Translational Cancer Mechanisms and Therapy

Serum Nardilysin, a Surrogate Marker forEpithelial–Mesenchymal Transition, PredictsPrognosis of Intrahepatic Cholangiocarcinomaafter Surgical ResectionTomoaki Yoh1, Etsuro Hatano2, Yosuke Kasai1, Hiroaki Fuji1, Kiyoto Nishi3,Kan Toriguchi2, Hideaki Sueoka2, Mikiko Ohno4, Satoru Seo1, Keiko Iwaisako5,Kojiro Taura1, Rina Yamaguchi6, Masato Kurokawa6, Jiro Fujimoto2, Takeshi Kimura3,Shinji Uemoto1, and Eiichiro Nishi4

Abstract

Purpose: Few studies have investigated prognostic biomar-kers in patients with intrahepatic cholangiocarcinoma (ICC).Nardilysin (NRDC), a metalloendopeptidase of the M16family, has been suggested to play important roles in inflam-mation and several cancer types. We herein examined theclinical significance and biological function of NRDC in ICC.

Experimental Design:Wemeasured serum NRDC levels in98 patients with ICC who underwent surgical resection in twoindependent cohorts to assess its prognostic impact. We alsoanalyzed NRDC mRNA levels in cancerous tissue specimensfrom 43 patients with ICC. We investigated the roles of NRDCin cell proliferation, migration, gemcitabine sensitivity, andgene expression in ICC cell lines using gene silencing.

Results: High serum NRDC levels were associated withshorter overall survival and disease-free survival in the primary

(n ¼ 79) and validation (n ¼ 19) cohorts. A correlation wasobserved between serum protein levels and cancerous tissuemRNA levels ofNRDC (Spearman r¼ 0.413; P¼ 0.006). Thegene knockdown of NRDC in ICC cell lines attenuated cellproliferation, migration, and tumor growth in xenografts, andincreased sensitivity to gemcitabine. The gene knockdown ofNRDC was also accompanied by significant changes in theexpression of several epithelial–mesenchymal transition(EMT)-related genes. Strong correlations were observedbetween the mRNA levels of NRDC and EMT-inducing tran-scription factors, ZEB1 and SNAI1, in surgical specimens frompatients with ICC.

Conclusions: Serum NRDC, a possible surrogate markerreflecting the EMT state in primary tumors, predicts theoutcome of ICC after surgical resection.

IntroductionIntrahepatic cholangiocarcinoma (ICC) is the second most

commonprimary liver cancer followinghepatocellular carcinoma(HCC), accounting for 5% to 15% of all primary liver cancers(1–3). There are marked geographic variations in the incidence ofICC, with a higher incidence in East Asia, whereas the number ofpatients with ICC has been reported to be increasing worldwide(3, 4). The survival rate of patients with ICC is poor because of thelate presentation of the disease and limited therapies. Although

surgical resection is the only curative treatment, 30% to 40% ofpatients with ICC have surgical indications (3). Moreover, therecurrence rate after surgical resection is 50% to 60% and the5-year overall survival (OS) rate after surgical resection is only25% to 31% (3, 5, 6), highlighting the need to optimize adjuvantstrategies. Recent evidence has suggested that adjuvant chemo-therapy is associatedwithprolonged survival, particularly in someadvanced cases (7, 8). However, there are no establishedmethodsto define patient subgroups that need adjuvant strategies. Thepreoperative measurement of serum tumor markers may identifyhigh-risk patients; however, few studies have investigated bio-markers in patients with ICC possibly due to the difficultiesassociated with collecting large numbers of serum samples frompatients with ICC (9–11).

Nardilysin (N-arginine dibasic convertase, NRDC) is a zincpeptidase of the M16 family that selectively cleaves dibasic sites(12, 13). NRDC exhibits widespread expression throughout thebody, and regulates multiple biological processes, such as mye-lination (14), body temperature homeostasis (15), and insulinsecretion (16). Although NRDC is a soluble cytosolic proteinwithout anobvious signal peptide or nuclear localization signal, itshuttles between the cytoplasm and nucleus and is secreted via anas yet unknown mechanism (17). We identified NRDC as aspecific binding partner of heparin-binding epidermal growthfactor-like growth factor (HB-EGF). Our subsequent studies

1Department of Surgery, Graduate School of Medicine, Kyoto University, Kyoto,Japan. 2Department of Surgery, Hyogo College ofMedicine, Nishinomiya, Japan.3Department of Cardiovascular Medicine, Graduate School of Medicine, KyotoUniversity, Kyoto, Japan. 4Department of Pharmacology, Shiga University ofMedical Science, Otsu, Japan. 5Department of Medical Life Systems, Faculty ofLife and Medical Sciences, Doshisha University, Kyotanabe, Japan. 6SanyoChemical Industries Ltd., Kyoto, Japan.

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Corresponding Authors: Eiichiro Nishi, Shiga University of Medical Science,Shiga, 520-2192, Japan. Phone: 81-77-548-2181; E-mail:enishi@belle.shiga-med.ac.jp; and Etsuro Hatano, shatano@hyo-med.ac.jp

doi: 10.1158/1078-0432.CCR-18-0124

�2018 American Association for Cancer Research.

ClinicalCancerResearch

www.aacrjournals.org 619

on March 18, 2020. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 23, 2018; DOI: 10.1158/1078-0432.CCR-18-0124

demonstrated that NRDC enhanced the ectodomain shedding ofHB-EGF andothermembraneproteins, such as TNFa, through theactivation of disintegrin and metalloproteinase (ADAM) pro-teases (18, 19). In addition to its extracellular functions, werecently clarified the nuclear functions of NRDC as a transcrip-tional coregulator, whichmodulates the transcriptional activity ofhistone deacetylase 3 (HDAC3; ref. 20), peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1a; ref. 15),and islet-1 (16). Furthermore, NRDC is strongly expressed inseveral cancer types, promotes tumor growth (19, 21–23), and isassociated with a poor prognosis (19, 22, 23), suggesting impor-tant roles for NRDC in tumor biology. Our recent findings alsoindicated the clinical usefulness of serum NRDC in specificclinical settings. For example, serumNRDC levels were associatedwith the survival outcomes of postoperative patients with HCCwith hepatitis C (23).

In this study, we retrospectively investigated serum expressionlevels of NRDC in patients with ICC who underwent surgicalresection to clarify whether serum NRDC has potential as apostoperative prognostic indicator. We also examined NRDCmRNA expression in surgical specimens resected from patientsand the pathophysiologic role of NRDC in ICC using an RNAinterference method in ICC cell lines.

