Identification of serum CCL15 in hepatocellular carcinoma.

0.8 MB·English

by  Li, Y

FULL PAPER
British Journal of Cancer (2013) 108, 99–106 | doi: 10.1038/bjc.2012.494
Keywords: CCL15; biomarker; HCC; migration; invasion
Identification of serum CCL15
in hepatocellular carcinoma
Y Li1, J Wu2, W Zhang2, N Zhang*,3,4,2 and H Guo*,3,4
1Clinical laboratory, Tianjin Medical University Cancer Institute and Hospital, Tianjin 300060, China; 2Department of Cancer Cell
Biology, Tianjin Medical University Cancer Institute and Hospital, Tianjin 300060, China; 3Key Laboratory of Breast Cancer
Prevention and Therapy, Ministry of Education, Tianjin 300060, China and 4Research Center of Basic Medical Sciences, Tianjin
Medical University,Tianjin 300070,China
Background:Earlyserumdetectionisofcriticalimportancetoimprovethetherapyforhepatocellularcarcinoma(HCC),oneofthe
most deadlycancers. Hepatitis infection isa leading causeof HCC.
Methods: In the present study, we collected total serum samples with informed consent from 80 HCC patients with HBV (þ)/
cirrhosis (þ),80patientswithbenigndiseases (50livercirrhosispatientsand30HBV-infectedpatients)and60healthycontrols.
Analysiswasbyusingsurface-enhancedlaserdesorption/ionisation-time-of-flightmassspectroscopy(SELDI-TOF-MS)tofindnew
serummarkersofHCC.SELDIpeakswereisolatedbySDS–PAGE,identifiedbyLC-MS/MSandvalidatedbyimmunohistochem-
istry (IHC)inlivertissues.Migration andinvasion assaywereperformed totest theability ofcellmigration andinvasion invitro.
Results:SELDI-TOF-MSrevealedabandat7777M/ZintheserumsamplesfromHCCpatientsbutnotfromhealthycontrolsor
patients with benign diseases. The protein (7777.27 M/Z) in the proteomic signature was identified as C-C motif chemokine 15
(CCL15)bypeptidemassfingerprinting.AsignificantincreaseinserumCCL15wasdetectedinHCCpatients.Functionalanalysis
showed that HCC cellexpressed CCL15,whichinturn promotedHCC cell migration andinvasion.
Conclusion: CCL15 may be a specific proteomic biomarker of HCC, which has an important role in tumorigenesis and tumour
invasion.
Primaryhepatocellularcarcinoma(HCC)isoneofthemostlethal detectingthepatientswithHCC(Johnson,2001).AlthoughDCPis
malignancies worldwide. It is the third leading cause of cancer believedtobeabettermarker,elevatedDCPactivityisonlypresent
death in China and the sixth most common cancer in the world in 44–47% of HCCs o3cm in size (Grazi et al, 1995; Soga et al,
(Jemaletal,2011).PrognosisofHCCremainspoor,mainlydueto 1998).Thus,bothAFPandDCPareinsufficientforearlydiagnosis
the failure of early diagnosis of the disease in symptom-free of HCC.
patients(Fujiyama et al, 2002; Marrero and Lok, 2004). Early There is a pressing need to find new serum biomarkers to
detectionofHCCbeforetheonsetofclinicalsymptomscanleadto improve the efficacy of HCC detection. Advances in proteomic
curativetreatment,significantlyimprovingprognosis(Witjesetal, analysis have offered exciting opportunities for finding novel
2010). biomarkers. Two major strategies are widely used to discover
At present, alpha-fetoprotein (AFP) and des-g-carboxy pro- clinically useful biomarkers. One utilises a surface-enhanced laser
thrombin (DCP) are the two most widely used tests to aid in the desorption ionisation-time of flight-mass spectrometry (SELDI-
diagnosis of HCC. However, AFP is found to be normal in TOF-MS)system(Petricoinetal,2002;Blancetal,2005;Leeetal,
approximately one-third of patients with small (o3cm) HCC 2006;Aivadoetal,2007),andtheotherisbasedon2DSDS–PAGE
(Jiangetal,2011),andthespecificityofAFPisonlyabout68.2%in coupledwithMS(Seowetal,2000;Wardetal,2006).TheSELDI-
*Correspondence: Dr N Zhang; E-mail: [email protected] or
HGuo;E-mail:[email protected]
Received16May2012;revised12October2012;accepted15October2012
&2013CancerResearchUK.Allrightsreserved0007–0920/13
www.bjcancer.com|DOI:10.1038/bjc.2012.494 99
BRITISHJOURNAL OF CANCER CCL15 andHCC
TOF, which generates the protein patterns by MS, has been Control patients: With patient consent, serum samples were
consideredasapowerfultoolforthediscoveryofnewbiomarkers obtained from the Tianjin Medical University General Hospital,
(Paradisetal,2005;Wardetal,2006).However,theproteinpeaks China and are as follows: benign diseases: liver cirrhosis (n¼25;
detected by SELDI are not easily identified. Thus, these findings mean age 61 years); calculus of bile duct (n¼20; mean age 60
cannot provide any information about the biological roles of the years); B hepatitis (n¼20; mean age 59 years); and 60 blood
marked proteins in the pathogenesis of a disease. On the other donors (mean age 61 years).
hand, the 2D SDS–PAGE-based method provides a lot of SerumsamplesfromHCCpatientswereimmunodepressedwith
information about proteins, including expression levels, actual p/ CCL15 antibody, which was bought from German Roche (Berlin,
s and molecular weights (Rocken et al, 2004). However, the 2D Germany) and then prepared for SELDI-TOF-MS analysis. All
SDS–PAGEmethodisnotsensitiveforresolvinghydrophobic,low serumsampleswereseparatedanddividedinto10mlaliquots,and
abundant or low molecular weight proteins. Liquid chromatogra- stored at (cid:2)801C until being assayed.
