O

Uhp

TIa

b

c

Id

e

f

g

h

i

a

ARRA

KUPBEM

I

nrP[o

h0

Pathology – Research and Practice 210 (2014) 733–738

Contents lists available at ScienceDirect

Pathology – Research and Practice

jou rn al hom epage: www.elsev ier .com/ locate /prp

riginal Article

p-regulation of urinary-type plasminogen activator correlates withigh-risk papillary thyroid carcinoma with BRAFV600E mutation and itsossible molecular mechanism

omoko Wakasaa,1, Yaqiong Lib,1, Yanhua Baic, Zhiyan Liud, Takashi Ozakie,chiro Mori f, Akira Miyauchig, Kennichi Kakudoh, Misa Nakamura i,∗

Department of Pathology and Laboratory Medicine, Hara Hospital, Kinki University School of Medicine, 1248-1, Otoda, Ikoma, Nara 630-0293, JapanDepartment of Pathology, Taishan Medical University, 2# Yingsheng Dong Road, Tai’an City, Shandong 271000, PR ChinaKey Laboratory of Carcinogenesis and Translational Research (Ministry of Education), Department of Pathology, Peking University Cancer Hospital &

nstitute, Beijing, PR ChinaDepartment of Pathology and Pathophysiology, Shandong University School of Medicine, Shandong 250012, PR ChinaSocial Insurance Kinan Hospital, 46-70 Shinjo-cho, Tanabe, Wakayama 646-858, JapanDepartment of Pathology, International University of Health and Welfare Mita Hospital, 143 Mita, Minato-ku, Tokyo 108-8329, JapanDepartment of Surgery, Kuma Hospital, 8-2-35 Shimoyamate-dori, Chuo-ku, Kobe 650-0011, JapanDepartment of Medical Technology, Tokiwa University, 2-6-2 Ootani, Nagata-ku, Kobe 653-0838, JapanDepartment of Rehabilitation, Osaka Kawasaki Rehabilitation University, 158, Mizuma, Kaizuka-City, Osaka 597-0104, Japan

r t i c l e i n f o

rticle history:eceived 21 January 2014eceived in revised form 19 June 2014ccepted 21 June 2014

eywords:rinary-type plasminogen activatorapillary thyroid cancerRAFRK1/2EK inhibitor

a b s t r a c t

The aim of the present study is to investigate the relationship between urinary-type plasminogen acti-vator (uPA) expression and clinicopathological features in papillary thyroid carcinoma (PTC) and todetermine the signal transduction of PTC cells in vitro.

PTC tissues from 42 patients were analyzed for the expression of uPA and the BRAFV600E mutation.BCPAP, a PTC cell line harboring the BRAFV600E mutation, was used to study MAPK signaling. PCR anddirect sequencing were applied to analyze BRAFV600E mutation status. uPA mRNA expression was mea-sured using a quantitative RT-PCR method, and uPA protein was localized using an immunohistochemicalmethod. The ERK protein status was detected by Western blot analysis.

uPA gene expression was significantly increased in PTC tissues as compared to the correspondingnon-tumor tissues. Furthermore, the up-regulation of uPA mRNAs was correlated with high-risk clini-

copathological features, including extrathyroid invasion, loss of cellular polarity/cohesiveness, and theBRAFV600E mutation. Marked dephosphorylation of ERK1/2 and down-regulation of uPA expression weredetected when BCPAP was treated with a MEK inhibitor, U0126.

MEK inhibitors might be a potential treatment strategy for aggressive PTC with BRAFV600E throughinhibition of uPA expression.

© 2014 Elsevier GmbH. All rights reserved.

ntroduction

Papillary thyroid carcinoma (PTC) is the most common malig-ant tumor of the human endocrine system. Although PTC iselatively indolent and highly curable, up to 10% of patients with

TC develop cancer recurrence and eventually die from this disease10]. There are several clinicopathological features that are rec-gnized as classical high-risk factors, including large tumor size,

∗ Corresponding author. Tel.: +81 72 421 3347; fax: +81 72 421 3347.E-mail address: nakamuram@kawasakigakuen.ac.jp (M. Nakamura).

