NGAL Can Alternately Mediate Sunitinib Resistance in Renal Cell Carcinoma

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657

5859606162636465666768697071

NGAL Can Alternately Mediate Sunitinib Resistance in RenalCell Carcinoma

Dah-Shyong Yu,* Chia-Lun Wu, Szu-Yuan Ping, Yi-Ling Huang and Kun-Hung Shen*

From the Uro-Oncology Laboratory, Division of Urology, Department of Surgery, Tri-Service General Hospital (DSY, SYP, YLH)

and Graduate Institute of Life Science (CLW), National Defense Medical Center, Division of Urology, Department of Surgery,

Chi-Mei Medical Center (KHS) and Department of Optometry, Chung Hwa University of Medical Technology (KHS), Tainan,

Taiwan, Republic of China

Abbreviations

and Acronyms

Caki-1/NGAL ¼ NGAL transfectedCaki-1

ELISA ¼ enzyme-linkedimmunosorbent assay

Erk1/2 ¼ extracellularsignal-regulated kinase

GTP ¼ guanosine triphosphate

HRP ¼ horseradish peroxidase

MMP ¼ matrix metalloproteinase

NGAL ¼ neutrophilgelatinase-associated lipocalin

PCR ¼ polymerase chain reaction

Ras ¼ small GTP

RCC ¼ renal cell carcinoma

STAT ¼ signal transducer andactivator of transcription

VEGF ¼ vascular endothelialgrowth factor

VEGFR ¼ VEGF receptor

Accepted for publication December 30, 2013.Study received National Defense Medical

Center Laboratory Animal Center institutionalanimal care and use committee approval.

Supported by Grants NSC97-2314-B-016-022MY and NSC-100-2314-B-016-007 from theNational Science Council, Republic of China,TSGH-C100-069 from Tri-Service General Hospi-

72737475767778798081828384858687888990919293949596979899

100101102

Purpose: Serum NGAL is highly expressed in patients with advanced renalcancer treated with sunitinib. We investigated the role of NGAL in sunitinibresistance in renal cell carcinoma to identify potential tactics to overcome it.

Materials and Methods: NGAL expression was correlated with sunitinib sensi-tivity. Vascular endothelial growth factor related upstream Ras, Erk1/2 andSTAT1 phosphorylation activity in Caki-1 and NGAL transfected Caki-1 cellsafter sunitinib treatment was analyzed using Western blot. NGAL and vascularendothelial growth factor-A interaction with sunitinib therapeutic efficacy wasmonitored in renal cell carcinoma tumor xenografted mice by tumor growthinhibition, serum NGAL and vascular endothelial growth factor-a levels, andmicroscopic examination of tumor microvascular density.

Results: Sunitinib cytotoxicity in various renal cell carcinoma cell lines wasreversibly related to NGAL expression. Sunitinib showed the lowest 50%inhibitory concentration (5.53 mM) in Caki-1 cells, which had the lowest NGALexpression of these renal cell carcinoma cell lines. After sunitinib treatmentadding NGAL inhibited Ras and Erk1/2 phosphorylation but activated STAT1aphosphorylation in Caki-1 cells and Caki-1 cells transfected with NGAL. In axenograft mouse model sunitinib significantly inhibited tumor growth in Caki-1mice. NGAL transfected Caki-1 mice had higher serum NGAL and lowervascular endothelial growth factor-A than Caki-1 mice. Microvascular densitywas decreased in Caki-1 mice with sunitinib treatment.

Conclusions: NGAL in tumor cells may show crosstalk with vascular endothelialgrowth factor-a and alternative activation in stimulating tumor growth duringsunitinib treatment. It may become a therapeutic target to reverse sunitinibresistance in renal cell carcinoma.

Key Words: kidney; carcinoma, renal cell; LCN2 protein,

human; sunitinib; drug resistance

tal and CMNDMC 10110 from the Chi-MeResearch Foundation.

* Correspondence: Division of Urology,Department of Surgery, Tri-Service General Hos-pital, National DefenseMedical Center, Taipei 104,Republic of China (telephone: þ886-2-8793157;e-mail: [email protected]) or 901 ZhonghuaRd., Yongkang District, Tainan City 710, Taiwan,Republic of China (telephone: þ886-6-2812811-53379; e-mail: [email protected]).