phy-mass spectrometry (LC-MS) is a powerful identification tool
SELDI-TOF-MSanalysis. Proteinprofilingofserumsampleswas
for proteins, but it is not suitable for analysing proteins directly
performed using the eight-spot format WCX-2 (weak cationic
(LimandElenitoba-Johnson,2004).Ithasbeensuggestedthatthe
exchange) Protein Chip Arrays (Ciphergen Biosystems, Fremont,
combinationofSELDI-TOF,2-DandLC-MSmayprovideabetter
CA, USA). Frozen serum samples were thawed and spun at
solution to identify disease-associated proteomic biomarkers
10000r.p.m.for5minat41C.Inall,20mlofU9bufferwasadded
(Adkins et al, 2002; Sheng et al, 2006).
to10mlaliquotsofeachserumsampleandplacedonicefor30min
HCC is closely associated with chronic inflammation, which
before adding 360ml WCX-2 buffers. Arrays were prepared as
produces a repertoire of cytokines and chemokines at low
follows:eacharraywaspre-equilibrated2(cid:3)5minin200mlWCX-
molecular weights. We hypothesise that these low molecular
2 buffer on a horizontal shaker (MSI Mini-shaker, Benchmark,
peptides may provide a clue for HCC early detection. The aim of
Brownstown, PA, USA) before sample addition. The sample
this study was to identify low molecular weight serum protein
supernatant wasaddedand incubatedfor 1hontheshaker. After
biomarkersinHCC.TheanalysiswasperformedbyusingSELDI-
incubation, the sample was removed, and each spot was washed
TOF-MS technology to screen potential protein patterns specific
with 200ml WCX-2 buffers for 2(cid:3)5min with agitation. After
for HCC. Candidate protein peaks were then separated by
washing, the array was carefully separated from the bioprocessor
SDS–PAGE followed by trypsin digestion and identified by
and washed briefly with deionized water. In all, 0.5ml sinapinic
LC-MS analysis.
acid was deposited on the array spots and allowed to air dry.
The ProteinChip Arrays were read by SELDI-TOF-MS (Pro-
teinChip PBS II reader, Ciphergen). This was calibrated using
MATERIALS AND METHODS
NP20chipsthathadbeenboundwithall-in-onestandardproteins
tosetuptheparameters.Theoptimaldetectionparameterofmass/
Collection and preparation of samples for SELDI-TOF-MS
charge size range was set between 5000 and 20000 M/Z with a
analysis
maximum of 50000 M/Z. The laser intensity was set at 175 and
detectorsensitivitywassetat5.Anaveragevalueof130spotswas
Samples for SELDI-TOF-MS analysis. With patient consent,
presented for each sample. All samples were detected with the
serum samples were obtained from 80 HCC patients (60 male,
same parameters. All the raw data was normalised with the
20 female; mean age, 65 years; range, 35–82 years) with HBV/
ProteinChipSoftwareversion3.1(homogenisationofthetotalion
cirrhosisfromtheTianjinMedicalUniversityCancerInstituteand
strengthandM/Z).TheM/Zsamplepeakswith42000M/Zwere
Hospital, China from January 2005 to November 2007. All HCC
normalisedwithbiomarkerwizardofProteinChipSoftwareversion
patientswerediagnosedaccordingtothecombinedclinicalcriteria,
3.1fornoisefiltering.Thefirstthresholdfornoisefilteringwasset
including imaging data and serum tumour markers, and further
at 5, and the second was set at 2. The minimum threshold for
confirmedbyhistopathologicalanalysis.Samplesfrom50patients
clustering wasset at10%.Spectrum analysiswasperformed using
with liver cirrhosis (35 male, 15 female; mean 63 years) and 30
the Biomarker Patterns Software (Ciphergen, Sparta, NJ, USA).
HBV-infected patients (20 male, 10 female; mean 60 years) were
kindly provided by the Department of Infection, Tianjin Medical Tricine-SDS–PAGE and LC-MS analysis. Pooled serum samples
University General Hospital, China. Diagnosis of liver cirrhosis (n¼15)fromHCCpatientswithhighSELDIintensitiesat7777.27
was mainly dependent on clinical history, physical examination, M/Z were selected. These samples were diluted with 9M urea,
laboratory findings, ultrasonography and/or computed tomogra- 10mM Tris/HCI (pH 7.4) and applied to AKTA Purifier T-900
phy,withorwithoutliverbiopsy.SerumsamplesofHBV-infected column system (Amersham, Piscataway, NJ, USA). After sample
patientswerecollected based onthepresence ofHBsAg(hepatitis purification,albuminandimmunoglobulinwereremovedfromthe
Bsurfaceantigen),HBeAg(hepatitisBeantigen)andHBVDNA. serum by 3GA and then by Protein A. The rest of the fractions
None of thepatients had received any treatment before collection were loaded onto a Tricine–SDS–PAGE gel using ETTAN II
ofbloodsamples.Inaddition,60blooddonors(45male,15female; (Amersham Pharmacia, Piscataway, NJ, USA) gel electrophoresis
mean age 61 years; range 50–76 years) without liver neoplasia, system. Electrophoresis was run at 20mA for 3h. Gels were then
alcoholic cirrhosis, hepatitis B or hepatitis C infection were stainedwithCoomassieBrilliantBlue.Thebandscorrespondingto
recruited from routine health examination at the Tianjin Medical the 8000 M/Z markers were excised and then destained with two
University Cancer Institute and Hospital, China. washesof50mldeionisedwater,followedby50mlACN/50mMl(cid:2)1