1 These authors contributed equally to this work.

ttp://dx.doi.org/10.1016/j.prp.2014.06.025344-0338/© 2014 Elsevier GmbH. All rights reserved.

extrathyroid invasion, lymph node metastasis, distant metastasisand advanced tumor stages [6]. For PTC, histological subtype is alsoan important factor in the risk evaluation of this cancer. Recently,loss of cellular polarity/cohesiveness (LOP/C) in PTC, particularlyin the invasive front, was proposed by Kakudo and colleaguesto describe a unique subgroup of PTC, in which the cancer cellsdisplay a hobnail-like pattern or loose arrangement without well-formed papillary or follicular structures. A subsequent study fromthe same group demonstrated that this subtype can predict a high-

risk for PTC recurrence [2]. Furthermore, PTC is associated withnon-overlapping activating mutations of RET, NTRK, RAS, and BRAF[1,22]. Among these genetic defects, the BRAFV600E mutation isthe most common oncogene identified in sporadic PTC. Although

7 earch a

tBap

rel[hnbpurtptop

onuew

M

P

lwHcp

wsfth

H

aWPsLwohn

C

BoPpct

34 T. Wakasa et al. / Pathology – Res

here is still controversy, many studies have demonstrated thatRAFV600E mutation is correlated with the high-risk factors listedbove and may be a driving force behind the aggressive clinico-athological characteristics of PTC [14].

On the other hand, in the complex tumor progression, a keyequirement is the ability of tumor cells to produce proteolyticnzymes within the tumor cell environment to promote extracellu-ar matrix degeneration to facilitate tumor invasion and metastasis7]. One such protease, urinary-type plasminogen activator (uPA),as strongly been implicated in the progression of several malig-ancy, including thyroid cancer [8,11,21]. However, the correlationetween increased expression of uPA and the high-risk clinico-athological features with genetic alterations of PTC is still poorlynderstood. Furthermore, in vitro studies suggest a significant cor-elation between uPA up-regulation and constitutive activation ofhe RAS-RAF-MEK-ERK/MAPK pathway (referred to as the MAPKathway) in breast cancer [17]. All these observations have led uso hypothesize that BRAFV600E mutation could drive up-regulationf uPA through constitutive activation of the MAPK pathway andarticipate in the progression of high-risk group of PTC.

In the present study, we examined the expression differencef uPA in clinical samples of PTC and the corresponding non-eoplastic tissues and evaluated its clinical correlation. We thensed a PTC cell line with BRAFV600E mutation to investigate uPAxpression and its changes upon suppression of the MAPK pathwayith a MEK-1 inhibitor.

aterials and methods

atients and tissue samples

Matched tumor tissues and non-neoplastic tissues were col-ected from 42 patients (mean age: 54.7 ± 17.0 years, 7 men and 35

omen) with primary PTC who underwent thyroidectomy at Kumaospital (Kobe, Japan). All patients gave their written informedonsent to the ethics committee of Kuma Hospital. None of the 42atients had distant metastasis at the time of surgery.

After surgical resection, each thyroid tissue was handled in twoays: one part of thyroid tissue was snap-frozen immediately and

tored at −80 ◦C for RNA extraction; the remnants were fixed inormalin and embedded in paraffin for histopathological examina-ion. Serial sections were cut from paraffin blocks and prepared forematoxylin and eosin (HE) staining and immunohistochemistry.

istological evaluation and immunohistochemistry

All HE sections were reviewed by three pathologists (Y.B., Y.L.nd K.K.) to confirm the histological diagnosis according to theHO classification of thyroid tumors (2004). Only common-type

TCs were included in the current study, and aggressive variants,uch as tall cell, columnar cell, and solid variants, were excluded [6].oss of cellular polarity/cohesiveness (LOP/C) in the invasive frontas evaluated according to the criteria previously proposed by

ur group [3,13,16]. Immunohistochemistry was performed usinguman anti-uPA antibody (No. 3689, 1:25 dilution, American Diag-ostica, CT).

ell culture

The human PTC-derived BCPAP cell line, which harbors theRAFV600E mutation, was purchased from the German Collectionf Microorganisms and Cell Cultures (DSMZ, Germany); the other

TC-derived TPC-1 cell line with RET/PTC-1 rearrangement wasrovided by Dr. R. Katoh (University of Yamanashi, Japan). Theseell lines were routinely maintained in RPMI1640 medium (Invi-rogen, Japan) supplemented with 10% fetal bovine serum (FBS)

nd Practice 210 (2014) 733–738

(HyClone, UT) in 5% CO2 at 37 ◦C. For in vitro experiments, 80%confluent cells were cultured in 0.5% serum-containing mediumfor 18 h and then treated with or without 10 �mol/L U0126 (Cal-biochem, CA) for different time periods before extraction of RNAand protein [15,17].

