E XP ER I ME NTAL C E LL RE S E ARCH

24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40Q1 41 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 73 74 75 76 77 78 79 80 81 82 83

] (]]]]) ]]]–]]]

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/yexcr

Research Article

CUB-domain containing protein 1 represses the epithelial phenotype of pancreatic cancer cells Shin Miuraa, Shin Hamadaa,n, Atsushi Masamunea, Kennichi Satohb, Tooru Shimosegawaa a

Division of Gastroenterology, Tohoku University Graduate School of Medicine, 1-1, Seiryo-machi, Aobaku, Sendai city, Miyagi 980-8574, Japan Division of Cancer stem cell, Miyagi Cancer Center Research Institute, Natori city, 47-1 Nodayama, Shiote, Medeshima, Natori city, Miyagi 9811293, Japan b

article information

abstract

Article Chronology:

The prognosis of pancreatic cancer is dismal due to the frequent metastasis and invasion to

Received 8 August 2013

surrounding organs. Numerous molecules are involved in the malignant behavior of pancreatic

Received in revised form

cancer cells, but the entire process remains unclear. Several reports have suggested that CUB-

17 December 2013

domain containing protein-1 (CDCP1) is highly expressed in pancreatic cancer, but its impact on

Accepted 18 December 2013

the invasive growth and the upstream regulator remain elusive. To clarify the role of CDCP1 in pancreatic cancer, we here examined the effects of CDCP1 knockdown on the cell behaviors of

Keywords: CUB-domain containing protein 1 Bone morphogenetic protein 4 Transforming growth factor-β Epithelial-mesenchymal transition Extracellular signal-regulated kinase

pancreatic cancer cells. Knockdown of CDCP1 expression in Panc-1 resulted in reduced cellular migration accompanied by the increased expression of E-cadherin and decreased expression of N-cadherin. Knockdown of CDCP1 attenuated the spheroid formation and resistance against gemcitabine, which are some of the cancer stem cell-related phenotypes. Bone morphogenetic protein 4 (BMP4) was found to induce CDCP1 expression via the extracellular signal regulated kinase pathway, suggesting that CDCP1 has a substantial role in the BMP4-induced epithelialmesenchymal transition. These results indicate that CDCP1 represses the epithelial phenotype of pancreatic cancer cells. & 2014 Published by Elsevier Inc.

Introduction Pancreatic cancer has a poor prognosis due to its rapid dissemination, invasive growth and chemo-resistance. In 2013, there were an estimated 45,220 new cases of pancreatic cancer in the United States, and 38,460 patients are expected to die from the disease [1]. Unfortunately, the therapeutic strategy against

pancreatic cancer has not been greatly improved for a long time. Thus, the identification of promising therapeutic target molecules is urgently needed. CUB-domain containing protein-1 (CDCP1) is a type I transmembrane glycoprotein [2–6]. The molecular architecture of CDCP1 consists of an extracellular domain including three CUBdomains, a trans-membrane domain and cytoplasmic domain

Abbreviations: BMP4, bone morphogenetic protein 4; BSA, bovine serum albumin; CDCP1, CUB-domain containing protein 1; CSC, cancer stem cell; DMEM, Dulbecco0 s modified Eagle0 s medium; DMSO, dimethylsulfoxide; EGF, epidermal growth factor; EMT, epithelialmesenchymal transition; ERK, extracellular signal-regulated kinase; FBS, fetal bovine serum; FGF, fibroblast growth factor; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HGF, hepatocyte growth factor; IL-6, interleukin-6; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide; PBS, phosphate-buffered saline; PKC, protein kinase c; TGF-β, transforming growth factor-β n

Corresponding author. Fax: þ81 22 717 7177. E-mail address: [email protected] (S. Hamada).

0014-4827/$ - see front matter & 2014 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.yexcr.2013.12.019

Please cite this article as: S. Miura, et al., CUB-domain containing protein 1 represses the epithelial phenotype of pancreatic cancer cells, Exp Cell Res (2014), http://dx.doi.org/10.1016/j.yexcr.2013.12.019

2

84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143

E XP E RI ME N TAL CE L L R ES E ARC H

including five tyrosine residues, which are potential phosphorylation sites [7]. The extracellular domain of 135-kDa full length CDCP1 can be cleaved by serine protease, producing 70-kDa cleaved CDCP1 [8–10]. Recent studies reported that cleaved CDCP1 can activate the Src/protein kinase c (PKC) δ/Akt pathway [11], which contributes to increased cell survival, cell migration, anoikis resistance and hypoxia resistance in several types of cancer cells [2,5,12–15]. Elevated CDCP1 expression was reported to correlate with poor prognosis in a variety of solid cancers such as renal cell carcinoma and pancreatic cancer [2,3]. However, the functional roles and the upstream regulators of CDCP1 expression in pancreatic cancer remain unknown. To clarify the role of CDCP1 in pancreatic cancer, we here examined the effects of CDCP1 knockdown on the cell behavior of pancreatic cancer cells.

Material and methods

] (]]]]) ]]]–]]]

incubated with goat anti-CDCP1 antibody overnight at 4 1C. The slides were incubated with biotinylated rabbit anti-goat IgG antibody, followed by peroxidase-conjugated streptavidin. Finally, the color was developed by incubating the slides for several minutes with diaminobenzidine (Dojindo). The degree of staining was defined as follows: negative, o20% positive cells were found; positive, Z20% positive cells were found.

