsewhere [22, 23]. PAR-2-activating synthetic peptide (P2AP) SLIGRL-NH2,

which is a more potent activator of human PAR-2 compared to SLIGKV-NH2 [24, 25],

was from Sigma. Anti-FLAG antibody was from Sigma and anti-phospho p44/p42

MAPK from Cell Signaling. All other antibodies were from Santa Cruz Biotechnology

and were used in 1/1000 dilution.

Cell culture and plasmid preparations

All cells, except HEK293, were cultured in Roswell Park Memorial Institute (RPMI;

Lonza) medium supplemented with Ultraglutamine 1, 10% foetal calf serum (FCS;

Biosera) and 1% antibiotic mixture (penicillin and streptomycin; Cambrex) at 37oC in a

humidified environment (5% CO2). Serum deprivation was performed in the same

medium but without FCS and for 12-16 hours. HEK293 were cultured in Dulbeco’s

Modified Eagle’s Medium (DMEM; Cambrex) with L-glutamine, supplemented with

10% FCS and 1% antibiotic mixture in 37oC and 5% CO2. All cell lines were used at a

low passage number. Bacterial (XL1-blue) cultures were grown in Luria Broth (LB)

medium and/or LB agar supplemented with ampicillin (100 g/ml; Sigma) or kanamycin

(50 g/ml; Sigma) with agitation at 250 rpm at 37oC. Bacterial transformations were

conducted according to Inoue et al., [26]. Plasmids were prepared by using commercial

kits (Qiagen or Fisher Scientific).

Western Blotting

Protein samples were analysed by SDS-PAGE in 10% home-cast polyacrylamide gels

with 4% stacking gels, according to Laemmli [27], and proteins were transferred onto

PVDF membranes (Millipore). The membrane was blocked with 5% milk or Odyssey

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

blocking buffer for 1 h. at room temperature (RT) and incubated with the appropriate

primary antibodies for 12-16 hours at 4oC. After 3 washes with a PBS solution

supplemented with Tween 20 (1%), the membrane was incubated with Alexa580- or

Alexa680- conjugated antibodies for 45 min. at RT. Finally, the membrane was scanned

by the Odyssey Infrared Imaging system (LI-COR). Band intensity was determined by

the Odyssey software and normalized to the loading control. Bar charts were created in

Microsoft Excel 2007.

Transfections

HEK293 cells were transfected with 10 g arr+FLAG by the calcium-phosphate method

in 90 mm tissue culture dishes. MCF-7 cells were transfected by using Lipofectamine

LTX following the manufacturer's instructions (Invitrogen). Fresh medium was added

after 24 hours and the cells were used for experiments or collected 48 hours post-

transfection. With these techniques, 90-100% transfection efficiency of HEK293 and 70-

80% transfection efficiency of MCF-7 was routinely achieved.

Under-agarose migration assay

The wells in the agarose gels were created by a special glass mould kindly provided by

Dr. Y. Mousseau [28]. Briefly, the glass mould was placed in the middle of a 60 mm

tissue culture dish and a 10 mL solution of 1% (w/v) agarose in PBS and serum-free

RPMI (1:1 ratio) was poured in the dish around the glass and left to polymerize for 8-12

min in a Class II tissue culture hood. The glass mould was removed and the 10 mL of

medium were added to equilibrate the gel. Before the experiment, 400 L of serum-free

RPMI were added in the central well and left in the incubator for at least one hour to

verify that there is not any leakage from the wells. Only agarose gels that did not leak

were subsequently used in the agarose migration assay. MCF-7 (4-8x104) cells were

seeded in the middle well, left to attach and serum-starved. After 16-20 hours, 50 M

P2AP was placed in a side well and serum-free medium alone was placed in the opposite

side well. Cell migration was monitored on an Olympus IMT-2 Phase contrast inverted

microscope and images were captured with an Olympus OM-2N camera.

Pseudopodia purification

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

For the separation of the pseudopodia from the cell bodies, both sides of several 24 mm

polycarbonate membranes with 3 m pore size were coated with collagen IV (6 g/cm2;

Sigma) and serum-starved MCF-7 cells (1.5x104) were seeded on the upper side. In the

lower chamber, 50 M P2AP was added for 0-120 minutes. Formed pseudopodia or cell

bodies were collected in SDS lysis buffer according to published protocols [4]. Lysates

were cleared by centrifugation and stored at -80oC.

Phosphopeptide enrichment

Equal amounts of CB and PD samples were loaded on an SDS-PAGE gel and the

electrophoresis was stopped as soon as the proteins migrated ≈0.5 mm in the resolving

gel. The gel was stained with Zinc sulfate, the band was sliced and subjected to in-gel

trypsin digestion. Peptide solutions were enriched for phosphorylated peptides with the

Magnetic Phosphopeptide Enrichment Kit (Clontech) on a magnetic separator, following

manufacturer's instructions.

Immunoprecipitations

MCF-7 or HEK293 cells were harvested with a cell scraper in 1% Triton X-100, 50 mM

Tris-HCl pH 6.8 supplemented with protease (PMSF, pepstatin, aprotinin) and

phosphatase (NaF, Na2VO3, Na4P2O7) inhibitors after rinsing with ice-cold PBS. Cells

were left on ice for 30 min. with periodical vortexing and sonicated for complete lysis

before being cleared by centrifugation at 14.000 rpm, 4oC for 20 min. Lysates were

mixed with agarose beads anti-FLAG M2 (Sigma) or protein A/G beads and incubated

for 12-16 hours at 4oC on a rotator. Unbound proteins were removed with 3-5 washes and

bound proteins were eluted twice with 3x FLAG peptide, according to manufacturer's

instructions, and subsequently with Laemmli buffer. For Western blot analysis, proteins

were seperated by 10% SDS-PAGE and were processed as described above. For MS/MS

analysis, the eluted proteins were separated by SDS-PAGE, the lanes were cut in 10

slices and in-gel digestion was performed with 20 g/mL trypsin (Trypsin Gold, Mass

Spectrometry Grade, Promega) according to Shevchenko et al., [29].