Materials and MethodsStudy design and population

This study was designed to investigate whether serum NRDChas the ability to predict the outcomes of patients with ICC aftersurgical resection. We analyzed serum NRDC levels in two inde-pendent cohorts: a primary cohort at Kyoto University and anexternal validation cohort at Hyogo College of Medicine. In theprimary cohort, we analyzed 79 consecutive patients with ICCwho underwent surgical resection at Kyoto University Hospital(Kyoto, Japan) between January 2006 and July 2013. In thevalidation cohort, we analyzed 19 consecutive patients with ICCwho underwent surgical resection at Hyogo College of Medicine(Nishinomiya, Japan) between January 2009 and January 2015.The final diagnosis of ICC was histologically confirmed. Thefollow-up data of the primary and validation cohorts wereupdated in April 2017 and February 2018, respectively. The

surgical procedure in the primary cohort was reported previously(6, 7, 9). Serum samples from patients with ICC (n ¼ 98) wereobtained preoperatively at the time of admission. Fresh frozencancer tissues were obtained from 43 of 79 patients with ICC inthe primary cohort, from which mRNA was isolated. Surgicalspecimens from 20 of the 43 patients with ICC were also assessedby IHC.

Clinicopathologic and survival data were extracted from aprospectively maintained institutional database. Clinicopatho-logic data, including gender, age, hepatitis virus markers, theChild–Pugh classification, primary tumor characteristics, andtreatment-related variables, were collected. Tumor characteristicsand resection margins were ascertained based on a final patho-logical assessment. The tumor stage was assessed by the seventhedition of the American Joint Committee on Cancer (AJCC)classification (24). Post-surgical adjuvant chemotherapy wasadministered using gemcitabine (from 2006) and/or tegafur-gimeracil-oteracil potassium (S-1; from 2007) for tumors classi-fied as stage II to IV according to the AJCC classification (from2006). Recurrence was diagnosed based on imaging studies andtumor markers.

Written informed consent for the use of serum and resectedtissue samples was obtained from all patients in accordance withthe Declaration of Helsinki, and this study was approved by theinstitutional review committee of the Graduate School of Med-icine, Kyoto University (approval code: R803-1) and HyogoCollege of Medicine (approval code: 201807-010).

Measurement of serum NRDC levelsSerum was isolated from whole blood and stored at �80�C

until analyzed. Toquantify serumNRDC levels, an enzyme-linkedimmunosorbent assay (ELISA) was performed according to apreviously describedmethod (25). Briefly, to establish a sandwichELISA system, all combinations of the 7 monoclonal antibodiesfor NRDC were tested, and the optimum combination of clone#231 for coating and #304 for detection was selected. An auto-mated analyzer for the chemiluminescent enzyme immunoassay,SphereLight 180 (Olympus, Tokyo, Japan), was utilized to mea-sure serumNRDC levels according to themanufacturer's protocol.

Cell culture and preparation of condition mediumThe human ICC cell lines HuCCT-1 andHuH28 were provided

by the Japanese Collection of Research Bioresources Cell Bank(Osaka, Japan) and SSP-25 by the RIKEN Bioresource Center(Tsukuba, Japan). ICC cells were grown in RPMI1640 mediasupplementedwith 10%FBS and1%penicillin and streptomycin.The human colorectal cancer cell line HCT116 was provided bythe ATCC and 293 T cells by the RIKEN Bioresource Center. Thesecellswere growth inDMEMmediumsupplementedwith 10%FBSand antibiotics. Cells were cultured at 37�C under 5% CO2 and95% relative humidity. These cells were incubated in serum-freemedium for 24 hours before the initiation of experiments. Toprepare condition medium (CM), confluent monolayers of ICCcells were incubated in RPMI1640 medium supplemented with0.1% BSA for 48 hours. Cultured media were collected andcentrifuged at 3,000 rpm at 4�C for 10 minutes, and the super-natant was harvested.

Gene knockdown of NRDCThe gene knockdown (KD) of NRDC in ICC cells was per-

formed by the transfection of lentiviral vectors expressing miRNA

Translational Relevance

Few studies have investigated prognostic biomarkers inintrahepatic cholangiocarcinoma (ICC) that may contributeto establishing adjuvant strategies. Nardilysin (NRDC), ametalloendopeptidase of the M16 family, has been suggestedto play important roles in inflammation and several cancertypes. The present results revealed (i) a correlation betweenserum NRDC levels and cancerous tissue mRNA levels ofNRDC, (ii) that preoperative serum NRDC levels were asso-ciated with survival and recurrence, and (iii) strong correla-tions between the mRNA levels of NRDC and EMT-inducingtranscription factors, ZEB1 and SNAI1, in ICC cell lines andcancerous tissue. Based on the potential relationship betweenNRDC and EMT, the preoperative evaluation of serum NRDChas potential as a clinical tool for predicting the postoperativeoutcomes of patients with ICC undergoing surgical resection.

Yoh et al.

Clin Cancer Res; 25(2) January 15, 2019 Clinical Cancer Research620

on March 18, 2020. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 23, 2018; DOI: 10.1158/1078-0432.CCR-18-0124

as previously reported (19, 23). The sequences targeting theNRDC gene were as follows: NRDC-KD1: 50-CTGATGCAAACA-GAAAGGAAA-30; NRDC-KD2: 50-GAGAAATGGTTTGGAACT-CAA-30. We used a control vector that contains a nontargetingsequence for any vertebrate gene as a negative control (NC). Theefficiency of the gene KD of NRDC was evaluated by Westernblotting and qRT-PCR analyses.

Cell proliferation assayCell proliferation was examined by a tetrazolium salt-based

proliferation assay (WST-8 assay) using Cell Counting Kit-8(CCK8; Dojindo Laboratories, Tokyo, Japan). Briefly, 96-wellplates were seeded with cells at a density of 5,000 to 10,000 cellsper well (HuCCT-1: 1.0� 104 cells, SSP-25: 5� 103 cell per well)and cultured for 72 hours. Ten microliters of CCK8 solution wasadded to each well and incubated at 37�C for 2 hours. The cellviability ratio was defined as the ratio of absorbance at 450 nm ofKD and NC cells. All assays were performed in quadruplicate andrepeated at least three times.

Migration assayThe migration of ICC cells was examined using a wound

healing assay and chemotaxis assay as previously reported(26). In the wound healing assay, confluent ICC cells in 24-wellplates were wounded using a sterile 200-mL pipette tip. Cells werethen grown for an additional 12 and 24 hours with serum-freemedia.Wound closure was observedwith an invertedmicroscope(Keyence) at 40�magnification. The cell migration distance wasmeasured using ImageJ software (NIH) and compared withbaseline measurements. All assays were repeated at least threetimes.

A chemotaxis assay was performed in 8-mm pore Transwellchambers (Corning Costar). Briefly, ICC cells (HuCCT-1: 5.0 �104 cells, SSP-25: 2.0� 104 cells) in serum-freemediawere placedin the upper chamber. The lower chamber was filled with 750mLof RPMI1640 supplemented with 10% FBS as a chemoattractant.After the incubation at 37�C for 24 hours (HuCCT-1) and 12hours (SSP-25), cells were fixed with 4% PFA at 20 minutes andstained with hematoxylin and eosin. Cells that migrated throughthe pores to the lower surface of the filter were counted under amicroscope. A total of three random fields were counted induplicate assays.

