The multifunctional growth factor midkine promotes proliferation and migration in pancreatic cancer.

Published OnlineFirst February 24, 2014; DOI: 10.1158/1541-7786.MCR-13-0467

Molecular Cancer Research

Cell Cycle and Senescence

The Multifunctional Growth Factor Midkine Promotes Proliferation and Migration in Pancreatic Cancer Tamina Rawnaq1, Luisa Dietrich1, Gerrit Wolters-Eisfeld1, Faik G. Uzunoglu1, Yogesh K. Vashist1, € r1 €ngo Kai Bachmann1, Ronald Simon2, Jakob R. Izbicki1, Maximilian Bockhorn1, and Cenap Gu

Abstract Pancreatic ductal adenocarcinoma (PDAC) has a devastating prognosis among solid tumors and despite increased knowledge of the molecular mechanisms contributing to progression and metastasis, minimal progress has been done in establishing new targeted therapies for this deadly disease. The expression of the multifunctional growth/ differentiation factor midkine (MK) promotes a variety of cellular functions leading to increased angiogenesis, proliferation, migration, and survival. Moreover, MK is intensively discussed as a potential new-therapy target and as biomarker for cancer progression and chemotherapeutic resistance in multiple cancers. Therefore, the present study investigated the molecular role of MK in pancreatic cancer. It was found that MK is elevated in PDAC and differentially expressed in other histologic subtypes of pancreatic cancer, whereas normal pancreatic cells did not express MK, thus making it an attractive candidate for targeted therapies. As a secreted growth/differentiation factor, MK was investigated as a biomarker in clinical serum specimens using ELISA. In addition, knockdown studies of MK revealed a link to proliferation and migration status in vitro. Finally, upstream signaling pathways were analyzed, with TNF-a and EGF being the main inductors of MK expression in PDAC. Implications: This study presents novel MK functions and new upstream signaling effectors that induce its expression to promote PDAC and therefore defines an attractive new therapeutic target in pancreatic cancer. Mol Cancer Res; 12(5); 670–80. 2014 AACR.

Introduction An unusual aggressiveness and early metastatic locoregional as well as distant spread of pancreatic cancer cells reflects the urgent necessity of new therapeutic options for this deadly disease as its incidence still nearly equals mortality in the Western countries (1). The clinical treatment failure of patients is often attributed to the early metastatic growth, an unmet high drug resistance to standard therapy options, and high rates of local recurrence (2–4). However, insufficient diagnostic tools and therapeutic options for pancreatic ductal adenocarcinoma (PDAC) still substantiate its ranking as fourth leading cause of cancer-related death (5). Therefore, a better understanding of newly identified and cancer-specific key molecules that promote proliferation, migration and survival of cancer cells may lead to the development of more effective therapeutic strategies.

Authors' Affiliations: 1Department of General, Visceral and Thoracic Surgery, Experimental Oncology and 2Institute for Pathology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany € r, Department of General, Visceral €ngo Corresponding Author: Cenap Gu and Thoracic Surgery, University Medical Center Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany. Phone: 49-4074-1051956; Fax: 49-4074-1053-496; E-mail: [email protected] doi: 10.1158/1541-7786.MCR-13-0467 2014 American Association for Cancer Research.

670

Previous studies have revealed that expression of midkine (MK) plays an important role in different aspects of tumor progression in diverse solid tumors (6–8). We have recently shown that high mRNA and protein expression of MK in PDAC protected cancer cells from chemotherapy-induced apoptosis. Moreover, we have identified that gemcitabine dose-dependently induced expression and secretion of MK followed by activation of Notch signaling to promote epithelial–mesenchymal transition (EMT) and intrinsic as well as acquired chemotherapy resistance through upregulation of NF-kB. In addition, our results explained for the first time how MK may trigger cancer cells to become more resistant against cell death and how it promotes cell viability (9). Recently, we and others have shown that the secretion of MK in tumor cells may serve as biomarker for gastrointestinal tumors and glioblastomas (10, 11). Therefore, the development of MK inhibitors has been considered as an attractive idea to prevent tumor growth and antiapoptosis in cancer cells (12, 13). In this study, we addressed the question of whether MK is also expressed in other histologic subtypes of PDAC and that siRNA- and short hairpin RNA (shRNA)–mediated depletion of MK may have dramatic effects on cancer cell proliferation and migration in vitro. The analysis of MK upstream signaling may provide reasonable explanations for the observed high MK expression levels in different cancers and therefore link therapeutic benefits with targeting of the multifunctional growth factor MK.

Mol Cancer Res; 12(5) May 2014

Downloaded from mcr.aacrjournals.org on February 18, 2015. © 2014 American Association for Cancer Research.


Relevance of MK in Pancreatic Cancer

Materials and Methods Study design and patients The study was approved by Ethics Committee of Chamber Physicians in Hamburg, Germany. Written informed consent was obtained from patients using tissue and serum. We analyzed 103 pancreatic cancer tissue spots and 37 serum samples for MK expression, retrospectively. As healthy controls, 148 blood bank donors were included. A tissue microarray (TMA), including 56 ductal, 28 papillar adenocarcinoma, and 19 neuroendocrine resected pancreatic cancer was stained with antibodies against human MK (Abcam). ELISA For quantification of human MK in serum of patients with cancer, we used MK ELISA (Antigenix America). Briefly, microtiter wells precoated with anti-human MK antibodies were incubated with patient's serum. After washing, biotin-labeled anti-human MK antibody was added and plates were incubated for 1 hour with captured MK. After thorough wash, streptavidin–HRP (horseradish peroxidase) conjugate was added to convert substrate H2O2–tetramethylbenzidine. TMA construction and immunohistochemistry Tissue samples were fixed in 4% formalin, paraffin embedded, and used for TMA construction as described elsewhere (14). Hematoxylin-eosin–stained sections were made from selected primary tumor blocks (donor blocks) to define representative tumor regions. Immunohistochemistry (IHC) and scoring were judged by two independent board-certified pathologists (T. Rawnaq and R. Simon). For each tissue sample spot, the fraction of positive-immunostained tumor cells was recorded, and staining intensity was estimated on a four-step scale (0, 1þ, 2þ, and 3þ). A final score was then built from these two parameters according to the following score: negative scores had absence of MK staining; weak scores had staining intensity of 1þ in 70% of tumor cells or staining intensity of 2þ in 30% of tumor cells; moderate scores had staining intensity of 1þ in >70% of tumor cells, staining intensity of 2þ in >30% but in 70% of tumor cells or staining intensity of 3þ in 30% of tumor cells; and strong scores had staining intensity of 2þ in >70% of tumor cells or staining intensity of 3þ in >30% of tumor cells. Cell lines, conditions, and growth factors Tumor tissues for establishment of primary chemoresistant PDAC cell line PaCa 5061 was taken from a patient who underwent total pancreaticoduodenectomy for advanced PDAC in 2010. The procedure of cell line establishment from tumor tissue was previously published (15). Primary cells were cultured in complete TUM medium (15). PANC1 cells were cultured in Dulbecco's Modified Eagle Medium (DMEM; Invitrogen) supplemented with 10% fetal calf serum (FCS; Lonza), 200 IU/mL Pen-Strep (Sigma-Aldrich) at 37 C, 5% CO2. Establishment of HPDE was described