Total RNA isolation, RT-PCR, and detection of BRAF mutation

Total RNA was isolated from each cell line and tissue sam-ples using the ULTRASPEC RNA Isolation System (Biotecx, TX) [18].cDNA was synthesized using the SuperScript First-Strand SynthesisSystem for RT-PCR (Invitrogen, CA). The PCR reaction primers foruPA and BRAF were as follows: uPA forward 5′-AAGGACTACAGCG-CTGACAC-3′, reverse 5′-AACTCCTGCAGGCTTCAGTC-3′; BRAF for-ward 5′-GCACAGGGCATGGATTACTT-3′, reverse 5′-GATGACTTC-TGGTGCCATCC-3′. RT-PCR analysis with ˇ-actin primers was usedas an internal quality control [18]. In addition, the purified PCRproducts of cDNA for the BRAF gene (exon 15) were sequencedusing PCR primers as previously described [15].

Quantitative real-time PCR

Quantitative real-time PCR was carried out in an ABI PRISM7000 sequence detection system apparatus (Applied Biosystems,CA) according to Taqman Protocol. Primer and probe sets used foranalysis of uPA (Hs00170182 m1) and the endogenous reference,ˇ-actin (Catalog #4310881E) transcripts were purchased fromApplied Biosystems (Applied Biosystems, CA). The PCR amplifica-tion was performed in a final volume of 20 �l containing: TaqManGene Expression Assay (uPA), cDNA template and Taqman Univer-sal PCR Master Mix. Each sample was measured in triplicate, and theexpression level of the tested gene was normalized to the house-keeping gene (ˇ-actin). Differences in gene expression betweenthe PTC tissues and the corresponding non-neoplastic tissueswere calculated using the formula 2exp (�Ct tumor − �Ct normal)as previously described [3]. The results are expressed as foldchanges.

Western blot analysis and ELISA assay

Thirty micrograms of total protein from each sample was ana-lyzed by 10% SDS-PAGE gels and transferred to polyvinylidenefluoride membranes (Millipore, MA). Membranes were incubatedwith anti-ERK antibody (Cell signaling, MA; #9102, 1:1000) or anti-pERK antibody (Cell signaling, MA; #9101, 1:1000) overnight at 4 ◦C[18]. This was followed by incubation with HRP-coupled anti-rabbitIgG secondary antibody (Dako, Denmark; 1:2000) at room tem-perature for 1 h, and the signals were detected with an enhancedchemiluminescence (ECL) kit (Amersham, UK). Three independentexperiments were performed.

For determining the secreted uPA levels in the medium of cul-tured cells, ELISA assay was also performed. Briefly, BCPAP cellswere pre-cultured in 0.5% serum-containing medium for 18 h,plated at 5 × 104 cells in 60-mm dishes, and treated with or with-out 10 �mol/L U0126. After incubation for different time-periods,the medium was collected. uPA ELISA kits (American Diagnostica,CT) were used to measure the concentrations of the uPA in cell cul-ture supernatants. The data were expressed as the mean ± SE of thedeterminations from three independent experiments.

Statistical analysis

The comparison of uPA mRNA expression levels between tumortissues and matched non-neoplastic tissues was performed usingpaired t-test. The Correlation between uPA mRNA expressionand clinicopathological variables was analyzed statistically using

T. Wakasa et al. / Pathology – Research and Practice 210 (2014) 733–738 735

Faˇ

ttmA5c

R

up

ct(afm

1rbdseob

Ffi(

ig. 1. mRNA expression of uPA in tumors and corresponding non-tumor tissuesnalyzed by quantitative RT-PCR. The expression levels of uPA were normalized to-actin expression. The bars represent mean ± SE.