Cell line and cell culture Human pancreatic cancer cell lines AsPC-1 and Panc-1 were purchased from American Type Culture Collection (Manassas, VA). AsPC-1, Panc-1 and its derivative cell lines were maintained in Dulbecco0 s modified Eagle0 s medium (DMEM) supplemented with 10% fetal bovine serum (FBS). Cell lines were incubated at 37 1C in a humidified incubator with 5% CO2.

Materials RNA extraction and quantitative RT-PCR 3-(4,5-Dimethylthiazole-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was obtained from Dojindo (Kumamoto, Japan). Recombinant human bone morphogenetic protein 4 (BMP4) and transforming growth factor-β (TGF-β) were purchased from R&D Systems (Minneapolis, MN). Recombinant human hepatocyte growth factor (HGF), interleukin-6 (IL-6), fibroblast growth factor (FGF) and epidermal growth factor (EGF) were purchased from Life Technologies (Carlsbad, CA). These cytokines were dissolved in phosphate-buffered saline (PBS) supplemented with 0.1% bovine serum albumin (BSA). Insulin–transferrin–selenium Supplement was purchased from Life Technologies. U0126, a specific MEK inhibitor was purchased from Merck (SanDiego, CA). U0126 was dissolved in dimethylsulfoxide (DMSO) and used at the indicated concentration. Gemcitabine was purchased from Wako (Osaka, Japan). Gemcitabine was dissolved in PBS and used at the indicated concentrations. Goat anti-CDCP1 antibody (ab1377) was from Abcam (Cambridge, MA). Mouse anti-α-tubulin antibody (T5168) was from Sigma-Aldrich (St. Louis, MO). Mouse antiE-cadherin antibody (610181) and anti-N-cadherin antibody (610920) were from BD Biosciences (Bedford, MA). Rabbit antiERK antibody (4695), anti phospho-ERK antibody (4370), antiPKCδ antibody (9616), anti-SRC antibody (2108), anti-CDCP1 antibody (4115) and peroxidase-conjugated anti-rabbit antibody (7074) were from Cell Signaling Technology (Beverly, MA). Peroxidase-conjugated anti-mouse antibody was from GE Healthcare (Piscataway, NJ). Other reagents were purchased from SigmaAldrich unless specifically described otherwise.

Immunohistochemistry Normal pancreatic tissues and pancreatic cancer tissues collected at the time of surgery were fixed in 10% paraformaldehyde overnight and embedded in paraffin wax. Immunohistochemical staining for CDCP1 was performed using a streptavidin–biotin– peroxidase complex detection kit (Histofine Kit, Nichirei, Tokyo, Japan). Briefly, tissue sections were deparaffinized and rehydrated in PBS. Following antigen retrieval with the target retrieval solution (Dako, Glostrup, Denmark), endogenous peroxidase activity was blocked by incubation with 0.3% H2O2. After the blocking with 10% normal rabbit serum, the sections were

Total cellular RNA was extracted using the mirVana™ miRNA Isolation Kit (Life Technologies) according to the manufacturer0 s protocol. The real-time RT-PCR analyses were performed using StepOnePlus real-time PCR system (Life Technologies). The primer sequences for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were described previously [16]. The primer sequences were as follows: GAPDH F; 50 -GGCGTCTTCACCACCATGGAG-30 GAPDH R; 50 -AAGTTGTCATGGATGACCTTGGC-30 , CDCP1 F; 50 -CTGAACTGCGGGGTCTCTATC-30 , CDCP1 R; 50 -GTCCCCAGCTTTATGAGAACTG-30 . The level of CDCP1 expression in each sample was normalized by the respective GAPDH expression level.

Establishment of CDCP1-knockdown cell lines The CDCP1-knockdown cell lines Di-5 and Di-13 were established by introducing the siRNA-expressing pBAsi-U6 Neo DNA vector (Takara Bio Inc., Ohtu, Japan) which targets 50 -GCAGCATTGACACATATCT-30 , corresponding to the nucleotide sequence of human CDCP1 in Panc-1. The control cell line PHU was established in a similar manner by introducing a non-targeting control siRNAexpressing vector that targets 50 -TCTTAATCGCGTATAAGGC-30 , which does not correspond to any known human mRNA.

Transient knockdown of CDCP1 The transient knockdown of CDCP1 in AsPC-1 was performed using ON-TARGET plus SMARTpool siRNA against human CDCP1 (Dharmacon, Chicago, IL) at a final concentration of 100 nM. As a negative control, the ON-TARGET plus non-targeting pool siRNA (Dharmacon) was used. The siRNA transfection was performed using lipofectamine 2000 (Life Technologies).

Cell growth assay One thousand cells per well were plated in 6-well plates in normal growth media. Cells were trypsinized and counted at days 1, 3, 5 and 7 after plating.

Please cite this article as: S. Miura, et al., CUB-domain containing protein 1 represses the epithelial phenotype of pancreatic cancer cells, Exp Cell Res (2014), http://dx.doi.org/10.1016/j.yexcr.2013.12.019

E X PE R IM EN TA L C ELL R E S EA RC H

144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203

] (]]]]) ]]]–]]]

3

Scratch assay

Cell viability assay

Cells were cultured until 100% confluence in 12-well plates and serum-starved for 24 h prior to the assay. A monolayer of the cells was scratched by a sterile tip, and cellular migration towards the scratched area was assessed at the indicated time points.