www.aacrjournals.org

elsewhere (16). HPDE cells were cultured in K-SFM (Invitrogen) supplemented with bovine pituitary extract, 5 ng/mL EGF and 5% FCS (Lonza), and 200 IU/mL PenStrep (Sigma-Aldrich). All cell lines were authenticated using the STR GenePrint System (Promega) in 2012. For MK induction studies, cells were grown overnight (12–15 hours) in serum-free media and were stimulated using 10% FCS (Lonza), recombinant TNF-a (10 ng/mL), EGF (10 ng/mL), insulin growth factor I (IGF-I; 10 ng/mL), basic fibroblast growth factor (bFGF; 10 ng/mL; R&D Systems) and (all-trans-) retinoic acid (1 mmol/L; Sigma-Aldrich). For analysis/quantification of secreted MK (S-MK), we used human MK-ELISA kit (Antigenix America). Recombinant human MK (rh-MK; R&D Systems) was used at concentrations of 20 ng/mL. siRNA/shRNA transfections and Western blot analyses Cells were transfected with siRNAs against MK or control-siRNA (Eurofins Scientific) using Lipofectamine (Invitrogen). PANC-1 and PaCa 5061 cells were transfected with indicated siRNAs. PaCa 5061 cells were transfected using shRNA-plasmids for stable MK downregulation. Briefly, cells were transfected either with sh-control or specific shMK plasmids (RHS4533-EG4192; Open Biosystems). After 72 hours, cells were incubated in the presence of puromycin and positive cell clones were subcloned using glass cylinders (Sigma-Aldrich). Viable cell clones were subsequently lysed and analyzed using SDS-PAGE and Western blot analysis for specific MK knockdown using anti-MK antibody. Following transfection, cells were lysed in RIPA buffer (Sigma-Aldrich) containing 1 protease inhibitor cocktail (Roche). For analyses of S-MK, cell culture media were collected and filtered (0.22 mM). Supernatants were resolved by SDS-PAGE and followed by immunoblotting as previously described (9). Western blot analyses and immunostainings were performed using MK antibodies (rab. anti-MK, Abcam; mouse anti-MK, Abnova). Real-time reverse transcription PCR Real-time reverse transcription PCR (real-time RT-PCR) was conducted to quantify gene expression or to verify siRNA-mediated downregulation. Of note, 1 mg of total RNA was reverse-transcribed using the Transcriptor cDNA Kit (Roche). Data were analyzed according to the comparative CT method and were normalized for cyclophilin expression in each sample. Immunocytochemistry and subcellular protein fractionation Cells were grown on coverslips in 24-well plates. Following fixation using paraformaldehyde (4%), cells were blocked with 1% BSA/0.1% Triton-X. Cells were then incubated using anti-MK antibody overnight and washed extensively using PBS (Invitrogen). For visualization, Cy3coupled secondary antibodies (Invitrogen) were used. Image acquisition was done using confocal microscopy (Leica). For analyses of subcellular MK distribution, we stepwise separated cytoplasmic, membrane, and nuclear soluble proteins



671


Rawnaq et al.

by using the Subcellular Fractionation Kit (Pierce). As marker proteins for different compartments, we used anticalpain, anti-calnexin, and anti-lamin antibodies (Santa Cruz Biotechnology). Scratch and Transwell migration assay We performed scratch-/wound healing assay by plating Paca 5061 and PANC-1 wild-type, shControl- and sh-MK– transfected subclones to create a confluent monolayer. Monolayer was scraped in a straight line to create a "scratch" in similar size with tips. Debris was washed out using PBS. For image acquisition, we used phase-contrast microscopy (Leica) at different time points (0, 16 hours). For Transwell assay, corresponding knockdown clones of PaCa 5061 cells and PANC-1 cells were transferred to migration assay for indicated time points. Cell migration was analyzed using 24well inserts (8 mmol/L) according to the manufacturer's protocol (BD Biosciences). Briefly, cancer cells remaining on top-side of the membrane were carefully removed, whereas cancer cells that had migrated to lower-side of the chamber were fixed and stained with crystal violet. At least 5 randomly selected fields per insert were photographically documented and average counted. Cell proliferation Cell proliferation was determined by CellTiter 96 AQueous One Solution Assay according to the manufacturer's protocol (Promega). Each experiment was performed in quadruplicate. The absorbance was measured at 490 nm. Values for control cells were considered as 100% viability. Every measurement was performed at least in three independent experiments. Data analyses Expression analyses and achieved data were analyzed using GCOS 1.4 and scaled to default target signal value of 150.