Chemosensitivity assayNCandNRDC-KD ICC cells (HuCCT-1: 1.0�104 cells, SSP-25:

5.0� 103 cells per well) were seeded on 96-well plates in normalgrowthmedia. Twelve hours after seeding, gemcitabinewas addedat the indicated concentrations. After a 72-hour treatment, cellviability was examined using the WST-8 assay. The cell viabilityratio was defined as absorbance at 450 nm of the sample dividedby the absorbance of the control for each cell. All assays wereperformed in quadruplicate and repeated at least three times.

Subcutaneous tumor xenograftSeven- to nine-week-old male nude mice were used as recipi-

ents for xenotransplantation. A total of 1.0�106ofHuCCT-1-NC,-KD1, or –KD2 cells were suspended in 100 mL PBS and subcu-taneously injected into the right (NC) and left (KD1or KD2)flankof nude mice (n ¼ 7, of each). Tumor sizes were measured everyweek after the inoculation. Tumor volumes were calculated usingthe formula: length� (width)2 � 0.52. Animal experiments were

performed in accordance with the protocols approved by theInstitutional Animal Care and Use Committee of Kyoto Univer-sity. Animal experiments were performed in accordance with theprotocols approved by the Institutional Animal Care and UseCommittee of Kyoto University.

IHCFour-micrometer-thick sections were incubated with an anti-

humanNRDCmousemonoclonal antibody (#102, established inour laboratory) at 4�C overnight. Envision polymer (DAKO),which is a horseradish peroxidase-labeled polymer conjugatedwith an anti-mouse IgG antibody, was used as a secondaryantibody according to the manufacturer's protocol. Color wasdeveloped with diaminobenzidine solution (DAKO), followedby counterstaining with hematoxylin.

qRT-PCRTotal RNAwas isolated fromHuCCT-1 and SSP-25 cells and 43

fresh frozen cancer tissues using TRIzol reagent (Thermo FisherScientific) and cleaned using the DNase set and RNeasy Mini Kit(Qiagen). cDNA generated by reverse transcription with theOmniscript RT Kit (Qiagen) was subjected to a RT-PCR assayusing the Step One Plus Real-Time PCR System (Applied Biosys-tems) and Fast SYBR Green Master Mix (Applied Biosystems)according to themanufacturer's protocols. The average expressionlevels of the target genes were normalized against GAPDH usingthe 2�DDCt method. The ICC cell line (SSP-25) was used as areference sample (i.e., 2�DDCt value of SSP-25 cells ¼ 1) whenanalyzing human samples. The primers used in this experimentare listed in Supplementary Table S1.

Western blottingTwenty micrograms of cell lysates were electrophoresed on a

10% SDS-PAGE (SDS fromWako; acrylamide from Bio-Rad) andtransferred to a nitrocellulose membrane (GE Healthcare). Theprimary antibodies are listed in Supplementary Table S2. The blotwas observed with EZ-Capture II (ATTO) visualized by ECL Prime(GE Healthcare). b-Actin was used as the loading control. Whenanalyzing protein levels in CM, Ponceau 3R Stain Solution(Wako) was used as the loading control. In figures, representativeimageswere selected fromat least three independent experiments,in which similar results were obtained.

Statistical analysisData from human clinical samples were expressed as median

values (range). Regarding continuous variables, data wereexpressed as a median (range), and compared using the Mann–Whitney U test. Categorical variables were expressed as a number(%) and compared using the x2 test or Fisher exact test whereappropriate. A ROC analysis was performed to evaluate thediscriminatory power of predictors. The area under the ROC curve(AUC)was calculated. DeLong test was used to compare the AUC.The cut-off values, sensitivity, and specificity of serum NRDCvariables were assessed by the Youden index. Other cut-off valueswere evaluated by clinically relevant values (5, 6). Relationshipsbetween continuous variables were evaluated by the Spearmancorrelation test (the value was expressed as r). To label thestrength of the relationship, for absolute values of r, 0 to 0.19is regarded as very weak, 0.2 to 0.39 as weak, 0.40 to 0.59 asmoderate, 0.6 to 0.79 as strong, and 0.8 to 1 as very strong (27).OS was calculated from the day of surgical resection to the date of

Nardilysin Is a Novel Prognostic Biomarker for ICC

www.aacrjournals.org Clin Cancer Res; 25(2) January 15, 2019 621

on March 18, 2020. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 23, 2018; DOI: 10.1158/1078-0432.CCR-18-0124

deathor endof the follow-upperiod,whereas disease-free survival(DFS) was calculated using the date of death or recurrence as thetime of the terminal event according to the Kaplan–Meier meth-od. Survival was compared using a generalized Wilcoxon test. Amultivariate analysis was performed by Cox's regression (Step-wise backwardmodel) for variables identified as significant in theunivariate analysis. When collinearity was encountered, a choicewasmade based on the P value and clinical reasoning. All analyseswere two-sided, and differences were considered significant whenP < 0.05. Statistical analyses were performed using JMP ver. 12.1software.

Data from in vitro and in vivo experiments were analyzed usingthe Student t test and expressed as means � SD. All statisticalanalyses were performed using JMP ver. 12.1 software (SASInstitute).

ResultsPrognostic impact of serum NRDC in patients with ICC aftersurgical resection

In the primary cohort, serumNRDC levels weremeasured in 79preoperative patients with ICC who consecutively underwentsurgical resection at Kyoto University Hospital between 2006 and2013. In this study population, most patients had advanced stage

disease [AJCC stage III/IV, n ¼ 52 (65.8%); ref. 24]. R0 resectionwas performed on 63 patients (79.7%) (Supplementary TableS3). SerumNRDC levels were significantly higher in patients withICC than inhealthy controls (HC) [median; 1627.4pg/mL (range:351.9–11318.7) vs. 539.8 pg/mL (range: 5.9–1184.2), P <0.001; Fig. 1A].

We then examined whether serumNRDC levels had prognosticvalue in patients with ICC after surgical resection. The medianfollow-up period was 41.6 months (range: 0.1–127.5 months).ThemedianOS timewas 47.6months, with 3- and 5-yearOS ratesof 57.0% and 42.3%, respectively. The cut-off value for serumNRDC was selected as 1627.4 pg/mL based on the highestaccuracy in relation to an outcome (death) using an ROC analysis(AUC values of 0.688). Patients with ICC were divided intotwo groups according to this cut-off value: high serum NRDC(n ¼ 40) and low serum NRDC groups (n ¼ 39). High serumNRDC levels correlated with the presence of multiple tumors(P¼ 0.006; Table 1). OS was significantly shorter in patients withhigh serum NRDC levels than in those with low serum NRDClevels (P ¼ 0.002, Fig. 1B). The median OS and 3- and 5-yearsurvival rates of high and low serum NRDC groups were 31.0months versus 85.6 months, 42.5% versus 71.8%, and 22.3%versus 62.6%, respectively. Moreover, DFS was significantlyshorter in patients with high serum NRDC levels than in those

Figure 1.