NH HCO (1:1, v/v), and dried in a SpeedVac concentrator
4 3
SamplesforCCL15analysis. Patientswithcarcinoma:withpatient (FisherScientific, Pittsburgh, PA, USA). The dried gel slices were
consent and approval by the ethical committee, serum samples rehydrated with 10mM DTT followed by 50mM IAM (45min at
were obtained from the Tianjin Medical University Cancer room temperature in the dark). After several washes with 25mM
InstituteandHospital,China.Thisgroupof125patientscontained NH HCO and 100% CAN, 20mgl(cid:2)1 solution of trypsin was
4 3
55patients withhepatitisB-relatedHCC(thepresenceofHBsAg, added to the gel slices and digestion was allowed to proceed at
HBeAg and HBV DNA; mean age 58 years), 25 with hepatitis 371Cfor12h.Thetrypsin-digestedsamplewasloadedontoaC18
C-relatedHCC(thepresenceofHCVRNA;meanage59years),15 reversed-phase column (5mm(cid:3)250mm, PepMapC18, LC Pack-
withlungcancer(meanage56years),15withgastriccancer(mean ings, Amsterdam, The Netherlands), and the peptides were
age60years)and15carcinomaofgallbladder(meanage57years). separated by electrospray ionisation (ESI, Bruker Esquire 3000,
100 www.bjcancer.com|DOI:10.1038/bjc.2012.494
CCL15 andHCC BRITISH JOURNALOF CANCER
Bruker Daltonik, Bremen, Germany). Proteins were identified by RT–PCR. Total RNA from cells was extracted using Trizol
an automated searching algorithm against the SWISS-Protand (Invitrogen). Then the RNA was transcribed using the RT–PCR
NCBI protein database. KitfromTaKaRaBiotechnologyCo.Ltd.(Dalian,China).Specific
primers for GAPDH (forward, 50-GCACCGTCAAGGCTGAGAA
Immunohistochemistry (IHC) assay. Normal (n¼50), adjacent C-30; reverse, 50-TGGTGAAGACGCCAGTGGA-30), CCL15 (for-
liver tissues (n¼80) and HCC tissues (n¼80) were processed ward,50-CGCGAATTCATGAAGGTCTCCGTGGCTG-30;reverse,50-
according to the standard approaches. The anti-CCL15 serum GCGCTCGAGTTATATTGAGTAGGGCTTCAG-30),andCCR1(for-
(1:1600,Immunechem Pharmaceuticals INC, Beijing, China) was ward, 50-CTCTTCCTGTTCACGCTTCC-30; reverse, 50-GCCTGAAA
applied to the slides of three groups and incubated in a moist CAGCTTCCACTC-30)werefromSangon(Shanghai,China).
chamber at 41C overnight. In all, 0.01ml PBS was used as the
Westernblottingassay. Westernblottingassaywasperformedas
negativecontrolinalltheexperiments.Slidescutinparalleltothe
described bySun et al (2005). Cells and clones were washed twice
IHC-treated sections were stained by hematoxylin and eosin for
withicecoldPBSpH7.4andthenlysedby1(cid:3)SDSsamplebuffer
better identification of the different tissue areas. To avoid
interindividualbiasofIHCstainingdifferentiations,allslideswere (Tris-HCl, pH 6.8, 62.5mM, 2% SDS, 10% glycerol.). Equal
amounts of cell lysates (20mg total protein per lane) were loaded
determined by an experienced pathologist.
onto 10% or 14% SDS–PAGE gels, transferred onto PVDF
Western blotting analysis in serum. The serum samples from membranes (Millipore, Billerica, MA, USA), probed with anti-
mixtureofHCCandmixtureofLCwereusedinwesternblotting CCR1 or anti-CCL15 primary antibody, followed by HRP-
analysis.Theproteinconcentrationsofthesepooledsampleswere conjugated secondary antibody and visualised by enhanced
measuredbyDCproteinassay(BioRad,Chicago,IL,USA).Fifteen chemiluminescence reagents ECL (Pierce, Rockford, IL, USA).
micrograms of total protein of each sample was separated in 4– Western blotting analysis of b-actin was used as loading control.
20%Tris-glycineSDS–PAGEgel(Invitrogen,Carlsbad,CA,USA)
Wound-healing assay. Cells were seeded in 35mm dishes for 2
andtransferredtoPVDFmembranes.Membraneswereblockedby
days,starvedinserum-freemediumovernight,afterwhich,scrape
5% nonfat milk in PBS-0.1% Tween 20 buffer (PBS-T) for 1h at
wounds were created with a 10-ml sterile plastic pipette tip in the
roomtemperatureandthenincubatedovernightat41CwithmAb
confluent cell monolayers. After washing with PBS, serum-free
ofCCL15,followedbyHRP-conjugatedsecondaryantibody.Signal
medium (to prevent cell proliferation) was added. The widths of
was developed using ECL plus Western Blotting Detection
the wounds were recorded at different time points (0, 3, 6, 9, 12
Reagents (Amersham Bioscience, Piscataway, NJ). The PVDF
and 24h) using the number of grids in the ocular of our
membrane was dried and stained with SimplyBlue Safe Stain
microscope. Ten grids represented 1mm in length, and we could
(Invitrogen) after western blot transferring.
therefore calculate the migration distances by the changes of the
gridnumber.Threemeasurementsatdifferentpositionsalongeach
Enzyme-linked immunosorbent assay (ELISA) analysis.
scratchinglinewererecordedateachtimepoints.Andinorderto
Patients’ samples were collected as serum, centrifuged for 15min
at 1500g and then stored until analysis at (cid:2)801C. CCL15 was test the effect of CCL15 on the motility of hepatocarcinoma cells,
determined using a commercially available sandwich enzyme- theserum-freemediumcontainingCCL15(1mM)wasaddedtothe
scraped cells and cultured for 24h.