Absolute and comparative analyses were performed using MAS 5.0 algorithm. Annotations were further analyzed with interactive query analysis. Experiments presented in figures are representative of three different repetitions. The data are presented as the mean values SE. Statistical analysis SPSS for Windows (v21.0; SPSS Inc.) was used for statistical analysis. Statistical significance was evaluated by the Mann–Whitney U test and the Kruskal–Wallis test. Significance statements refer to P values of less than 0.05. Results MK expression in normal pancreatic and cancer cells First, the MK gene and protein expression were evaluated in PDAC cell lines PANC-1 and PaCa 5061, whereas HPDE cells served as control. HPDE cells were previously established from normal pancreatic tissue (16). PaCa 5061 cells were previously established from patient-derived PDAC tissue, operated in our clinic (15). Because MK was already shown to be significantly overexpressed in cancer cells compared with corresponding normal cells (9), we investigated and quantified MK mRNA in the investigated lines using real-time RT-PCR (Fig. 1A). We observed a statistically significant mRNA overexpression (P < 0.05), whereas no expression was detected in HPDE cells. In a previously published report, we already showed that MK expression in patients with PDAC is significantly upregulated in more than 50% of the investigated samples (9). To verify that the MK mRNA expression resulted in robust protein expression, we investigated the MK protein expression in cancer cell lines and HPDE cells using specific antibodies (Fig. 1B). Because MK is a secreted growth factor involved in paracrine/autocrine regulation of growth and differentiation, we investigated whether the MK secretion is increased in PDAC compared with HPDE cells. Indeed, we

Figure 1. MK expression in normal pancreatic and cancer cells. A, differential MK mRNA expression in PaCa 5061 and PANC-1 cells compared with normal pancreatic cells (HPDE) obtained by real-time RT-PCR. MK is significantly overrepresented in PDAC cell lines, more than 5-fold compared with HPDE cells. Results are expressed as fold over control; , P < 0.05. B, Western blot analyses showing steady-state MK expression (top) as well as S-MK levels (bottom) in PaCa 5061 and PANC-1 cells, compared with HPDE cells.

672






found robust released MK levels in PDAC cells, whereas no increased MK expression was detectable in HPDE cells at all (Fig. 1B).

patient's serum using ELISA. This analysis showed significantly increased serum MK levels in patients with PDAC compared with healthy donors (Fig. 2B; Table 2).

Elevated MK expression in pancreatic cancer tissue and serum samples To gain relevant insight into expression of MK in PDAC patient samples, we analyzed the MK protein expression using TMA. Because the vast majority of published information about MK expression in cancer cells resulted from evaluation of ductal adenocarcinomas, we hypothesized that MK may also be differentially expressed in other histologic subtypes. We, therefore, conducted TMA stainings from 103 patients with pancreatic cancer and 10 normal pancreatic tissue samples that were spotted on the TMA. From these 103 samples, 56 were ductal adenocarcinoma, 28 intraductal papillary-mucinous tissue samples, and 19 neuroendocrine tumors and all of them were stained with MK antibodies (Fig. 2A). Consistent with previous observations where MK was shown to be significantly overexpressed in ductal adenocarcinomas compared with normal control cells (9), we found moderate to strong MK expression in 36% (20 of 56) and 38% (21 of 56) of ductal adenocarcinomas, respectively. More interestingly, we found also strong MK expression in 64% (18 of 28) of papillary adenocarcinoma tissue samples as well as in 47% (9 of 19) of neuroendocrine patient samples (Table 1). Because MK is a secreted growth factor and its detection may serve as biomarker, we also quantified MK expression in

MK is primarily expressed in the cytoplasm of PDAC cells The MK expression was previously shown to promote survival through influencing expression of apoptosis-related proteins such as Bcl-2 (17). In addition, MK was shown to localize to the nucleus and nucleolus of HepG2 cells (18), whereas other studies showed different MK localizations. The nucleolus is a known hotspot for rRNA transcription, and the authors conducted siRNA-mediated knockdown of MK, which resulted in lower rRNA transcription rates. The authors claimed that nucleolar MK expression influenced the transcription of rRNA in HepG2 cells and therefore affecting cell proliferation and apoptosis. The intracellular expression pattern of MK in cancer cells is still enigmatic and confusing. We, therefore, analyzed intracellular localization of endogenous MK in PDAC cells. First, we performed immunocytochemical stainings with PANC-1 and PaCa 5061 cells using specific MK antibodies. The subcellular localization of MK was examined using laser confocal microscope (Fig. 3A). MK was clearly expressed and localized exclusively in the cytoplasm. Second, we separated subcellular fractions from PANC-1 and PaCa 5061 cells and analyzed MK expression in comparison with expression of cell compartment–relevant protein markers. Strikingly, we found that MK is predominantly expressed in the

A

A

B

B

S-MK concentration (pg/mL)

Figure 2. Elevated MK expression in pancreatic cancer tissue and serum samples. A, single representative tissue spots of MK expression in PDAC (A), pancreatic neuroendocrine carcinoma (B), and pancreatic papillar adenocarcinoma (C; magnification, 100). B, boxplot analysis of S-MK protein concentration of healthy controls (n ¼ 148) and patients with pancreatic cancer (n ¼ 37) determined by ELISA. Statistical significance was determined by Mann–Whitney U test (P < 0.001).

C

2,000 1,500 1,000 500 0 Controls


Pancreatic carcinoma



673


Rawnaq et al.

Table 1. MK expression in pancreatic cancer TMA MK immunostaining (%) Tumor subtype

N

Negative

Weak

Moderate

Strong

Pancreas, adenocarcinoma ductal Pancreas, adenocarcinoma papillary Pancreas, neuroendocrine

56 28 19

5 (9) 1 (4) 4 (21)

10 (18) 4 (14) 2 (11)

20 (36) 5 (18) 4 (21)