Ten thousand cells per well were plated in 96-well plates. Cells were grown in normal growth media for 24 h and then treated with gemcitabine or CPT-11 at the indicated concentrations. After 72 h of incubation, cell viability was measured by the MTT assay. Cells were treated with 5 μg/ml of MTT solution for 2 h and then solubilized in DMSO. Optical density was measured by a spectrophotometer at a wavelength of 570 nm.

Two chamber assay A two-chamber assay was performed using an 8 μm pore 24-well BD FALCON™ Cell Culture Insert (BD). Five thousand cells per upper chamber were plated into the Cell Culture Insert in normal growth media and the lower chambers were also filled with normal growth media. Cellular migration was assessed 24 h after plating. Migrated cells were counted in five random fields (X200). For the BMP4-induced cellular migration, each cell line was treated with 50 ng/ml of BMP4 for 72 h and then subjected to the two-chamber assay in a similar manner (two-chamber assay was performed using normal growth media).

Quantification of cell-to-cell contact The number of cells without cell-to-cell contact (single cell) and the total number of cells were counted in random 5 high power field. The ratio of the single cell number to the total cell number in each field was calculated.

Spheroid culture Cells were cultured in low-adhesion coated EZ-BindShut II (ASAHI TECHNO GLASS CORP, Tokyo, Japan) 6-well plates in serum-free media supplemented with 10 ng/mL FGF, 20 ng/mL EGF and 2.75 ng/mL selenium. After 7 days of culture, the numbers of formed spheroids were counted in five random fields (X40).

Western blotting Cells were lysed, and total cell lysates ( 100 μg) were subjected to electrophoresis using NuPAGE 4–12% Bis–Tris Gel (Life Technologies). Electrophoresed samples were transferred to an Immobilon-P membrane (Merck Millipore, Billerica, MA), and the membrane was incubated overnight at 4 1C with primary antibody. After incubation with peroxidase-conjugated antibody, reactive bands were detected using ECL™ western blotting detection reagents (GE Healthcare). The specific bands were subjected to densitometry analysis using Image J software (http://rsbweb.nih.gov/ij/index.html).

Statistical analysis Statistical analysis was performed using JMP (SAS Institute, Cary, NC). Each experiment was repeated at least three times, and representative data are shown. The differences between two groups were analyzed by Student0 s t-test. The differences between multiple groups were analyzed using the Tukey–Kramer method. A p value of o0.05 was regarded as statistically significant. Data are presented as meanþSEM.

Fig. 1 – (A–C) Immunohistochemistry of CDCP1 in normal pancreatic tissues. Arrow indicates pancreatic duct. Original magnification is X200. (D–F) Immunohistochemistry of CDCP1 in pancreatic cancer tissues. Arrowhead indicates cancer cells. Original magnification is  200. Please cite this article as: S. Miura, et al., CUB-domain containing protein 1 represses the epithelial phenotype of pancreatic cancer cells, Exp Cell Res (2014), http://dx.doi.org/10.1016/j.yexcr.2013.12.019

4

204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263

E XP E RI ME N TAL CE L L R ES E ARC H

] (]]]]) ]]]–]]]

Fig. 2 – (A) CDCP1 induction by the stimulation of various cytokines in Panc-1 was assessed by quantitative real-time RT-PCR (n¼3). BMP4 treatment showed the highest induction of CDCP1. ** indicates po0.01. * indicates po0.05. n.s., Not significant. (B) BMP4 increased the protein level expression of CDCP1 in Panc-1. BMP4 treatment showed the highest induction of CDCP1. The α-tubulin is displayed as a loading control. Arrow indicates full-length CDCP1 and arrowhead indicates cleaved form CDCP1. Lower panel shows densitometry analysis (n¼3). ** indicates po0.01. * indicates po0.05. n.s., not significant. (C) Effect of ERK pathway inhibition by U0126 on the BMP4-induced CDCP1 induction was assessed by Western blot. Pre-treatment of Panc-1 by 10 μM U0126 reduced the expression level of CDCP1 after the BMP4 treatment (50 ng/ml). Upper left panel shows inhibition of ERK phosphorylation by U0126. Upper right panel shows inhibition of CDCP1 induction by U0126. The α-tubulin is displayed as a loading control. Arrow indicates full-length CDCP1 and arrowhead indicates the cleaved form of CDCP1. Lower panel shows densitometry analysis (n¼3). ** indicates po0.01.

Results CDCP1 is highly expressed in pancreatic cancer tissues We examined the expression of CDCP1 in normal and pancreatic cancer tissues by immunohistochemistry. As shown in Fig. 1A–C, CDCP1 expression was barely detectable in normal pancreatic tissues. In contrast, pancreatic cancer tissues showed high expression of CDCP1 (Fig. 1D–F). We assessed 9 normal and 42 cancerous pancreatic tissues. CDCP1 expression was not detectable in

normal pancreatic tissues (0/9, 0%) and 36 cancerous pancreatic tissues showed positive staining for CDCP1 (36/42, 85.7%). These findings agree with previous reports showing elevated expression of CDCP1 in colorectal tumors compared with normal tissues [17].