Mass spectrometric analysis

Tryptic peptides were injected onto an Ultimate II nano-LC (Dionex) coupled to Esquire

HCT quadrupole ion trap mass spectrometer (Bruker Daltonics) via online nano

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

electrospray ion source. Peptides were separated in a Pepmap 75, 75 m i.d., 150 mm

long reverse-phase column packed with 3 m C18 silica particles (Dionex). The elution

from the column and the data acquisition have been previously described [20]. The

conditions for MS/MS analysis with the Orbitrap Velos have been recently described

[21].

Bioinformatic analysis

The generated data were converted to mzXML files by using the compassXport software

(Bruker Daltonics) and MS/MS data were uploaded on and analysed by the

Computational Proteomics Analysis System (CPAS) pipeline [30] with Mascot or X!

Tandem as search engines. Peptide Prophet [31] and Protein Prophet [32] scores were

generated to provide statistical analysis of protein identification by the Transproteomic

Pipeline (TPP) from the Seattle Institute of Systems Biology, which was integrated into

the CPAS. The following criteria were used to establish the confidence of the identified

proteins: protein prophet ≥0.9, high peptide prophet scores (≥0.7), low Expect value,

assignment of most prominent peaks as b- or y- ions, experimental sequence-theoretical

sequence alignment, K or R as the most C-terminal residue, aspartic acid- or proline-

directed fragmentation. To avoid false-positive hits, only proteins matched with at least 2

peptides or 1 unique peptide were considered as true hits while all MS/MS spectra were

manually inspected. Relative abundance was estimated by the spectral counting method

[33]. For the pseudopodial proteins, pathway analysis, molecular function and biological

processes of proteins were obtained from GeneOntology (www.geneontology.org) and

pie charts were created in PANTHER classification system [34]. The trial version of the

Ingenuity Pathway Analysis program was used for functional annotations, signalling

pathway analysis and creation of bar charts [35]. For the -arrestin-interacting proteins,

we performed two biological replicates and two technical replicates. The PANTHER

classification system was used to classify proteins based on their molecular function and

PANTHER protein class. In the analyses, we used only proteins that were not identified

in the control experiment (transfection with an empty vector or untransfected cells) or

were identified with significantly higher (ten-fold) spectral counts in the transfected

conditions.

Results

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

Endogenous expression of PAR-2 in MCF-7 cells and activation of ERK cascade by

uniform concentrations of P2AP

In order to study PAR-2 signalling in its physiological context, we used MCF7 cells, a

breast cancer cell line which expresses PAR-2 endogenously (Fig. 2A). Signalling

downstream of PAR-2 is strongly associated with ERKs and has important roles in

migration [18]. The effect of chemotactic agents at uniform or gradient concentrations on

the ERK pathway is cell-type dependent [36]. In order to test whether uniform

concentrations of the PAR-2-activating peptide SLIGRL (P2AP) induce ERK

phosphorylation, MCF-7 (low invasive) and MDA-MB-231 (highly invasive) cells were

deprived of serum and then treated with P2AP or FCS for 0-120 min. As expected, the

addition of P2AP for 30 min in culture significantly induced the ERK pathway in both

cell lines and the pERK levels remained higher than basal over a time course of at least 2

hours (Fig. 2B). P2AP did not induce any change in the low levels of basal pERK in the

non-invasive CAMA and ZR-75 cells (data not shown). These results confirmed that

PAR-2 activation by uniform concentrations of P2AP induced the ERK pathway in MCF-

7 cells.

Migratory effect of P2AP on MCF-7 cells

Since MCF-7 cells responded well to uniform concentrations of P2AP with respect to

ERK activity upon PAR-2 activation and as this cell line is considered as low invasive

[37], we tested whether PAR-2 could sense gradient concentration of its ligand and

trigger the ERK pathway as well as cell migration. PAR-2 activation is known to enhance

migration of other breast cancer cells via the ERK pathway, including MDA-MB-231 and

MDA-MB-468, as well as of mouse embryonic fibroblast cells NIH3T3 [10, 18] and

there is one study that tested this in MCF-7 [14]. We studied the effect of PAR-2

activation on cell migration with the agarose migration assay. Indeed, even though the

cells did not migrate under the gel in the agarose migration assay, probably due to their

large size or their slow migration rate and their need for longer incubation times in order

to continue migrating under the agarose, they accumulated at the side of the well towards

P2AP (Fig. 3A). These results are consistent with Kamath et al. [14] and suggest that

MCF-7 cells sensed the gradient of P2AP, and the subsequent activation of PAR-2

promoted their migration.

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

Proteins and signalling pathways enriched in the pseudopodia of breast cancer cells.

The migration of cells is characterized by the extension of a leading edge and the

formation of a pseudopodium [3]. Since activation of PAR-2 promoted the migration of

MCF-7, we aimed to identify the proteins that accumulate at the pseudopodia of

migrating MCF-7 cells upon activation of PAR-2, by the method developed by Klemke

and colleagues [4]. First, we found the optimum timepoint to harvest the dynamic

structure of pseudopodia at the time they were formed. The P2AP was left in the lower

compartment for 0, 30, 60, 90 or 120 min and the PD were harvested. The protein

amounts in each time point were: low at 30 and 90 min. and almost minimal at 0 and 120

min. The highest amount of proteins was harvested at the 60 min suggesting that this is

the timepoint of pseudopodial formation (data not shown). Therefore, for the rest of the

study we stimulated the cells for 60 min before PD purification.

