Research Article Transcriptomic and Immunohistochemical...

Research ArticleTranscriptomic and Immunohistochemical Profiling ofSLC6A14 in Pancreatic Ductal Adenocarcinoma

Alan R Penheiter1 Sibel Erdogan2 Stephen J Murphy1 Steven N Hart3

Joema Felipe Lima1 Fariborz Rakhshan Rohakhtar4 Daniel R OrsquoBrien3

William R Bamlet3 Ryan E Wuertz5 Thomas C Smyrk6 Fergus J Couch7

George Vasmatzis1 Claire E Bender8 and Stephanie K Carlson18

1Department of Molecular Medicine Mayo Clinic 200 First Street SW Rochester MN 55905 USA2Department of Biochemistry and Molecular Biology Mayo Clinic 200 First Street SW Rochester MN 55905 USA3Biomedical Statistics and Informatics Mayo Clinic 200 First Street SW Rochester MN 55905 USA4Medical Genome Facility Mayo Clinic 200 First Street SW Rochester MN 55905 USA5Department of Epidemiology Mayo Clinic 200 First Street SW Rochester MN 55905 USA6Department of Anatomic Pathology Mayo Clinic 200 First Street SW Rochester MN 55905 USA7Department of Laboratory Medicine and Pathology Mayo Clinic 200 First Street SW Rochester MN 55905 USA8Department of Radiology Mayo Clinic 200 First Street SW Rochester MN 55905 USA

Correspondence should be addressed to Stephanie K Carlson scarlsonmayoedu

Received 19 December 2014 Accepted 24 April 2015

Academic Editor Ernesto Picardi

Copyright copy 2015 Alan R Penheiter et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

We used a target-centric strategy to identify transporter proteins upregulated in pancreatic ductal adenocarcinoma (PDAC)as potential targets for a functional imaging probe to complement existing anatomical imaging approaches We performedtranscriptomic profiling (microarray and RNASeq) on histologically confirmed primary PDAC tumors and normal pancreas tissuefrom 33 patients including five patients whose tumors were not visible on computed tomography Target expression was confirmedwith immunohistochemistry on tissue microarrays from 94 PDAC patients The best imaging target identified was SLC6A14 (aneutral and basic amino acid transporter) SLC6A14 was overexpressed at the transcriptional level in all patients and expressed atthe protein level in 95 of PDAC tumors Very little is known about the role of SLC6A14 in PDAC and our results demonstrate thatthis target merits further investigation as a candidate transporter for functional imaging of PDAC

1 Introduction

Early detection and surgical resection of pancreatic ductaladenocarcinoma (PDAC) confined to the pancreas offers thebest hope for cure or extension of lifespan Recent break-throughs in serum profiling most notably mass spectraland antibody array technologies provide hope for screeningpatients with asymptomatic disease [1 2] However evenwithscreening two critical problems remain (1) localization ofsmall or diffusely infiltrative occult lesions in the pancreasand (2) detection of small metastatic deposits

Themajority of large PDACs are detectedwith anatomicalimaging techniques such as computed tomography (CT)

magnetic resonance imaging (MRI) and ultrasound Multi-detector helical CT with intravenous administration of con-trast material is the most commonly used imaging procedureto detect and stage suspected PDAC Diagnostic accuracydecreases however with decreasing tumor size [3ndash5] andin patients with chronic pancreatitis [6 7] There is alsoa subgroup of tumors (about 10) that are isoattenuatingto normal pancreas These are typically diffusely infiltrativerather than mass-forming which renders them invisible onCT despite tumor dimensions greater than the expectedsize for detection [8 9] Functional imaging with 2-deoxy-2-[18F]fluoro-D-glucose positron emission tomogra-phy (18FDG-PET) combined with CT or MRI is a highly

Hindawi Publishing CorporationBioMed Research InternationalVolume 2015 Article ID 593572 10 pageshttpdxdoiorg1011552015593572

2 BioMed Research International

sensitive diagnostic tool for many tumor types but its utilityin PDAC is hampered by low tumor signal-to-backgroundratios that limit its sensitivity for detection of lesions belowthe realized resolution of PET (approximately 1 cm) A newfunctional imaging probe that selectively targets PDAC withhigh sensitivity is a critical unmet need in PETCT or PETMRI that could transform patient management by allowingearlier PDAC detection and surgical intervention and thatcould improve preoperative staging of disease to decreasethe number of unwarranted surgeries in patients who mightbenefit from experimental systemic therapy

One of the greatest success stories in functional imagingis sodium iodide symporter- (NIS-) mediated imaging forthyroid cancer NIS is a membrane transporter from the SLCfamily (SLC5A5) that is responsible for the uptake of iodinein thyroid follicular cells as the first step in the synthesis ofthyroid hormone The combined action of NIS and a secondtrapping (or organification) step allow thyroid cancer cellsto accumulate radiolabeled iodine gt1000-fold above bloodlevels at 48 hours after administration [10] This efficient2-compartment system permits highly sensitive detectionof primary and metastatic thyroid cancer deposits withgamma-camera single-photon emission computed tomogra-phy (SPECT) and PET imaging as well as the effective useof 131I (iodine 131) radiotherapy for thyroid cancer Anotherwell-characterized SLC family member useful in functionalimaging is SLC2A1 (GLUT1) a major glucose membranetransporter that is upregulated in tumor cells (glycolytic ratesin tumors can be more than 30-fold higher than in normalcells) [11 12] 18FDG is a glucose analog that is taken up byGLUT 1 in tumor cells and phosphorylated by hexokinasethus trapping it in the cell This accumulation of 18FDGwithin tumor cells serves as the basis for PET imaging of awide variety of cancers

Our previous work with NIS [13ndash15] and the success ofSLC transporter-mediated functional imaging in other tumormodels led us to investigate the potential of SLC-mediatedfunctional imaging for PDAC In this study we performedgene expression profiling in human PDAC samples usinglaser capture microdissection (LCM) and RNA sequencing(RNASeq) to search for upregulated SLC transporters inPDAC compared with normal pancreatic tissue and normalliver tissue (the major site of PDAC metastases) Our tran-scriptomic results were validated at the protein expressionlevel using immunohistochemistry and tissue microarrayanalysis The top candidate transporter (upregulated at thetranscriptional level in all patients studied) identified in oursearch was SLC6A14