103104105106107108109110

RENAL cell carcinoma is a highly vas-cularized tumor thought to dependhighly on VEGF mediated angiogen-esis. Sunitinib (Sutent�) is a smallmolecule tyrosine kinase inhibitorwith targets that include VEGFR,

Dochead: Investigative Urology FLA 5.2.0 D

0022-5347/14/1922-0001/0

THE JOURNAL OF UROLOGY®

© 2014 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESE

platelet-derived growth factorreceptor-a and b, and stem cell factorreceptor.1 Sunitinib has beenapproved multinationally as first linetreatment of clear cell RCC,2 gastro-intestinal stromal tumor after disease

111112113114

TD � JURO11004_proof � 25 April 2014 � 1:59 pm � EO: JU-13-2126

ARCH, INC.

http://dx.doi.org/10.1016/j.juro.2013.12.049

Vol. 192, 1-8, August 2014

Printed in U.S.A.www.jurology.com j 1

mailto:[email protected]

mailto:[email protected]

http://dx.doi.org/10.1016/j.juro.2013.12.049

http://www.jurology.com

2 NGAL CAN ALTERNATELY MEDIATE SUNITINIB RESISTANCE IN RENAL CELL CANCER

115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171

172173174175176177178179180181182183184185186187188189190191192193194

progression or intolerance to imatinib treatmentand metastatic pancreatic neuroendocrine tumor.3e5

Regulation of the VEGF pathway for RCC growthis affected by the high frequency of VHL genemutations in RCC.6,7 The VHL recognition compo-nent of a ubiquitin ligase complex targets thetranscription factor hypoxia-inducible factor andalters VEGF tumor over expression, which arecritical for tumor angiogenesis, cell proliferationand migration.8e10 NGAL is a member of the lip-ocalin family present in human neutrophil specificgranules, in part covalently associated with gelati-nase B/MMP-9, which is closely related to the met-astatic process of tumor cells.10e13 NGAL has dualeffects on tumor growth. It can stimulate tumorgrowth or decrease cell apoptosis caused by chemo-therapy drugs in breast, prostate, liver or braincancer.14e19 However, it also can inhibit angiogen-esis and decrease growth in pancreatic cancer.20 Wepropose that NGAL may also have a key role inregulating RCC growth during sunitinib treatmentby inhibiting VEGF. We elucidated the relationshipbetween NGAL and VEGF in sunitinib resistance.

195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228

MATERIALS AND METHODS

Cell Line CultureThe Caki-1, A498 and ACHN human renal carcinoma celllines were cultured in HyClone� Dulbecco’s modifiedEagle’s medium. Cultured cells were incubated in 5% CO2

at 37C.

Chemicals and AntibodiesSunitinib and recombinant NGAL were obtained fromPfizer, New York, New York and Sigma-Aldrich�,respectively. We used primary antibodies against total orphosphorylated Ras, Erk1/2 (Thr 202/204), VEGFA(GeneTex, Irvine, California) and NGAL (Novus Bio-logicals�). Anti-rabbit IgG horseradish peroxidase andmouse IgG were obtained from Santa Cruz Biotechnology,Santa Cruz, California. Phosphatase inhibitor cocktail,TransAM� STAT Kits and Restore� Western BlotStripping Buffer for Western blot were obtained fromPierce, Rockford, Illinois.

Reverse Transcriptase-PCRTotal RNA was extracted from RCC cells using TRIzol�reagent. Reverse transcription was performed usingLightCycler� 480 SYBR� Green I Master and oligo(deox-ythymidine) primers, including forward ATGAGCTGAAAGAAGACAAGAGCTA and reverse GAAGTTCTCCTTTAGTTCCGAAGTC. PCR amplification was done at60C annealing temperature for 20 to 30 cycles using theLightCycler 480.