The efficiency of the purification was controlled with Western blots against the

poly[ADP-ribose]polymerase 1 (PARP-1). Accordingly, PARP-1 was enriched in the cell

body fractions and only trace amounts were detected in the pseudopodia fraction (Fig. 3C

upper panel). Similarly, the band that corresponds to histones was absent from the PD

fraction (Fig. 3B). These results showed the high efficiency of pseudopodia purification

from migratory MCF-7 cells. Next, the ERK levels in each compartment were measured.

Phospho-ERK1/2 levels were increased in both CB and PD upon PAR-2 activation by

2.4- and 2.8-fold, respectively (Fig. 3C lower panel and Fig. 3D). More detailed, pERK1

levels showed an increase of similar magnitude (2.95 and 3.05 times, respectively) while

pERK2 levels showed higher increase in the PD (2.16-fold) than in the CB (1.57-fold).

Two complementary workflows were used in order to identify PAR-2-regulated

pseudopodial proteins and pathways during MCF-7 cell migration: (i) the pools of

proteins from the pseudopodia and cell bodies were analysed by SDS-PAGE and the

entire lanes were sliced into 10 regions (Fig. 3B); (ii) the pools of proteins from the

pseudopodia and cell bodies were digested by trypsin and phosphopeptides were

enriched. Relative quantitation was performed with the spectral counting method;

proteins with PD/CB ratio ≥ 1.5 were considered enriched in the pseudopodium, whereas

those with a ratio ≤ 0.66 were considered to be rather excluded from the PD [4, 20, 33].

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

The sample fractionation at the protein level in workflow 1 was crucial for the

identification of 3.2 times more proteins compared to workflow 2. In total, 250 non-

redundant proteins were identified, of which 207 with workflow 1 and 63 with workflow

2 while 20 proteins were identified in both (that is 32% of the total proteins identified in

workflow 2). Comparative analysis of CB and PD fractions showed a total of 93 proteins

enriched in the PD, 117 proteins were rather depleted from PD, while 60 did not show

any significant change. To increase the confidence of the identification, of the 20 proteins

that overlapped in workflows 1 and 2, only those that showed identical enrichment

profiles in both workflows were considered as PD enriched proteins (Fig. 4 and Table 1).

Protein pathway analysis by the PANTHER Classification System [34] revealed the

pathways that are most represented by the enriched proteins. In the PD, some of those

were the apoptosis, interleukin or Huntington disease signalling pathways. Interestingly,

several other receptor signalling pathways, such as dopamine receptor, EGF, FGF and

PDGF, were also enriched in the PAR-2 mediated pseudopodia. In the CB, DNA

replication-related pathways as well as biosynthetic and metabolic pathways (arginine, O-

antigen, pyruvate and others) were enriched (Fig. 4). Furthermore, the breakdown of

biological processes of the PD-enriched proteins was as follows: cellular processes

(34.6%), transport (23.5%), cell communication (12.3%), cellular component

organization (21%), metabolic process (49.4%), developmental processes (22.2%) and

others (SI Fig. 3).

Identification of -arrestin-interacting proteins

Among the proteins found enriched in the pseudopodia of MCF-7 cells were the two

large adaptins of the clathrin-associated adaptor protein (AP) complex (AP-2 complex

subunit alpha-1 and subunit beta) as well as the clathrin heavy chain and coatomer

subunits, indicating the significance of clathrin-mediated endocytosis to pseudopodia

formation. The AP complex regulates the formation of the clathrin-coated vesicles, which

are sequestered by -arrestins, the important mediators of 7-TMR endocytosis. As PAR-2

is dependent on the scaffolding ability of -arrestins for PAR-2 endocytosis as well as for

downstream signalling, and the PAR-2-regulated -arrestin interactome is far from

complete, we aimed to identify the proteins that are sequestered by -arrestins and hence

to reveal novel -arrestin-dependent functions triggered by PAR-2 activation.

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

In order to achieve this, we transiently transfected MCF-7 cells with FLAG-tagged -

arrestin-1 or -arrestin-2 in MCF-7 cells with endogenous levels of PAR-2, and after

treatment with either P2AP, FCS or serum-free medium, the -arrestins and their

interacting partners were immunoprecipitated by using the FLAG purification system. As

controls, we performed parallel experiments with cells transfected with an empty vector

or untransfected. The proteins identified in these negative controls were considered as

background and excluded from further analyses (SI table 11). Furthermore, proteins that

are usual contaminants in proteomics experiments, such as keratins, albumin,

glyceraldehyde-3-phosphate dehydrogenase, heat shock protein, desmoplakin, or others

that interact unspecifically with the anti-FLAG beads, were excluded from our analyses.

With the above stringent criteria, we identified 418 proteins that co-immunoprecipitate

with -arrestins in all conditions. Of these, 223 interacted with -arrestin-1, 325 with -

arrestin-2 while several of them interacted with both. More specifically, 115 proteins

interacted with -arrestin-1 in uninduced conditions, and 35 of those interactions were

lost after treatment; 28 interacted only in FCS-treatment; 57 only in PAR-2 activating

conditions, while 23 interacted after both treatments (table 2 and SI table 9). Similarly,

with -arrestin-2, 186 proteins interacted in uninduced conditions and 84 of these were no

longer bound after treatment; 55 bound to -arrestin-2 only upon FCS treatment; 65 only

in PAR-2 activating conditions, while 19 interacted after both treatments (table 3 and SI

table 9). Finally, a number of proteins were found associated with one -arrestin in

untreated conditions, but upon treatment were found associated with the other -arrestin,

and vice versa.

In order to increase the confidence of our proteomics screen, we show by immunoblotting

that two of the identified protein (nucleophosmin and heat shock protein 70) co-

immunoprecipitate specifically with -arrestin-2 upon PAR-2 activation.