SLC6A14 (also known as ATB0 +) is a Na+ and Clminusdependent solute transport system in the SLC6 family thatis capable of active transport of all neutral and basic aminoacids except glutamate and aspartate which are nonessential[16 17] Because of its broad substrate specificity and the factthat the transporter is expressed only at low levels in manynormal tissues SLC6A14 has received considerable attentionfor its potential as a drug delivery system Recent studieshave shown the anticancer potential of inhibitors of thistransporter system [12 18ndash21] SLC6A14 has been shown to beupregulated in primary and metastatic colorectal cancer [21]

cervical cancer [22] and estrogen receptor-positive breastcancer [18] Kandaswamy et al recently determined thatSLC6A14 is upregulated severalfold in cultured pancreaticcancer cell lines (compared with the normal ductal cell line)and that blockade of the transporter with 120572-methyltryp-tophan led to decreased proliferation of PDAC cells in vitroand in a mouse xenograft model [23] However very little isknown about SLC6A14 expression subcellular localizationand transporter function at the protein level in any of thesetumor types The goal of our study was to further investigatethe expression of SLC6A14 in primary human PDAC tissuesand determine its potential as a target for the development ofa new functional diagnostic imaging probe

2 Materials and Methods

In addition to our retrospective review of patientsrsquo medicalrecords we analyzed PDAC specimens contained in theinstitutional pancreatic cancer tissue registry after approvalof our study protocol by theMayo Clinic Institutional ReviewBoard Between January 1 2002 and September 30 2012 atotal of 616 patients with surgically resected PDAC had con-sented to be included in the pancreatic cancer tissue registrywhich is maintained by the Mayo Clinic SPORE (SpecializedProgram of Research Excellence (National Cancer Institute))in Pancreatic Cancer Tissues from a total of 33 patients wereincluded in our transcriptomic studies and tissues from atotal of 94 patients were included in our protein expression-profiling studies

21 Microarray Analysis We first performed microarrayanalysis on frozen bulk PDAC tumor samples from 12 patientsand 5 (matched) normal human pancreas samples Tissueswere mounted on glass slides and macrodissected with razorblades RNA was isolated and characterized as describedbelow with the LCM-captured samples ComplementaryDNA (cDNA) libraries constructed from the purified RNAwere run on a U133 plus 20 Affymetrix (Santa Clara CA)array

22 Laser Capture Microdissection (LCM) LCM was per-formed on frozen pancreatic tissues from an additional21 patients including 5 with CT-isoattenuating tumors(described below) from the pancreatic cancer tissue registrywith histologically confirmed PDAC for whom both tumorand normal pancreas tissues were available Tissues were cutinto 10 120583m sections on nuclease-free polyethylene naphtha-late (PEN) membrane slides (Carl Zeiss Thornword NY)prechilled at 4∘C After tissues were sectioned the slides wereplaced on dry ice and then stored at minus80∘C Slides to bemicrodissected were removed from freezer storage stainedwith cresyl violet in accordance with the Ambion LCMStaining Kit protocol (Life Technologies Grand Island NY)dehydrated with two 90-second xylene washes and allowedto air-dry for 5 minutes Regions of interest from the dehy-drated tissue sectionswere subsequentlymicrodissected fromthe PEN membrane slides using an Arcturus Veritas LaserCapture Microdissection instrument (Life Technologies)

BioMed Research International 3

utilizing both its ultraviolet cutting laser and its infraredcapture lasers Captured cells were collected onto CapSureMacro LCM caps (Life Technologies) The amount of tissueon each cap was approximately one-third to one-half the areaof the LCM cap Multiple LCM caps were used as neededfor each patient case After microdissection each LCM capwas immediately placed on top of a 05mL polymerase chainreaction (PCR) tube containing 125 120583L ofQIAzol lysis reagent(QIAGEN Germantown MD) which was then inverted toallow the reagent to cover the captured tissue before it wasplaced on dry ice These samples were then stored at minus80∘Cuntil processed for RNA extraction

23 Patient and ImagingDataCollection for CT-IsoattenuatingTumors To ensure that our results were applicable to CT-isoattenuating tumors and that these tumors were similar atthe transcriptional level to the classic hypoattenuating PDACtumors we searched the pancreatic cancer tissue registry toidentify a subset of CT-isoattenuating cases We searchedfor patients who had histological confirmation of pancreaticadenocarcinoma on resected tissue but did not have a massevident on a preoperative contrast-enhanced CT scan Of the616 patients in the registry 49 had both frozen tumor andfrozen normal pancreatic tissue as well as a CT scan prior totreatment which were available for review

All 49 patients had undergone high-resolution thin-section dynamic intravenous contrast-enhanced imaging onmultidetector row CT scanners using a pancreatic protocolthat included biphasic imaging in the pancreatic phase(approximately 45 seconds after contrast injection) and in theportal venous phase (70 seconds after contrast injection) Ini-tial review of the CT images by an abdominal radiologist (SK C) on our institutionrsquos imaging archive software identified5 patients (3 men 2 women mean age 59 years (range 47to 76 years)) with pancreatic tumors that appeared to be CT-isoattenuating (imperceptible changes in contrast to the adja-cent pancreatic parenchyma upon visual inspection in boththe pancreatic and portal venous phases of contrast enhance-ment) so they were included in this study

24 RNA Isolation RNA was extracted from all dissectedtissues using the miRNeasy Mini Kit protocol and the auto-mated QIAcube instrument (QIAGEN) For cases in whichmultiple LCM caps were used to collect an identical cell typethe extracted material from those LCM caps was consol-idated

25 mRNASeq Library Preparation and Sequencing of Forma-lin-Fixed Paraffin-Embedded-Derived RNA RNA librarieswere prepared according to the manufacturerrsquos instructionsfor the NuGEN Ovation RNASeq FFPE (formalin-fixedparaffin-embedded) and Ultralow Library Systems (San Car-los CA) Briefly first strand cDNA was generated from 10 ngof total RNA using DNARNA chimeric primers and reversetranscriptase to create a cDNARNA hybrid Second strandcDNA was then synthesized containing a DNARNA duplexThe resulting double-stranded cDNAmolecule was amplifiedby Single Primer Isothermal Amplification (SPIA) using

a chimeric SPIA primer DNA polymerase and RNase H(NuGEN) After amplification the products were modifiedby random priming and extension to create double-strandedproducts suitable for generating libraries for sequencingThedouble-stranded products underwent blunt-end repair andunique index molecules were ligated to the 51015840 and 31015840 ends ofeach fragment to facilitate PCRamplification of the fragmentsand to produce the final library The concentration and sizedistribution of the resulting libraries were determined on anAgilent Bioanalyzer DNA 1000 chip (Santa Clara CA) andthen confirmed by Qubit fluorometry (Life Technologies)Libraries were loaded onto paired-end flow cells with 3samples per lane at concentrations of 8 to 10 pM to generatecluster densities of 700000mm2 following the standardprotocol using the Illumina cBot and cBot paired-end clusterkit version 3 (San Diego CA) The flow cells were sequencedas 51 times 2 paired-end reads on an Illumina HiSeq 2000 usingTruSeq SBS sequencing kit version 3 and SCS version 148data collection software Base-calling was performed usingIlluminarsquos RTA version 11242