NGAL cDNA Cloning and TransfectionThe cDNA synthesis reaction was performed using totalRNAs isolated from Caki-1 cells. The entire codingregion of NGAL cDNA sequences of sense or antisenseorientation was PCR amplified from cDNAs using target

Dochead: Investigative Urology FLA 5.2.0 DTD � JURO110

sequence specific primers and Gateway� Cloning Tech-nology. PCR primer sequences used for cloning were senseNGAL forward 50- ATAAGCTTGCCTGCTTCCTCGGCCCTGA-30 and sense NGAL reverse 50-ATCTCGAGGCACCTGTGCACTCAGCCGT-30. PCR products were clonedinto the pCR 3.1vector (Invitrogen�) and sequences wereconfirmed (Macrogen, Seoul, South Korea). Caki-1 cells in6-well plates were transfected with 4 m of PCR 3.1/NGALusing PolyJet� DNA In Vitro Transfection Reagent.Stable transfectants were selected in the presence of G418(200 mg/ml) 2 days after transfection. Up-regulation ordown-regulation of NGAL mRNA or protein in stabletransfectants was confirmed by reverse transcriptase-PCR or Western blot, respectively.

Cell Viability MeasurementCell viability was determined by MTT assay. Cells wereseeded in triplicate at 2 � 104 per well on 96-well platesincubated with or without NGAL (100 mM) for 72 hours.At the end of drug treatment culture medium wasremoved and MTT (0.5 mg/ml) was added, followed byincubation at 37C for 3 hours in a CO2 incubator. Afterinsoluble crystals were completely dissolved in dimethylsulfoxide absorbance was measured at 570 nm using amicroplate reader (Thermo Scientific Instruments, Salz-burg, Austria).

Western BlotProtein expression ofNGAL,VEGFA,Ras, Erk1/2 andRas-GTPwasmeasured byWestern blot. Briefly, 30 mg proteinsextracted from Caki-1 and Caki-1/NGAL or Caki-1 cellsplus NGAL (100 mM) after sunitinib (5 mM) treatment wereloaded in sodium dodecyl sulfate-polyacrylamide gel elec-trophoresis with 12% resolution and 4% stacking gel.Sodium dodecyl sulfate-polyacrylamide gel electrophoresiswas treated with 90V for 45 minutes and 120 V for 90minutes in running buffer. Proteins were transferred topolyvinylidene fluoride membrane in transfer buffer at100 V at 4C for 1 hour. The membrane was rinsed andshaken on a 40 rpm shaker in 5%milk in 0.1% tris-bufferedsaline-Tween� 20 at room temperature for 1 hour and thenrinsed in specific antibodies at room temperature for1 hour. After washing the membrane was rinsed in HRPconjugated secondary antibodies for 1 hour. Enhancedchemiluminescence was added to rinse the membrane andsignals were collected and analyzed using a BioSpectrum�Imaging System�.

STAT Activation AssayActivation of STAT1a, 3, 5A and 5B was determined usingTransAM STAT Family transcription factor assay ac-cording to manufacturer recommendations. Cell nucleiwere prepared from transformed lymphocytes by hypo-tonic lysis. Nuclear lysates (8 mg total protein) wereincubated in 96-well dishes containing immobilized oli-gonucleotides, which contained a STAT consensus DNAbinding site (50-TTCCCGGAA-30), for 1 hour at roomtemperature. Wells were washed 3 times and 100 mlmonoclonal antibody (1:1,000) were added to each well for1 hour at room temperature. Wells were washed 3 timesand 100 ml HRP conjugated secondary antibody (1:1,000)were added to each well for 1 hour at room temperature.Wells were washed 4 times and 100 ml developing solution

04_proof � 25 April 2014 � 1:59 pm � EO: JU-13-2126

NGAL CAN ALTERNATELY MEDIATE SUNITINIB RESISTANCE IN RENAL CELL CANCER 3

229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285

286287288289290291292293294295296297298299300301302303

D

were added to each well for 10 minutes at room temper-ature. Stop solution (100 ml) was added to each well andabsorbance at 450 nm was determined using an ELISAreader set to 450 nm. Binding specificity was determinedby competition with wild-type and mutated STAT oligo-nucleotides as positive and negative controls, respectively.

Nude MiceFour-week-old male athymic nu/nu mice were obtainedfrom the breeding colony at the National Laboratory An-imal Center, Tainan, Taiwan, Republic of China. Beforeinitiating the experiment all mice were acclimatized to apulverized diet for 3 days. Our experimental protocol wasreviewed and approved by the National Defense MedicalCenter Laboratory Animal Center.