Canonical pathway analysis of -arrestin interactors

Using Ingenuity Pathway Analysis, we generated and compared the canonical pathways

of the proteins interacting with either -arrestin (SI Fig. 3). Interestingly, after PAR-2

activation, -arrestin-1 interactions are more abundant than -arrestin-2 in pathways of

cell junction signalling, gap junction signalling, intrinsic prothrombin activation pathway,

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

dendritic cell maturation. On the other hand, -arrestin-2 appears to be more involved in

interleukin signalling, actin cytoskeleton signalling, clathrin-mediated endocytosis or

other endocytic pathways, regulation of actin based motility by Rho, RhoA signalling,

RhoGDI signalling, protein kinase A signalling, granzyme A signalling, PI3K/AKT

signalling and others. We also used PANTHER to categorize proteins based on their GO

Molecular Function and Biological Process (SI Fig. 5A-I) as well as PANTHER Protein

Class (Tables 2, 3). In terms of GO Molecular Function, -arrestin-1-interacting proteins

in untreated cells have binding, catalytic, enzyme regulator, transcription factor and

structural molecule activities and when treated with P2AP, they are additionally involved

in receptor, motor and transporter activities. For β-arrestin-2-interacting proteins in

untreated cells, they are involved in the same activities as for -arrestin-1 in addition to

receptor and transporter activities and in PAR-2-activating conditions they are

additionally engaged with ion channel and translation regulator activities.

Discussion

Protease inhibitors have been used in the treatment of several pathologies, such as

asthma, emphysema, and others [8, 38], and over the past 10 years the possibility of using

protease inhibitors or PAR-2 antagonists as therapeutics in tumour metastasis has been

investigated [8]. Taking into consideration the recent finding that actively migrating

cancer cells secrete trypsin proteases [10] as well as the heterogeneous nature of cancer,

it is conceivable that final stage tumour cells, with an acquired ability to invade, secrete

proteases for ECM degradation as well as for promoting migration of other earlier stage

tumour cells. Therefore, identifying the molecular mechanisms that govern the PAR-2-

mediated metastasis of breast cancer cells is critical for the complete understanding of

this process.

By identifying proteins enriched in the pseudopodia of MCF-7 breast cancer cells, which

endogenously express the PAR-2 receptor, we cataloged the most important signalling

pathways represented in this cell line (Table 1 and Fig. 4). In addition, we identified a

number of putative -arrestin interactors downstream of the PAR-2 receptor (Tables 2 and

3). Several of these identified proteins interact with -arrestin upon angiotensin II type 1a

receptor (AT1A-R) activation in HEK293 cells, but the majority are novel suggesting that

they might be cancer-specific interactions during cancer metastasis.

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

It is generally accepted that a migrating cell requires spatiotemporal re-arrangement of

proteins that lead to the formation of a leading edge and a trailing back [2, 3]. It is shown

here that, in MCF-7 breast cancer cells, the ERK pathway is activated by uniform

concentrations of PAR-2-activating peptide (P2AP) and the levels remain high over a

course of at least 2 hours (Fig. 2B). This is also the case with the highly invasive MDA-

MB-231, consistent with the findings of Ge et al., [10] (Fig. 2B). Upon gradient

concentrations of P2AP, an increase in the pERK1/2 levels is observed in the entire cell

but there is no spatial difference in either total pERK1/2 or pERK1 activity in cell bodies

and pseudopodia, while pERK2 is slightly enriched in the PD and thus preferred in the

regulation of the pseudopodia formation in MCF-7 cells (Fig. 3C, D). This suggests that

ERK1 and ERK2 have diverse functions and target distinct substrates leading to

cytoskeletal re-arrangments and subsequent formation of the pseudopodium. Whether

these functions are interchangeable is something that has to be studied further. The

relatively weak increase of pERK2 in PD could possibly explain the low ability of MCF-

7 to migrate and invade; perhaps a bigger increase would be observed in highly

metastatic cells.

Proteomic screens of the pseudopodial proteome have been published from experiments

in COS-7 cells in response to lipopolysaccharide (LPS) [4, 5] and in human astrocytoma

cells [39]. In the present study we used a cell line with low metastatic ability and we

induced migration by activating the endogenous PAR-2 receptor (Fig. 3A). We analysed

the proteins involved in the formation and extension of the pseudopodium after activation

of PAR-2. Subsequent analyses were focused on the proteins found enriched in the PD

fraction as these proteins are more likely to contribute to the formation of the

pseudopodia. Other proteins that are equally distributed in the PD and CB might also play

a role and they can be the subject of further studies, especially of phosphoproteomic

studies [40]. All the proteins confidently found enriched in the pseudopodium are

presented in Table 1 and none of them have been previously described as PAR-2

regulated. Despite our phosphopeptide enrichment step, we did not detect

phosphorylation sites, possibly due to the small size of pseudopodia formed by the MCF-

7 cells producing low protein abundant pseudopodial samples as starting material. Two of

the proteins that are candidates for further studies are nidogen-1 and nidogen-2. After

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

careful inspection of our mass spectrometry data we found that the peptides

corresponding to these proteins were not proteotypic. Therefore, it was not possible to

distinguish whether it is one of the two specific isoforms or both that were identified. In

the tables we have included only nidogen-1 but it is highly likely that both are involved in

pseudopodia formation as they have around 50% sequence similarity and they perform

similar or perhaps identical functions [57]. Kubota et al., proposed a possible function for

nidogen-1 in migration of C. elegans, during the interaction of integrins with the

basement membrane, and that this function is evolutionarily conserved in mammals [41].

Furthermore, nidogen-1 knockout mice presented impaired wound healing [42]. What

makes our finding more intriguing is the recent establishment of nidogen-2 as a

biomarker for ovarian cancer, since it was overexpressed in the serum of women with

ovarian cancer [43].