26 RNASeq Data Acquisition and Analysis TheMayo Anal-ysis Pipeline for RNA Sequencing (MAP-RSeq) developedby the Bioinformatics Core at Mayo Clinic was used toperform the thorough processing of the paired-end RNASeqdata MAP-RSeq integrates a suite of open source bioinfor-matics tools with in-house developed methods to analyzepaired-end RNASeq data The application processes thereads produced by the sequencer in the following mannerquality control genomic alignments reference and noveltranscriptomic junction alignments alignment cleanupidentification of genomic features per sample and summaryof data across samples Postanalysis quality control measureswere taken computationally and manually to verify thatthe analysis was successful Such computational measuresincluded estimating the distance between paired-end readssequencing depths at alternate splicing sites the rate ofduplicate reads and evaluating coverage across genes Theseand other metrics were measured using RSeQC software[24] Paired-end reads were aligned by TopHat 206 [25]against the hg19 genome build with the Bowtie1 aligner [26]as a backbone Gene counts for evaluating RNA fold changewithin and across samples were generated using HTseq soft-ware (httpwwwhuberembldeusersandersHTSeqdocoverviewhtml) and the gene annotation files used wereobtained from Illumina (httpcole-trapnell-labgithubiocufflinks)

27 Tissue Microarray Screening An adenocarcinoma tissuemicroarray (TMA) containing samples from 140 patientswithout prior chemotherapy was stained with hematoxylin-eosin or colorimetrically developed with antibodies againstSLC6A14 or KRT19 (positive control cytokeratin 19) in thePathology Research Core of Mayo Clinic Rochester MNTMA slides were placed in the BOND III (Leica BiosystemsChicago IL USA) stainer for online processing Slides weretreated with Epitope Retrieval 2 solution for 20 minutesstained with rabbit polyclonal anti-SLC6A14 (Sigma St


SLC16A6SLC30A2SLC41A1

SLC1A2SLCO4A1SLC6A14SLC11A1

SLC1A2SLC16A6SLC22A3SLC43A1SLC16A3

SLC2A3SLC16A1

SLC2A1SLC7A5

SLC39A5

Nor

mal

1

Nor

mal

2

Nor

mal

3

Nor

mal

4

Nor

mal

5

Tum

or 1

Tum

or 2

Tum

or 3

Tum

or 4

Tum

or 5

Tum

or 6

Tum

or 7

Tum

or 8

Tum

or 9

Tum

or 1

0

Tum

or 1

1

Tum

or 1

2

Tum

or 1

3

Figure 1Heatmap of transcriptomic analysis Analysis shows results for pancreatic ductal adenocarcinoma samples from 12 patients (samples5 and 6 are from 1 patient) and 5 matched bulk normal pancreas patient samples Samples were macrodissected and run on an AffymetrixU133 plus 20 array Black indicates low level of transcripts and white indicates high level of transcripts SLC1A2 and SLC6A16 were detectedby two probes in the array All SLC transporters shown were significantly (119875 lt 005) upregulated or downregulated compared to normalpancreas

Louis MO HPA003193) at 1 100 dilution (in Bkg reducingdiluent Dako Carpinteria CA S3022) for 30 minutes ormouse monoclonal anti-KRT19 (Dako RCK108) at 1 200 for15 minutes Detection was achieved using the Polymer RefineDetection kit following the manufacturerrsquos instructions(Leica Biosystems) Counterstaining was performed for 5minutes with hematoxylin-eosin Slides were dehydratedthrough increasing concentrations of alcohol cleared inxylene and coverslipped in xylene-based mounting mediaThe TMAs were evaluated microscopically at 400x magnifi-cation for SLC6A14 expression by a trained pancreatic pathol-ogist and were scored as strong moderate weak or absentSubcellular localization of the staining was noted for eachcore In the final analysis 174 tumor cores from 94 uniquepatients retained sufficient tissue on the slide with a largeenough region of histologically identifiable adenocarcinomato permit scoring Information across the multiple evaluablecores per patient was reduced to one observation per uniquesubject by using the core which stained with the highestexpression

3 Results

31 Transcriptomic Profiling Studies Our work with NIS ledus to investigate the differential expression of othermetabolicgenes in the SLC family that may be upregulated in PDACcells compared with normal pancreas We first performed apilot study of 12 macrodissected frozen human PDAC sam-ples and 5 (matched) normal human pancreas samples We

isolated RNA from these samples and ran libraries on a U133plus 20 Affymetrix array In these samples we found 15 SLCtransporters that were overexpressed in bulk tumor tis-sue versus normal tissue (Figure 1) The most promisingupregulated candidate gene was the amino acid transporterSLC6A14 This technique also identified the glucose trans-porter SLC2A1 and the lactate exporter SLC16A3 (Figure 1)SLC2A1 and SLC16A3 along with hexokinase II facilitatetrapping of 18FDG However PDAC tumors have a largestromal cell component and it was not clear which of theSLCs were specifically upregulated in tumor epithelial cells orwhich were simply present on expanding or invading stromalcells such as myofibroblasts macrophages and lymphocytes

To specifically address the issue of cell-type contributionwe used LCM to isolate PDAC cells and control cells in anadditional 21 PDAC patient samples including 5 tumor andmatched normal pancreas frozen specimens from patientswith CT-isoattenuating tumors that were not visualized onpreoperative CT imaging studies (Figure 2)We then purifiedRNA from the samples and performed RNASeq analysis Aswe found in the bulk tumor samples SLC6A14 was highlyoverexpressed in this patient group after LCM (Table 1)SLC6A14 was overexpressed at least 2-fold in all 21 patientsversus normal pancreas (Table 1) For the comparison withtumor tissue we harvested 3 different control tissues Adja-cent normal ducts were collected from all 16 patients notselected for CT-isoattenuating characteristics Acinar tissueadjacent to the tumor was also recovered from the slides of 5patientsThese tumor-adjacent controls provide some insight


(a) (b)

Figure 2 Contrast-enhanced computed tomography images obtained during the pancreatic phase (a) Axial CT image of the pancreatic headshows abrupt termination of amarkedly dilated pancreatic duct (arrow) due to a surgically proven pancreatic cancer causing duct obstructionNo mass is visualized in the region of the duct-obstructing tumor (arrowhead) (b) Image of the pancreatic head 1 cm inferior to image (a)shows homogeneous enhancement without evidence of hypoattenuation changes ormass effect despite the presence of large (39 cm) diffuselyinfiltrating pancreatic ductal adenocarcinoma tumor

into whether a gene is upregulated specifically in adenocarci-noma cells or is simply a response of epithelial cells to thetumor microenvironment For the CT-isoattenuating caseswe used acinar tissue from a region of the resected pancreasdistant from the tumor The distant acinar control wasincluded because it best reflects the background one wouldencounter when imaging a pancreas from a patient withoutcancer