linked immunosorbent assay (ELISA, Schebo Tech, Giessen,
German). Serum determination of AFP was performed by means Chemotaxis assay. Chemotaxis assays were performed essentially
ofanenzymeimmunoassaywithacommerciallyavailablekitonly as previously described using Boyden chambers equipped with
for hepatitis B-related HCC. The measurements of two tumour 8mm porosity polyvinylpyrrolidone-free polycarbonate filters
markers were performed blindly. (Carloni et al, 1998). Briefly, polycarbonate filters were coated on
the lower surface with 20mgml(cid:2)1 human type I collagen
Cellcultureandreagents. ThreehumanHCCcelllines,including
(Collaborative Biomedical Products; Bedford, MA, USA) for
HepG2 (American Type Culture Collection) and Huh-7, HLE 30minat371Candplacedbetweenthelowerchamberandupper
(Human Sciences Research Resources Bank, Tokyo, Japan), were
chamber. The lower chamber was filled with MEM media
usedinthisstudy.HepG wasculturedinMEMmediacontaining
2 containing CCL15 as chemoattractants. Serum-deprived HepG ,
2
10%foetalbovineserum(FBS),1%non-essentialaminoacidsand
Huh-7 and HLE cells were separately washed, trypsinized,
1% sodium pyruvate (Thermo Scientific Hyclone, Logan, VT,
resuspended in serum-free medium containing 1% albumin at a
USA), whereas Huh-7 and HLE were cultured in DMEM media concentration of 3(cid:3)105cellsml(cid:2)1 and placed in the upper
with 10% FBS and RPMI1640 with 10% FBS, respectively. chamber. The chambers were incubated at 371C in a cell-culture
Chemotaxis chambers and membranes were from Neuroprobe
incubatorfor6h.Thenthefiltermembranewaswashed,andcells
(Gaithersburg, MD, USA). Recombinant Human CCL15/68 and
attachedwerefixedandstained.Thenumberofmigratingcellswas
antibody to CCL15 were from RD Systems (USA). Antibody to
countedinsixrandomlychosenfields((cid:3)400)bylightmicroscope,
CCR1wasfromAbcamInc.(Cambridge,MA,USA).Antibodyto andthecountswereaveraged(means±s.d.).Chemotaxisindex¼
b-actin, goat anti-mouse IgG-FITC and donkey anti-goat IgG-
the migrating cell number in a chemoattractant gradient/the
FITC were from Santa Cruz Biotechnology, Inc. (Santa Cruz,
migrating cell number in a medium control.
CA, USA).
Matrigel invasion assay. For invasion assays, 105 cells in 400ml
Gene transfection. 2(cid:3)105 cells were cultured in 12-well plates 1 serum-free MEM medium were plated in the top chamber of
day before transfection in MEM containing 10% FBS, 1% non- Transwells with a Matrigel-coated membrane (Corning Incorpo-
essentialaminoacidsand1%sodiumpyruvate.Afterstarvationfor rated, Corning, NY, USA) containing 8-mm diameter pores. The
3h, transfection was performed by using Lipofectamine 2000 inserts were placed into the bottom chamber wells of a 24-well
(Invitrogen) according to the manufacturer’s instructions. To plate containing MEM with CCL15 as a chemoattractant. After
establishstablecelllines,thecellswereenrichedusing0.6mgml(cid:2)1 36h of incubation, cells remaining on the insert top layers were
G418 sulphate (Gibco/Invitrogen, Grand Island, NY, USA) or removed by cotton swab scrubbing, as recommended by the
0.4mgml(cid:2)1 Hygromycin B from BD Biosciences (Bedford, MA, manufacturer. Cells on the lower surface of the membrane were
USA). After the cells grew up, single cell expressing GFP was fixedandstained.Insertswerewashedforseveraltimesindistilled
picked out and cultured into 96-well plates, and then each clone water before air drying. The membranes were photographed, and
was validated by RT–PCR and western blotting analysis. cell counts were determined after totalling five random fields.
www.bjcancer.com|DOI:10.1038/bjc.2012.494 101
BRITISHJOURNAL OF CANCER CCL15 andHCC
Statistical analysis. All statistical analyses were carried out with MS revealed that expression of CCL15 was elevated in HCC
the use of SPSS software Version 13.0 (Barbecana, Houston, TX, patients.
USA). Data were expressed as the mean±s.e. from at least three
CCL15asabiomarkerforHCC. Immunohistochemistryanalysis
separate experiments. Statistical significance of the results was
wasusedtoassessCCL15expression inlivertissues. Asshownin
analysed bytwo-wayANOVA and Po0.05was considered to be
Figure 2A, CCL15 were detected on tumour tissues from HCC
statistically significant.
patients but not in tissues from healthy samples. Among 80
tumour samples, CCL15 were detected in 64 tumour tissues and
only in 16 tissues adjacent to tumour (Table 2). Only 1 out of 50
samplesfromhealthytissuesshowedpositivestainingwithCCL15.
RESULTS
CCL15expressioninlivertumourseraandlivercirrhosisserawas
analysed by western blotting. Interestingly, strong staining signals
Identification of CCL15 from serum samples of hepatoma
were found in HCC sera (Figure 2B-c,d), whereas no CCL15
patient. SELDI-MS was used to investigate the expression of low
molecular weight peptides in serum samples of HCC patients. As
shown in Figure 1A, four peptides, designated as P1 at 6489
Dalton, P2 at 6662 Dalton, P4 at 8593 Dalton, and P5 at 8720 Table1.ComparativecontentofsevendifferentproteinsatM/Zin
Dalton, were detected in the pooled serum from healthy controls WCX-2
(Figure 1A, Lane a), but not in the serum from patients with
Peptides/M/Z
benign diseases (Lane b), or HCC (Lane c and d, Table 1). P7 at Comparativecontentofproteins(%)
(Da)
16200 Dalton only appeared in serum from patients with benign
diseases, including liver cirrhosis and HBV-carriers (Figure 1A, Proteins Healthycontrol Benigndiseases HCC
Lane b). Two bands, P3 at 7777 Dalton and P6 at 9250 Dalton, P1 6489.48 19.77±3.54a 6.46±2.18 6.87±2.30
were detected only in HCC patients (Figure 1A, Lane c).
P2 6662.34 22.82±4.15a 9.96±2.93 8.06±2.47
To isolate the proteins of interest and to determine candidate
proteinidentities,15serumsamplesfromHCCpatientscontaining P3 7777.27 4.09±1.14 6.87±2.18 15.87±4.30a
highSELDIintensityof7777.27M/Zwerepooledandseparatedby P4 8593.75 11.45±2.73a 2.13±0.89 3.21±1.21
Tricine-SDS–PAGE (Figure 1B). The band at 8kDa was excised, P5 8720.23 13.69±2.30a 4.57±2.12 4.32±2.89
trypsinized and analysed by LC-MS/MS. As shown in Figure 1C,
P6 9250.00 4.51±1.32 4.87±1.82 23.15±3.81a
twopeptidesequenceswereidentified,whichmatchedwithhuman
CCL15,ata31%ofcoverage,suggestingthatexpressionofCCL15 P7 16200.17 8.83±2.76 26.72±4.51a 7.63±2.48
was enhanced in serum from HCC patients. CCL15 antibody was
used to treat serums from HCC patients, and the band P3 Abbreviations:HCC¼hepatocellularcarcinoma;WCX-2¼weakcationicexchange.Benign
diseasesincludelivercirrhosisandhepatitisB.
disappeared after immuno-depletion, showing that P3 at 7777 aThisvaluewassignificantlyhigherthanthatintheothertwogroups(Po0.01).