The PAR-2-regulated signalling pathways represented by the CB- and PD-enriched

proteins are listed in Fig. 4. The results are consistent with the findings of Wang et al. [4]

study, in which EGF, PDGF and Huntington's disease signalling were also enriched in the

PD. In addition, here we show that clathrin-mediated endocytosis as well as heme

biosynthesis, interleukin and PI3 kinase pathways signalling together with dopamine

receptor and JAK/STAT signalling are also important pathways in cancer cell migration.

These new findings are likely to be attributed to the migration of cancer cells specifically.

With respect to the CB-enriched pathways, we found cell cycle-related and various

metabolic pathways, similarly to the Wang et al study [4]. The AP-2 complex large

subunits alpha-1 and beta were found only in PD. The AP-2 complex is localized at the

plasma membrane and is associated with -arrestin and clathrin-mediated endocytosis but

here we show that upon PAR-2-mediated pseudopium formation AP-2 complex and

clathrin signalling are enriched in the PD. The latter strongly suggests that receptor

internalization is a critical process for the promotion of migratory signals. This is in

agreement with previous studies [10, 19] and raises the question of whether blocking

receptor internalization could be used in therapies against metastasis. Furthermore,

although the actin cytoskeleton is heavily regulated in both compartments, the Rho

signalling pathway was strongly enriched in PD showing the importance of this pathway

in cell polarization and invasion, in agreement with others [44].

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

It has been reported by several groups that -arrestin is an important scaffold protein

initiating signalling pathways during receptor internalization via clathrin-coated pits.

Hence, we pulled-down -arrestins after activating the PAR-2 receptor and we identified

-arrestin-interacting proteins. The -arrestin interactome has been studied by Lefkowitz

and colleagues using proteomics downstream of the angiotensin II type 1a receptor in

human embryonics kidney cells (HEK293) [45] and by other groups using other methods

(reviewed in [46-48]), therefore, here we report the first interactome in cancer cells and

upon stimulation of PAR-2.

Several proteins identified in our study were previously known as -arrestin interactors,

such as 14-3-3, serine/threonine kinase 38, clathrin heavy chain 1 and 2, copine, annexin,

fatty acid synthase, ATP synthase and , phosphatases, actins, tubulins, tubulin -3, -8,

F-actin capping protein, myosin-9, filamin, kinesin, dynein, titin, calpain-1 catalytic

subunit, dehydrogenase, phospholipase, tropomyosin, histone H1, heat shock protein 70,

70.1, 90, T-complex protein 1, carboxypeptidase, treacle, nucleophosmin, nucleolin,

hnRNP, ribosomal proteins S3, S7, L7, L19, L22, P0, P1, P2 and others (SI table 4) [45].

In addition, several isoforms of proteins that are known to interact with -arrestins, such

as phosphatases and ribosomal proteins, were also found here. We also excluded proteins

that were found in other proteomic screens as -arrestin-interactors but in our strategy

were considered as background due to their identification in the negative controls (SI

table 11). Also note that in this study, we did not over-express the receptor, reducing the

possibility of unspecific binding due to exogenous protein expression.

Importantly, we identify new interactions, such as TRIM29, glutaredoxin, periplakin and

others, and these interactions might play a role in PAR-2 signalling in cancer. The finding

of alpha-actinin as well as several other actin isoforms in our screens suggests that actin

cytoskeleton and -arrestins interact to manage receptor endocytosis [49] and leading

edge formation [50]. Similarly, their interaction with S100A7 increases the role of -

arrestins in metastatic pathways [51]. It is also shown here that PAR-2 regulates the

activity of kinases such as cyclin-dependent kinase (cdk) 13, N-acetyl-D-glucosamine

kinase and microtubule-associated serine/threonine-protein kinase 2, in a -arrestin-

dependent manner. More specifically, cdk13 and N-acetyl-D-glucosamine kinase are

pulled down with both -arrestins in uninduced conditions but in induced conditions this

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

428

429

430

431

432

433

434

435

436

437

438

association is lost. The opposite seems to be the case for microtubule-associated

serine/threonine-protein kinase 2 (Tables 2, 3 and SI Table 9). Another two proteins we

showed here by mass spectrometry and independent co-IPs that are genuine -arrestin-

interacting proteins downstream of PAR-2 are nucleophosmin B23 (NPM1) and heat

shock protein 70 (HSP70) (Fig. 5). This is specifically interesting in the case of NPM1, as

it is a multifunctional protein that is over-expressed in cancer cells, compared to normal

cells, and it is considered a putative therapeutic target for hematological diseases [52-54].

Another protein found in our proteomic screen, the eukaryotic translation initiation factor

3 (eIF3), has been recently found as a -arrestin-2 interactor by another study [55]. This

interaction, which is mediated by epidermal growth factor (EGF) stimulation, promotes

binding of eIF3a to SHC and Raf-1 and subsequent suppression of the ERK pathway

[55]. Whether all these interactions take place in other types of cancer or are also

regulated by other GPCRs is a question that requires further study.

Furthermore, we compared the overlap between the -arrestin-interactors and the PD-

enriched protein screens in order to identify other proteins of special interest (Table 4).

These proteins are very likely to spatiotemporally translocate to the pseudopodium or

retained away from it in a -arrestin-dependent manner. For example, copine, T-plastin,

coatomer subunit beta and LDH-A were found associated with -arrestin-2 in untreated

conditions but when PAR-2 is activated this interaction is lost and the proteins are found

in pseudopodia. Alpha-actinin, AHNAK and elongation factor 1-alpha 1 are likely to be

recruited to the pseudopodia by -arrestins when PAR-2 is activated. Very importantly,

RNA helicase DDX5 has also been found in this overlap. DDX5 is a nuclear DEAD-box

containing protein with roles in development, miRNA pathway, in regulation of

transcription and RNA processing and ribosome biogenesis. It is mainly nuclear but it

was reported to translocate to the cytoplasm in order to recruit beta-catenin back to the

nucleus and control transcription [56]. In our screens, it was found as -arrestin-2 partner

only when cells were treated with FCS. Therefore, as the pseudopodia in our study were

formed upon PAR-2 activation, this shows that DDX5 is enriched at the PD when not

sequestered by -arrestin-2. As an additional validation of our screens, AP-2 complex

subunits alpha-1 and beta were found in both screens.