32 Protein Expression Profiling Studies Next we sought todeterminewhether SLC6A14 transcript upregulation resultedin relevant upregulation at the protein level For this analysiswe used a pancreatic cancer TMA and immunohistochem-istry with an antibody against SLC6A14 (Figure 3) A total of174 tumor cores from 94 patients were evaluated for SLC6A14staining intensity percentage of tumor cells stained forSLC6A14 and subcellular localization of SLC6A14 stainingOf the 174 cores 16 were scored as strong staining 41 as mod-erate 97 as weak and 20 as negative for SLC6A14 The 154cores with weak to strong staining were then scored for thepercentage of tumor cells within the core that were stained forSLC6A14 Of these 116 cores had gt75 of their tumor cellsstained whereas 16 14 and 8 cores exhibited staining in50ndash75 25ndash50 and lt25 of tumor cells respectivelyStaining was predominantly cytoplasmic in 153 cores distinctplasma membrane staining was found in only 1 core Of the153 cores with predominantly cytoplasmic staining 3 werenoted to also have staining of the plasma membrane and10 were noted to have additional nuclear staining Figure 3illustrates a core that was scored as strong for SLC6A14 alongwith an adjacent section of the TMA developed for KRT19(a marker for all pancreatic ductal cells) immunohistochem-istry KRT19 staining was also scored as strong

Whole sections from the 5 CT-isoattenuating cases werealso processed for SLC6A14 immunohistochemistry Fourcases were found to have moderate cytoplasmic staining and1 case was found to have strong cytoplasmic staining A rep-resentative case of moderate cytoplasmic staining is shown inFigure 4 This case shows the diffusely infiltrating pathology

Table 1 Gene normalized levels of SLC6A14 mRNA

Patient Adjacentnormalduct

Adjacentnormalacinar

Distantnormalacinar

Tumor Tumornormal1

1 254 13868 54602 084 3616 43053 017 151 8884 116 11989 103355 1031 013 6244 6066 134 010 1182 8827 029 783 27008 110 2973 27039 666 037 5433 81610 022 34269 15576811 038 013 2201 579212 260 2722 104713 226 1124 49714 894 1026 114815 036 3456 960016 444 003 1000 22517 (CT-iso) 003 1310 4366718 (CT-iso) 003 2256 7520019 (CT-iso) 010 10303 10303020 (CT-iso) 002 43930 219650021 (CT-iso) 004 1549 38725CT-iso computed tomography-isoattenuating1Adjacent normal duct was used for the calculation in patients 1ndash16 Distantnormal acinar was used for patients 17ndash21 Values are in RPKM (mappedreads per kilobase per million mapped reads)

that is a hallmark of CT-isoattenuating pancreatic adenocar-cinoma histology [9 27]

No staining of pancreatic epithelial cells (acinar cells orduct cells) or hepatocytes was observed on control tissuecores included on the TMA (3 cores of normal pancreas and 3cores of normal liver)Occasionalmoderate to strong staining


KRT19

SLC6A14

Figure 3 Representative core staining of pancreatic adenocarcinoma by tissue microarray Top panel was stained with an antibody to KRT19(a marker for all pancreatic ductal cells) Bottom panel was stained with an antibody to SLC6A14

(a) (b)

(c) (d)

Figure 4 Histology of CT-isoattenuating tumor (andashc) Note the diffusively infiltrative invasive adenocarcinoma with ductal morphology(open arrows) the relatively intact lobular architecture of the residual normal pancreas (solid arrows) and the regions of fibrosis with sparsecellularity (area between arrows) Scale bar equals 1mm (a) KRT19 immunohistochemistry of a formalin-fixed paraffin-embedded (FFPE)section (b) Hematoxylin-eosin stain of a frozen section from the same patient An adjacent section from this same frozen block was used forlaser capturemicrodissection (c) An adjacent section from the FFPE block shown in panel (a) stained for SLC6A14 (d) Higher magnificationof region of panel (c) (rectangle) showsmoderate cytoplasmic SLC6A14 staining in adenocarcinoma cells (solid arrowhead) while the adjacentnormal acinar cells and a normal duct (open arrowhead) are unstained Strong cytoplasmicmembranous staining was also observed inscattered stromal cells Scale bar equals 200 120583m


Table 2 SLC6A14 IHC staining intensity

0 none (119873 = 5) 1 weak (119873 = 50) 2 moderate (119873 = 28) 3 strong (119873 = 11) Total (119873 = 94) 119901

valueSurvival 01181119873 5 42 24 8 79Events 5 36 20 7 68Median survivaldays 6330 (4870ndash25320) 7330 (4050ndash12470) 5950 (3340ndash12810) 3660 (2530ndash6410) 5790 (4190ndash9620)

2 Yr survival rate 400 (00ndash829) 512 (359ndash665) 458 (259ndash658) 00 442 (331ndash553)Year 2119873 at risk 2 21 11 0 34

Age of onset 01099119873 5 37 22 8 72Mean (SD) 730 (70) 676 (106) 665 (92) 611 (90) 669 (100)Median 750 690 700 595 690Q1 Q3 740 760 610 750 600 750 540 680 600 750Range (610ndash790) (370ndash850) (480ndash760) (500ndash760) (370ndash850)Median age ofonset 750 (610ndash790) 690 (640ndash730) 700 (600ndash740) 595 (500ndash690) 690 (660ndash720) 00978

Sex 02142Missing 0 8 4 3 15Female 1 (200) 25 (595) 10 (417) 3 (375) 39 (494)Male 4 (800) 17 (405) 14 (583) 5 (625) 40 (506)

Obesity 02358Missing 1 21 9 5 36BMI lt 30 2 (500) 24 (828) 17 (895) 4 (667) 47 (810)BMI 30+ 2 (500) 5 (172) 2 (105) 2 (333) 11 (190)

Diabetes self-reported 06592Missing 0 8 4 3 15No DM 5 (1000) 33 (786) 18 (750) 6 (750) 62 (785)DM 0 (00) 9 (214) 6 (250) 2 (250) 17 (215)

Pancreatitisself-reported 08622

Missing 0 8 4 3 15No pancreatitis 4 (800) 31 (738) 18 (750) 7 (875) 60 (759)Pancreatitis 1 (200) 11 (262) 6 (250) 1 (125) 19 (241)