Prognostic impact of serum NRDC levels in patients with ICC after surgical resection. A, Comparison of serum NRDC levels in patients with HC (n ¼ 112) andICC in the primary (n ¼ 79) and validation (n ¼ 19) cohorts. B and C, Kaplan–Meier analyses for OS (B) and DFS (C) in 79 patients with ICC in the primary cohortaccording to serum NRDC levels. D and E, Kaplan–Meier analyses for OS (B) and DFS (C) in 19 patients with ICC in the external validation cohort accordingto serum NRDC levels.

Yoh et al.

Clin Cancer Res; 25(2) January 15, 2019 Clinical Cancer Research622

on March 18, 2020. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 23, 2018; DOI: 10.1158/1078-0432.CCR-18-0124

with low serum NRDC levels (P ¼ 0.002, Fig. 1C). The medianDFS and 1- and 3-year DFS rates of the high and low serumNRDCgroupswere 9.4months versus 23.0months, 42.5%versus 71.8%,and 15.0% versus 43.6%, respectively.

To confirm the prognostic relevance of serum NRDC as abiomarker, we performed univariate and multivariate analysesby Cox's hazardmodel using six potential confounders (Table 2).Among knownprognostic factors (4–6), lymphnode (LN)metas-tasis (P ¼ 0.001), vascular invasion (P ¼ 0.033), and multipletumors (P < 0.001) were poor prognostic factors for OS. LNmetastasis (P ¼ 0.042), multiple tumors (P < 0.001), and poordifferentiation (P ¼ 0.026) were poor prognostic factors for DFS.After the multivariate analysis, serum NRDC levels were main-

tained as independent prognostic factors for OS (P ¼ 0.019) andDFS (P ¼ 0.009).

To validate the prognostic impact of NRDC in patients withICC, we analyzed serum NRDC levels in an external independentcohort of 19 patients with ICC who underwent surgical resectionat Hyogo College of Medicine between 2009 and 2015 (Supple-mentary Table S4). The median follow-up period was 25.4months (range: 3.8–84.3 months) and the median OS was27.9 months, with 3- and 5-year OS rates of 35.1% and 23.4%,respectively. Serum NRDC levels were significantly higher in thevalidation cohort (median; 1330, range: 528–14597 pg/mL) thanin HC, but were not significantly different from those in theprimary cohort (Fig. 1A). OS and DFS were significantly stratifiedaccording to cut-off values (1295 pg/mL: AUC values of 0.729),which were selected independently in the validation cohort by anROCanalysis. The clinical backgrounds of the high and low serumNRDC groups and Kaplan–Meier curves for OS and DFS areshown in accordance with this independent cut-off value (Sup-plementary Table S5; Fig. 1D and E). These results reinforced thepredictive value of serum NRDC in postoperative patientswith ICC.

Comparison of serum NRDC levels with other tumor markersCarcinoembryonic antigen (CEA) and carbohydrate antigen

19-9 (CA19-9) are serum tumor markers that are commonlymeasured in patients with ICC (3, 5, 6). A correlation was notobserved between serum NRDC and CEA or CA19-9 levels (Sup-plementary Fig. S1). A Kaplan–Meier curve analysis revealed thatthe elevated serum CEA (�5 ng/mL) and CA19-9 (�37 IU/ml)levels in patients with ICC correlated with a poor prognosis in theprimary cohort (Supplementary Fig. S2). An ROC analysis inrelation to the outcome (death) showed that the prognosticability of preoperative serum NRDC levels (AUC: 0.688) wasequivalent to that of serum CEA (0.569) and CA19-9 (0.671)levels (Supplementary Fig. S3). We additionally analyzed theprognostic value of the combination of 3 markers (Supplemen-tary Table S6) and found that the combination of NRDC andCA19-9 provided the highest AUC value (0.756), which had asignificantly stronger prognostic value than CA19-9 alone (Sup-plementary Fig. S4).

Table 1. Clinicopathologic characteristics and surgical outcomes according toserum NRDC levels

NRDC low NRDC highVariables n ¼ 39 n ¼ 40 P value

Clinical factorsAge (years) 69 (32–84) 67.5 (37–83) 0.677Gender (male) 23 (58.9%) 26 (65.0%) 0.581Hepatitis Ba 3 (7.7%) 1 (2.5%) 0.356Hepatitis Ca 7 (17.9%) 3 (7.5%) 0.193CA19-9 (IU/mL) 35.2 (0–5461.1) 75.4 (0–1788.0) 0.111CEA (ng/mL) 2.1 (0.4–23.7) 2.95 (0–116.6) 0.310

Histologic factorsTumor diameter (cm) 4.1 (1.0–9.0) 4.6 (1.0–14.0) 0.157Multiple tumors 3 (7.7%) 13 (32.5%) 0.006b

LN metastasis 9 (23.1%) 12 (30.0%) 0.486Poor differentiationa 2 (5.1%) 7 (17.5%) 0.154Vascular invasion 20 (51.3%) 25 (62.5%) 0.314Biliary invasion 18 (46.2%) 17 (42.5%) 0.744AJCC T3/T4 22 (56.4%) 21 (52.5%) 0.727AJCC stage III/IV 25 (64.1%) 27 (67.5%) 0.750

Surgical outcomesR0 resection 32 (82.1%) 31 (77.5%) 0.615Major hepatectomy(�3 segments)

34 (87.2%) 35 (87.5%) 1.000

Adjuvant chemotherapy 16 (41.0%) 23 (57.5%) 0.143Morbidity 17 (43.6%) 18 (45.0%) 0.900Mortality (<30 days)a 0 (0%) 1 (2.5%) 1.000

Abbreviation: R0, no residual tumor.aFisher's exact test and the c2 test were used for all other categorical variables.bSignificant difference P < 0.05.

Table 2. Univariate and multivariate analyses of factors predicting postoperative prognosis

Univariate analysisMultivariate analysis

Variables P valueHazard ratio (95%confidence interval) P value

SurvivalHigh serum NRDC (vs. low) 0.002a 2.047 (1.126–3.821) 0.019a

LN metastasis (vs. N0/Nx) 0.001a 1.942 (1.032–3.584) 0.040a

Vascular invasion (vs. negative) 0.033a — —

Multiple tumors (vs. solitary) <0.001a 2.169 (1.066–4.226) 0.033a

Poor differentiation (vs. well/moderate) 0.176 — —

Tumor size �5 cm (vs. <5 cm) 0.873 — —

RecurrenceHigh serum NRDC (vs. low) <0.001a 2.017 (1.189–3.470) 0.009a

LN metastasis (versus N0/Nx) 0.042a — —

Vascular invasion (vs. negative) 0.619 — —

Multiple tumors (vs. solitary) <0.001a 2.397 (1.279–4.289) 0.008a

Poor differentiation (vs. well/moderate) 0.026a — —

Tumor size �5 cm (vs. <5 cm) 0.617 — —

Abbreviations: N0, negative for nodal metastasis; Nx, nodal metastasis status undetermined.aSignificant difference P < 0.05.