Conclusion

439

440

441

442

443

444

445

446

447

448

449

450

451

452

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

In conclusion, this study presents the proteins and the signalling pathways enriched in the

pseudopodia of migrating breast cancer cells by using pseudopodia purification and mass

spectrometry-based proteomics. Furthermore, by using immunoprecipitation and mass

spectrometry, several PAR-2-regulated -arrestin-1 and -arrestin-2 interactors were

identified and revealed novel functions. This study therefore expanded the possible

functions of -arrestins, especially during breast cancer cell migration, and provides

directions for future research.

Acknowledgment

We acknowledge all members of the Metodiev laboratory for support and useful

discussions. We are grateful to Robert Lefkowitz, for kindly providing the -arrestin-

expressing vectors used in the study, Yoanne Mousseau for technical help and critical

advice in setting up the agarose migration assay. NP is grateful to Daniel Fisher for his

continuous support. We also acknowledge Daniel Fisher, Liliana Krasinska and James

Hutchins for proofreading and for critical recommendations while writing the first

version of the manuscript.

470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

Figure legends.Figure 1.

Schematic representation of the experimental strategies performed in this study.

Figure 2.

Endogenous expression of PAR-2 in MCF7 cells and ERK activity in untreated, FCS

(10%) or PAR-2 (50 M) treated conditions. (A) Detection of PAR-2 mRNA in MCF7

cells by reverse transcription PCR. (B) Levels of ERK activity upon treatment with P2AP

(50 M) or FCS. Cells were serum-starved and then treated or not with the indicated

agents for the indicated times and pERK was detected by anti-phospho p44/p42 MAPK

(Cell Signaling) in clear lysates by western blotting. Tubulin was used as loading control

Figure 3.

Effect of gradient concentrations of P2AP on chemotaxis and on ERK activity in MCF-7

cells. (A) Under agarose migration assay: Three wells were generated essentially as

described in [28]. cells were seeded in the middle well while 50 M P2AP or RPMI only

were added in the side wells. The chemoattractant diffused through the agarose creating a

gradient to which cells responded by migrating towards the side with the chemoattractant.

The image is representative of 3 individual experiments. Bar: 100 m; (B) The pools of

proteins from cell bodies (CB) or pseudopodia (PD) separated by the pseudopodia

purification assay were analysed by 10% SDS-PAGE. Arrow indicates the histone

proteins. Lines indicate the sliced gel pieces from which proteins were digested and

analysed by mass spectrometry. (C) Immunoblots of PARP-1 for validation of PD

purification (upper panel) and pERK (lower panel) in PD and CB. (D) Levels of

pERK1/2, pERK1 and pERK2 in PD and CB.

Figure 4.

Comparison of the protein identifications from the 2 different proteomic workflows used

to identify PD and CB enriched proteins (see text for details). In parallel, proteins are

categorized based on the PD/CB ratio, with the PD enriched proteins to the right-hand

side and CB-enriched to the left-hand side. The most enriched signalling pathways

represented by these proteins in each fraction are listed; the complete PANTHER

pathway analysis is shown in SI figure 2.

485486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

Figure 5.

Validation of nucleophosmin (A) and heat shock protein 70 (HSP70; B) as -arrestin

interacting proteins by co-immunoprecipitation (co-IP). Control experiments were co-IPs

with -FLAG beads from untransfected cells or with unreactive protein A/G beads from

FLAG-tagged -arrestin-transfected cells.

SI Figure 1.

PANTHER pathway analysis of proteins enriched in the pseudopodia (left-hand panel) or

cell bodies (right-hand panel) [34]. Numbers in square brackets indicate the percentage of

the number of proteins corresponding to this pathway.

SI Figure 2

Gene Ontology (GO) terms of PD- and CD-enriched proteins in pie charts generated by

PANTHER Classification System (http://www.pantherdb.org/).

SI Figure 3

Ingenuity Pathway Analysis of proteins that co-immunoprecipitate with -arrestin-1 (dark

blue bars) or -arrestin-2 (light blue bars).

SI Figure 4 A-I.

Gene Ontology (GO) terms Molecular Function and Biological Process of -arrestin-

interacting proteins, upon treatment with FCS, P2AP or not treated, in pie charts

generated by PANTHER Classification System (http://www.pantherdb.org/).

References

[1] Ridley AJ, Schwartz MA, Burridge K, Firtel RA, Ginsberg MH, Borisy G, et al. Cell migration: integrating signals from front to back. Science. 2003;302:1704-9.[2] Devreotes P, Janetopoulos C. Eukaryotic chemotaxis: distinctions between directional sensing and polarization. Journal of Biological Chemistry. 2003;278:20445-8.[3] Chodniewicz D, Klemke RL. Guiding cell migration through directed extension and stabilization of pseudopodia. Experimental Cell Research. 2004;301:31-7.[4] Wang Y, Ding SJ, Wang W, Jacobs JM, Qian WJ, Moore RJ, et al. Profiling signaling polarity in chemotactic cells. Proceedings of the National Academy of Sciences of the United States of America. 2007;104:8328-33.