Dalton, not P6, was CCL15 (Figure1A, Lane d). Taken together,
A B
a b
5000 7500 10 000 12 500 15 000 17 500
20
ce100 P1 P2P4 P5 a 10kD
ndan2100 P7 b 8kD
bu 0
a20
ative 100 P3 P6 c 4kD
Rel20
10 P6 d
0 2kD
M/Z
C 10 20 30 40 50
MKVSVAALSC LMLVAVLGSQAQFINDAETELMMSKLPLENPVVLNSFHFA
60 70 80 90 100
ADCCTSYISQ SIPCSLMKSY FETSSECSKPGVIFLTKKGR QVCAKPSGPG
110
VQDCMKKLKPYSI
Accession Names MW/pI Score Peptide Coverage Matched Sequences
numbers matches
CCL15_HUMAN C-C motif 7773/9 61 2 31% SY FETSSECSKP GVIELTK
chemokine QVCAKPSGPGVQDCMK
15
Figure1.IdentificationofCCL15inHCCpatientserum.(A)DifferentialexpressionoftheSELDIpeaksinthecomparisonsoffourgroups.a:the
samplefrommixtureofhealthycontrols;b:thesamplefrommixtureofbenigndiseases(livercirrhosisorhepatitis);c:thesamplefrommixtureof
HBV-HCC;andd:thesamplefrommixtureofHBV-HCCafterimmunodepressionassayusingCCL15antibody.P1-P7showsdifferentpeptidesor
proteins.(B)Isolationofthe7777.27M/Zpeak.Arrowindicatesthebandof7777.27M/Zprotein.a:marker;b:samplefrommixtureofHCC.
(C)The7777.27M/Zprotein-matchedCCL15protein(MW¼7773.10M/Z,pI¼9)withtwoconsistentsequences(Score¼61).
102 www.bjcancer.com|DOI:10.1038/bjc.2012.494
CCL15 andHCC BRITISH JOURNALOF CANCER
Table3.Therelationshipbetweenclinicopathologicaldataofpatients
A 10× 20× 40× withHCCandtheserumlevelsofCCL15
HCC CCL15(pgml(cid:2)1)
Normal Total(beforetreatment)(55) 26.6±1.2
tissue
Subtype
Poorlydifferentiated(15) 30.3±2.2a
Moderatelydifferentiated(20) 24.3±0.6
Welldifferentiated(20) 19.1±0.6
HCC
Tumoursize
X3cm(35) 27.6±1.2b
o3cm(20) 19.6±2.1
B a b c d
Metastasis
CCL15
Yes(30) 27.9±3.6c
(cid:2)-Actin No(25) 20.6±1.1
Aftertreatment 9.7±1.2d
C 100 Operation(25) 9.9±0.8
Arterialembolization(20) 10.0±1.4
Percutaneoustumourablation(10) 9.0±1.6
80 Healthycontrols(60) 1.2±0.1
y Abbreviations:CCL15¼C-Cmotifchemokine15;HCC¼hepatocellularcarcinoma.
nsitivit 60 AFP abSShhoowweeddPPoo00..0055bbeettwweeeennptuomorolyurdsifizfeereXn3tiaatneddoan3dcwme.lldifferentiated.
Se 40 CCL15 dShowedPo0.05betweenpatientswithmetastasisandwithoutmetastasis.
dShowedPo0.05betweenpatientsbeforeandaftertreatments.
20
0
0 20 40 60 80 100 Table4.DifferentiallevelsofCCL15invariousgroups
100-Specificity
Groups CCL15(pgml(cid:2)1)
Figure2.TheexpressionofCCL15inlivertissuesandsera,andits
Cancer
comparisonwithAFPasaproteomicbiomarker.(A)Immuno-
histochemicalstainingofCCL15expressioninHCCtissuesand B-relatedHCC(55) 26.6±1.2
adjacentlivertissues.ThepositivesignalsforCCL15wereobservedin C-relatedHCC(25) 23.9±3.1a
brown.NopositivesignalofCCL15wasdetectedinadjacentnormal Lungcancer(15) 4.3±1.7b
livertissues.(B)WesternblottinganalysisofCCL15expressioninHBV- Gastriccancer(15) 4.1±0.5c
HCCseraandlivercirrhosissera.ArrowindicatesthatCCL15were Carcinomaofgallbladder(15) 3.6±1.5d
observedinHCC.a,b:samplefromtwobatchesofmixtureofliver Benigndiseases(65) 2.0±0.9e,f
cirrhosis;c,d;samplefromtwobatchesofmixtureofHCC;(C)Receiver Healthycontrols(60) 1.2±0.1g
operationcharacteristic(ROC)curvesinpatientswithB-relatedHCC
andB-hepatitis. Abbreviations:CCL15¼C-Cmotifchemokine15;HCC¼hepatocellularcarcinoma.
aShowedP40.05betweenB-relatedHCCandC-relatedHCC.
bShowedPo0.05betweenB-relatedHCCandlungcancer.
cShowedPo0.05betweenB-relatedHCCandgastriccancer.
Table2.TheexpressionofCCL15inthelivertissues dShowedPo0.05betweenB-relatedHCCandcarcinomaofgallbladder.
eShowedPo0.05betweenB-relatedHCCandbenigndiseases.
Livertissue Samples Positiverate(%) Score fShowedP40.05betweenbenigndiseasesandhealthycontrols.
gShowedPo0.05betweenB-relatedHCCandhealthycontrols.