BMP4 induces CDCP1 expression in pancreatic cancer cells A previous report suggested that CDCP1 has a tumor-promoting role by enhancing the invasive growth of cancer cells [2]. Since the upstream

Please cite this article as: S. Miura, et al., CUB-domain containing protein 1 represses the epithelial phenotype of pancreatic cancer cells, Exp Cell Res (2014), http://dx.doi.org/10.1016/j.yexcr.2013.12.019

E X PE R IM EN TA L C ELL R E S EA RC H

264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323

] (]]]]) ]]]–]]]

5

Fig. 3 – (A) Expression levels of CDCP1 mRNA in control cell line PHU and CDCP1-knockdown cell lines Di-5 and Di-13 were assessed by quantitative real-time RT-PCR (n ¼3). ** indicates po0.01. CDCP1 expression in Di-5 or Di-13 was reduced compared with PHU. (B) Protein level expression of CDCP1 in each cell line was assessed by Western blot. The α-tubulin is displayed as a loading control. Arrow indicates full-length CDCP1 and arrowhead indicates the cleaved form of CDCP1. CDCP1 expression in Di-5 and Di-13 was reduced compared with PHU. SRC and PKCδ expression was not altered. Right panel shows densitometry analysis (n ¼3). ** indicates po0.01. n.s., not significant. (C) Cell proliferation was assessed by direct cell count. PHU, Di-5 and Di-13 revealed similar proliferation (n¼ 6). n.s., not significant. (D) Cellular migration of each cell line was assessed by the scratch assay. Cellular migration of Di-5 or Di-13 toward the scratched area was reduced. Original magnification is  100. (E) Cellular migration of each cell line was assessed by two chamber assay. Cellular migration of Di-5 and Di-13 was significantly reduced compared with PHU (n¼ 5, ** indicates po0.01). Original magnification is  200. (F) Di-5 and Di-13 showed tight cell-to-cell contact compared with PHU (left panel). Original magnification is X200. Right panel shows the quantification of cell-to-cell contact in random high power field (n¼ 5, ** indicates po0.01). E-cadherin and N-cadherin expressions in each cell line were assessed by Western blot (lower panel). The α-tubulin is displayed as a loading control. E-cadherin expression in Di-5 or Di-13 was increased compared with PHU. In contrast, N-cadherin expression in Di-5 or Di-13 was decreased compared with PHU. Please cite this article as: S. Miura, et al., CUB-domain containing protein 1 represses the epithelial phenotype of pancreatic cancer cells, Exp Cell Res (2014), http://dx.doi.org/10.1016/j.yexcr.2013.12.019

6

324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383

E XP E RI ME N TAL CE L L R ES E ARC H

] (]]]]) ]]]–]]]

Fig. 4 – (A) Protein level expression of CDCP1 was assessed by Western blot. The α-tubulin is displayed as a loading control. Arrow indicates full-length CDCP1 and arrowhead indicates the cleaved form CDCP1. Transient knockdown of CDCP1 efficiently decreased CDCP1 expression in AsPC-1. Lower panel shows densitometry analysis (n¼3). ** indicates po0.01. (B) Cellular migration of AsPC-1 after the introduction of control siRNA or CDCP1 siRNA was assessed by two chamber assay. Transient knockdown of CDCP1 decreased cellular migration (n¼ 5, ** indicates po0.01). Original magnification is  200. (C) E-cadherin expression in control siRNA- or CDCP1 siRNAintroduced AsPC-1 was assessed by Western blot. The α-tubulin is displayed as a loading control. Transient knockdown of CDCP1 increased E-cadherin expression in AsPC-1. Lower panel shows densitometry analysis (n¼ 3). ** indicates po0.01.

E X PE R IM EN TA L C ELL R E S EA RC H

384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443

] (]]]]) ]]]–]]]

7

Fig. 5 – (A) Spheroid formation of Di-5 and Di-13 was significantly reduced compared with PHU (n¼ 5, ** indicates po0.01. * indicates po0.05). Original magnification is  40. B. Cell viability of each cell line after the gemcitabine treatment was assessed by the MTT assay (left panel). Di-5 and Di-13 revealed significantly lower cell viability after the gemcitabine treatment than PHU (n¼6, ** indicates po0.01). In contrast, the cell viability of Di-5 and Di-13 after the CPT-11 treatment was not decreased compared with PHU (right panel, n¼6).

regulator of CDCP1 has not yet been identified entirely, we tested several cytokines and growth factors that promote EMT in pancreatic cancer cells [18–22]. Among the ligands tested, BMP4 showed the highest induction of CDCP1 mRNA expression, while TGF-β and HGF showed significant induction (Fig. 2A). BMP4 also induced CDCP1 expression at the protein level (Fig. 2B). These results indicate that CDCP1 is a downstream target of the BMP signal in pancreatic cancer cells. We next tried to identify the downstream pathway responsible for BMP4-induced CDCP1 expression. A recent report described that CDCP1 is also a downstream target of the EGF signal, which increases cellular migration through extracellular signal-regulated kinase (ERK) activation [23]. Since BMP4 is known to activate the ERK pathway [19], the effect of ERK pathway inhibition was examined. Treatment by U0126 efficiently decreased the phosphorylation of ERK in the basal and BMP4-stimulated condition (Fig. 2C). The induction of CDCP1 by BMP4 was attenuated significantly by the pretreatment of Panc-1 with U0126 (Fig. 2C). Thus, BMP4 induced CDCP1 expression at least in part through the ERK pathway in Panc-1 cells.