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536537538539540541542543544

[5] Wang Y, Kelber JA, Tran Cao HS, Cantin GT, Lin R, Wang W, et al. Pseudopodium-enriched atypical kinase 1 regulates the cytoskeleton and cancer progression [corrected]. Proc Natl Acad Sci U S A. 2010;107:10920-5.[6] Nystedt S, Emilsson K, Larsson AK, Strombeck B, Sundelin J. Molecular cloning and functional expression of the gene encoding the human proteinase-activated receptor 2. European Journal of Biochemistry. 1995;232:84-9.[7] Dery O, Corvera CU, Steinhoff M, Bunnett NW. Proteinase-activated receptors: novel mechanisms of signaling by serine proteases. American Journal of Physiology. 1998;274:C1429-52.[8] Ramachandran R, Noorbakhsh F, Defea K, Hollenberg MD. Targeting proteinase-activated receptors: therapeutic potential and challenges. Nat Rev Drug Discov. 2012;11:69-86.[9] D'Andrea MR, Derian CK, Santulli RJ, Andrade-Gordon P. Differential expression of protease-activated receptors-1 and -2 in stromal fibroblasts of normal, benign, and malignant human tissues. American Journal of Pathology. 2001;158:2031-41.[10] Ge L, Shenoy SK, Lefkowitz RJ, DeFea K. Constitutive protease-activated receptor-2-mediated migration of MDA MB-231 breast cancer cells requires both beta-arrestin-1 and -2. Journal of Biological Chemistry. 2004;279:55419-24.[11] Ikeda O, Egami H, Ishiko T, Ishikawa S, Kamohara H, Hidaka H, et al. Signal of proteinase-activated receptor-2 contributes to highly malignant potential of human pancreatic cancer by up-regulation of interleukin-8 release. International Journal of Oncology. 2006;28:939-46.[12] Shi X, Gangadharan B, Brass LF, Ruf W, Mueller BM. Protease-activated receptors (PAR1 and PAR2) contribute to tumor cell motility and metastasis. Mol Cancer Res. 2004;2:395-402.[13] Hjortoe GM, Petersen LC, Albrektsen T, Sorensen BB, Norby PL, Mandal SK, et al. Tissue factor-factor VIIa-specific up-regulation of IL-8 expression in MDA-MB-231 cells is mediated by PAR-2 and results in increased cell migration. Blood. 2004;103:3029-37.[14] Kamath L, Meydani A, Foss F, Kuliopulos A. Signaling from protease-activated receptor-1 inhibits migration and invasion of breast cancer cells. Cancer Research. 2001;61:5933-40.[15] Morris DR, Ding Y, Ricks TK, Gullapalli A, Wolfe BL, Trejo J. Protease-activated receptor-2 is essential for factor VIIa and Xa-induced signaling, migration, and invasion of breast cancer cells. Cancer Research. 2006;66:307-14.[16] Krupnick JG, Benovic JL. The role of receptor kinases and arrestins in G protein-coupled receptor regulation. Annual Review of Pharmacology and Toxicology. 1998;38:289-319.[17] Ossovskaya VS, Bunnett NW. Protease-activated receptors: contribution to physiology and disease. Physiological Reviews. 2004;84:579-621.[18] DeFea KA, Zalevsky J, Thoma MS, Dery O, Mullins RD, Bunnett NW. beta-arrestin-dependent endocytosis of proteinase-activated receptor 2 is required for intracellular targeting of activated ERK1/2. Journal of Cell Biology. 2000;148:1267-81.[19] Ge L, Ly Y, Hollenberg M, DeFea K. A beta-arrestin-dependent scaffold is associated with prolonged MAPK activation in pseudopodia during protease-activated receptor-2-induced chemotaxis. Journal of Biological Chemistry. 2003;278:34418-26.[20] Alldridge L, Metodieva G, Greenwood C, Al-Janabi K, Thwaites L, Sauven P, et al. Proteome profiling of breast tumors by gel electrophoresis and nanoscale electrospray ionization mass spectrometry. J Proteome Res. 2008;7:1458-69.[21] Greenwood C, Metodieva G, Al-Janabi K, Lausen B, Alldridge L, Leng L, et al. Stat1 and CD74 overexpression is co-dependent and linked to increased invasion and lymph node metastasis in triple-negative breast cancer. J Proteomics. in print.[22] Laporte SA, Oakley RH, Zhang J, Holt JA, Ferguson SS, Caron MG, et al. The beta2-adrenergic receptor/betaarrestin complex recruits the clathrin adaptor AP-2 during endocytosis.