Xenograft Tumor GrowthCaki-1 cells were transfected with NGAL and corre-sponding control vector. Nude mice were divided into2 groups of 5 each. Group 1 was injected with Caki-1 cells

Figure 1. NGAL expression and external addition correlated with sun

treatment for 72 hours (A). NGAL mRNA (B) and protein (C ) express

indicate p <0.01. MTT assay shows results of adding 100 mM rNGAL a

ochead: Investigative Urology FLA 5.2.0 DTD � JURO11004

transfected with vector only and group 2 was injectedwith Caki-1/NGAL. Mice were anesthetized withAvertin� solution. Tumor cells (1 � 106) in 100 ml phos-phate buffered saline were injected in the subcutaneousflank region using a 27 gauge needle. The day of cellimplantation was designated day 0. Subcutaneous tumorsin the group treated with sunitinib (dosing strategy40 mg/kg daily, 3 weeks on and 1 week off) and thecontrol group were measured daily with vernier calipersto determine the long and short dimensions after 6 daysof cell injections. Body weight was assessed daily aftercell injection. At 32 days the animals were sacrificed.The primary tumors were excised, weighed and preservedfor further study.

Immunohistochemistry StainingImmunohistochemistry studies were performed using theavidin-biotin technique with the LSAB�þ System-HRP.Paraffin sections of harvested tumor specimens weredeparaffinized in xylene, rehydrated in graded ethanol

itinib treatment in RCC cell lines. Cell viability under sunitinib

ion in RCC. Single asterisk indicates p <0.05. Double asterisks

fter sunitinib in Caki-1 (D), A498 (E ) and ACHN (F ) for 48 hours.

304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342

_proof � 25 April 2014 � 1:59 pm � EO: JU-13-2126

½F1�

½F2�


343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399

400401402403404405406407408409410411412413414415416417418419420421422423424

and treated for 10 minutes with 3% H2O2 in methanol toblock endogenous peroxidase. Sections were incubatedin a moist chamber with primary anti-NGAL, VEGFAand CD34 goat IgG (5 mg/ml, R&D Systems�) for 30 mi-nutes at room temperature, followed by incubation withHRP and biotinylated conjugated secondary antibody(DakoCytomation, Glostrup, Denmark) for 30 minutes.Reaction product was visualized using diaminobenzidinechromogen solution (DakoCytomation). Sections werecounterstained with hematoxylin and mounted in Far-amount Aqueous Mounting Medium (DakoCytomation).Representative microphotographs were recorded using adigital camera (Nikon, Tokyo, Japan).

Serum NGAL and VEGFA ELISA AssayMouse serum NGAL and VEGFA levels were measuredquantitatively by sandwich ELISA using RauBop� NGALand VEGF-A Rapid ELISA Kits in accordance withmanufacturer instructions. ELISA plates were read at450 nm and collected data were analyzed using theElx808� Absorbance Microplate Reader.

Statistical AnalysisData are shown as the mean � SD of at least independentexperiments. To compare 3 or more groups we used 1-wayANOVA followed by the Student method for multiplepairwise comparisons. Differences were considered sta-tistically significant at p <0.05.

½F3�

½F4�

Figure 2. NGAL and VEGFA in Caki-1 and Caki-1/NGAL cells

were measured by Western blot (A). Normalization with Caki-1

(B). Asterisk indicates p <0.01.

425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456

RESULTS

RCC Cell Line NGAL Expression and Sunitinib

Sensitivity

RCC cell lines had a wide range of cytotoxicity re-sponses to sunitinib (fig. 1, A). The 50% inhibitoryconcentration of sunitinib in Caki-1, ACHN andA498 cells was 5.53, 8.14 and 8.46 mM, respectively.NGAL mRNA and protein levels were lower in Caki-1 than in ACHN and A498 cells (fig. 1, B and C ).

When rNGAL (100 mM) was added during suni-tinib treatment for RCC cells, Caki-1 cells demon-strated marked decreased sensitivity to sunitinib.The 50% inhibitory concentration increased to16.26 mM compared to 8.28 mM for ACHN and 8.21mM for A498 cells (fig. 1, D). These results indicatethat NGAL has a role in regulating RCC cellsensitivity to sunitinib.