he unpaired t-test with Welch’s correction for comparisons ofwo groups. The results of in vitro experiments are expressed as

eans ± SE, and values were statistically evaluated using two-wayNOVA. All analyses were performed using GraphPad Prism version.0 (GraphPad Software, CA), and p < 0.05 was considered statisti-ally significant.

esults

PA expression levels are increased in tumor samples of PTCatients and PTC cell lines

Quantitative expression of uPA mRNA was analyzed in all 42ases of PTC. The level of uPA transcripts was significantly higher inumor tissues than in the non-neoplastic counterparts (p = 0.0323)Fig. 1). The differences in expression between tumor tissuesnd paired non-neoplastic tissues (T/N ratio) were expressed asold changes. The mean difference was 7.981 ± 1.970 (means ± SE,

edian = 2.800).To confirm the uPA expression results at the protein level,

0 cases of PTC were examined by immunohistochemistry. Theemaining cases were not available for immunohistochemistryecause the tissues were treated with decalcification solutionue to calcification and/or bone formation. The degree of expres-

ion was graded as negative, weak, intermediate, or strong. In allxamined cases, intermediate to strong uPA immunoreactivity wasbserved in the cytoplasm of tumor cells. Although normal andenign thyroid tissues adjacent to the tumor area were sometimes

ig. 2. Immunostaining of uPA. Interface of normal thyroid (upper left) and PTC (lower

brous stroma is shown in B. Note the positive stains of PTC tumor cells for uPA (B), paLOP/C) (Arrow).

Fig. 3. RT-PCR analyses of uPA expression in PTC cell lines and a pair of matchednormal and tumor tissues from case No. 16. The endogenous ˇ-actin transcriptsserve as control.

positive for the antibody, the staining was less intense than thestaining in tumors (Fig. 2A).

To evaluate the uPA mRNA expression levels in both BCPAP andTPC-1 cells, semi-quantitative RT-PCR was performed. The cDNAsamples from case No. 16 were used as a control. Compared withthat of non-neoplastic thyroid tissue, more intense bands at 789 bpfor uPA transcripts of tumor tissue and PTC cell lines were noted ingel electrophoresis (Fig. 3), which indicated almost the same levelover-expression of uPA mRNA in PTC cell lines as that found incancerous tissue.

Increased expression of uPA is correlated with high-riskclinicopathological features and BRAFV600E mutation in PTC

Based on a retrospective review of the cases, loss of cellularpolarity/cohesiveness (LOP/C) in the invasive front was presentin 24 of 42 (57.1%) cases. A genetic assessment for the BRAFV600E

mutation revealed that 31 of 42 (73.8%) cases were positive forthis mutation. The association of uPA mRNA expression levels with

clinicopathological parameters was demonstrated in Table 1. Nosignificant correlation was noted between the levels of T/N ratioand gender, age, tumor size or lymph node metastasis. However,significant higher expression of uPA was observed in cases with pEx

right) is shown in A, which shows expansive growth. Invasive growth of PTC intorticularly of those in the invasive front with loss of cellular polarity/cohesiveness

736 T. Wakasa et al. / Pathology – Research a

Table 1The correlation between uPA expression and clinicopathological parameters in PTC.

Parameters Patients uPA expression(T/N ratio)

p-Value

No. % Mean ± SE

GenderMale 7 16.7 8.357 ± 5.727 0.9432Female 35 83.3 7.906 ± 2.112

Age (years)<45 11 26.2 3.874 ± 1.452 0.0678≥45 31 73.8 9.439 ± 2.582

Tumor size (cm)<4.0 28 66.7 9.475 ± 2.838 0.1707≥4.0 14 33.3 4.993 ± 1.496

Extrathyroid invasion (pEx)pEx0 11 26.2 3.446 ± 1.521 0.0392*pEx1/pEx2 31 73.8 9.810 ± 2.565

Tumor stage (pT)pT1/pT2 10 23.8 2.931 ± 1.378 0.0254*pT3/pT4 32 76.2 9.559 ± 2.496

Lymph node metastasis (LN-M)LN-M (−) 8 19.0 8.418 ± 2.368 0.5160LN-M (+) 34 81.0 6.125 ± 2.537

Loss of cellular polarity/cohesiveness (LOP/C)LOP/C (–) 18 42.9 11.27 ± 3.197 0.0273*LOP/C (+) 24 57.1 3.351 ± 1.161