545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592

Proceedings of the National Academy of Sciences of the United States of America. 1999;96:3712-7.[23] Luttrell LM, Ferguson SS, Daaka Y, Miller WE, Maudsley S, Della Rocca GJ, et al. Beta-arrestin-dependent formation of beta2 adrenergic receptor-Src protein kinase complexes. Science. 1999;283:655-61.[24] Blackhart BD, Emilsson K, Nguyen D, Teng W, Martelli AJ, Nystedt S, et al. Ligand cross-reactivity within the protease-activated receptor family. Journal of Biological Chemistry. 1996;271:16466-71.[25] Compton SJ, Cairns JA, Palmer KJ, Al-Ani B, Hollenberg MD, Walls AF. A polymorphic protease-activated receptor 2 (PAR2) displaying reduced sensitivity to trypsin and differential responses to PAR agonists. Journal of Biological Chemistry. 2000;275:39207-12.[26] Inoue H, Nojima H, Okayama H. High efficiency transformation of Escherichia coli with plasmids. Gene. 1990;96:23-8.[27] Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227:680-5.[28] Mousseau Y, Leclers D, Faucher-Durand K, Cook-Moreau J, Lia-Baldini AS, Rigaud M, et al. Improved agarose gel assay for quantification of growth factor-induced cell motility. Biotechniques. 2007;43:509-16.[29] Shevchenko A, Tomas H, Havlis J, Olsen JV, Mann M. In-gel digestion for mass spectrometric characterization of proteins and proteomes. Nat Protoc. 2006;1:2856-60.[30] Rauch A, Bellew M, Eng J, Fitzgibbon M, Holzman T, Hussey P, et al. Computational Proteomics Analysis System (CPAS): an extensible, open-source analytic system for evaluating and publishing proteomic data and high throughput biological experiments. J Proteome Res. 2006;5:112-21.[31] Keller A, Eng J, Zhang N, Li XJ, Aebersold R. A uniform proteomics MS/MS analysis platform utilizing open XML file formats. Mol Syst Biol. 2005;1:2005 0017.[32] Nesvizhskii AI, Keller A, Kolker E, Aebersold R. A statistical model for identifying proteins by tandem mass spectrometry. Anal Chem. 2003;75:4646-58.[33] Liu H, Sadygov RG, Yates JR, 3rd. A model for random sampling and estimation of relative protein abundance in shotgun proteomics. Analytical Chemistry. 2004;76:4193-201.[34] Thomas PD, Kejariwal A, Campbell MJ, Mi H, Diemer K, Guo N, et al. PANTHER: a browsable database of gene products organized by biological function, using curated protein family and subfamily classification. Nucleic Acids Res. 2003;31:334-41.[35] Calvano SE, Xiao W, Richards DR, Felciano RM, Baker HV, Cho RJ, et al. A network-based analysis of systemic inflammation in humans. Nature. 2005;437:1032-7.[36] Brahmbhatt AA, Klemke RL. ERK and RhoA differentially regulate pseudopodia growth and retraction during chemotaxis. Journal of Biological Chemistry. 2003;278:13016-25.[37] Thompson EW, Paik S, Brunner N, Sommers CL, Zugmaier G, Clarke R, et al. Association of increased basement membrane invasiveness with absence of estrogen receptor and expression of vimentin in human breast cancer cell lines. Journal of Cellular Physiology. 1992;150:534-44.[38] Eden E, Hammel J, Rouhani FN, Brantly ML, Barker AF, Buist AS, et al. Asthma features in severe alpha1-antitrypsin deficiency: experience of the National Heart, Lung, and Blood Institute Registry. Chest. 2003;123:765-71.[39] Beckner ME, Chen X, An J, Day BW, Pollack IF. Proteomic characterization of harvested pseudopodia with differential gel electrophoresis and specific antibodies. Lab Invest. 2005;85:316-27.[40] Thingholm TE, Jensen ON, Larsen MR. Analytical strategies for phosphoproteomics. Proteomics. 2009;9:1451-68.

593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640

[41] Kubota Y, Ohkura K, Tamai KK, Nagata K, Nishiwaki K. MIG-17/ADAMTS controls cell migration by recruiting nidogen to the basement membrane in C. elegans. Proceedings of the National Academy of Sciences of the United States of America. 2008;105:20804-9.[42] Baranowsky A, Mokkapati S, Bechtel M, Krugel J, Miosge N, Wickenhauser C, et al. Impaired wound healing in mice lacking the basement membrane protein nidogen 1. Matrix Biology. 2008;29:15-21.[43] Kuk C, Gunawardana CG, Soosaipillai A, Kobayashi H, Li L, Zheng Y, et al. Nidogen-2: a new serum biomarker for ovarian cancer. Clinical Biochemistry. 2010;43:355-61.[44] Vishnubhotla R, Sun S, Huq J, Bulic M, Ramesh A, Guzman G, et al. ROCK-II mediates colon cancer invasion via regulation of MMP-2 and MMP-13 at the site of invadopodia as revealed by multiphoton imaging. Lab Invest. 2007;87:1149-58.[45] Xiao K, McClatchy DB, Shukla AK, Zhao Y, Chen M, Shenoy SK, et al. Functional specialization of beta-arrestin interactions revealed by proteomic analysis. Proc Natl Acad Sci U S A. 2007;104:12011-6.[46] Lefkowitz RJ, Rajagopal K, Whalen EJ. New roles for beta-arrestins in cell signaling: not just for seven-transmembrane receptors. Mol Cell. 2006;24:643-52.[47] Lefkowitz RJ, Shenoy SK. Transduction of receptor signals by beta-arrestins. Science. 2005;308:512-7.[48] Min J, Defea K. beta-arrestin-dependent actin reorganization: bringing the right players together at the leading edge. Mol Pharmacol.80:760-8.[49] Lamaze C, Fujimoto LM, Yin HL, Schmid SL. The actin cytoskeleton is required for receptor-mediated endocytosis in mammalian cells. J Biol Chem. 1997;272:20332-5.[50] Min J, Defea K. beta-arrestin-dependent actin reorganization: bringing the right players together at the leading edge. Mol Pharmacol. 2011;80:760-8.[51] Nasser MW, Qamri Z, Deol YS, Ravi J, Powell CA, Trikha P, et al. S100A7 enhances mammary tumorigenesis through upregulation of inflammatory pathways. Cancer Res.72:604-15.[52] Colombo E, Alcalay M, Pelicci PG. Nucleophosmin and its complex network: a possible therapeutic target in hematological diseases. Oncogene. 2011;30:2595-609.[53] Okuwaki M. The structure and functions of NPM1/Nucleophsmin/B23, a multifunctional nucleolar acidic protein. J Biochem. 2008;143:441-8.[54] Yung BY. Oncogenic role of nucleophosmin/B23. Chang Gung Med J. 2007;30:285-93.[55] Xu TR, Lu RF, Romano D, Pitt A, Houslay MD, Milligan G, et al. Eukaryotic translation initiation factor 3, subunit a, regulates the extracellular signal-regulated kinase pathway. Mol Cell Biol.32:88-95.[56] Yang L, Lin C, Liu ZR. P68 RNA helicase mediates PDGF-induced epithelial mesenchymal transition by displacing Axin from beta-catenin. Cell. 2006;127:139-55.[57] Ulazzi L, Sabbioni S, Miotto E, Veronese A, Angusti A, Gafa R, et al. Nidogen 1 and 2 gene promoters are abberantly methylated in human gastointestinal cancer. Mol Cancer. 2007;6:17

641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679

680