NGAL and VEGFA Expression in Caki-1

and Caki-1/NGAL Cells

NGAL and VEGFA expression was the opposite inCaki-1 and Caki-1/NGAL cells (fig. 2). NGAL andVEGFA were moderately expressed in Caki-1 cellsbut VEGFA was significantly decreased in Caki-1/NGAL cells (p <0.01).

NGAL and VEGFA Activity of Ras, Erk1/2

and STAT1 in Caki-1 Cells

Caki-1 cells showed a 54% decrease in Ras-GTP/Rasand a 76% decrease in phospho-Erk1/2 /Erk1/2 after


sunitinib treatment (fig. 3, A). In contrast, Ras-GTP/Ras and phospho-Erk1/2 /Erk1/2 did not change inA498 cells after sunitinib treatment (fig. 3, B).Further study of sunitinib co-treatment with NGAL(30 and 100 mM) in Caki-1 and Caki-1/NGAL cellsshowed that 30 mM up-regulated Ras-GTP/Ras andphospho-Erk1/2 /Erk1/2 (fig. 3, C ).

STAT1a but not STAT3, 5A and 5B was inacti-vated in Caki-1 cells after 5 mM sunitinib treatmentfor 12 hours (fig. 3, D). Interestingly, increasedNGAL expression caused by exogenously added100 mM NGAL or that in Caki-1/NGAL cellsrestored STAT1a phosphorylation after sunitinibtreatment (fig. 3, E ). This indicates that NGALcould enhance RCC cell survival signals throughSTAT1a activation after sunitinib treatment.

Xenograft

Tumor suppression by sunitinib. Caki-1 tumor growthsteadily increased throughout the study period inthe control group but growth was suppressed step-wise in sunitinib treated groups (fig. 4, A). From day24 after treatment the relative tumor size (treatedtumor volume/control tumor volume) of Caki-1 andCaki-1/NGAL cells significantly decreased (mean0.35 � 0.07 vs 0.81 � 0.16 fold, p <0.05).


½F5�

Figure 3. Intracellular Ras and Erk1/2 signal pathways changes in Caki-1 (A) and A498 (B) cells after sunitinib treatment at different

intervals. Single asterisk indicates p <0.05. Double asterisks indicate p <0.01. Effect of NGAL addition and transfection in Ras-GTP/

Ras and phospho-Erk1/2 /Erk1/2 changes in Caki-1 cells (C ). Single asterisk indicates p <0.01. Double asterisks indicate p <0.001.

STAT family changes in Caki-1 cells, Caki-1 cells with 30 or 100 mM NGAL and Caki-1/NGAL cells after 5 mM sunitinib treatment for

12 hours (D). Asterisk indicates p <0.01. Phospho-STAT1a/STAT1 changes in Caki-1 cells, Caki-1 cells with 30 or 100 mM NGAL and

Caki-1/NGAL cells after 5 mM sunitinib treatment for 12 hours (E ). Asterisk indicates p <0.001.


457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513

514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570

D

NGAL, VEGF and microvessel density in tissue. NGALwas highly expressed but VEGFA was inverselyexpressed in control and sunitinib treated Caki-1/NGAL groups (fig. 4, B and C ). CD34 stainingrevealed that microvessel density decreased in theCaki-1 group more than in the Caki-1/NGALgroup after sunitinib treatment (fig. 4, D).

Changes in serum VEGFA and NGAL in mice. Caki-1/NGAL mice had the highest serum NGAL (controlvs sunitinib mean 105.3 � 1.9 vs 35.9 � 1.5 and119.9 � 7.3 vs 91.3 � 3.4 pg/ml, p <0.01) and the


lowest VEGFA (67.2 � 6.2 vs 121.0 � 17.4 and62.1 � 7.2 vs 109.2 � 26.7 pg/ml, p <0.01, respec-tively) compared to Caki-1 mice (fig. 5).

DISCUSSIONNGAL is a member of the lipocalin family, whichbinds or transports lipid and other hydrophobicmolecules, and has a potent prometastatic fac-tor.15,20 NGAL up-regulation may activate proan-giogenic pathways and cause tumors to evade the


Figure 4. Relative fold of Caki-1 and Caki-1/NGAL xenograft tumor sizes in treatment vs control group (A). Asterisk indicates p <0.05.