BRAFV600E mutationMutation (−) 11 26.2 2.455 ± 1.156 0.0105*

(abdiwci

inhibition of pERK at 6 h after treatment (Fig. 4C). Furthermore, uPA

FEe

Mutation (+) 31 73.8 9.997 ± 2.552

p = 0.0392), advanced tumor stages (p = 0.0254), LOP/C (p = 0.0273)nd BRAFV600E mutation (p = 0.0105). Furthermore, a correlationetween uPA expression levels and LOP/C was to some degreeemonstrated by immunostaining at the protein level. Strong pos-

tive staining for uPA (Fig. 2B) was identified in the PTC tumor area

ith LOP/C (Arrow) in 8 of 10 cases subjected to immunohisto-

hemical examination, while only intermediate staining was notedn the tumor cell area without LOP/C in the same specimens.

ig. 4. Effect of U0126 on phosphorylation of ERK1/2 and expression levels of uPA in BCRK1/2 antibody in cells treated with or without U0126 (A and C). The blots were also nxamine the mRNA levels of uPA in cells treated with or without U0126 (B and D).

nd Practice 210 (2014) 733–738

MEK inhibitor, U0126, suppresses uPA expression in a BCPAP cellline harboring BRAFV600E mutation through the MAPK pathway

We found that the up-regulation of uPA occurred more fre-quently in BRAFV600E mutation-positive tumors than in BRAFV600E

mutation-negative tumors. Therefore, we speculated that theMAPK signaling pathway may play an important role in the regula-tion of uPA expression in a PTC cell line with BRAFV600E mutation.To test this hypothesis, we treated the PTC cell line BCPAP and TPC-1 with U0126, a MEK-specific inhibitor. BCPAP cells and TPC-1 cellsshowed constitutive activation of ERK1/2 (Fig. 4A, 4C) and over-expression of uPA (Fig. 3). Western blot analysis was performedto assess the expression of phosphorylated ERK1/2. During the cellculture process, some extent of pERK decreasing was noted after12 h of inhibitor treatment. This phenomenon may be due to lack ofnutrition, because these PTC cell lines were primarily maintained inRPMI1640 medium supplemented with 10% FBS. During the obser-vation period (24 h), no significant change regarding morphology,cell proliferation activity or motility has been identified in bothBCPAP and TPC-1 cell line, after treatment of the inhibitor U0126.A time-course analysis showed that in BCPAP cells, 10 �mol/L ofU0126 strongly inhibited ERK1/2 phosphorylation at least 3 h aftertreatment and that the inhibition lasted for 24 h (Fig. 4A). To deter-mine whether the levels of uPA expression showed a correspondingchange, we performed real-time PCR and ELISA analysis. Consis-tent with the observation of ERK1/2 phosphorylation, uPA mRNAlevels in BCPAP cells were decreased to an average of one-halfof the levels in untreated BCPAP cells at 3, 6, 12 and 24 h afterU0126 treatment (Fig. 4B). Similarly, U0126 significantly inhibitedthe secretion of uPA in BCPAP cell supernatants, as compared withthat exhibited by untreated cells (Fig. 5A). On the contrary, TPC-1cell line was relatively resistant to U0126, which exhibited slight

mRNA level and protein secretion in the supernatant from TPC-1 cells receiving U0126 treatment revealed no significant change(Figs. 4D and 5B).

PAP and TPC-1 cells. Western blots were performed using an antiphospho-specificormalized to total ERK1/2 in each sample. Quantitative RT-PCR was performed to

T. Wakasa et al. / Pathology – Research a

Fqe

D

pdtisliiTilcaupbgd

ttpTots

ig. 5. uPA protein secretion levels in BCPAP and TPC-1 cell supernatants wereuantified by ELISA. The symbols represent the mean ± SE of three independentxperiments.

iscussion

In recent years, several reports appeared documenting theresence of over-expression of uPA in human thyroid carcinomaerived cell lines and PTC tissues [11,21]. In agreement withhese reports, we found that uPA gene expression is significantlyncreased in PTC. On the contrary, in normal and benign thyroid tis-ues adjacent to the tumor, expression of uPA mRNA was found atower level. There results were further confirmed at protein level bymmunohistochemistry. Significantly stronger staining was notedn tumor tissues compared to normal and benign thyroid tissues.he increase in uPA production further supports the key role of uPAn promoting thyroid cancer progression. Furthermore, we corre-ated the up-regulation of uPA mRNAs with conventional high-risklinicopathological features of PTC, including extrathyroid invasionnd advanced tumor stages. These results suggest that uPA could besed as a molecular biomarker to indicate the risk stratification andrognosis of PTC patients. This speculation is strongly supportedy the work of Ulisse and colleagues, who showed that increasedene expression of uPA was significantly associated with a shorterisease-free interval [20].