Nardilysin Is a Novel Prognostic Biomarker for ICC

www.aacrjournals.org Clin Cancer Res; 25(2) January 15, 2019 623

on March 18, 2020. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 23, 2018; DOI: 10.1158/1078-0432.CCR-18-0124

NRDCmRNA levels in resected cancerous tissue correlatedwiththe prognosis of patients with ICC after surgical resection

The upregulated expression of NRDC in cancer tissue has beenreported to have a poor prognostic impact in several cancer types(19, 22, 23). In addition to serum NRDC levels, we examinedNRDC expression in surgical specimens resected from patientswith ICC. An IHC analysis showed the membranous, cytosolic,and nuclear expression of NRDC in the cancer epithelium(Fig. 2A), which was consistent with previous findings (12–20). We quantified the mRNA expression levels of NRDC incancerous tissues from 43 patients with ICC and analyzed therelationship with matched serum NRDC levels. As expected, acorrelation was observed betweenmRNA and serumNRDC levels(r ¼ 0.413, P ¼ 0.006, Fig. 2B), suggesting that ICC tumors are apotential source of serum NRDC.

We also assessed the prognostic value ofNRDCmRNA levels intumors. The cut-off values of mRNA levels were selected based onthe highest accuracy in relation to the outcome of death (cut-off2�DDCt value: 1.708, AUC value of 0.634) and were used to dividepatients into two groups: low NRDC mRNA (n ¼ 11) and high

NRDCmRNA (n¼ 32) groups. Using these cut-off values (patientcharacteristics are shown in Supplementary Table S7), OS andtime to recurrence were significantly shorter in patients with highNRDC mRNA levels than in those with low NRDC mRNA levels(Fig. 2C and D).

Effects of the gene KDofNRDC on the proliferation, migration,and chemosensitivity of ICC cells

We examined NRDC expression in three ICC cell lines andfound that all cell lines strongly expressed NRDC (Supple-mentary Fig. S5). NRDC protein levels in ICC cells weresimilar to those in colon cancer HCT116 cells (28) and higherthan those in 293T cells (13). Therefore, to examine thepathophysiologic roles of NRDC in ICC cells, we performedgene KD experiments using two ICC cell lines with highmalignant potential: HuCCT-1 (obtained from metastaticascites; ref. 29) and SSP-25 (spindle cell-type with mesenchy-mal features; ref. 30). The sufficient silencing of NRDCexpression in both cells was confirmed by Western blottingand qRT-PCR analyses (Fig. 3A). Secreted NRDC in CM was

Figure 2.

NRDC mRNA levels in resected cancerous tissue correlate with the prognosis of patients with ICC after surgical resection. A, IHC staining of NRDC in surgicalspecimens of ICC. The scale bar represents 100 mm. B, Relationship between serum NRDC and NRDC mRNA expression in cancer tissue. C and D, Kaplan–Meieranalyses for OS (C) and DFS (D) in 43 patients with ICC according to NRDC mRNA levels in resected cancerous tissue.

Yoh et al.

Clin Cancer Res; 25(2) January 15, 2019 Clinical Cancer Research624

on March 18, 2020. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 23, 2018; DOI: 10.1158/1078-0432.CCR-18-0124

also clearly decreased by the gene KD (Fig. 3B). We initiallyevaluated cell proliferation using the tetrazolium salt assayand found that the proliferation of HuCCT-1 and SSP-25 cellswas significantly decreased by the gene KD of NRDC (Fig. 3C).To further assess the impact of NRDC on cell growth, controland NRDC-knocked down HuCCT-1 cells were used in tumorxenograft experiments. Tumor growth after subcutaneousimplantation was markedly less in cells with the gene KD ofNRDC than in control cells (Fig. 3D). Because we confirmedthe similar effects of two different siRNAs on in vitro and invivo cell proliferation, one (KD2) was selected for furtherexperiments. In two different assays (wound healing assayand chemotaxis assay), HuCCT-1 cells in which NRDC wasknocked down (NRDC-KD2) showed significantly less migra-tory potential than control cells (NC; Fig. 3E and F). Thesimilar inhibitory effect of NRDC KD on cell migrationwas also confirmed in SSP-25 cells (Supplementary Fig.S6). Furthermore, the influence of NRDC levels on the che-mosensitivity of HuCCT-1 and SSP-25 cells to gemcitabinewas investigated. In both cell lines, NRDC-KD2 cells exhibitedgreater sensitivity to gemcitabine than NC cells (Fig. 3G;Supplementary Fig. S6).

Relationship between mRNA levels of NRDC and epithelial–mesenchymal transition-related genes in tumor tissues fromICC patients

To gain mechanical insights into the tumor-promoting andchemoresistant potential of NRDC in ICC, we investigated geneprofiles in ICC cell lines. An analysis by qRT-PCR revealed thatseveral epithelial–mesenchymal transition (EMT)- and cancerstem cell (CSC)-related genes were downregulated by the geneKD ofNRDC in HuCCT-1 (Fig. 4A) and SSP-25 cells (Fig. 4B). Forexample, the mRNA levels of vimentin (VIM), a marker of mes-enchymal cells, zincfinger E-box-binding homeobox 1 (ZEB1), anEMT-inducing transcription factor (EMT-TF), sex-determiningregion Y-box 2 (SOX2), a marker of CSC, and hypoxia-induciblefactor-1a (HIF1A), a trigger of the EMT pathway, were down-regulated by the gene KD of NRDC. Other EMT-TF, SNAI1 andTWIST1, were also significantly decreased byNRDCKD in SSP-25cells (Fig. 4B). The Western blot analysis revealed that the proteinexpression levels of these genes were markedly reduced (Fig. 4Aand B), whereas E-cadherin (CDH1), a marker of epithelial cells,was increased by NRDC KD in HuCCT-1 cells (Fig. 4A).

We then investigatedwhether relationships existed between themRNA levels of NRDC and EMT-related genes in tumor tissues

Figure 3.

Effects of thegene knockdownofNRDCon the proliferation,migration, and chemosensitivity of ICC cells.A,StableNRDC-KD ICC cellswere established. qRT-PCRandWestern blotting were then performed to confirm the expression of NRDC. B, NRDC secreted in CM was decreased by the gene KD of NRDC in HuCCT-1andSSP-25 cells.C andD,SilencingNRDC attenuated cell proliferation invitro and tumor growth in vivo (D).E andF,Woundhealing assays (E) and chemotaxis assays(F) demonstrated that the migratory ability of HuCCT-1 cells was decreased by the gene KD of NRDC. G, Effects of gemcitabine concentrations on the viability ofHuCCT-1 cells. Silencing NRDC increased chemosensitivity. Data represent the mean � SD of at least three independent experiments; � , P < 0.05; ��, P < 0.01.