Normal 50 1(2.00) 0.12±0.13a
Adjacentcancer 80 16(20.00) 0.55±1.23b
HCC 80 64(80.00) 2.25±1.29c
detected in HCC patients and treatment reduced serum levels of
Abbreviation:CCL15¼C-Cmotifchemokine15. CCL15.
aNormalvshepatocellularcarcinoma(HCC);Po0.05.
bAdjacentcancervsnormal,P40.05. Receiver operation characteristic curves were plotted to assess
cAdjacentcancervsHCC,Po0.05. thespecificityandsensitivityofCCL15asabiomarkerforHCCin
distinctiontochronichepatitisB(Figure2C).TheAUC(95%CI)
was 0.964 (0.812–0.989), 0.723(0.555–0.861) for CCL15 and AFP,
respectively. A significant difference was noted between the AUC
for CCL15 and those for AFP (Po0.05), suggesting that CCL15
staining was found in liver cirrhosis sera (Figure 2B-a,b). ELISA wasabetterbiomarker toAFP.Basedonthemaximisationofthe
wasusedtoassessserumCCL15levelsinHCCpatients(Table3), Youdenindex,theoptimalcutoffvaluewas16ngml(cid:2)1forCCL15
inothercancerpatientsandcontrols(benignpatientsandhealthy (sensitivity, 88.2%; specificity, 93%) and 25mgl(cid:2)1 for AFP
controls; Table 4). A significant increase in serum CCL15 was (sensitivity, 72.4%; specificity, 81%).
www.bjcancer.com|DOI:10.1038/bjc.2012.494 103
BRITISHJOURNAL OF CANCER CCL15 andHCC
A B
CCl15 (1 nM)
0
80 0.01 HuH7 HLE HepG2
PF 0.1 CCL15
H 60 1 mRNA
s/ 10
mber 40 100 nM GAPDH
u
n
ell 20 CCL15
C
Protein
0 (cid:2)-Actin
HuH7 HLE HepG2
C D
0.3 0.3
HepG2
HuH7 HLE (CCL15 1nM)
m) HLE m) HLE (control)
m m
ance ( 0.2 ** *** ance ( 0.2 **
dist * dist
n n
atio 0.1 atio 0.1
gr gr
Mi Mi
0 0
0 3 6 9 12 24 0 3 6 9 12 24
Times (h) Times (h)
Figure3.EffectsofCCL15onhepatocarcinomacellschemotaxisandmigration.(A)Chemotaxisassayofthreekindsofhepatocarcinomacells
withindicatedamountofCCL15aschemoattractantfor6h.(B)RT–PCRandwesternblottinganalysisofCCL15inHuh-7,HLEandHepG cells.
2
(C)Wound-healingassayofthethreekindofhepatocarcinomacells.Woundwidthswererecordedatindicatedtimepoints.***Po0.0005.
(D)Wound-healingassayofHLEcellswithorwithoutCCL15(1nM)treatment.***Po0.005.
A HepG2 Control siClone1 siClone2 C100 *
CCL15 80
s
er
b 60
(cid:2)-Actin m
u
n
B 0.3 Control Cell 40
20
m) siClone1
m
e ( 0.2 0 Control siClone1 siClone1
anc ** +CCL15
st
di
n
o 0.1
ati
gr
Mi
0 Control siClone1 siClone1
0 3 6 9 12 24 +CCL15
Times (h)
Figure4.DownregulationofCCL15impairedHepG cellsmigrationandinvasion.(A)WesternblottinganalysisofHepG cellsanditsclones
2 2
withdownregulationofCCL15.(B)Wound-healingassayofHepG cellsandsiCCL15/HepG cellsatindicatedtimepoints.(C)Invasionassaysof
2 2
HepG cellsandsiCCL15/HepG cells.CCL15coatedinMatrigel-enhancedHepG /siClone2cellsinvasiveability,withpicturesshown((cid:3)200;
2 2 2
*Po0.05,**Po0.005).
CCL15 promoted HCC cell migration and invasion. CCL15, a confirmthisresult,weexaminedtheexpressionofCCL15inHCC
chemokine,inducescellmigrationbybindingtoCCR1,areceptor cell lines. CCL15mRNAand proteinweredetected inbothHuh7
expressed in most liver cells (Brown et al, 2007). We hypothesise and HepG2, two HCC cell lines, but not in HLE, less malignant
that elevated CCL15 promoted HCC cell migration and invasion. cells (Figure 3B). A scratch assay was used to further assess HCC
To investigate the role of CCL15, a micro-Boyden chamber assay migration.AsshowninFigure3C,bothHepG2andHuh7showed
wasperformedonHCCcells.CCL15inducedrobustchemotaxisof a stronger migration capacity than HLE cells. Treatment with
Huh7, HLE and HepG2 cells, indicating that CCL15 mediated CCL15 induced a marked increase in HLE cell migration
HCC cell migration (Figure 3A). Immunohistochemistry results (Figure 3D). Taken together, the results clearly showed that
suggestedthatCCL15wasexpressedbyHCCcells(Figure2A).To CCL15 promoted HCC cell migration.
104 www.bjcancer.com|DOI:10.1038/bjc.2012.494
CCL15 andHCC BRITISH JOURNALOF CANCER
KnockdownCCL15impairedHCCcellinvasion. CCL15maybe Scientific Foundation of China (NSFC No. 81201646, No.
secretedfromtwomajorsources,tumourinfiltratedleucocytesand 81072160,No.81101754)andChangjiangScholarsandInnovative
HCCcells.CCL15fromHCCmaybesufficienttopromotetumour Research Team in University in China (Grant IRT1076).
invasion and micrometastasis. To further confirm that CCL15
fromHCChadanactiveroleininvasion,expressionofCCL15was
knocked down by small RNA interference (Figure 4A). A scratch REFERENCES
assay showed a decrease in HepG2 migration, suggesting that
CCL15 from HepG2 functioned in an autocrine way to promote AdkinsJN,VarnumSM,AuberryKJ,MooreRJ,AngellNH,SmithRD,
HepG2 migration (Figure 4B). A Matrigel analysis showed that SpringerDL,PoundsJG(2002)Towardahumanbloodserumproteome:
knockdown of CCL15 impaired HepG2 cell invasion, while analysisbymultidimensionalseparationcoupledwithmassspectrometry.
supplement with CCL15 restored the invasiveness (Figure 4C). MolCellProteomics1(12):947–955.