Recently, immunohistochemical analysis by Liu et al. revealedhat the tumor area with LOP/C showed decreased expression ofhyroid transcription factor-1 and E-cadherin, aberrant ˇ-cateninolarity (cellular localization), and increased vimentin expression.

hey suggested that LOP/C may be a useful morphological featuref epithelial–mesenchymal transition (EMT) under HE observa-ion [16]. It is important to emphasize, in the present study, wehowed for the first time that up-regulation of uPA mRNA is

nd Practice 210 (2014) 733–738 737

significantly correlated with the presence of LOP/C in the inva-sive front of PTC. Using immunostaining for uPA, a strong positivereaction to uPA was revealed in the invasive front of PTC withpronounced LOP/C or tumor cells showing a hobnail-like pattern,while only intermediate uPA staining was found in a relatively cen-tral region of the tumor, as shown in Fig. 2. Such observationsindicate that uPA over-expression in PTC may play an importantrole in cancer cell migration and invasion through promoting theEMT.

Although there is still controversy, the majority of studies withrelatively long periods of follow-up suggest that BRAFV600E muta-tion is significantly correlated with high-risk clinicopathologicalfeatures and poor prognosis in PTC [11]. It is possible that mutantBRAF has significant effects on the expression patterns of markersthat could be correlated with tumor progression in PTC. Molecu-lar markers involved in BRAFV600E mutation-promoted progressionand aggressiveness of PTC include vascular endothelial growthfactor and matrix metalloproteinases [12]. In the present study,we show that significant higher expression level of uPA gene isidentified in BRAFV600E mutation-positive PTC, which is consis-tent with a previous report [19]. These results suggest that mutantBRAF may contribute to up-regulation of uPA. Since BRAF is aserine–threonine kinase that mediates the signal transduction ofthe MAPK pathway, we further investigated whether this path-way is involved in uPA expression regulation in BCPAP cells, a PTCcell line harboring the BRAFV600E mutation. After treatment withU0126, a specific MEK1/2 inhibitor, BCPAP cells rapidly exhibiteddramatic dephosphorylation of ERK1/2. Most importantly, corre-sponding decreases in uPA mRNA expression and protein secretionwere demonstrated in the same time period. Although there werea few studies that demonstrated an up-regulation of uPA in PTCspecimens with RET/PTC-1 arrangement [5], our results indicatedthat the cells with a BRAFV600E mutation may be more sensitiveto some MEK inhibitors than cells with RET/PTC1 rearrangementespecially in terms of down-regulation of uPA expression. Suchobservations led to the concepts that uPA expression can be reg-ulated through the MAPK pathway in PTC cell lines harboring theBRAFV600E mutation and that indirect inhibition of the uPA systemwith a MEK1/2 inhibitor provides a novel avenue for controlling andtargeting aggressive PTC. Despite the inhibitory effect of U0126 onPTC cells, this commercially available inhibitor is only commonlyused for in vitro studies due to its inactivity in vivo [9].

Genetically, there are two upstream regions of uPA gene regulat-ing its transcription: the minimal promoter (MP) and the enhancerelement. The MP of uPA gene, which extends approximately 86 bpupstream of the transcription start site, plays a very importantrole in uPA gene expression. Binding of a phosphorylated Sp1transcription factor is essential for the activation of MP. A recentstudy has demonstrated that the MAPK pathway activates uPAgene expression through Sp1 phosphorylation in HeLa, LNCaP andCCL39-derivative cells [4]. Therefore, we hypothesized that a sim-ilar mechanism may contribute to the over-expression of uPA inthyroid cancer cells. It will be critical to explore in future studies.