Nardilysin Is a Novel Prognostic Biomarker for ICC

www.aacrjournals.org Clin Cancer Res; 25(2) January 15, 2019 625

on March 18, 2020. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 23, 2018; DOI: 10.1158/1078-0432.CCR-18-0124

from patients with ICC. Notably, the expression levels of NRDCpositively and strongly correlated with two EMT-TFs, ZEB1(r ¼ 0.679, P < 0.001) and SNAI1 (r ¼ 0.647, P < 0.001), andHIF1A expression (r ¼ 0.721, P < 0.001; Fig. 4C). NRDC mRNAlevels correlated with SOX2 expression, but not with TWIST1expression or the VIM/CDH1 ratio (Supplementary Fig. S7). Wethen examined the relationship between serumNRDCandmRNAlevels of EMT-related genes in tumor tissues. Serum NRDC pos-itively correlated with SNAI1 (r ¼ 0.327, P ¼ 0.032) and HIF1A(r ¼ 0.301, P ¼ 0.050) mRNA levels in surgical specimens. ZEB1mRNA levels were also positively associated with serum NRDClevels (r¼ 0.287,P¼0.062; Fig. 4D). Therefore, serumNRDChaspotential as a surrogate marker for EMT in the tumors of patientswith ICC.

DiscussionThis study highlights the clinical implications of preoperative

serum NRDC measurements in patients with ICC, which maycontribute to the identification of patients with a poor prognosis.OS and DFS were significantly stratified by serumNRDC levels in

the primary (development) and validation cohorts. The expressionlevels of NRDC in sera and cancerous tissues were significantlylinked; therefore, an evaluation of pathophysiologic features inresected tissues provided an insight into the clinical significance ofserum NRDC. NRDC mRNA levels correlated with EMT-relatedgenes in primary tumors. Moreover, direct correlations wereobserved between serum NRDC and mRNA levels of majorEMT-TF, suggesting that serum NRDC has potential as a surrogatemarker for EMT features in primary tumors. This hypothesis wassupported by a cell analysis because the gene silencing ofNRDC inICC cell lines reduced EMT-related gene expression. Functionally,the gene KD of NRDC was accompanied by attenuated prolifera-tion/migration and increased chemosensitivity to gemcitabine.Importantly, a correlation was not noted between serum NRDCandCA19-9orCEA, currently availableprognosticmarkers for ICC.Furthermore, the combination of NRDC and CA19-9 had strongerprognostic value than eithermarker analyzed individually, suggest-ing that serum NRDC is a unique prognostic marker for patientswith ICC after surgical resection.

Based on the prognostic impact of serum NRDC levels inpatients with ICC after surgical resection, the source of NRDC in

Figure 4.

Relationship between NRDC and EMT-related genes in ICC. A and B, qRT-PCR and Western blot analyses were performed to assess the mRNA and protein levels ofEMT-related genes in NC versus NRDC KD2 in HuCCT-1 cells (A) and SSP-25 cells (B). Data represent the mean � SD of at least three independent experiments;� , P < 0.05; �� , P < 0.01. In Western blotting, representative images were selected from three independent experiments, in which similar results were obtained.C, Correlation analysis between the mRNA levels of NRDC and EMT-related genes (ZEB1, SNAI1, and HIF1A) in surgical specimens from patients with ICC.D, Correlation analysis between the serum NRDC and mRNA levels of EMT-related genes (ZEB1, SNAI1, and HIF1A) in surgical specimens from patients with ICC.

Yoh et al.

Clin Cancer Res; 25(2) January 15, 2019 Clinical Cancer Research626

on March 18, 2020. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 23, 2018; DOI: 10.1158/1078-0432.CCR-18-0124

serum needs to be identified. According to the following findingsof clinical studies: (i) serumNRDC levels were significantly higherin patients with ICC than in HC, (ii) a correlation between serumNRDC and NRDC expression levels in resected cancer tissue, wespeculate that a potential source of serum NRDCmay be the ICCtumor itself. Experiments using ICC cell lines also revealed thatNRDC is secreted into CM, the amount of which correlated withthe intracellular expression level of NRDC. Our previous analysisof patients with HCC also suggested that serum NRDC reflectedthe amount of NRDC in cancer tissues (23). Another possiblesource of serum NRDC is inflammatory cells adjacent to tumors.We recently reported that NRDC levels were increased in thesynovial fluid of patients with rheumatoid arthritis (31). NRDCin macrophages regulates arthritis via the control of TNFa secre-tionbecause themacrophage-specific deletionofNRDCmarkedlyameliorated arthritis (31). We also demonstrated that inflamma-tory cells infiltrating the infarcted myocardium strongly expressNRDC (25). However, clinical data from this study do notstrongly support this hypothesis because preoperative C-reactiveprotein levels in ICC patients did not correlate with serumNRDClevels (data not shown). In any case, serial measurements ofserum NRDC levels after surgical resection are needed to identifythe real source, which will also further clarify the relationshipbetween serum NRDC levels and tumor recurrence.

The significant link between patient prognosis and NRDCmRNA expression levels in cancerous tissues prompted us toexamine the pathophysiologic functions of NRDC in ICC cells.The geneKDofNRDC in twodifferent ICC cell lines attenuated cellproliferation andmigration, indicating important roles for NRDCin ICC progression. Moreover, the gene KD ofNRDC in HuCCT-1cells resulted in increased sensitivity to gemcitabine. Because thepoor prognosis of patients with ICC is mainly attributed to itshighly metastatic characteristics, EMT in the pathogenesis of ICChas been attracting increasing attention from researchers (32).Accumulated evidence also suggests a relationship between che-moresistance and the acquisition of the EMT phenotype and/orexistenceofCSCwithin the tumor (33).AmongEMT-relatedgenes,the significance of EMT-TF, such as ZEB1, SNAI1, and TWIST1, hasbeen emphasized (32–36). An analysis of surgical specimens alsodemonstrated that the strong expression of EMT-TF in resectedtissue is associated with the poor prognosis of patients with ICCafter surgical resection (32). We herein showed that the gene KDofNRDCwas accompanied bymarked reductions in several EMT-related genes. Moreover, strong correlations between NRDC andEMT-related genes (ZEB1, SNAI1, and HIF1A) were recapitulatedin resected cancerous tissue from patients with ICC. Park andcolleagues very recently demonstrated that NRDC regulatedEMT-related genes, including SNAI1, in colon cancer cells (37);NRDCwas responsible for the insulin-like growth factor-1 (IGF-1)-induced regulation of EMT-related genes. Together with our in vitroand in vivo data from ICC cells, NRDC appears to play importantand general roles in the regulation of EMT.