21 (38) 18 (64) 9 (47)

cytoplasm of pancreatic cancer cells (Fig. 3B), which is in contrast with previously published results (18). siRNA-mediated knockdown of MK is associated with decreased migration of pancreatic cancer cells in vitro Early metastasis formation and invasion is a hallmark of pancreatic cancer and reflects the urgent need of new therapeutic strategies for this deadly disease. Because MK was earlier discussed to be involved in migration of embryonic neuronal cells (19), we hypothesized that MK expression is linked to the highly migratory potential of PDAC cells. To pinpoint whether overexpressed MK promotes migration of PDAC cells in vitro, we first performed transient siRNA and stable shRNA-mediated downregulation of MK in PaCa 5061 (Fig. 4A) and PANC-1 cells (Fig. 4E). Effective MK downregulation was proven by Western blotting whole-cell lysates and cell culture medium using specific antibodies (Fig. 4B and F) and real-time RT-PCR (Fig. 4C and G). In a second approach, knockdown and corresponding control cells were transferred to a scratch-induced migration assay. After PANC-1 and PaCa 5061 cells had grown for 16 hours under normal conditions, we compared the respective cell migration. Strikingly, downregulation of MK is linked to markedly impaired migratory potential of both cell lines and the injured area is nearly unchanged after 16 hours (Fig. 4D and H). On the other hand, the respective control cells showed no impairment in migration and filled the injured area. In addition, we performed Transwell assays and quantified the migrative potential of PDAC cell lines lacking

MK expression, compared with corresponding control siRNA transfections (Fig. 4I). Our results clearly demonstrate that MK is indeed linked to the high migratory potential of PDAC cells in vitro. MK depletion is linked with decreased proliferation of pancreatic cancer cells in vitro MK is a heparin-binding growth factor and was recently shown to promote growth of various cancer cells in vitro and in vivo. The MK overexpression and/or extracellular treatment of tumor cells with recombinant MK was shown to promote and accelerate the growth of osteosarcoma cells (20), neuroblastoma (21) and gastric cancer cells (22). To examine whether MK promotes proliferation of PDAC cells in vitro, we first downregulated MK expression through

Table 2. Elevated S-MK concentration in pancreatic cancer compared with healthy controls

Total Mean age, y SD Sex, No. (%) Male Female S-MK, y SD (pg/mL) S-MK, median (pg/mL) [25th and 75th percentile]

674

Controls

Pancreas carcinoma

148 58 11

37 64 10

98 (66) 50 (34) 128 115 99 [33 and 198]

19 (51) 18 (49) 1,200 1,712 721 [409 and 1,197]


Figure 3. MK is primarily expressed in the cytoplasm of PDAC cells. A, intracellular localization of endogenous MK (red) in PaCa 5061 and PANC-1 cells. Cells were grown on cover slips followed by fixation, blocking, and antibody treatment. Appropriate secondary Cy3-labeled antibodies were used. DAPI (40 ,6-diamidino-2-phenylindole) was used for costaining nuclei (blue) and immunoreactions were documented on a confocal microscope. B, subcellular fractions from PANC-1 and PaCa 5061 cells were analyzed for MK expression in comparison with expression of cell compartment–relevant protein markers. MK is predominantly expressed in cytoplasm of pancreatic cancer cells.





Figure 4. siRNA-mediated knockdown of MK is associated with decreased migration of pancreatic cancer cells in vitro. A, shRNA-mediated knockdown of MK in PaCa 5061 cells efficiently downregulated MK as shown by Western blotting total cell lysates of two independent cell clones, compared with control-transfected cells. Actin served as loading control. B, shRNA-mediated knockdown of MK in PaCa 5061 cells resulted in undetectable SMK levels as shown by Western blotting cell culture medium from two knockdown cell clones, compared with control-transfected cells. C, effective MK downregulation on mRNA level was proven by real-time RT-PCR. D, PaCa 5061 wild-type, controltransfected, and two independent MK-knockdown cell clones were transferred to scratch-induced migration assay. After placing the scratch, cells were grown for additional 0 and 16 hours, respectively, and cell migration was captured using high-power field microscope. E, siRNAmediated knockdown of MK in PANC-1 cells using two independent and specific siRNAs efficiently downregulated MK as shown by Western blotting total cell lysates. F, siRNA-mediated knockdown of MK in PANC-1 cells resulted in decreased S-MK levels as shown by Western blotting cell culture medium. G, effective MK downregulation on mRNA level was proven by real-time RT-PCR. H, PANC-1 wild-type, controltransfected, and two MKknockdown cell clones were transferred to scratch-induced migration assay. After placing the scratch, cells were grown for additional 0 and 16 hours, respectively, and cell migration was captured using high-power field microscope. I, Transwell migration assay was performed with PaCa 5061 and PANC-1 MK knockdown cells. The graphs indicate the average number of cells per field at the indicated time points. , P < 0.01; , P < 0.05.

transient transfections of specific siRNA or control siRNA into PaCa 5061 (Fig. 5A) and PANC-1 cells (Fig. 5C). We then analyzed consequences of depleted MK and performed proliferation assays. Corresponding untransfected wild-type


cells were used as independent control. Interestingly, we found that MK depletion is strongly linked with impaired growth of PaCa 5061 (Fig. 5B) and PANC-1 cells (Fig. 5D). In contrast, proliferation of control transfected cells was not



675


Rawnaq et al.

Figure 5. MK-depletion is linked with decreased proliferation of pancreatic cancer cells in vitro. A, PaCa 5061 cells were transiently transfected with two independent siRNAs against MK or with sicontrol for 48 hours. Protein lysates were extracted. Actin served as loading control. B, MTT assay of PaCa 5061 wild-type (WT), controltransfected, and MK knockdown cells transfected with two independent siRNAs. In rescue experiments, MK-knockdown cells were also treated with rh-MK (20 ng/mL) for indicated time points. C, PANC-1 cells were transiently transfected with two independent siRNAs against MK or with si-control for 48 hours. Protein lysates were extracted. Actin served as loading control. D, MTT assay of PANC-1 wildtype, control-transfected, and MKknockdown cells. In rescue experiments, MK-knockdown cells were also treated with rh-MK (20 ng/mL) for indicated time points.