CDCP1 knockdown decreases cellular migration and enhances epithelial phenotypes of pancreatic cancer cells To further evaluate the functional role of CDCP1 expression in pancreatic cancer cells, we established the stable CDCP1-knockdown

cell lines Di-5 and Di-13 and a control cell line, PHU. As shown in Fig. 3A and B, the CDCP1 expression levels were efficiently reduced in Di-5 and Di-13 compared with PHU both at the mRNA and protein levels. Knockdown of CDCP1 did not affect the expression level of total SRC or PKCδ (Fig. 3B). Cell proliferation of Di-5 or Di-13 was not decreased compared with PHU (Fig. 3C). The scratch assay indicated that the migration of cells toward the scratched edge remarkably decreased in Di-5 and Di-13 compared with PHU after 24 h and 48 h (Fig. 3D). The two chamber assay confirmed that the number of migrated cells decreased in Di-5 and Di-13 compared with PHU (Fig. 3E). These results suggested that knockdown of CDCP1 decreased cellular migration, in agreement with a previous report describing that CDCP1 promoted cancer-cell migration in gastric cancer and pancreatic cancer [7]. Decreased cellular migration is a characteristic feature of differentiated epithelial cells. We assessed the morphological changes and expression status of an epithelial marker and mesenchymal marker in CDCP1 knockdown cell lines. Di-5 and Di-13 showed tight cell-to-cell contact compared with PHU (Fig. 3F). The expression of E-cadherin, an epithelial marker that prevents cancer cell migration by mediating cell to cell contact [24], was increased in Di-5 and Di-13 compared with PHU, whereas the mesenchymal marker N-cadherin [25] expression was attenuated in Di-5 and Di-13 compared with PHU (Fig. 3F).

Please cite this article as: S. Miura, et al., CUB-domain containing protein 1 represses the epithelial phenotype of pancreatic cancer cells, Exp Cell Res (2014), http://dx.doi.org/10.1016/j.yexcr.2013.12.019

8

444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503

E XP E RI ME N TAL CE L L R ES E ARC H

] (]]]]) ]]]–]]]

Fig. 6 – (A) Expression of E-cadherin after the BMP4 treatment in each cell line was assessed by Western blot. Reduction of E-cadherin expression by BMP4 was attenuated in Di-5 and Di-13. The α-tubulin is displayed as a loading control. Lower panel shows densitometry analysis (n¼ 3). * indicates po0.05. n.s., not significant. (B) Expression of E-cadherin after the TGF-β treatment in each cell line was assessed by Western blot. Reduction of E-cadherin expression by TGF-β was retained in PHU, Di-5 and Di-13. The α-tubulin is displayed as a loading control. Lower panel shows densitometry analysis (n¼ 3). ** indicates po0.01. * indicates po0.05. (C) Cellular migration of each cell line after the BMP4 treatment was assessed by the two chamber assay. BMP4-induced cellular migration was significantly attenuated in Di-5 or Di-13. The basal migration was also reduced in Di-5 and Di-13 (n¼ 5, ** indicates po0.01, n.s., not significant). Original magnification is  200.

Together with the decreased cellular migration, these results indicate that CDCP1 knockdown causes the reversal of EMT. We also tested the transient knockdown of CDCP1 in another pancreatic cancer cell line, AsPC-1. Introduction of siRNA against human CDCP1 efficiently suppressed CDCP1 expression (Fig. 4A). The cellular migration after the introduction of CDCP1 siRNA or control siRNA was assessed by the two chamber assay, revealing decreased cellular migration of AsPC-1 into which CDCP1 siRNA had been introduced (Fig. 4B). E-cadherin expression was increased by

the CDCP1 knockdown (Fig. 4C), but no N-cadherin reduction was observed (data not shown). The alteration of N-cadherin expression by CDCP1 knockdown might be a long-term effect.

CDCP1 partly maintains the cancer stem cell (CSC)-related phenotypes The EMT of cancer cells is itself one of the CSC-related phenotypes [26]. We therefore pursued other CSC-related phenotypes affected

Please cite this article as: S. Miura, et al., CUB-domain containing protein 1 represses the epithelial phenotype of pancreatic cancer cells, Exp Cell Res (2014), http://dx.doi.org/10.1016/j.yexcr.2013.12.019

E X PE R IM EN TA L C ELL R E S EA RC H

504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563

by CDCP1 knockdown. The spheroid-forming ability of Di-5 and Di-13 was significantly reduced compared to PHU (Fig. 5A). Furthermore, the cell viability of Di-5 and Di-13 after the gemcitabine treatment decreased by 20% (Fig. 5B). We also tested a standard chemotherapeutic agent for colon cancer, CPT-11 [27], and found that CDCP1-knockdown did not sensitize the cells to CPT-11 (Fig. 5B). These results indicate that CDCP1 contributes to the maintenance of CSC-related cell functions in pancreatic cancer cells.

CDCP1 contributes to the E-cadherin repression during BMP4-induced EMT Finally, we examined the effect of CDCP1 knockdown in BMP4induced EMT. The reduction of E-cadherin expression by BMP4 was attenuated in Di-5 and Di-13 (Fig. 6A). However, the effect of TGF-β on E-cadherin expression was not affected by CDCP1 knockdown (Fig. 6B). Along with the retained E-cadherin expression, the cellular migration of Di-5 or Di-13 after the BMP4 treatment was significantly attenuated, while the cellular migration of PHU was significantly enhanced by BMP4 (Fig. 6C). These results suggested that CDCP1 contributes to the E-cadherin repression by BMP4, which might mediate distinct EMT signal from TGF-β in pancreatic cancer cells.