Several signaling pathways triggered, for example, by TGFb andEGF as well as hypoxia may induce EMT (35, 36). These signals

lead to the activation of EMT-TF, such as ZEB1, SNAI1, andTWIST1, which directly or indirectly control key EMT-relatedgenes. Because of the multiple functions of NRDC, there areseveral possibilities by which NRDC is involved in EMT. Extra-cellular NRDC may activate EGF or TNF receptor signaling byenhancing the ectodomain shedding of EGF receptor ligands orTNFa, respectively (12, 13, 18). Furthermore, nuclear NRDCmayregulate the transcription of EMT-related genes, including EMT-TF(38, 39). Although the underlying mechanisms have not yet beenelucidated in detail, the present results suggest that elevatedNRDC levels in cancer tissue are associated with EMT programs.Based on the positive correlation between serum and tumorNRDC levels, it may be possible to assess the level of EMT featuresin primary tumors by measuring serum NRDCs.

In conclusion, this is the first study to demonstrate that serumNRDC has potential as a novel prognostic biomarker for ICC,which may reflect EMT features in primary tumor regions. Wepropose that the preoperative evaluation of serum NRDC is apotential clinical tool for predicting tumor recurrence in and theoverall prognosis of patients after curative-intent surgery.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: T. Yoh, E. Hatano, S. Seo, E. NishiDevelopment of methodology: T. Yoh, E. Hatano, Y. Kasai, R. Yamaguchi,M. Kurokawa, E. NishiAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): T. Yoh, E. Hatano, H. Fuji, H. Sueoka, K. Iwaisako,K. Taura, R. Yamaguchi, J. FujimotoAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): T. Yoh, E. Hatano, Y. Kasai, M. Ohno, K. Iwaisako,K. Taura, E. NishiWriting, review, and/or revisionof themanuscript: T. Yoh, E.Hatano, Y. Kasai,K. Taura, E. NishiAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): E. Hatano, K. Nishi, M. Ohno, J. Fujimoto,T. KimuraStudy supervision: E. Hatano, K. Toriguchi, K. Taura, T. Kimura, S. Uemoto

AcknowledgmentsThe authors gratefully acknowledge the patients who consented to blood

collection for this study. The detection of serum NRDC was supported byYoshiyuki Amano (Sanyo Chemical Industries). The authors thank TakayukiKawai, Takahiro Nishio, Masayuki Okuno, SeidaiWada, Asahi Sato, YoshinobuIkeno, Yusuke Morita, Shintaro Matsuda, and Hiromi Iwai (Kyoto University)for their excellent help. This study was financially supported by Grants-in-AidKAKENHI (17H04048, 17K09575, 17K16147, and 18H04694) and AMED(JP17cm0106608). It was also supported by the Takeda Science Foundation,SENSHIN Medical Research Foundation, and Suzuken Memorial Foundation.

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received January 23, 2018; revised July 21, 2018; accepted October 19, 2018;published first October 23, 2018.

References1. Aljiffry M, Abdulelah A, Walsh M, Peltekian K, Alwayn I, Molinari M.

Evidence-based approach to cholangiocarcinoma: a systematic review ofthe current literature. J Am Coll Surg 2009;208:134–47.

2. Kudo M, Izumi N, Ichida T, Ku Y, Kokudo N, Sakamoto M, et al. Report ofthe 19th follow-up survey of primary liver cancer in Japan. Hepatol Res2016;46:372–90.

Nardilysin Is a Novel Prognostic Biomarker for ICC

www.aacrjournals.org Clin Cancer Res; 25(2) January 15, 2019 627

on March 18, 2020. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 23, 2018; DOI: 10.1158/1078-0432.CCR-18-0124

3. Bridgewater J, Galle PR, Khan SA, Llovet JM, Park JW, Patel T, et al.Guidelines for the diagnosis and management of intrahepatic cholangio-carcinoma. J Hepatol 2014;60:1268–89.

4. Maithel SK, Gamblin TC, Kamel I, Corona-Villalobos CP, Thomas M,Pawlik TM. Multidisciplinary approaches to intrahepatic cholangiocarci-noma. Cancer 2013;119:3929–42.

5. Mavros MN, Economopoulos KP, Alexiou VG, Pawlik TM. Treatment andprognosis for patients with intrahepatic cholangiocarcinoma: systematicreview and meta-analysis. JAMA Surg 2014;149:565–74.

6. Yoh T, Hatano E, Yamanaka K, Nishio T, Seo S, Taura K, et al. Is surgicalresection justified for advanced intrahepatic cholangiocarcinoma? LiverCancer 2016;5:280–9.

7. Yoh T, Hatano E, Nishio T, Seo S, Taura K, Yasuchika K, et al. Significantimprovement in outcomes of patients with intrahepatic cholangiocarci-noma after surgery. World J Surg 2016;9:2229–36.

8. Miura JT, Johnston FM, Tsai S, George B, Thomas J, Eastwood D, et al.Chemotherapy for surgically resected intrahepatic cholangiocarcinoma.Ann Surg Oncol 2015;22:3716–23

9. Yoh T, Seo S, Hatano E, Taura K, Fuji H, Ikeno Y, et al. A novel biomarker-based preoperative prognostic grading system for predicting survival aftersurgery for intrahepatic cholangiocarcinoma. Ann Surg Oncol 2017;24:1351–7.

10. Yamashita S, Passot G, Aloia TA, Chun YS, Javle M, Lee JE, et al. Prognosticvalue of carbohydrate antigen 19-9 in patients undergoing resection ofbiliary tract cancer. Br J Surg 2017;104:267–77.

11. Bergquist JR, Ivanics T, Storlie CB, Groeschl RT, Tee MC, Habermann EB,et al. Implications of CA19-9 elevation for survival, staging, and treatmentsequencing in intrahepatic cholangiocarcinoma: a national cohort analy-sis. J Surg Oncol 2016;114:475–82.

12. Nishi E, Prat A, Hospital V, Elenius K, Klagsbrun M. N-arginine dibasicconvertase is a specific receptor for heparin-binding EGF-like growth factorthat mediates cell migration. EMBO J 2001;20:3342–50.

13. Nishi E, Hiraoka Y, Yoshida K, Okawa K, Kita T. Nardilysin enhancesectodomain shedding of heparin-binding epidermal growth factor-likegrowth factor through activationof tumor necrosis factor-alpha-convertingenzyme. J Biol Chem 2006;281:31164–72.

14. Ohno M, Hiraoka Y, Lichtenthaler SF, Nishi K, Saijo S, Matsuoka T, et al.Nardilysin prevents amyloid plaque formation by enhancing a-secretaseactivity in an Alzheimer's disease mouse model. Neurobiol Aging2014;35:213–22.

15. Hiraoka Y, Matsuoka T, OhnoM, Nakamura K, Saijo S, Matsumura S, et al.Critical roles of nardilysin in the maintenance of body temperaturehomoeostasis. Nat Commun 2014;5:3224.

16. Nishi K, Sato Y, OhnoM,Hiraoka Y, Saijo S, Sakamoto J, et al. Nardilysin isrequired for maintaining pancreatic b-cell function. Diabetes 2016;65:3015–27.