impaired. In proof-of-concept studies, we performed rescue experiments via exogenous treatment of MK-depleted PaCa 5061 and PANC-1 cells with recombinant human MK (rh-MK; 20 ng/mL; Fig. 5B and D). Here we observed substantially increased proliferation in rh-MK–treated MK-depleted cells, compared with untreated cells and proliferation reached almost similar levels as control-transfected and wild-type cells. Because proliferation of PANC-1 and PaCa 5061 cells was strongly impaired in MK-depleted cells compared with control transfected and wild-type cells, these results suggest that high expression and secretion of MK positively regulate proliferation of PDAC cells in vitro. Interestingly, exogenous treatment of MK-depleted PANC-1 and PaCa 5061 cells with rh-MK rescued the observed impaired proliferation of MK-depleted pancreatic cancer cells. MK expression is inducible by EGF and TNF-a in pancreatic cancer cells in vitro Although much effort has been done on catching cellular MK functions during embryonic development and in neoplastic cells, very little information is available about upstream signaling that controls expression of this interesting multifunctional growth factor. A single report has investigated the role of various growth factors and cytokines that may induce MK transcription in prostate cancer cells (23). The authors claimed that EGF, androgen, IGF-I, hepatocyte growth factor (HGF), and cytokines TNF-a and interleukin-1beta (IL1-b) are inductors of MK expres-

676


sion. To identify factors that may induce MK expression in PDAC, we treated PaCa 5061 and PANC-1 cells with serum, IGF-I, bFGF, TNF-a, and EGF or left them untreated. Because MK was first identified as a retinoic acid–inducible gene during embryonal carcinoma cell differentiation (24, 25), we also treated cells with retinoic acid. First, we serum starved PaCa 5061 and PANC-1 cells for 12 to 15 hours and treated cells with various growth factors/ cytokines or serum (10%) for 24 hours or left them untreated and analyzed MK expression by Western blotting using specific antibodies (Fig. 6B and D). In a second attempt, we quantified MK expression in PaCa 5061 and PANC-1 cells following growth factor/cytokine treatment using a specific MK ELISA (Fig. 6A and C). Interestingly, we were able to induce strong MK secretion with extracellular treatment of TNF-a (10 ng/mL) and EGF (10 ng/mL) in both cell lines. Although the MK expression in PaCa 5061 cells was induced to a lesser content by bFGF (10 ng/mL) and IGF-I (10 ng/mL), we detected very faint MK expression after retinoic acid treatment (1 mmol/L) in both cell lines (Fig. 6A). In contrast, treatment of PANC-1 cells with bFGF showed undetectable MK expression levels. Moreover, serum-treated PaCa 5061 and PANC-1 cells displayed comparable MK expression levels as those resulted from single TNF-a and EGF treatments, whereas control cells showed almost no MK expression at all (Fig. 6C). These results suggested that MK is indeed inducible by TNF-a, EGF, and serum, whereas retinoic acid, IGF-I, and serum starvation (12–15 hours) failed to induce MK expression in





Figure 6. MK expression is inducible by EGF and TNF-a in pancreatic cancer cells in vitro. A, serum-starved PaCa 5061 cells were treated with various growth factors/cytokines or treated with serum (10%) as indicated for 96 hours or left them untreated (S/f) and quantified MK expression/secretion levels by ELISA. B, similar results were observed in Western blotting experiments, reflecting consistency. Actin served as loading control. C, same experiment as in A was performed using PANC-1 cells. D, comparable results were obtained using Western blotting. Actin served as loading control. MK mRNA expression levels in PaCa 5061 (E) and PANC-1 cells (F) treated with indicated growth factors/cytokines were quantified using real-time RT-PCR.

PDAC cells. Similar results were also observed in Western blotting experiments, reflecting consistency (Fig. 6B and D). Moreover, it is known that the MK signaling is posttranslationally extinguished by ubiquitination and proteasomal degradation. Therefore, we examined whether exogenously applied growth factor–induced MK upregulation occurs at the mRNA level. Indeed, we found that MK mRNA


expression is robustly induced by TNF-a, EGF, and serum (10%; Fig. 6E and F). Discussion There are still no satisfactory treatment options for pancreatic cancer. As a matter of fact, the clinical reality regarding



677


Rawnaq et al.

successful therapeutic intervention of PDAC still reflects that only very few patients benefit from current therapeutic options and the majority of patients still cannot be cured in 90% to 95% (26). PDAC is often identified in an advanced stage in which it is no longer possible to operate. This is due to rapid growth and invasive phenotypes leading to early onset metastatic spread, accompanied by frequent occurrence of chemotherapeutic resistance. Even surgical complete tumor– resected patients develop relapse disease, which leads to death despite receiving chemotherapy. Early detection and novel therapeutic targets are urgently needed to improve the outcome. Therefore, an intensive investigation of molecular mechanisms contributing to tumor proliferation and migration of PDAC should be conducted as a basis for successful establishment of new targeted therapies. Elevated MK expression levels have been detected in various tumors (27). MK is described as a potential prognostic marker in several malignancies (10, 28–32). The present study confirmed the frequent overexpression of MK in patient-derived PDAC tissue samples. More interestingly, analyses of histologic subclasses revealed strong MK expression in the majority of patient-derived papillary adenocarcinoma and neuroendocrine tumors. Furthermore, we demonstrated that serum MK concentrations were significantly elevated in patients with PDAC compared with healthy controls. These results encourage us to examine serum MK levels as a novel prognostic tumor marker for PDAC, which can be easily detected in peripheral blood using commercially available ELISA. Moreover, downregulation of MK by shRNA and siRNA strategies resulted in substantially reduced proliferation rates compared with control cells. Interestingly, exogenous treatment of PDAC cells with rh-MK rescued proliferation deficiency, suggesting that MK may trigger extracellular signaling to accelerate proliferation of PDAC cells. In addition, MK is known to be a secreted growth factor and high expression levels may contribute to increased proliferation by autocrine/paracrine signaling. A pronounced stromal response referred to as desmoplastic reaction is a hallmark of PDAC that is characterized by a complex interplay between the host epithelial cells, invading tumor cells, stromal fibroblasts, pancreatic stellate cells, immune cells, proliferating endothelial cells, and an altered extracellular matrix. Elevated expression and secretion of MK by the tumor might affect the surrounding stromal cells that express MK receptors in a paracrine mechanism and thereby contribute to the formation of the latter. Because it was shown that the stromal compartment in PDAC negatively affects radiation therapy (33, 34) and creates a milieu for chemotherapy resistance (35, 36), a fundamental issue in PDAC treatment, the role of MK in tumor stroma interplay and its effects on chemotherapy resistance requires further investigation. Because several studies have already shown that MK may interact with various extracellular receptors (27), it is challenging to identify the specific receptors contributing to this effect. The list of MK interacting receptors on cell surface is still growing, since we were recently able to identify the Notch-2 receptor as a new MK interactor in PDAC cells