Tumor grade 059172 0 (00) 8 (160) 1 (36) 1 (91) 10 (106)3 4 (800) 31 (620) 20 (714) 6 (545) 61 (649)4 1 (200) 11 (220) 7 (250) 4 (364) 23 (245)

was observed in other cell types in the control tissues includ-ing stellate cells in the pancreas and Kupffer cells in the liverLymphocytes within vessels in the control tissues were alsostained for SLC6A14 A complete catalog of normal tissuestaining with the SLC6A14 antibody we used can be foundon the website of the Human Protein Atlas (httpwwwproteinatlasorg) [28]

We also examined the relationship between SLC6A14staining intensity and patient demographics including sur-vival after resection tumor stage and tumor grade (Table 2)For this analysis patients with more than 1 core on the TMAwere grouped by the core with the highest intensity of theSLC6A14 stainingMost (95 (8994)) of the patients had 1 or

more cores exhibiting weak or greater SLC6A14 staining Nosignificant relationship between SLC6A14 staining intensityand patient or histological parameters was found

4 Discussion

Traditional strategies for imaging SLCs (metabolic imag-ing) have followed a probe-centric paradigm Radiolabelledldquonutrientsrdquo were synthesized based on the concept that tumorcells are metabolically active and must upregulate transportand metabolic machinery Trial-and-error biodistributionstudies were used to determine (1) which tumor types werebest visualized by a particular probe (2) whether background


uptake might interfere with localization or (3) whether anunacceptable dose of radiation had been administered to anormal tissue in extreme cases Some minimal mechanisticstudieswith inhibitors (primarily at the cell culture level) havebeen used to identify transporters or transporter familiesresponsible for metabolic imaging probe accumulation butin many cases the transporters responsible for success-ful metabolic imaging are not known This probe-centricapproach has been quite successful Notable achievements areL-[18F-fluoro]dihydroxyphenylalanine [29 30] for detectionof neuroendocrine tumors L-[11C-methyl]methionine [3132] for many types of extracranial tumors and 11C-choline or18F-choline [33 34] for detection of primary and metastaticprostate cancer However for pancreatic cancer none of thesestrategies has been successful enough to reach routine use inclinical practice The main limitation is the high metabolicand biosynthetic activity of the normal pancreas whichresults in massive accumulation of these probes [35ndash38]

Here we used a target-centric strategy using transcrip-tomics and protein expression profiling from large numbersof resected patient tissues The aim was to find a particulartransporter that is specifically upregulated in pancreatic can-cer versus normal pancreas that could serve as a target for thefuture development of a PDAC-selective functional imagingprobe Surprisingly we found that SLC6A14 was overex-pressed at the transcriptional level in all the patient sampleswe studied when these were compared to either microdis-sected normal ducts or macro- or microdissected acinartissue Additionally at the protein level SLC6A14 staining ofweak or greater intensity was observed in 95 of patientsamples with moderate to strong staining in 41 of patientsamples

We also confirmed that SLC6A14 is overexpressed in CT-isoattenuating PDAC a subset of tumors with histologicaland enhancement characteristics that make them notoriouslydifficult to detect with anatomical imaging techniquesWhilegt90 of PDAC tumors (in symptomatic patients) present ata late stage and are readily detectable by dual-phase contrast-enhanced CT 5 to 10 of patients present with tumormasses that are not visualized on CT [8 9 27] In most casesthe CT-isoattenuating tumors on gross pathology are muchlarger than the resolution of CT but are diffusely infiltratingwith sparse tumor tissue and abundant normal tissue andthus they do not appear as a distinct locus of hypoattenuation[9 27]

For the purposes of functional imaging a transporterprotein needs to be expressed on the plasma membranein order to facilitate entry of a probe into the target cellsWhile the function of SLC6A14 is being a plasma membranetransporter staining was predominantly cytoplasmic in themajority of resected PDAC tissues in our study A character-istic of this and other related plasma membrane transportersis a predominantly cytoplasmic localization (most likelysequestered in vesicles) with translocation to the plasmamembrane after a stimulus [39] For SLC6A14 protein kinaseC alpha activation results in a dramatic translocation to themembrane and subsequent increase in transport activity [39]At this point we can only speculate as to why SLC6A14was predominantly cytoplasmic in our PDAC samples One

possibility is that the loss of blood supply prior to surgicalresection may have resulted in the downregulation of signalsrequired for membrane localization of SLC6A14 Interest-ingly in other tumor types (thyroid cervical and squamouscell carcinoma of the head and neck) strong membranestaining has been observed with the same SLC6A14 antibody(httpwwwproteinatlasorg) [28]

Further understanding of the function substrate speci-ficities and tissue distribution of SLC6A14 will aid in thepotential future development of a selective functional imag-ing probe or the discovery of targeted therapeutic agents Anexciting development towards this goal was the recent syn-thesis ofO-2((2-[18F]fluoroethyl)methylamino)ethyltyrosineor ([18F]FEMAET) a cationic amino acid PET probe thatdemonstrates positive uptake in SLC6A14-positive xenografts[40 41] Anticancer strategies (transporter inhibitors tostarve tumor cells of essential amino acids or cytotoxic mole-cules that could be selectively transported by SLC6A14) arealso being actively explored for other tumor types which havebeen shown to overexpress SLC6A14 [18 42]

5 Conclusion

SLC6A14 appears to be a good candidate transporter for fur-ther exploration as a functional imaging target for PDAC thatcould complement existing anatomical imaging techniquesfor both diagnosis and staging

Abbreviations

cDNA Complementary DNACT Computed tomographyFFPE Formalin-fixed paraffin-embedded18FDG 2-Deoxy-2-[18F]fluoro-D-glucose18FDG-PET 2-Deoxy-2-[18F]fluoro-D-glucose

positron emission tomographyLCM Laser capture microdissectionMAP-RSeq Mayo Analysis Pipeline for RNA

SequencingMRI Magnetic resonance imagingNIS Sodium iodide symporterPDAC Pancreatic ductal adenocarcinomaPET Positron emission tomographyRNase H Ribonuclease HRNASeq RNA sequencingSLC Solute carrierSPECT Single-photon emission computed

tomographySPIA Single Primer Isothermal AmplificationSPORE Specialized Program of Research

ExcellenceTMA Tissue microarray

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper


Acknowledgments

This work was supported in part by generous benefactor giftsfrom Ted L Shabert and Louellen Essex PhD and from theDel Droogsma Family an award from the National CancerInstitute (Award no P50CA102701 (Mayo Clinic SPORE inPancreatic Cancer)) the Imaging Biomarker Program withinthe Center for Individualized Medicine at Mayo Clinic andthe Minnesota Partnership for Biotechnology and MedicalGenomics The authors would like to thank Anthony JBlahnik LouAnn A Gross Eileen L Holicky Lorna KLubinski Marie R Passow Nathan G Faiman and Peggy MGosse Rahnenfuehrer for their technical expertise and KarlaJ Kopp Christopher P Kolbert and Dr Gloria M Petersenfor helpful discussion