17. Ma Z, Chow KM, Yao J, Hersh LB. Nuclear shuttling of the peptidasenardilysin. Arch Biochem Biophys 2004;422:153–60.

18. Hiraoka Y, Yoshida K, Ohno M, Matsuoka T, Kita T, Nishi E. Ectodomainshedding of TNF-alpha is enhanced by nardilysin via activation of ADAMproteases. Biochem Biophys Res Common 2008;370:154–8.

19. Kanda K, KomekadoH, Sawabu T, Ishizu S,Nakanishi Y, Nakatsuji M, et al.Nardilysin and ADAM proteases promote gastric cancer cell growth byactivating intrinsic cytokine signalling via enhanced ectodomain sheddingof TNF-a. EMBO Mol Med 2012;4:396–411.

20. Li J, Chu M, Wang S, Chan D, Qi S, Wu M, et al. Identification andcharacterization of nardilysin as a novel di-methyl H3K4-bindingprotein involved in transcriptional regulation. J Biol Chem 2012;287:10089–98.

21. Choong LY, Lim SK, Chen Y, Loh MC, Toy W, Wong CY, et al. ElevatedNRD1metalloprotease expression plays a role in breast cancer growth andproliferation. Genes Chromosomes Cancer 2011;50:837–47.

22. Uraoka N, Oue N, Sakamoto N, Sentani K, Oo HZ, Naito Y, et al. NRD1,which encodes nardilysin protein, promotes esophageal cancer cell inva-sion through induction of MMP2 and MMP3 expression. Cancer Sci2014;105:134–40.

23. Kasai Y, Toriguchi K, Hatano E, Nishi K, Ohno M, Yoh T, et al. Nardilysinpromotes hepatocellular carcinoma through activation of signal transducerand activator of transcription 3. Cancer Sci 2017;108:910–7.

24. Edge SB, Compton CC. The American Joint Committee on Cancer: the 7thEdition of theAJCC cancer stagingmanual and the future of TNM.AnnSurgOncol 2010;17:1471–4.

25. Chen PM,OhnoM,Hiwasa T,Nishi K, Saijo S, Sakamoto J, et al. Nardilysinis a promising biomarker for the early diagnosis of acute coronary syn-drome. Int J Cardiol 2017;243:1–8.

26. Kawai T, Yasuchika K, Ishii T, Katayama H, Yoshitoshi EY, Ogiso S, et al.Keratin 19, a cancer stem cell marker in human hepatocellular carcinoma.Clin Cancer Res 2015;21:3081–91.

27. Evans JD. Straightforward statistics for the behavioral sciences. 1996.Brooks/Cole, Pacific Grove.

28. Kanda K, Sakamoto J, Matsumoto Y, Ikuta K, Goto N, Morita Y, et al.Nardilysin controls intestinal tumorigenesis through HDAC1/p53-dependent transcriptional regulation. JCI Insight 2018;3:e91316.

29. Miyagiwa M, Ichida T, Tokiwa T, Sato J, Sasaki H. A new human cholan-giocellular carcinoma cell line (HuCC-T1) producing carbohydrate antigen19/9 in serum-free medium. In Vitro Cell Dev Biol 1989;25:503–10

30. FukutomiM, EnjojiM, IguchiH, YokotaM, IwamotoH,NakamutaM, et al.Telomerase activity is repressed during differentiation along the hepato-cytic and biliary epithelial lineages: verification on immortal cell lines fromthe same origin. Cell Biochem Funct 2001;19:65–8.

31. Fujii T, Nishi E, Ito H, Yoshitomi H, Furu M, Okabe N, et al. Nardilysin isinvolved in autoimmune arthritis via the regulation of tumour necrosisfactor alpha secretion. RMD Open 2017;3:e000436.

32. Vaquero J, Guedj N, Clap�eron A, Nguyen Ho-Bouldoires TH, Paradis V,Fouassier L. Epithelial-mesenchymal transition in cholangiocarcinoma:from clinical evidence to regulatory networks. J Hepatol 2017;66:424–41.

33. Krebs AM,Mitschke J, Lasierra LosadaM, SchmalhoferO, BoerriesM, BuschH, et al. The EMT-activator Zeb1 is a key factor for cell plasticity andpromotes metastasis in pancreaticcancer. Nat Cell Biol 2017;19:518–29.

34. Liu Y, Lu X, Huang L,WangW, Jiang G, Dean KC, et al. Different thresholdsof ZEB1 are required for Ras-mediated tumour initiation and metastasis.Nat. Commun 2014;5:5660.

35. Polyak K, Weinberg RA. Epithelial- Transitions between epithelial andmesenchymal states: acquisition of malignant and stem cell traits. Nat RevCancer 2009;9:265–73.

36. Thiery JP, Acloque H, Huang RY, Nieto MA. Epithelial-mesenchymaltransitions in development and disease. Cell 2009;139:871–90.

37. Park GB, Kim D. Insulin-like growth factor-1 activates different catalyticsubunits p110 of PI3K in a cell-type-dependent manner to induce lipo-genesis-dependent epithelial-mesenchymal transition through the regula-tion of ADAM10 and ADAM17. Mol Cell Biochem 2018;439:199–211.

38. Eom S, Kim Y, Park D, Lee H, Lee YS, Choe J, et al. Histone deacetylase-3mediates positive feedback relationship between anaphylaxis and tumormetastasis. J Biol Chem 2014;289:12126–44.

39. LeBleu VS, O'Connell JT, Gonzalez Herrera KN, Wikman H, Pantel K,HaigisMC, et al. PGC-1amediatesmitochondrial biogenesis andoxidativephosphorylation in cancer cells to promote metastasis. Nat Cell Biol2014;10:992–1003.

Clin Cancer Res; 25(2) January 15, 2019 Clinical Cancer Research628

Yoh et al.

on March 18, 2020. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 23, 2018; DOI: 10.1158/1078-0432.CCR-18-0124

2019;25:619-628. Published OnlineFirst October 23, 2018.Clin Cancer Res   Tomoaki Yoh, Etsuro Hatano, Yosuke Kasai, et al.   after Surgical ResectionTransition, Predicts Prognosis of Intrahepatic Cholangiocarcinoma

Mesenchymal−Serum Nardilysin, a Surrogate Marker for Epithelial

  Updated version

  10.1158/1078-0432.CCR-18-0124doi:

Access the most recent version of this article at:

  Material

Supplementary

  http://clincancerres.aacrjournals.org/content/suppl/2018/10/23/1078-0432.CCR-18-0124.DC1

Access the most recent supplemental material at:

   

   

  Cited articles

  http://clincancerres.aacrjournals.org/content/25/2/619.full#ref-list-1

This article cites 38 articles, 8 of which you can access for free at:

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.org

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/25/2/619To request permission to re-use all or part of this article, use this link

on March 18, 2020. © 2019 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 23, 2018; DOI: 10.1158/1078-0432.CCR-18-0124