NGAL (B), VEGFA (C ) and CD34 (D) expression in Caki-1 and Caki-1/NGAL xenograft tissue in control and sunitinib treated groups

was analyzed by immunohistochemistry staining. Arrows indicate blood vessels. Scale bars represent 20 mm (B to D).


571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627

628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684

antiangiogenic effect of sunitinib mediated VEGFblockade in RCC.

Identifying prognostic and predictive factors forantiangiogenic treatment of sunitinib resistanceRCC has become a priority. High serum VEGFand NGAL often show a poor response to sunitinibtreatment in patients with advanced RCC.21 To datethe underlying mechanism of NGAL and VEGFin patients with sunitinib resistance RCC is stillunclear.

Our major aim has been to clarify the relation-ship between VEGF and NGAL in sunitinib sensi-tivity in RCC cell lines. Our observations of thecorrelation of NGAL and VEGFA with sunitinibsensitivity in different RCC cell lines prompted us tofurther elucidate the underlying mechanisms ofNGAL involved in renal cancer cell survival/apoptotic pathways. These pathways include tran-scription factors, apoptosis mediators and cell cycleregulators.

We first noted that sunitinib cytotoxicity invarious RCC cell lines showed various degrees of


NGAL expression. We observed extrinsic additionof NGAL to Caki-1 or Caki-1/NGAL cells coulddecrease sunitinib cytotoxicity. We further notedthat NGAL could enhance the activation ofRas-GTP, Erk1/2 and STAT1a signal pathways toincrease RCC cellular proliferation and simulta-neously affect sunitinib cytotoxicity negatively. Onthe other hand, NGAL secreted by tumor cells alsoenhanced vascular endothelial cell proliferation topromote angiogenesis in a RCC xenograft model.To our knowledge this study is the first to showthat NGAL activation of angiogenic signaling inRCC cells may functionally compensate for suniti-nib antiangiogenic activity through VEGF/VEGFRmediation.

Interestingly VEGF up-regulation was not asso-ciated with extrinsic addition of NGAL or NGALtransfected Caki-1 cells. Our results indicate thatNGAL may decrease VEGFA expression throughother signal pathways and subsequently promoteangiogenetic activity in Caki-1 cells. The underlyingmechanism of NGAL and VEGF interaction through


Figure 5. Serum NGAL (A) and (B) VEGFA in 5 Caki-1 and

5 Caki-1/NGAL xenograft mice with and without 40 mg/kg

daily sunitinib treatment, respectively, at day 32. Single

asterisk indicates p <0.05. Double asterisks indicate p <0.01.


685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741

742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798

D

crosstalk with VEGFA transcription inhibition ordirect VEGFA suppression by NGAL still needsfurther study.

NGAL is considered a cytokine that also canpromote angiogenetic activity in cancer, a functionsimilar to that of interleukin-8.22 While sunitinibinhibits VEGF activity, the NGAL pathway istriggered through a crosstalk mechanism in RCCcells. In addition, NGAL is not only a growth factorbut it also increases sunitinib cytotoxicity andprevents sunitinib resistance in RCC cells. Thus,we speculate that alternate proangiogenic path-ways adopted after VEGF/VEGFR blockade in RCCcells may be due to NGAL activation of othersignaling pathway, such as Ras-GTP, Erk1/2 andSTAT1a, to compensate for tumor angiogenesis.Others reported that Ras/Raf and Erk signalingpathways were activated by retinoid acid and thesepathways were related to NGAL expression.23,24

This supports our findings that NGAL overexpression in RCC cells can increase Ras-GTP andErk1/2 phosphorylation, and activate the signalingpathways of mitogenic activity and STAT3 phos-phorylation in tumor cell growth regulation andoxidative stress accommodation. We also found thatNGAL over expression in Caki-1 cells increasedSTAT1a phosphorylation, which to our knowledge


is a novel apoptosis regulator. These signalingpathways are often considered important cell sur-vival signals.25

In our animal study the Caki-1/NGAL group hadless tumor growth inhibition after sunitinibtreatment than the Caki-1 group (fig. 4, A). Liaoet al reported that NGAL can serve as a survivalfactor for endometrial carcinoma by stimulatingcellular growth and migration ability.26 Based onthose findings we speculated that NGAL may alsoenhance RCC growth in the initial stage of tumordevelopment and it may subsequently be resistantto sunitinib treatment after VEGF inhibition andalternative pathway activation.