References

[1] S Tonack C Jenkinson T Cox et al ldquoiTRAQ reveals candidatepancreatic cancer serum biomarkers influence of obstructivejaundice on their performancerdquo British Journal of Cancer vol108 no 9 pp 1846ndash1853 2013

[2] D A Pirman E Efuet X-P Ding et al ldquoChanges in cancercell metabolism revealed by direct sample analysis withMALDImass spectrometryrdquo PLoS ONE vol 8 no 4 Article ID e613792013

[3] K Okano K Kakinoki S Akamoto et al ldquo18F-fuorodeox-yglucose positron emission tomography in the diagnosis ofsmall pancreatic cancerrdquoWorld Journal of Gastroenterology vol17 no 2 pp 231ndash235 2011

[4] Y L Bronstein E M Loyer H Kaur et al ldquoDetection ofsmall pancreatic tumors withmultiphasic helical CTrdquoAmericanJournal of Roentgenology vol 182 no 3 pp 619ndash623 2004

[5] E P Tamm EM Loyer S C Faria D B Evans R AWolff andCCharnsangavej ldquoRetrospective analysis of dual-phaseMDCTand follow-up EUSEUS-FNA in the diagnosis of pancreaticcancerrdquo Abdominal Imaging vol 32 no 5 pp 660ndash667 2007

[6] Y Yamada H Mori S Matsumoto H Kiyosue Y Hori andN Hongo ldquoPancreatic adenocarcinoma versus chronic pancre-atitis differentiation with triple-phase helical CTrdquo AbdominalImaging vol 35 no 2 pp 163ndash171 2010

[7] B Taylor ldquoCarcinoma of the head of the pancreas versuschronic pancreatitis diagnostic dilemma with significant con-sequencesrdquo World Journal of Surgery vol 27 no 11 pp 1249ndash1257 2003

[8] RW Prokesch L C Chow C F Beaulieu R Bammer andR BJeffrey Jr ldquoIsoattenuating pancreatic adenocarcinoma at multi-detector rowCT secondary signsrdquoRadiology vol 224 no 3 pp764ndash768 2002

[9] J H Kim S H Park E S Yu et al ldquoVisually isoattenuating pan-creatic adenocarcinoma at dynamic-enhanced CT frequencyclinical and pathologic characteristics and diagnosis at imagingexaminationsrdquo Radiology vol 257 no 1 pp 87ndash96 2010

[10] J C Sisson ldquoPractical dosimetry of 131I in patients with thyroidcarcinomardquo Cancer Biotherapy and Radiopharmaceuticals vol17 no 1 pp 101ndash105 2002

[11] L He K Vasiliou and D W Nebert ldquoAnalysis and updateof the human solute carrier (SLC) gene superfamilyrdquo HumanGenomics vol 3 no 2 pp 195ndash206 2009

[12] V Ganapathy MThangaraju and P D Prasad ldquoNutrient trans-porters in cancer relevance towarburg hypothesis and beyondrdquoPharmacology andTherapeutics vol 121 no 1 pp 29ndash40 2009

[13] A R Penheiter S J Russell and S K Carlson ldquoThe sodiumiodide symporter (NIS) as an imaging reporter for gene viraland cell-based therapiesrdquo Current Gene Therapy vol 12 no 1pp 33ndash47 2012

[14] A R Penheiter D Dingli C E Bender S J Russell and SK Carlson ldquoMonitoring the initial delivery of an oncolyticmeasles virus encoding the human sodium iodide symporter tosolid tumors using contrast-enhanced computed tomographyrdquoJournal of Gene Medicine vol 14 no 9-10 pp 590ndash597 2012

[15] A R Penheiter G E Griesmann M J Federspiel D DingliS J Russell and S K Carlson ldquoPinhole micro-SPECTCTfor noninvasive monitoring and quantitation of oncolytic virusdispersion andpercent infection in solid tumorsrdquoGeneTherapyvol 19 no 3 pp 279ndash287 2012

[16] A B Pramod J Foster L Carvelli and L K Henry ldquoSLC6transporters structure function regulation disease associationand therapeuticsrdquoMolecular Aspects of Medicine vol 34 no 2-3 pp 197ndash219 2013

[17] S Broer and U Gether ldquoThe solute carrier 6 family of trans-portersrdquoBritish Journal of Pharmacology vol 167 no 2 pp 256ndash278 2012

[18] S Karunakaran S Ramachandran V Coothankandaswamy etal ldquoSLC6A14 (ATB 0+) protein a highly concentrative andbroad specific amino acid transporter is a novel and effectivedrug target for treatment of estrogen receptor-positive breastcancerrdquo Journal of Biological Chemistry vol 286 no 36 pp31830ndash31838 2011

[19] M E Ganapathy and V Ganapathy ldquoAmino acid transporterATB0+ as a delivery system for drugs and prodrugsrdquo CurrentDrug Targets Immune Endocrine amp Metabolic Disorders vol 5no 4 pp 357ndash364 2005

[20] T Hatanaka T Nakanishi W Huang et al ldquoNa+- and Clminus-coupled active transport of nitric oxide synthase inhibitors viaamino acid transport system B0+rdquo Journal of Clinical Inves-tigation vol 107 no 8 pp 1035ndash1043 2001

[21] N Gupta S Miyauchi R G Martindale et al ldquoUpregulationof the amino acid transporter ATB0+ (SLC6A14) in colorectalcancer and metastasis in humansrdquo Biochimica et BiophysicaActamdashMolecular Basis of Disease vol 1741 no 1-2 pp 215ndash2232005

[22] N Gupta P D Prasad S Ghamande et al ldquoUp-regulation ofthe amino acid transporter ATB0+ (SLC6A14) in carcinoma ofthe cervixrdquoGynecologic Oncology vol 100 no 1 pp 8ndash13 2006

[23] V C Kandaswamy P D Puttur M Thangaraju and V Ganap-athy ldquoAn amino acid transporter is a novel and effective drugtarget for treatment of pancreatic cancerrdquo in Proceedings of theAACR Annual Meeting Lake Tahoe Nev USA April 2012

[24] L Wang S Wang and W Li ldquoRSeQC quality control of RNA-seq experimentsrdquo Bioinformatics vol 28 no 16 pp 2184ndash21852012

[25] C Trapnell L Pachter and S L Salzberg ldquoTopHat discoveringsplice junctions with RNA-Seqrdquo Bioinformatics vol 25 no 9pp 1105ndash1111 2009

[26] B Langmead C Trapnell M Pop and S L Salzberg ldquoUltrafastand memory-efficient alignment of short DNA sequences tothe human genomerdquo Genome Biology vol 10 no 3 article R252009