678


whose activation by extracellular MK ended in a highly chemoresistant phenotype accompanied by EMT (9). PDAC is known to have a devastating prognosis, which is primarily due to an early locoregional and distant metastatic spread. This process is supported by the ability of primary cancer cells that acquire invasive properties and gain access to blood and/or lymphatic vascular systems to establish a new deadly niche. In this study, we could clearly show that MK is involved in migration processes of PDAC cells. The migrative ability of PDAC cells in which MK is specifically and stably knocked down, revealed strong migration deficiency. The specificity of our observation was further supported by transient siRNA transfections targeting MK in other PDAC cell lines. In line with this, MK was recently shown to interact with various protein members of the TGF-b pathway in vitro, a pathway that is well accepted to be a central mediator of EMT processes and consequently navigate increased migration of cancer cells in vitro and in vivo (37). On the other hand, subcellular distribution of MK in cancer cells is still enigmatic. Therefore, we analyzed MK localization in PDAC cells using specific antibodies and detected MK in cytoplasm of all investigated PDAC cell lines, whereas no nuclear staining was observed. Moreover, we separated subcellular fractions of cultured cells and used specific compartment markers that corroborate our findings. This is in contrast with previously shown observations in which MK was found primarily localized to nucleus and nucleolus in HepG2 cells; the latter is a known hotspot for rRNA synthesis. The authors claimed that MK might interfere with rRNA transcription and ribosomal assembly and ultimately with increased protein synthesis in vitro (38). Therefore, MK intracellular distribution may be cell context–dependent and need further attention in future studies. The primary goal of this study was to gain relevant insight into mechanisms governing MK expression in PDAC and to further identify relevant new target structures that might interfere with MK expression. Recently, it was shown that growth factors EGF and TNF-a induced expression and secretion of MK in prostate cancer cells in vitro. Of all the cytokines and growth factors tested, TNF-a was the strongest inducer of MK expression in LNCaP cells. The authors claimed that MK expression is inducible by the NF-kB signaling pathway and elevated MK may partially inhibit TNF-a–induced apoptosis in prostate cancer cells (23). In line with this, our investigations revealed that TNF-a may also induce MK expression and secretion in all tested PDAC cell lines. Moreover, treatment with EGF resulted again in robust MK expression and secretion levels in an equal manner, suggesting that MK expression is in fact primarily inducible by TNF-a and EGF in PDAC. Because retinoic acid seems to be the main inductor of MK expression during mouse embryogenesis (39), our analyses showed almost no influence on MK expression in PDAC cells in vitro. There are accumulating evidences in the literature reflecting that cellular MK act as a multifunctional growth factor whose expression is highly linked to proliferation, migration, survival, angiogenesis, and EMT in various neoplasms. Interestingly, several studies have already discussed the





possibility of successfully neutralizing MK by using antibodies, antisense RNA, or aptamers and thereby blocking tumor growth or reversing chemotherapeutic resistance (9). Presumably, in case of PDAC, the appliance of TNF-a antagonists such as infliximab might be conceivable to counteract elevated MK expression in general or even during chemotherapeutical treatment modalities. Another strategy might be the utilization of tyrosine kinase inhibitors such as sunitinib and axitinib that have been designed to counteract angiogenesis. This might also reflect a new strategy to interrupt MK-induced angiogenesis in salivary gland tumors (40), in which inhibition of angiogenesis may limit metastasis formation, which is, especially for PDAC, a very early and challenging clinical event and the primary cause of cancer-related death worldwide. In conclusion, our study showed to our best knowledge for the first time that MK is, beyond PDAC, frequently upregulated in other histologic and neoplastic subtypes of the pancreas like neuroendocrine tumors or papillary adenocarcinoma. The MK expression is robustly inducible by TNF-a and EGF and high expression levels are strongly linked to the proliferative and migrative potential of PDAC cells in vitro. Therefore, it is of great interest to investigate how MK may influence proliferation, metastasis formation, and even chemotherapeutic resistance of PDAC cells in a mouse model and how currently available treatment modalities might circumvent MK expression to gain prospectively a benefit for patients suffering from this deadly disease.

Disclosure of Potential Conflicts of Interest No potential conflicts of interest were disclosed.

Authors' Contributions Conception and design: T. Rawnaq, Y.K. Vashist, K. Bachmann, J.R. Izbicki, M. Bockhorn, C. G€ ung€or Development of methodology: T. Rawnaq, L. Dietrich, C. G€ ung€or Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): T. Rawnaq, L. Dietrich, F.G. Uzunoglu, Y.K. Vashist, R. Simon, J.R. Izbicki, C. G€ ung€or Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): T. Rawnaq, G. Wolters-Eisfeld, Y.K. Vashist, K. Bachmann, C. G€ ung€or Writing, review, and/or revision of the manuscript: T. Rawnaq, G. Wolters-Eisfeld, F.G. Uzunoglu, Y.K. Vashist, K. Bachmann, R. Simon, J.R. Izbicki, M. Bockhorn, C. G€ ung€or Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): T. Rawnaq, F.G. Uzunoglu, C. G€ ung€or Study supervision: T. Rawnaq, J.R. Izbicki, M. Bockhorn, C. G€ ung€or

Acknowledgments The authors thank U. Eicke-Kohlmorgen, A. Heinecke, P. Merkert, and P. Schr€oder for excellent technical assistance. The authors also thank Prof. Sebens (University Kiel, Germany) for providing HPDE cells.

Grant Support This work was in part supported by the Roggenbuck Foundation (053/103) to C.G€ ung€or. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Received September 3, 2013; revised February 10, 2014; accepted February 10, 2014; published OnlineFirst February 24, 2014.

References 1.