According to our in vitro and in vivo studiessunitinib treatment of Caki-1 cells through VEGF/VEGFR proangiogenic signaling inhibition canlater result in drug resistance by turning on thealternative NGAL angiogenetic pathway. Therefore,manipulating NGAL expression merits furtherstudy to determine how to overcome sunitinibresistance in patients with RCC. This study alsoshows that intrinsic insensitivity to sunitinib de-velops if NGAL expression is increased in RCC.Simultaneous inhibition of VEGF and NGAL activ-ity in RCC cells may increase sunitinib cytotoxicityand result in longer efficacy. Also, the NGALexpression level in RCC cells may be useful as apredictive biomarker for sunitinib sensitivity beforetarget therapy in patients with metastatic RCC.Smith et al reported that NGAL, MMPs and VEGFmay be useful as urinary biomarkers to predictbrain tumors and the response to therapy.27

NGAL has pro and con effects in tumor cells.Bolignano et al observed that when NGAL acts asan intracellular iron carrier and protects MMP-9from proteolytic degradation, it has a clear protu-mor enhancing effect, which has been observed intumors of the human breast, stomach, esophagus,rectum, thyroid and brain.28 In contrast, Zhang et alreported that NGAL can inhibit proneoplasticfactor hypoxia-inducible factor-1a, focal adhesionkinase phosphorylation and VEGF synthesis sothat it paradoxically has antiumor and anti-metastatic effects on neoplasms of the colon, ovaryand pancreas.29

CONCLUSIONSOur study demonstrates that NGAL enhances RCCgrowth and becomes an alternative pathway forRCC to escape sunitinib treatment. NGAL has animportant role in the development of RCC suniti-nib resistance. It may potentially serve as a pre-dictor of sunitinib therapy, a means of prognosisand a therapeutic target for sunitinib resistancein RCC.



799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860

861
REFERENCES 862863864865866867868869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899900901902903904
1. Izzedine H, Buhaescu I, Rixe O et al: Sunitinibmalate. Cancer Chemother Pharmacol 2007; 60:357.

2. Motzer RJ and Bukowski RM: Targeted therapyfor metastatic renal cell carcinoma. J Clin Oncol2006; 24: 5601.

3. Joensuu H: Sunitinib for imatinib-resistant GIST.Lancet 2006; 368: 1303.

4. Delbaldo C, Faivre S, Dreyer C et al: Sunitinibin advanced pancreatic neuroendocrine tumors:latest evidence and clinical potential. Ther AdvMed Oncol 2012; 4: 9.

5. Antoniu SA and Kolb MR: Intedanib, a triplekinase inhibitor of VEGFR, FGFR and PDGFR forthe treatment of cancer and idiopathic pulmo-nary fibrosis. IDrugs 2010; 13: 332.

6. Chintala S, Najrana T, Toth K et al: Prolylhydroxylase 2 dependent and Von-Hippel-Lindauindependent degradation of Hypoxia-induciblefactor 1 and 2 alpha by selenium in clearcell renal cell carcinoma leads to tumor growthinhibition. BMC Cancer 2012; 12: 293.

7. Fu L, Wang G, Shevchuk MM et al: Generationof a mouse model of Von Hippel-Lindau kidneydisease leading to renal cancers by expressionof a constitutively active mutant of HIF1a.Cancer Res 2011; 71: 6848.

8. Knebelmann B, Ananth S, Cohen HT et al:Transforming growth factor alpha is a target forthe von Hippel-Lindau tumor suppressor. CancerRes 1998; 58: 226.

9. Cockman ME, Masson N, Mole DR et al: Hypoxiainducible factor-alpha binding and ubiquitylationby the von Hippel-Lindau tumor suppressorprotein. J Biol Chem 2000; 275: 25733.

10. Kim WY and Kaelin WG: Role of VHL genemutation in human cancer. J Clin Oncol 2004; 22:4991.

11. Kjeldsen L, Bainton DF, Sengeløv H et al:Identification of neutrophil gelatinase-associated

Dochead: Investigative Urology FLA

lipocalin as a novel matrix protein of specificgranules in human neutrophils. Blood 1994; 83:799.