[27] S H Yoon J M Lee J Y Cho et al ldquoSmall (le 20mm) pancre-atic adenocarcinomas Analysis of enhancement patterns andsecondary signs with multiphasic multidetector CTrdquo Radiologyvol 259 no 2 pp 442ndash452 2011

[28] M Uhlen P Oksvold L Fagerberg et al ldquoTowards aknowledge-based human protein atlasrdquo Nature Biotechnologyvol 28 no 12 pp 1248ndash1250 2010

[29] F Montravers K Kerrou V Nataf et al ldquoImpact of fluorodihy-droxyphenylalanine-(18F) positron emission tomography onmanagement of adult patients with documented or occultdigestive endocrine tumorsrdquo Journal of Clinical Endocrinologyamp Metabolism vol 94 no 4 pp 1295ndash1301 2009

[30] D Taıeb L Tessonnier F Sebag et al ldquoThe role of 18F-FDOPA and 18F-FDG-PET in the management of malignantand multifocal phaeochromocytomasrdquo Clinical Endocrinologyvol 69 no 4 pp 580ndash586 2008

[31] S Leskinen-Kallio P Lindholm M Lapela H Joensuu andE Nordman ldquoImaging of head and neck tumors with positronemission tomography and [11C]methioninerdquo International Jour-nal of Radiation Oncology Biology Physics vol 30 no 5 pp1195ndash1199 1994

[32] O S Nettelbladt A E Sundin S O Valind et al ldquoCombinedfluorine-18-FDG and carbon-11-methionine PET for diagno-sis of tumors in lung and mediastinumrdquo Journal of NuclearMedicine vol 39 no 4 pp 640ndash647 1998

[33] T Hara N Kosaka and H Kishi ldquoPET imaging of prostatecancer using carbon-11-cholinerdquo Journal of Nuclear Medicinevol 39 no 6 pp 990ndash995 1998

[34] T R DeGrado R E Coleman S Wang et al ldquoSynthesisand evaluation of 18F-labeled choline as an oncologic tracerfor positron emission tomography initial findings in prostatecancerrdquo Cancer Research vol 61 no 1 pp 110ndash117 2001

[35] L Tessonnier F Sebag C Ghander et al ldquoLimited value of 18F-F-DOPA PET to localize pancreatic insulin-secreting tumors inadults with hyperinsulinemic hypoglycemiardquo Journal of ClinicalEndocrinology amp Metabolism vol 95 no 1 pp 303ndash307 2010

[36] O Schillaci F Calabria M Tavolozza et al ldquo18F-cholinePETCT physiological distribution and pitfalls in image inter-pretation experience in 80 patients with prostate cancerrdquoNuclear Medicine Communications vol 31 no 1 pp 39ndash452010

[37] T Tolvanen T Yli-Kerttula T Ujula et al ldquoBiodistributionand radiation dosimetry of [11C]choline a comparison betweenrat and human datardquo European Journal of Nuclear Medicine ampMolecular Imaging vol 37 no 5 pp 874ndash883 2010

[38] C Plathow andW AWeber ldquoTumor cell metabolism imagingrdquoJournal of Nuclear Medicine vol 49 supplement 2 pp 3Sndash63S2008

[39] Ł Samluk M Czeredys K Skowronek and K A NałeczldquoProtein kinase C regulates amino acid transporter ATB0+rdquoBiochemical andBiophysical ResearchCommunications vol 422no 1 pp 64ndash69 2012

[40] A Chiotellis A Muller K Weyermann et al ldquoSynthesisand preliminary biological evaluation of O-2((2-[18F]fluor-oethyl)methylamino)ethyltyrosine ([18F]FEMAET) as a poten-tial cationic amino acid PET tracer for tumor imagingrdquo AminoAcids vol 46 no 8 pp 1947ndash1959 2014

[41] AMuller AChiotellis C Keller et al ldquoImaging tumourATB0+transport activity by PET with the cationic amino acid O-2((2-[18F]fluoroethyl)methyl-amino)ethyltyrosinerdquoMolecular Imag-ing and Biology vol 16 no 3 pp 412ndash420 2014

[42] B M Rotoli O Bussolati R Sala G C Gazzola and VDallrsquoAsta ldquoThe transport of cationic amino acids in human air-way cells expression of system y+L activity and transepithelialdelivery of NOS inhibitorsrdquoTheFASEB Journal vol 19 no 7 pp810ndash812 2005

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom


sensitive diagnostic tool for many tumor types but its utilityin PDAC is hampered by low tumor signal-to-backgroundratios that limit its sensitivity for detection of lesions belowthe realized resolution of PET (approximately 1 cm) A newfunctional imaging probe that selectively targets PDAC withhigh sensitivity is a critical unmet need in PETCT or PETMRI that could transform patient management by allowingearlier PDAC detection and surgical intervention and thatcould improve preoperative staging of disease to decreasethe number of unwarranted surgeries in patients who mightbenefit from experimental systemic therapy

One of the greatest success stories in functional imagingis sodium iodide symporter- (NIS-) mediated imaging forthyroid cancer NIS is a membrane transporter from the SLCfamily (SLC5A5) that is responsible for the uptake of iodinein thyroid follicular cells as the first step in the synthesis ofthyroid hormone The combined action of NIS and a secondtrapping (or organification) step allow thyroid cancer cellsto accumulate radiolabeled iodine gt1000-fold above bloodlevels at 48 hours after administration [10] This efficient2-compartment system permits highly sensitive detectionof primary and metastatic thyroid cancer deposits withgamma-camera single-photon emission computed tomogra-phy (SPECT) and PET imaging as well as the effective useof 131I (iodine 131) radiotherapy for thyroid cancer Anotherwell-characterized SLC family member useful in functionalimaging is SLC2A1 (GLUT1) a major glucose membranetransporter that is upregulated in tumor cells (glycolytic ratesin tumors can be more than 30-fold higher than in normalcells) [11 12] 18FDG is a glucose analog that is taken up byGLUT 1 in tumor cells and phosphorylated by hexokinasethus trapping it in the cell This accumulation of 18FDGwithin tumor cells serves as the basis for PET imaging of awide variety of cancers

Our previous work with NIS [13ndash15] and the success ofSLC transporter-mediated functional imaging in other tumormodels led us to investigate the potential of SLC-mediatedfunctional imaging for PDAC In this study we performedgene expression profiling in human PDAC samples usinglaser capture microdissection (LCM) and RNA sequencing(RNASeq) to search for upregulated SLC transporters inPDAC compared with normal pancreatic tissue and normalliver tissue (the major site of PDAC metastases) Our tran-scriptomic results were validated at the protein expressionlevel using immunohistochemistry and tissue microarrayanalysis The top candidate transporter (upregulated at thetranscriptional level in all patients studied) identified in oursearch was SLC6A14