Thus,theresultssuggestedthatCCL15fromHCCwassufficientto AivadoM,SpentzosD,GermingU,AlterovitzG,MengXY,GrallF,
promote tumour invasion in an autocrine mode. GiagounidisAA,KlementG,SteidlU,OtuHH,CzibereA,PrallWC,
Iking-KonertC,ShayneM,RamoniMF,GattermannN,HaasR,Mitsiades
CS,FungET,LibermannTA(2007)Serumproteomeprofilingdetects
myelodysplasticsyndromesandidentifiesCXCchemokineligands4and7
DISCUSSION asmarkersforadvanceddisease.ProcNatlAcadSciUSA104(4):
1307–1312.
SELDI has been successfully applied in defining early detection BlancJF,LalanneC,PlomionC,SchmitterJM,BathanyK,GionJM,Bioulac-
SageP,BalabaudC,BonneuM,RosenbaumJ(2005)Proteomicanalysisof
biomarkers in multiple cancers, including ovarian, prostate and
differentiallyexpressedproteinsinhepatocellularcarcinomadevelopedin
breast cancers (Zhang et al, 2004; Hu et al, 2005; Le et al, 2005).
patientswithchronicviralhepatitisC.Proteomics5(14):3778–3789.
However,itisratherachallengetoidentifythemolecularidentity
BrownMF,BahnckKB,BlumbergLC,BrissetteWH,BurrellSA,DriscollJP,
of these novel biomarkers. Our results revealed CCL15 as a HCC FedelesF,FisherMB,FotiRS,GladueRP,Guzman-MartinezA,Hayward
biomarker by using SELDI in combination with SDS–PAGE and MM,LiraPD,LillieBM,LuY,LundquistGD,McElroyEB,McGlynnMA,
LC-MS/MS. SELDI analysis showed two molecules at 7.8 and ParadisTJ,PossCS,RoacheJH,ShavnyaA,ShepardRM,TrevenaKA,
9.3kD specifically expressed in serum from HCC patients. SDS– TylaskaLA(2007)PiperazinylCCR1antagonists–optimizationofhuman
PAGE and LC-MS/MS analysis showed that CCL15 was specifi- livermicrosomestability.BioorgMedChemLett17(11):3109–3112.
cally expressed in serum of HCC patient. Treatment with anti- CarloniV,RomanelliRG,MercurioAM,PinzaniM,LaffiG,CotrozziG,
GentiliniP(1998)Knockoutofalpha6beta1-integrinexpressionreverses
CCL15 antibody eliminated SELDI signal at 7.8kD, confirming
thetransformedphenotypeofhepatocarcinomacells.Gastroenterology
that the CCL15 was specifically expressed in serum of HCC
115(2):433–442.
patients.Takentogether,bothSELDIandSDS–PAGE/LC-MS/MS
FujiyamaS,TanakaM,MaedaS,AshiharaH,HirataR,TomitaK(2002)
resultssuggestthatCCL15isaspecificserumbiomarkerforHCC
Tumormarkersinearlydiagnosis,follow-upandmanagementofpatients
patients. withhepatocellularcarcinoma.Oncology62(Suppl1):57–63.
Micrometastasis is amajorcause offailure inHCCtreatments. GraziGL,MazziottiA,LegnaniC,JovineE,MinieroR,GallucciA,PalaretiG,
Identification of CCL15 from patient tumour tissues provides a GozzettiG(1995)Theroleoftumormarkersinthediagnosisof
new insight into tumorigenesis and micrometastasis process of hepatocellularcarcinoma,withspecialreferencetothedes-gamma-
HCC.CCL15isapeptideinchemokinefamily,whichhasapivotal carboxyprothrombin.LiverTransplSurg1(4):249–255.
role in mediating leucocyte trafficking and essential for inflam- HuY,ZhangS,YuJ,LiuJ,ZhengS(2005)SELDI-TOF-MS:theproteomics
andbioinformaticsapproachesinthediagnosisofbreastcancer.Breast
matoryresponses(Richteretal,2005;Kwonetal,2008).Receptors
14(4):250–255.
ofCCL15,CCR1,wereexpressedinliver(Nomiyamaetal,2001).
HwangJ,KimCW,SonKN,HanKY,LeeKH,KleinmanHK,KoJ,NaDS,
Our results clearly showed that CCL15 enhanced migration and
KwonBS,GhoYS,KimJ(2004)AngiogenicactivityofhumanCC
invasion of HCC cells. Furthermore, it has been reported that chemokineCCL15invitroandinvivo.FEBSLett570(1-3):47–51.
CCL15 also increased the expression of matrix metalloproteinase JemalA,BrayF,CenterMM,FerlayJ,WardE,FormanD(2011)Global
expression and induced angiogenesis (Hwang et al, 2004; Kwon cancerstatistics.CACancerJClin61(2):69–90.
et al, 2008). Thus, CCL15/CCR1 may have a critical role in the JiangJ,WuC,ShenY,XuB,ZhengX,LiX,XuN(2011)Clinicalapplication
micrometastasis of HCC and can be used as a target for anti- ofdeterminingserumAFP-IgMcomplexesfordiagnosisofsmall
metastasis treatment. hepatocellularcarcinoma.AnticancerRes31(2):687–691.