Siegel R, DeSantis C, Virgo K, Stein K, Mariotto A, Smith T, et al. Cancer treatment and survivorship statistics 2012. CA Cancer J Clin 2012;62:220–41. 2. Burris HA III, Moore MJ, Andersen J, Green MR, Rothenberg ML, Modiano MR, et al. Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: a randomized trial. J Clin Oncol 1997;15:2403–13. € r C, Hofmann BT, Wolters-Eisfeld G, Bockhorn M. Pancreatic €ngo 3. Gu cancer. Br J Pharmacol 2014;171:849–58. 4. Strobel O, Hartwig W, Hackert T, Hinz U, Berens V, Grenacher L, et al. Re-resection for isolated local recurrence of pancreatic cancer is feasible, safe, and associated with encouraging survival. Ann Surg Oncol 2013;20:964–72. 5. Kloppe G, Uttges J. The pathology of ductal-type pancreatic carcinomas and pancreatic intraepithelial neoplasia: insights for clinicians. Curr Gastroenter Rep 2004;6:111–8. 6. Garver RI, Radford DM, Donis-Keller H, Wick MR, Milner PG. Midkine and pleiotrophin expression in normal and malignant breast tissue. Cancer 1994;74:1584–90. 7. Nakagawara A, Milbrandt J, Muramatsu T, Deuel TF, Zhao H, Cnaan A, et al. Differential expression of pleiotrophin and midkine in advanced neuroblastomas. Cancer Res 1995;15:1792–97. 8. Aridome K, Tsutsui J, Takao S, Kadomatsu K, Ozawa M, Aikou T, et al. Increased midkine gene expression in human gastrointestinal cancers. Jpn J Cancer Res 1995;86:655–61. € r C, Zander H, Effenberger KE, Vashist YK, Kalinina T, Izbicki €ngo 9. Gu JR, et al. Notch signaling activated by replication stress-induced expression of midkine drives epithelial-mesenchymal transition and chemoresistance in pancreatic cancer. Cancer Res 2011;71: 5009–19. 10. Rawnaq T, Kunkel M, Bachmann K, Simon R, Zander H, Brandl S, et al. Serum midkine correlates with tumor progression and imatinib


11.

12.

13.

14.

15.

16.

17.

18.

19.

response in gastrointestinal stromal tumors. Ann Surg Oncol 2011;18:559–65. Erguven M, Bilir A, Yazihan N, Ermis E, Sabanci A, Aktas E, et al. Decreased therapeutic effects of noscapine combined with imatinib mesylate on human glioblastoma in vitro and the effect of midkine. Cancer Cell Int 2011;11:18. Muramatsu T. Midkine: a promising molecule for drug development to treat diseases of the central nervous system. Curr Pharm Des 2011;17:410–23. Kishida S, Mu P, Miyakawa S, Fujiwara M, Abe T, Sakamoto K, et al. Midkine promotes neuroblastoma through Notch2 signaling. Cancer Res 2013;73:1318–27. Schraml P, Kononen J, Bubendorf L, Moch H, Bissig H, Nocito A, et al. Tissue microarrays for gene amplification surveys in many different tumor types. Clin Cancer Res 1995;5:1966–75. € r C, Thieltges S, Mo € ller-Krull M, Penas EM, Wicklein €ngo Kalinina T, Gu D, et al. Establishment and characterization of a new human pancreatic adenocarcinoma cell line with high metastatic potential to the lung. BMC Cancer 2010;10:295. Ouyang H, Mou Lj, Luk C, Liu N, Karaskova J, Squire J, et al. Immortal human pancreatic duct epithelial cell lines with near normal genotype and phenotype. Am J Pathol 2000;157:1623–31. Qi M, Ikematsu S, Ichihara-Tanaka K, Sakuma S, Muramatsu T, Kadomatsu K. Midkine rescues Wilms' tumor cells from cisplatininduced apoptosis: regulation of Bcl-2 expression by Midkine. J Biochem 2000;127:269–77. Dai L, Xu D, Yao X, Lu Y, Xu Z. Conformational determinants of the intracellular localization of midkine. Biochem Biophys Res Commun 2005;330:310–7. Maeda N, Ichihara-Tanaka K, Kimura T, Kadomatsu K, Muramatsu T, Noda M. A receptor-like protein-tyrosine phosphatase PTPzeta/ RPTPbeta binds a heparin-binding growth factor midkine. Involvement



679


Rawnaq et al.

20.

21.

22.

23.

24.

25.

26.

27.

28. 29.

680

of arginine 78 of midkine in the high affinity binding to PTPzeta. J Biol Chem 1999;274:12474–9. Sueyoshi T, Jono H, Shinriki S, Ota K, Ota T, Tasaki M. Therapeutic approaches targeting midkine suppress tumor growth and lung metastasis in osteosarcoma. Cancer Lett 2012;316:23–30. Reiff T, Huber L, Kramer M, Delattre O, Janoueix-Lerosey I, Rohrer H. Midkine and Alk signaling in sympathetic neuron proliferation and neuroblastoma predisposition. Development 2011;138: 4699–708. Xu Y, Qu X, Zhang X, Luo Y, Zhang Y, Luo Y. Midkine positively regulates the proliferation of human gastric cancer cells. Cancer Lett 2009;279:137–44. You Z, Dong Y, Kong X, Beckett LA, Gandour-Edwards R, Melamed J. Midkine is a NF-kB-inducible gene that supports prostate cancer cell survival. BMC Med Genomics 2008;1:6. Kadomatsu K, Tomomura M, Muramatsu T. cDNA cloning and sequencing of a new gene intensely expressed in early differentiation stages of embryonal carcinoma cells and in mid-gestation period of mouse embryogenesis. Biochem Biophys Res Commun 1988;151: 1312–8. Tomomura M, Kadomatsu K, Matsubara S, Muramatsu T. A retinoic acid-responsive gene, MK, found in the teratocarcinoma system. Heterogeneity of the transcript and the nature of the translation product. J Biol Chem 1990;265:10765–70. Bilimoria KY, Bentrem DJ, Ko CY, Ritchey J, Stewart AK, Winchester DP, et al. Validation of the 6th edition AJCC Pancreatic Cancer Staging System: report from the National Cancer Database. Cancer 2007; 110:738. Kadomatsu K, Kishida S, Tsubota S. The heparin-binding growth factor midkine: the biological activities and candidate receptors. J Biochem 2013;153:511–21. Kadomatsu K, Muramatsu T. Midkine and pleiotrophin in neural development and cancer. Cancer Lett 2004;204:127–43. Shimada H, Nabeya Y, Okazumi S, Matsubara H, Kadomatsu K, Muramatsu T, et al. Increased serum midkine concentration as a possible tumor marker in patients with superficial esophageal cancer. Oncol Rep 2003;10:411–4.