SLC6A14 (also known as ATB0 +) is a Na+ and Clminusdependent solute transport system in the SLC6 family thatis capable of active transport of all neutral and basic aminoacids except glutamate and aspartate which are nonessential[16 17] Because of its broad substrate specificity and the factthat the transporter is expressed only at low levels in manynormal tissues SLC6A14 has received considerable attentionfor its potential as a drug delivery system Recent studieshave shown the anticancer potential of inhibitors of thistransporter system [12 18ndash21] SLC6A14 has been shown to beupregulated in primary and metastatic colorectal cancer [21]

cervical cancer [22] and estrogen receptor-positive breastcancer [18] Kandaswamy et al recently determined thatSLC6A14 is upregulated severalfold in cultured pancreaticcancer cell lines (compared with the normal ductal cell line)and that blockade of the transporter with 120572-methyltryp-tophan led to decreased proliferation of PDAC cells in vitroand in a mouse xenograft model [23] However very little isknown about SLC6A14 expression subcellular localizationand transporter function at the protein level in any of thesetumor types The goal of our study was to further investigatethe expression of SLC6A14 in primary human PDAC tissuesand determine its potential as a target for the development ofa new functional diagnostic imaging probe

2 Materials and Methods

In addition to our retrospective review of patientsrsquo medicalrecords we analyzed PDAC specimens contained in theinstitutional pancreatic cancer tissue registry after approvalof our study protocol by theMayo Clinic Institutional ReviewBoard Between January 1 2002 and September 30 2012 atotal of 616 patients with surgically resected PDAC had con-sented to be included in the pancreatic cancer tissue registrywhich is maintained by the Mayo Clinic SPORE (SpecializedProgram of Research Excellence (National Cancer Institute))in Pancreatic Cancer Tissues from a total of 33 patients wereincluded in our transcriptomic studies and tissues from atotal of 94 patients were included in our protein expression-profiling studies

21 Microarray Analysis We first performed microarrayanalysis on frozen bulk PDAC tumor samples from 12 patientsand 5 (matched) normal human pancreas samples Tissueswere mounted on glass slides and macrodissected with razorblades RNA was isolated and characterized as describedbelow with the LCM-captured samples ComplementaryDNA (cDNA) libraries constructed from the purified RNAwere run on a U133 plus 20 Affymetrix (Santa Clara CA)array

22 Laser Capture Microdissection (LCM) LCM was per-formed on frozen pancreatic tissues from an additional21 patients including 5 with CT-isoattenuating tumors(described below) from the pancreatic cancer tissue registrywith histologically confirmed PDAC for whom both tumorand normal pancreas tissues were available Tissues were cutinto 10 120583m sections on nuclease-free polyethylene naphtha-late (PEN) membrane slides (Carl Zeiss Thornword NY)prechilled at 4∘C After tissues were sectioned the slides wereplaced on dry ice and then stored at minus80∘C Slides to bemicrodissected were removed from freezer storage stainedwith cresyl violet in accordance with the Ambion LCMStaining Kit protocol (Life Technologies Grand Island NY)dehydrated with two 90-second xylene washes and allowedto air-dry for 5 minutes Regions of interest from the dehy-drated tissue sectionswere subsequentlymicrodissected fromthe PEN membrane slides using an Arcturus Veritas LaserCapture Microdissection instrument (Life Technologies)














SLC2A3SLC16A1

SLC2A1SLC7A5

SLC39A5

Nor

mal

1

Nor

mal

2

Nor

mal

3

Nor

mal

4

Nor

mal

5

Tum

or 1

Tum

or 2

Tum

or 3

Tum

or 4

Tum

or 5

Tum

or 6

Tum

or 7

Tum

or 8

Tum

or 9

Tum

or 1

0

Tum

or 1

1

Tum

or 1

2

Tum

or 1

3



3 Results





(a) (b)








Distantnormalacinar

Tumor Tumornormal1





KRT19

SLC6A14


(a) (b)

(c) (d)













Tumor grade 059172 0 (00) 8 (160) 1 (36) 1 (91) 10 (106)3 4 (800) 31 (620) 20 (714) 6 (545) 61 (649)4 1 (200) 11 (220) 7 (250) 4 (364) 23 (245)




4 Discussion









5 Conclusion


Abbreviations









Acknowledgments


References

















































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of





























SLC2A3SLC16A1

SLC2A1SLC7A5

SLC39A5

Nor

mal

1

Nor

mal

2

Nor

mal

3

Nor

mal

4

Nor

mal

5

Tum

or 1

Tum

or 2

Tum

or 3

Tum

or 4

Tum

or 5

Tum

or 6

Tum

or 7

Tum

or 8

Tum

or 9

Tum

or 1

0

Tum

or 1

1

Tum

or 1

2

Tum

or 1

3



3 Results





(a) (b)








Distantnormalacinar

Tumor Tumornormal1





KRT19

SLC6A14


(a) (b)

(c) (d)













Tumor grade 059172 0 (00) 8 (160) 1 (36) 1 (91) 10 (106)3 4 (800) 31 (620) 20 (714) 6 (545) 61 (649)4 1 (200) 11 (220) 7 (250) 4 (364) 23 (245)




4 Discussion









5 Conclusion


Abbreviations









Acknowledgments


References

















































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















(a) (b)








Distantnormalacinar

Tumor Tumornormal1





KRT19

SLC6A14


(a) (b)

(c) (d)













Tumor grade 059172 0 (00) 8 (160) 1 (36) 1 (91) 10 (106)3 4 (800) 31 (620) 20 (714) 6 (545) 61 (649)4 1 (200) 11 (220) 7 (250) 4 (364) 23 (245)




4 Discussion









5 Conclusion


Abbreviations









Acknowledgments


References

















































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















KRT19

SLC6A14


(a) (b)

(c) (d)













Tumor grade 059172 0 (00) 8 (160) 1 (36) 1 (91) 10 (106)3 4 (800) 31 (620) 20 (714) 6 (545) 61 (649)4 1 (200) 11 (220) 7 (250) 4 (364) 23 (245)




4 Discussion









5 Conclusion


Abbreviations









Acknowledgments


References

















































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of



























Tumor grade 059172 0 (00) 8 (160) 1 (36) 1 (91) 10 (106)3 4 (800) 31 (620) 20 (714) 6 (545) 61 (649)4 1 (200) 11 (220) 7 (250) 4 (364) 23 (245)




4 Discussion









5 Conclusion


Abbreviations









Acknowledgments


References

















































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of























5 Conclusion


Abbreviations









Acknowledgments


References

















































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















Acknowledgments


References

















































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of






































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of
















Research Article Transcriptomic and Immunohistochemical...

Documents

Transcript of Research Article Transcriptomic and Immunohistochemical...