JohnsonPJ(2001)Theroleofserumalpha-fetoproteinestimationinthe
The most effective treatment of HCC is surgery, which is
diagnosisandmanagementofhepatocellularcarcinoma.ClinLiverDis
typicallyappliedtosmallHCC(o3cm).Thus,thekeytoimprove
5(1):145–159.
survival rate is to detect tumour at early stage. Current serum
KwonSH,JuSA,KangJH,KimCS,YooHM,YuR(2008)ChemokineLkn-1/
biomarker,AFP,hasitslimitationinsensitivityandapproximately
CCL15enhancesmatrixmetalloproteinase-9releasefromhuman
30% HCC patients do not show an elevation in AFP. Our macrophagesandmacrophage-derivedfoamcells.NutrResPract2(2):
preliminaryinvestigationssuggestthatCCL15maybeapromising 134–137.
candidate,whichshowsabetterAUC(95%)at0.964bytesting55 LeL,ChiK,TyldesleyS,FlibotteS,DiamondDL,KuzykMA,SadarMD
HCC patients and 60 health donors. Although, this investigation (2005)IdentificationofserumamyloidAasabiomarkertodistinguish
focused on HBV-HCC patients, based onour studies of HCC cell prostatecancerpatientswithbonelesions.ClinChem51(4):695–707.
lines,wespeculatethatCCL15mayalsohaveanimportantrolein LeeIN,ChenCH,SheuJC,LeeHS,HuangGT,ChenDS,YuCY,WenCL,Lu
FJ,ChowLP(2006)IdentificationofcomplementC3aasacandidate
other types of HCC. Further clinical investigation by using larger
biomarkerinhumanchronichepatitisCandHCV-relatedhepatocellular
samplesizesisrequiredtodeterminethediagnosisvalueofCCL15.
carcinomausingaproteomicsapproach.Proteomics6(9):2865–2873.
ItisalsopossiblethatacombinationofAFPandCCL15willbea
LimMS,Elenitoba-JohnsonKS(2004)Proteomicsinpathologyresearch.Lab
more sensitive and specific way to detect HCC at an early stage. Invest84(10):1227–1244.
MarreroJA,LokAS(2004)Newermarkersforhepatocellularcarcinoma.
Gastroenterology127(5Suppl1):S113–S119.
NomiyamaH,HieshimaK,NakayamaT,SakaguchiT,FujisawaR,TanaseS,
ACKNOWLEDGEMENTS
NishiuraH,MatsunoK,TakamoriH,TabiraY,YamamotoT,MiuraR,
YoshieO(2001)HumanCCchemokineliver-expressedchemokine/
Thisworkwassupportedbyresearchgrantsfrom973programme CCL16isafunctionalligandforCCR1,CCR2andCCR5,and
grants (No. 2011CB933100 and No. 2010CB933900), National constitutivelyexpressedbyhepatocytes.IntImmunol13(8):1021–1029.
www.bjcancer.com|DOI:10.1038/bjc.2012.494 105
BRITISHJOURNAL OF CANCER CCL15 andHCC
ParadisV,DegosF,DargereD,PhamN,BelghitiJ,DegottC,JaneauJL, insmallhepatocellularcarcinoma.Hepatogastroenterology45(23):
BezeaudA,DelforgeD,CubizollesM,LaurendeauI,BedossaP(2005) 1737–1741.
Identificationofanewmarkerofhepatocellularcarcinomabyserumprotein SunR,GaoP,ChenL,MaD,WangJ,OppenheimJJ,ZhangN(2005)
profilingofpatientswithchronicliverdiseases.Hepatology41(1):40–47. ProteinkinaseCzetaisrequiredforepidermalgrowthfactor-induced
PetricoinEF,ArdekaniAM,HittBA,LevinePJ,FusaroVA,SteinbergSM, chemotaxisofhumanbreastcancercells.CancerRes65(4):
MillsGB,SimoneC,FishmanDA,KohnEC,LiottaLA(2002)Useof 1433–1441.
proteomicpatternsinserumtoidentifyovariancancer.Lancet359(9306): WardDG,ChengY,N’KontchouG,TharTT,BargetN,WeiW,Billingham
572–577. LJ,MartinA,BeaugrandM,JohnsonPJ(2006)Changesintheserum
RichterR,BistrianR,EscherS,ForssmannWG,VakiliJ,HenschlerR, proteomeassociatedwiththedevelopmentofhepatocellularcarcinomain
SpodsbergN,Frimpong-BoatengA,ForssmannU(2005)Quantum hepatitisC-relatedcirrhosis.BrJCancer94(2):287–292.
proteolyticactivationofchemokineCCL15byneutrophilgranulocytes WitjesCD,deManRA,EskensFA,DwarkasingRS,ZondervanPE,Verhoef
modulatesmononuclearcelladhesiveness.JImmunol175(3):1599–1608. C,IjzermansJN(2010)[Hepatocellularcarcinoma:thesignificanceof
RockenC,EbertMP,RoessnerA(2004)Proteomicsinpathology,research cirrhosisfortreatmentandprognosis–retrospectivestudy].NedTijdschr
andpractice.PatholResPract200(2):69–82. Geneeskd154:A1747.
SeowTK,OngSE,LiangRC,RenEC,ChanL,OuK,ChungMC(2000)Two- ZhangZ,BastJr.RC,YuY,LiJ,SokollLJ,RaiAJ,RosenzweigJM,Cameron
dimensionalelectrophoresismapofthehumanhepatocellularcarcinoma B,WangYY,MengXY,BerchuckA,VanHaaften-DayC,HackerNF,de
cellline,HCC-M,andidentificationoftheseparatedproteinsbymass BruijnHW,vanderZeeAG,JacobsIJ,FungET,ChanDW(2004)Three
spectrometry.Electrophoresis21(9):1787–1813. biomarkersidentifiedfromserumproteomicanalysisforthedetectionof
ShengS,ChenD,VanEykJE(2006)Multidimensionalliquid earlystageovariancancer.CancerRes64(16):5882–5890.
chromatographyseparationofintactproteinsbychromatographic
focusingandreversedphaseofthehumanserumproteome:optimization
Thisworkispublishedunderthestandardlicensetopublishagree-
andproteindatabase.MolCellProteomics5(1):26–34.
SogaK,WatanabeT,AikawaK,ToshimaM,ShibasakiK,AoyagiY(1998) ment. After 12 months the work will become freely available and
Serumdes-gamma-carboxyprothrombinlevelbyamodifiedenzyme the license terms will switch to a Creative Commons Attribution-
immunoassaymethodinhepatocellularcarcinoma:clinicalsignificance NonCommercial-Share Alike 3.0 Unported License.
106 www.bjcancer.com|DOI:10.1038/bjc.2012.494