12. Flower DR, North AC and Attwood TK: Mouseoncogene protein 24p3 is a member of thelipocalin protein family. Biochem Biophys ResCommun 1991; 180: 69.

13. Beynon RJ and Hurst JL: Multiple roles ofmajor urinary proteins in the house mouse, Musdomesticus. Biochem Soc Trans 2003; 31: 142.

14. Nelson AM, Zhao W, Gilliland KL et al: Neutro-phil gelatinase-associated lipocalin mediates13-cis retinoic acid-induced apoptosis of humansebaceous gland cells. J Clin Invest 2008; 118:1468.

15. Bartsch S and Tschesche H: Cloning andexpression of human neutrophil lipocalin cDNAderived from bone marrow and ovarian cancercells. FEBS Lett 1995; 357: 255.

16. Furutani M, Arii S, Mizumoto M et al: Identifi-cation of a neutrophil gelatinase-associatedlipocalin mRNA in human pancreatic cancersusing a modified signal sequence trap method.Cancer Lett 1998; 122: 209.

17. Stoesz SP, Friedl A, Haag JD et al: Heteroge-neous expression of the lipocalin NGAL in pri-mary breast cancers. Int J Cancer 1998; 79: 565.

18. Friedl A, Stoesz SP, Buckley P et al: Neutrophilgelatinase-associated lipocalin in normal andneoplastic human tissues. Cell type-specificpattern of expression. Histochem J 1999; 31: 433.

19. Iannetti A, Pacifico F, Acquaviva R et al:The neutrophil gelatinase-associated lipocalin(NGAL), a NF-kappaB-regulated gene, is a sur-vival factor for thyroid neoplastic cells. Proc NatlAcad Sci U S A 2008; 105: 14058.

20. Tong Z, Kunnumakkara AB, Wang H et al:Neutrophil gelatinase-associated lipocalin: anovel suppressor of invasion and angiogenesisin pancreatic cancer. Cancer Res 2008; 68: 6100.

5.2.0 DTD � JURO11004_proof � 25 April 20

21. Porta C, Paglino C, De Amici M et al: Predictivevalue of baseline serum vascular endothelialgrowth factor and neutrophil gelatinase-associated lipocalin in advanced kidney cancerpatients receiving sunitinib. Kidney Int 2010; 77:809.

22. Yoshiji H, Harris SR and Thorgeirsson UP:Vascular endothelial growth factor is essentialfor initial but not continued in vivo growth ofhuman breast carcinoma cells. Cancer Res 1997;57: 3924.

23. Huang D, Ding Y, Zhou M et al: Interleukin-8mediates resistance to antiangiogenic agentsunitinib in renal cell carcinoma. Cancer Res2010; 70: 1063.

24. Videira PA, Piteira AR, Cabral MG et al: Effectsof bevacizumab on autocrine VEGF stimulation inbladder cancer cell lines. Urol Int 2011; 86: 95.

25. Cheepala SB, Yin W, Syed Z et al: Identificationof the B-Raf/Mek/Erk MAP kinase pathway asa target for all-trans retinoic acid during skincancer promotion. Mol Cancer 2009; 8: 27.

26. Liao CJ, Huang YH, Au HK et al: The cancermarker neutrophil gelatinase-associated lipocalinis highly expressed in human endometrial hy-perplasia. Mol Biol Rep 2012; 39: 1029.

27. Smith ER, Zurakowski D, Saad A et al: Urinarybiomarkers predict brain tumor presence andresponse to therapy. Clin Cancer Res 2008; 14:2378.

28. Bolignano D, Donato V, Lacquaniti A et al:Neutrophil gelatinase-associated lipocalin(NGAL) in human neoplasias: a new proteinenters the scene. Cancer Lett 2010; 288: 10.

29. Zhang XF, Zhang Y, Zhang XH et al: Clinicalsignificance of Neutrophil gelatinase-associatedlipocalin (NGAL) expression in primary rectalcancer. BMC Cancer 2009; 9: 134.

905906907908909910911912913914915916917918919920921922

14 � 1:59 pm � EO: JU-13-2126

Documents

NGAL Can Alternately Mediate Sunitinib Resistance in Renal Cell Carcinoma