Discussion The major findings of this study are as follows. (1) Increased expression of CDCP1 in pancreatic cancer tissue compared with normal pancreas. (2) BMP4 induced CDCP1 at least in part through the ERK pathway. (3) Knockdown of CDCP1 increased the epithelial phenotypes of pancreatic cancer cells. (4) The CSCrelated characteristics (gemcitabine resistance and spheroid formation) were also attenuated by CDCP1 knockdown. These results suggest the possible contribution of CDCP1 in maintaining some of the CSC-related phenotypes. Indeed, a recent study identified CDCP1 and CD110 as biomarkers of migrating CSCs in colorectal cancer [28]. In that report, CDCP1 promoted the adhesion of cancer cells to the lung endothelial cells. Another study also described that CDCP1 plays a crucial role in the metastatic colonization of prostate cancer cells [11]. Until now, the tumor-promoting effects of CDCP1 were mainly attributed to the activation of intracellular signals within cancer cells, such as the Src pathway [12]. The interaction between cancer cells and stromal cells could form a cancer-promoting microenvironment [29–31], and the contribution of CDCP1 to this process should be examined. We also found CDCP1 induction by BMP4 for the first time. The EMT-inducing effect of BMP4 was previously reported in pancreatic cancer [19,32], and CDCP1 is involved in this signaling pathway. However, another EMT-inducer, TGF-β, showed less induction of CDCP1. ERK pathway inhibition attenuated BMP4induced CDCP1 expression, but both BMP4 and TGF-β activate the ERK pathway [18,19]. This difference might be attributed to differences in the downstream signals of BMP4 and TGF-β. For example, BMP-bound receptor complex phosphorylates intracellular Smad1, 5 and 8 [33–35], which leads to the formation of a complex with Smad4. In contrast, TGF-β phosphorylates Smad2 and 3, which also form a complex with Smad4 [36].

] (]]]]) ]]]–]]]

9

The maintenance of the CSC functions is regulated by multiple factors in a context-specific manner. TGF-β is reported to decrease the side population cells, which contain a putative CSC fraction [37]. On the other hand, blocking of the BMP signal by the BMP type I receptor inhibitor decreased the clonogenicity of breast cancer, suggesting BMP has a role in supporting CSC [38]. The induction of CDCP1 by BMP4 might indicate some roles of this factor during the maintenance of CSC-related characteristics in the current context. Further study is required to clarify the contribution of CDCP1 to the CSC functions in pancreatic cancer cells.

Conclusion In conclusion, we confirmed that CDCP1 represses the epithelial phenotype in pancreatic cancer cells. BMP4 induces CDCP1 expression through ERK activation, and CDCP1 knockdown attenuated BMP4-induced EMT.

Acknowledgment This work was supported by the Grant from the Japan Society for the Promotion of Science (23591008, 24790674 and 23390194) and the Research Committee of Intractable Pancreatic Diseases (Principal investigator: Tooru Shimosegawa) provided by the Ministry of Health, Labor and Welfare of Japan.

references [1] R. Siegel, D. Naishadham, A. Jemal, Cancer statistics, CA Cancer J Clin. 63 (2013) 11–30. [2] Y. Miyazawa, T. Uekita, N. Hiraoka, S. Fujii, T. Kosuge, Y. Kanai, Y. Nojima, R. Sakai, CUB domain-containing protein 1, a prognostic factor for human pancreatic cancers, promotes cell migration and extracellular matrix degradation, Cancer Res. 70 (2010) 5136–5146. [3] Y. Awakura, E. Nakamura, T. Takahashi, H. Kotani, Y. Mikami, T. Kadowaki, A. Myoumoto, H. Akiyama, N. Ito, T. Kamoto, T. Manabe, H. Nobumasa, G. Tsujimoto, O. Ogawa, Microarray-based identification of CUB-domain containing protein 1 as a potential prognostic marker in conventional renal cell carcinoma, J. Cancer Res. Clin. Oncol. 134 (2008) 1363–1369. [4] M. Scherl-Mostageer, W. Sommergruber, R. Abseher, R. Hauptmann, P. Ambros, N. Schweifer, Identification of a novel gene, CDCP1, overexpressed in human colorectal cancer, Oncogene 20 (2001) 4402–4408. [5] T. Uekita, L. Jia, M. Narisawa-Saito, J. Yokota, T. Kiyono, R. Sakai, CUB domain-containing protein 1 is a novel regulator of anoikis resistance in lung adenocarcinoma, Mol. Cell. Biol. 27 (2007) 7649–7660. [6] J.D. Hooper, A. Zijlstra, R.T. Aimes, H. Liang, G.F. Claassen, D. Tarin, J.E. Testa, J.P. Quigley, Subtractive immunization using highly metastatic human tumor cells identifies SIMA135/CDCP1, a 135 kDa cell surface phosphorylated glycoprotein antigen, Oncogene 22 (2003) 1783–1794. [7] T. Uekita, R. Sakai, Roles of CUB domain-containing protein 1 signaling in cancer invasion and metastasis, Cancer Sci. 102 (2011) 1943–1948. [8] T.A. Brown, T.M. Yang, T. Zaitsevskaia, Y. Xia, C.A. Dunn, R.O. Sigle, B. Knudsen, W.G. Carter, Adhesion or plasmin regulates tyrosine phosphorylation of a novel membrane glycoprotein p80/gp140/