30. Tanabe K, Matsumoto M, Ikematsu S, Nagase S, Hatakeyama A, Takano T, et al. Midkine and its clinical significance in endometrial carcinoma. Cancer Sci 2008;99:1125–30. 31. Obata Y, Kikuchi S, Lin Y, Yagyu K, Muramatsu T, Kumai H. Serum midkine concentrations and gastric cancer. Cancer Sci 2005;96:54–6. 32. Kaifi JT, Fiegel HC, Rafnsdottir SL, Aridome K, Schurr PG, Reichelt U, et al. Midkine as a prognostic marker for gastrointestinal stromal tumors. J Cancer Res Clin Oncol 2007;133:431–5. 33. Ohuchida K, Mizumoto K, Murakami M, Qian LW, Sato N, Nagai E, et al. Radiation to stromal fibroblasts increases invasiveness of pancreatic cancer cells through tumor-stromal interactions. Cancer Res 2004;64: 3215–22. 34. Neoptolemos JP, Stocken DD, Friess H, Bassi C, Dunn JA, Hickey H, et al. A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. European Study Group for Pancreatic Cancer. N Engl J Med 2004;350:1200–10. € ster S, Wegehenkel K, Arlt A, Witt M, Sipos B, Kruse ML, et al. €erko 35. Mu Tumor stroma interactions induce chemoresistance in pancreatic ductal carcinoma cells involving increased secretion and paracrine effects of nitric oxide and interleukin-1beta. Cancer Res 2004;64: 1331–7. 36. Erkan M, Hausmann S, Michalski CW, Fingerle AA, Dobritz M, Kleeff J, et al. The role of stroma in pancreatic cancer: diagnostic and therapeutic implications. Nat Rev Gastroenterol Hepatol 2012;9:454–67. 37. Katsuno Y, Lamouille S, Derynck R. TGF-b signaling and epithelialmesenchymal transition in cancer progression. Curr Opin Oncol 2013;25:76–84. 38. Dai LC, Shao JZ, Min LS, Xiao YT, Xiang LX, Ma ZH. Midkine accumulated in nucleolus of HepG2 cells involved in rRNA transcription. World J Gastroenterol 2008;14:6249–53. 39. Chen Q, Yuan Y, Lin S, Chang Y, Zhuo X, Wei W, et al. Transiently truncated and differentially regulated expression of midkine during mouse embryogenesis. Biochem Biophys Res Commun 2005;330: 1230–6. 40. Ota T, Ota K, Jono H, Fujimori H, Ueda M, Shinriki S , et al. Midkine expression in malignant salivary gland tumors and its role in tumor angiogenesis. Oral Oncol 2010;46:657–61.




The Multifunctional Growth Factor Midkine Promotes Proliferation and Migration in Pancreatic Cancer Tamina Rawnaq, Luisa Dietrich, Gerrit Wolters-Eisfeld, et al. Mol Cancer Res 2014;12:670-680. Published OnlineFirst February 24, 2014.

Updated version

Cited Articles

E-mail alerts Reprints and Subscriptions Permissions

Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-13-0467

This article cites by 40 articles, 10 of which you can access for free at: http://mcr.aacrjournals.org/content/12/5/670.full.html#ref-list-1

Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at [email protected]. To request permission to re-use all or part of this article, contact the AACR Publications Department at [email protected].


Midkine promotes perineural invasion in human pancreatic cancer.

Endocrine gland‑derived vascular endothelial growth factor modulates proliferation, apoptosis and migration in pancreatic cancer cells.

Beta-Catenin Axis.

Nifedipine promotes the proliferation and migration of breast cancer cells.

miRNA-346 promotes proliferation, migration and invasion in liver cancer.

Eukaryotic Translation Initiation Factor 3a (eIF3a) Promotes Cell Proliferation and Motility in Pancreatic Cancer.

ZKSCAN3 promotes bladder cancer cell proliferation, migration, and invasion.

PREX2 promotes the proliferation, invasion and migration of pancreatic cancer cells by modulating the PI3K signaling pathway.

EphB4 regulates the growth and migration of pancreatic cancer cells.

Decreased TUSC3 Promotes Pancreatic Cancer Proliferation, Invasion and Metastasis.

NOV promoted the growth and migration of pancreatic cancer cells.

The advances of Midkine with peripheral invasion in pancreatic cancer.

S transition, migration and invasion of pancreatic cancer cells by targeting CHIP.

Twist promotes invasion and cisplatin resistance in pancreatic cancer cells through growth differentiation factor 15.

CLPTM1L promotes growth and enhances aneuploidy in pancreatic cancer cells.

CP-673451, a platelet-derived growth-factor receptor inhibitor, suppresses lung cancer cell proliferation and migration.

CCL18 promotes epithelial-mesenchymal transition, invasion and migration of pancreatic cancer cells in pancreatic ductal adenocarcinoma.

Placenta growth factor promotes migration through regulating epithelial-mesenchymal transition-related protein expression in cervical cancer.

Epidermal growth factor receptor in pancreatic cancer.

Inhibition of fibroblast growth factor receptor 2 attenuates proliferation and invasion of pancreatic cancer.

The Mu opioid receptor promotes opioid and growth factor-induced proliferation, migration and Epithelial Mesenchymal Transition (EMT) in human lung cancer.

Regulation of proliferation by fibroblast growth factor in a pancreatic cancer cell line.

Downregulation of CD9 promotes pancreatic cancer growth and metastasis through upregulation of epidermal growth factor on the cell surface.

Transcriptional factor specificity protein 1 (SP1) promotes the proliferation of glioma cells by up-regulating midkine (MDK).