In conclusion, we present evidence that some human PTC tissuesover-express uPA and that the up-regulation of uPA is correlatedwith high-risk clinicopathological features and BRAFV600E muta-tion.

Moreover, we demonstrate that the expression of uPA can beregulated through the MAPK pathway and effectively inhibited byMEK inhibitors in the PTC cell line with BRAFV600E mutation.

Acknowledgments

We thank Ms. Miyoko Higuchi, Kuma Hospital, for prepar-ing fresh samples; Dr. Yasuhiro Ito, Kuma Hospital, for providing

7 earch a

ci

R

[

[

[

[

[

[

[

[

[

[

[

[

papillary carcinomas, Eur. J. Cancer 42 (2006) 2631–2638.[22] H. Zuo, Y. Nakamura, H. Yasuoka, P. Zhang, M. Nakamura, I. Mori, A. Miyauchi,

38 T. Wakasa et al. / Pathology – Res

linical information; Ms. Emiko Taniguchi for her skillful assistancen immunohistochemistry.

eferences

[1] A.J. Adeniran, Z. Zhu, M. Gandhi, D.L. Steward, J.P. Fidler, T.J. Giordano, P.W. Bid-dinger, Y.E. Nikiforov, Correlation between genetic alterations and microscopicfeatures, clinical manifestations, and prognostic characteristics of thyroid pap-illary carcinomas, Am. J. Surg. Pathol. 30 (2006) 216–222.

[2] Y. Bai, K. Kakudo, Y. Li, Z. Liu, T. Ozaki, Y. Ito, M. Kihara, A. Miyauchi, Subclassifi-cation of non-solid-type papillary thyroid carcinoma identification of high-riskgroup in common type, Cancer Sci. 99 (2008) 1908–1915.

[3] Y. Bai, K. Kakudo, M. Nakamura, T. Ozaki, Y. Li, Z. Liu, I. Mori, A. Miyauchi, G.Zhou, Loss of cellular polarity/cohesiveness in the invasive front of papillarythyroid carcinoma and periostin expression, Cancer Lett. 281 (2009) 188–195.

[4] E. Benasciutti, G. Pagès, O. Kenzior, W. Folk, F. Blasi, M.P. Crippa, MAPK andJNK transduction pathways can phosphorylate Sp1 to activate the uPA mini-mal promoter element and endogenous gene transcription, Blood 104 (2004)256–262.

[5] M.G. Borrello, L. Alberti, A. Fischer, D. Deql’Innocenti, C. Ferrario, M. Gariboldi,F. Marchesi, P. Allavena, A. Greco, P. Collini, S. Pilotti, G. Cassinelli, P. Bressan, L.Fugazzola, A. Mantovani, M.A. Pierotti, Induction of a proinflammatory programin normal human thyrocytes by the RET/PTC1 oncogene, Proc. Natl. Acad. Sci.U. S. A. 102 (2005) 14825–14830.

[6] R.A. DeLellis, R.V. Lloyd, P.U. Heitz, C. Eng, Tumours of Endocrine Organs, WorldHealth Organization Classification of Tumours; Pathology and Genetics, IARCPress, Lyon, 2004.

[7] B. Han, M. Nakamura, I. Mori, Y. Nakamura, K. Kakudo, Urokinase-type plas-minogen activator system and breast cancer, Oncol. Rep. 14 (2005) 105–112(Review).

[8] S.R. Harvey, S.N. Sait, Y. Xu, J.L. Bailey, R.M. Penetrante, G. Markus, Demon-stration of urokinase expression in cancer cells of colon adenocarcinomas byimmunohistochemistry and in situ hybridization, Am. J. Pathol. 155 (1999)1115–1120.

[9] Y.C. Henderson, M.J. Fredrick, G.L. Clayman, Differential responses of humanpapillary thyroid cancer cell lines carrying the RET/PTC1 rearrangement or aBRAF mutation to MEK1/2 inhibitors, Arch. Otolaryngol. Head Neck Surg. 133(2007) 810–815.

10] S.A. Hundahl, I.D. Fleming, A.M. Fremgen, H.R. Menck, A national cancer database report on 53,856 cases of thyroid carcinoma treated in the U.S., 1985–1995,Cancer 83 (1998) 2638–2648 (see comments).