Please cite this article as: S. Miura, et al., CUB-domain containing protein 1 represses the epithelial phenotype of pancreatic cancer cells, Exp Cell Res (2014), http://dx.doi.org/10.1016/j.yexcr.2013.12.019

10

564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

E XP E RI ME N TAL CE L L R ES E ARC H

CUB domain-containing protein 1 in epithelia, J. Biol. Chem. 279 (2004) 14772–14783. A.S. Bhatt, H. Erdjument-Bromage, P. Tempst, C.S. Craik, M.M. Moasser, Adhesion signaling by a novel mitotic substrate of src kinases, Oncogene 24 (2005) 5333–5343. Y. He, A. Wortmann, L.J. Burke, J.C. Reid, M.N. Adams, I. AbdulJabbar, J.P. Quigley, R. Leduc, D. Kirchhofer, J.D. Hooper, Proteolysis-induced N-terminal ectodomain shedding of the integral membrane glycoprotein CUB domain-containing protein 1 (CDCP1) is accompanied by tyrosine phosphorylation of its C-terminal domain and recruitment of Src and PKCdelta, J. Biol. Chem. 285 (2010) 26162–26173. B. Casar, Y. He, M. Iconomou, J.D. Hooper, J.P. Quigley, E.I. Deryugina, Blocking of CDCP1 cleavage in vivo prevents Aktdependent survival and inhibits metastatic colonization through PARP1-mediated apoptosis of cancer cells, Oncogene 31 (2012) 3924–3938. C.H. Benes, G. Poulogiannis, L.C. Cantley, S.P. Soltoff, The SRCassociated protein CUB Domain-Containing Protein-1 regulates adhesion and motility, Oncogene 31 (2012) 653–663. E.I. Deryugina, E.M. Conn, A. Wortmann, J.J. Partridge, T.A. Kupriyanova, V.C. Ardi, J.D. Hooper, J.P. Quigley, Functional role of cell surface CUB domain-containing protein 1 in tumor cell dissemination, Mol. Cancer Res. 7 (2009) 1197–1211. B.M. Emerling, C.H. Benes, G. Poulogiannis, E.L. Bell, K. Courtney, H. Liu, R. Choo-Wing, G. Bellinger, K.S. Tsukazawa, V. Brown, S. Signoretti, S.P. Soltoff, L.C. Cantley, Identification of CDCP1 as a hypoxia-inducible factor 2alpha (HIF-2alpha) target gene that is associated with survival in clear cell renal cell carcinoma patients, Proc. Natl. Acad. Sci. USA 110 (2013) 3483–3488. T. Uekita, M. Tanaka, M. Takigahira, Y. Miyazawa, Y. Nakanishi, Y. Kanai, K. Yanagihara, R. Sakai, CUB-domain-containing protein 1 regulates peritoneal dissemination of gastric scirrhous carcinoma, Am. J. Pathol. 172 (2008) 1729–1739. S. Hamada, A. Masamune, T. Takikawa, N. Suzuki, K. Kikuta, M. Hirota, H. Hamada, M. Kobune, K. Satoh, T. Shimosegawa, Pancreatic stellate cells enhance stem cell-like phenotypes in pancreatic cancer cells, Biochem. Biophys. Res. Commun. 421 (2012) 349–354. S.E. Perry, P. Robinson, A. Melcher, P. Quirke, H.J. Buhring, G.P. Cook, G.E. Blair, Expression of the CUB domain containing protein 1 (CDCP1) gene in colorectal tumour cells, FEBS Lett. 581 (2007) 1137–1142. V. Ellenrieder, S.F. Hendler, W. Boeck, T. Seufferlein, A. Menke, C. Ruhland, G. Adler, T.M. Gress, Transforming growth factor beta1 treatment leads to an epithelial-mesenchymal transdifferentiation of pancreatic cancer cells requiring extracellular signalregulated kinase 2 activation, Cancer Res. 61 (2001) 4222–4228. S. Hamada, K. Satoh, M. Hirota, K. Kimura, A. Kanno, A. Masamune, T. Shimosegawa, Bone morphogenetic protein 4 induces epithelial-mesenchymal transition through MSX2 induction on pancreatic cancer cell line, J. Cell. Physiol. 213 (2007) 768–774. T. Shirakihara, K. Horiguchi, K. Miyazawa, S. Ehata, T. Shibata, I. Morita, K. Miyazono, M. Saitoh, TGF-beta regulates isoform switching of FGF receptors and epithelial-mesenchymal transition, EMBO J. 30 (2011) 783–795. K. Okitsu, T. Kanda, F. Imazeki, Y. Yonemitsu, R.B. Ray, C. Chang, O. Yokosuka, Involvement of interleukin-6 and androgen receptor signaling in pancreatic cancer, Genes Cancer 1 (2010) 859–867. D. Xu, Y. Matsuo, J. Ma, S. Koide, N. Ochi, A. Yasuda, H. Funahashi, Y. Okada, H. Takeyama, Cancer cell-derived IL-1alpha promotes HGF secretion by stromal cells and enhances metastatic potential in pancreatic cancer cells, J. Surg. Oncol. 102 (2010) 469–477.

] (]]]]) ]]]–]]]

[23] Y. Dong, Y. He, B.L. de, M.S. Stack, J.W. Lumley, J.A. Clements, J.D. Hooper, The cell surface glycoprotein CUB domain-containing protein 1 (CDCP1) contributes to epidermal growth factor receptor-mediated cell migration, J. Biol. Chem. 287 (2012) 9792–9803. [24] H. Furuyama, S. Arii, A. Mori, M. Imamura, Role of E-cadherin in peritoneal dissemination of the pancreatic cancer cell line, panc1, through regulation of cell to cell contact, Cancer Lett. 157 (2000) 201–209. [25] Y. Shintani, M.A. Hollingsworth, M.J. Wheelock, K.R. Johnson, Collagen I promotes metastasis in pancreatic cancer by activating c-Jun NH(2)-terminal kinase 1 and up-regulating N-cadherin expression, Cancer Res. 66 (2006) 11745–11753. [26] C. Scheel, R.A. Weinberg, Cancer stem cells and epithelialmesenchymal transition: concepts and molecular links, Semin. Cancer Biol. 22 (2012) 396–403. [27] B.M. Wolpin, R.J. Mayer, Systemic treatment of colorectal cancer, Gastroenterology 134 (2008) 1296–1310. [28] W. Gao, L. Chen, Z. Ma, Z. Du, Z. Zhao, Z. Hu, Q. Li, Isolation and phenotypic characterization of colorectal cancer stem cells with organ-specific metastatic potential, Gastroenterology 145 (2013) 636–646.e5. [29] M. Erkan, G. Adler, M.V. Apte, M.G. Bachem, M. Buchholz, S. Detlefsen, I. Esposito, H. Friess, T.M. Gress, H.J. Habisch, R.F. Hwang, R. Jaster, J. Kleeff, G. Kloppel, C. Kordes, C.D. Logsdon, A. Masamune, C.W. Michalski, J. Oh, P.A. Phillips, M. Pinzani, C. Reiser-Erkan, H. Tsukamoto, J. Wilson, StellaTUM: current consensus and discussion on pancreatic stellate cell research, Gut 61 (2012) 172–178. [30] A. Masamune, T. Shimosegawa, Pancreatic stellate cells—multifunctional cells in the pancreas, Pancreatology 13 (2013) 102– 105. [31] A. Masamune, T. Shimosegawa, Signal transduction in pancreatic stellate cells, J. Gastroenterol. 44 (2009) 249–260. [32] K.J. Gordon, K.C. Kirkbride, T. How, G.C. Blobe, Bone morphogenetic proteins induce pancreatic cancer cell invasiveness through a Smad1-dependent mechanism that involves matrix metalloproteinase-2, Carcinogenesis 30 (2009) 238–248. [33] S. Kawai, C. Faucheu, S. Gallea, S. Spinella-Jaegle, A. Atfi, R. Baron, S.R. Roman, Mouse smad8 phosphorylation downstream of BMP receptors ALK-2, ALK-3, and ALK-6 induces its association with Smad4 and transcriptional activity, Biochem. Biophys. Res. Commun. 271 (2000) 682–687. [34] M. Kretzschmar, J. Doody, J. Massague, Opposing BMP and EGF signalling pathways converge on the TGF-beta family mediator Smad1, Nature 389 (1997) 618–622. [35] R. Nishimura, Y. Kato, D. Chen, S.E. Harris, G.R. Mundy, T. Yoneda, Smad5 and DPC4 are key molecules in mediating BMP-2-induced osteoblastic differentiation of the pluripotent mesenchymal precursor cell line C2C12, J. Biol. Chem. 273 (1998) 1872–1879. [36] A. Moustakas, S. Souchelnytskyi, C.H. Heldin, Smad regulation in TGF-beta signal transduction, J. Cell Sci. 114 (2001) 4359–4369. [37] S. Ehata, E. Johansson, R. Katayama, S. Koike, A. Watanabe, Y. Hoshino, Y. Katsuno, A. Komuro, D. Koinuma, M.R. Kano, M. Yashiro, K. Hirakawa, H. Aburatani, N. Fujita, K. Miyazono, Transforming growth factor-beta decreases the cancer-initiating cell population within diffuse-type gastric carcinoma cells, Oncogene 30 (2011) 1693–1705. [38] A.L. Balboni, J.A. Hutchinson, A.J. DeCastro, P. Cherukuri, K. Liby, M.B. Sporn, G.N. Schwartz, W.A. Wells, L.F. Sempere, P.B. Yu, J. DiRenzo, DeltaNp63alpha-mediated activation of bone morphogenetic protein signaling governs stem cell activity and plasticity in normal and malignant mammary epithelial cells, Cancer Res. 73 (2013) 1020–1030.

Please cite this article as: S. Miura, et al., CUB-domain containing protein 1 represses the epithelial phenotype of pancreatic cancer cells, Exp Cell Res (2014), http://dx.doi.org/10.1016/j.yexcr.2013.12.019

Q2

CUB-domain containing protein 1 represses the epithelial phenotype of pancreatic cancer cells.

The prognosis of pancreatic cancer is dismal due to the frequent metastasis and invasion to surrounding organs. Numerous molecules are involved in the...
3MB Sizes 0 Downloads 0 Views