11] B. Jarzab, M. Wiench, K. Fujarewicz, K. Simek, M. Jarzab, M. Oczko-Wojciechowska, J. Wloch, A. Czarniecka, E. Chmielik, D. Lange, A. Pawlaczek,S. Szpak, E. Gubala, A. Swierniak, Gene expression profile of papillary thyroidcancer: sources of variability and diagnostic implications, Cancer Res. 65 (2005)1587–1597.

nd Practice 210 (2014) 733–738

12] Y.S. Jo, S. Li, J.H. Song, K.H. Kwon, J.C. Lee, S.Y. Rha, H.J. Lee, J.Y. Sul, G.R. Kweon,H.K. Ro, J.M. Kim, M. Shong, Influence of the BRAF V600E mutation on expres-sion of vascular endothelial growth factor in papillary thyroid cancer, J. Clin.Endocrinol. Metab. 91 (2006) 3667–3670.

13] K. Kakudo, W. Tang, Y. Ito, I. Mori, Y. Nakamura, A. Miyauchi, Papillarycarcinoma of the thyroid in Japan: subclassification of common type and iden-tification of low risk group, J. Clin. Pathol. 57 (2004) 1041–1046.

14] Y. Li, M. Nakamura, K. Kakudo, Targeting of the BRAF gene in papillary thyroidcarcinoma, Oncol. Rep. 22 (2009) 671–681 (Review).

15] D. Liu, S. Hu, P. Hou, D. Jiang, S. Condouris, M. Xing, Suppression ofBRAF/MEK/MAP kinase pathway restores expression of iodide-metabolizinggenes in thyroid cells expressing the V600E BRAF mutant, Clin. Cancer Res. 13(2007) 1341–1349.

16] Z. Liu, K. Kakudo, Y. Bai, Y. Liu, T. Ozaki, A. Miyauchi, E. Taniguchi, I. Mori, K.Kakudo, Loss of cellular polarity/cohesiveness in the invasive front of papil-lary thyroid carcinoma, a novel predictor for lymph node metastasis; possiblemorphological indicator of epithelial mesenchymal transition, J. Clin. Pathol.64 (2011) 325–329.

17] Z. Ma, D.J. Webb, M. Jo, S.L. Gonias, Endogenously produced urokinase-typeplasminogen activator is a major determinant of the basal level of activatedERK/MAP kinase and prevents apoptosis in MDA-MB-231 breast cancer cells, J.Cell Sci. 114 (2001) 3387–3396.

18] M. Nakamura, B. Han, T. Nishishita, Y. Bai, K. Kakudo, Calcitonin targets extra-cellular signal-regulated kinase signaling pathway in human cancers, J. Mol.Endocrinol. 39 (2007) 375–384.

19] S. Ulisse, E. Baldini, S. Sorrenti, S. Barollo, L. Gnessi, A. Catania, A. Nesca, L.Gnessi, M.R. Pelizzo, C. Mian, C. De Vito, A. Calvanese, S. Palermo, S. Persechino,E. De Antoni, M. D’Armiento, In papillary thyroid carcinoma BRAFV600E is asso-ciated with increased expression of the urokinase plasminogen activator andits cognate receptor, but not with disease-free interval, Clin. Endocrinol. (Oxf.)77 (2012) 780–786.

20] S. Ulisse, E. Baldini, S. Sorrenti, A. Catania, M.R. Pellizzo, F. Nardi, C. Mian, E. DeAntoni, M. D’Armiento, L. Frati, High expression of the urokinase plasminogenactivator and its cognate receptor associates with advanced stages and reduceddisease-free interval in papillary thyroid carcinoma, J. Clin. Endocrinol. Metab.96 (2011) 504–508.

21] S. Ulisse, E. Baldini, M. Toller, E. Marchioni, L. Giacomelli, E. De Antoni, E. Ferretti,A. Marzullo, F.M. Graziano, P. Trimboli, L. Biordi, F. Curcio, A. Gulino, F.S. Ambesi-Impiombato, M. D’Armiento, Differential expression of the components of theplasminogen activating system in human thyroid tumour derived cell lines and

K. Kakudo, Lack of association between BRAF V600E mutation and mitogen-activated protein kinase activation in papillary thyroid carcinoma, Pathol. Int.57 (2007) 12–20.