use of the proton pump inhibitor pantoprazole to modify the

Cancer Therapy: Preclinical

Use of the Proton Pump Inhibitor Pantoprazole to Modify theDistribution and Activity of Doxorubicin: A Potential Strategyto Improve the Therapy of Solid Tumors

Krupa J. Patel1, Carol Lee1, Qian Tan1, and Ian F. Tannock1,2

AbstractPurpose: Limited drug distribution within solid tumors is an important cause of drug resistance. Basic

drugs (e.g., doxorubicin)maybe sequestered in acidic organelles, thereby limiting drug distribution todistal

cells and diverting drugs from their target DNA. Here we investigate the effects of pantoprazole, a proton

pump inhibitor, on doxorubicin uptake, and doxorubicin distribution and activity using in vitro andmurine

models.

ExperimentalDesign:MurineEMT-6 andhumanMCF-7 cellswere treatedwithpantoprazole to evaluate

changes in endosomal pH using fluorescence spectroscopy, and uptake of doxorubicin using flow

cytometry. Effects of pantoprazole on tissue penetration of doxorubicin were evaluated in multilayered

cell cultures (MCC), and in solid tumors using immunohistochemistry. Effects of pantoprazole to influence

tumor growth delay and toxicity because of doxorubicin were evaluated in mice.

Results: Pantoprazole (>200 mmol/L) increased endosomal pH in cells, and also increased nuclear

uptake of doxorubicin. Pretreatment with pantoprazole increased tissue penetration of doxorubicin in

MCCs. Pantoprazole improved doxorubicin distribution from blood vessels in solid tumors. Pantoprazole

given before doxorubicin led to increased growth delay when given as single or multiple doses to mice

bearing MCF7 xenografts.

Conclusions: Use of pantoprazole to enhance the distribution and cytotoxicity of anticancer drugs in

solid tumors might be a novel treatment strategy to improve their therapeutic index. Clin Cancer Res; 1–11.

�2013 AACR.

IntroductionDrug resistance limits treatment of cancer by chemother-

apy. For a tumor to respond to chemotherapy, a drug mustleave tumorblood vessels efficiently anddistribute through-out tumor tissue to reach all cancer cells in concentrationsthat will lead to cytotoxicity (1, 2). The distribution ofanticancer drugs such as doxorubicin is limited in solidtumors and this limited distribution may be an importantmechanism of drug resistance (3–9).Many solid tumors develop regions of extracellular acid-

ity because of production and poor clearance of carbonicand lactic acid (10–13). The pH gradient between an acidicextracellular environment and a neutral–alkaline intracel-

lular pH may influence drug uptake and activity. Also, cellscontain acidic organelles such as lysosomes and endosomes(14–16). Because many chemotherapeutic drugs such asanthracyclines and vinca alkaloids are weak lipophilic bas-es, they readily enter acidic organelles where they areprotonated and sequestered (17).

Vacuolar-Hþ-ATPases (V-Hþ-ATPase) transport Hþ ionsacross the membranes of a wide array of intracellularcompartments and are themajor mechanism for regulationof endosomal pH (18). Agents that disrupt the pH gradientbetween the cytoplasm and endosomes in tumors mightdecrease the sequestration of basic anticancer drugs andrender cells more sensitive to the drug. A class of Hþ-ATPaseinhibitors called proton pump inhibitors (PPI) inhibitacidification of cells in the wall of the stomach and areused clinically for treating patients with peptic ulcer disease.These agents also inhibit V-Hþ-ATPase albeit at somewhathigher concentrations then is required to inhibit acidifica-tion in the stomach (19). PPIs accumulate selectively inacidic spaces and with the inhibition of V-Hþ-ATPase activ-ity, increase both extracellular pH and the pH of acidicorganelles (20). Pretreatment with a PPI might alter intra-cellular drug distribution by inhibiting drug sequestration,thereby allowing more drug to enter the nucleus and causecytotoxicity, and to exit the cell and be taken up by cells

Authors' Affiliations: 1Department of Medical Biophysics and 2Division ofMedical Oncology and Hematology, Princess Margaret Cancer Centre andUniversity of Toronto, Toronto, Ontario, Canada

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

Corresponding Author: Ian F. Tannock, Princess Margaret Hospital andUniversity of Toronto, 610 University Avenue, Suite 5-208, Toronto, ONM5G 2M9, Canada. Phone: 416-946-2245; Fax: 416-946-4563; E-mail:[email protected]

doi: 10.1158/1078-0432.CCR-13-0128

�2013 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org OF1

Research. on April 9, 2018. © 2013 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 18, 2013; DOI: 10.1158/1078-0432.CCR-13-0128

http://clincancerres.aacrjournals.org/

distant from blood vessels. We hypothesized that prevent-ing sequestration of drug in acidic endosomes wouldimprove extracellular drug distribution. In support of thishypothesis, modest improvements in drug distributionhave been observed in multilayered cell cultures (MCC)pretreatedwith chloroquine or thePPI, omeprazole (21). Aswell, PPIs have been shown to increase chemotherapyefficacy in vitro and improve tumor control in vivo (22–24).

Here we describe and evaluate the role of pantoprazole,which has been reported (or might be expected) to increaseendosomal pH, for its effects on uptake of doxorubicin intotumor cells, and its distribution within them. We thendescribe experiments to characterize the effects of panto-prazole to modify the distribution and activity of doxoru-bicin in solid tumors.

Materials and MethodsDrugs and reagents

Doxorubicin (Pharmacia) was purchased from the hos-pital pharmacy as a solution at a concentration of 2mg/mL.Purified rat anti-mouse CD31 (platelet/endothelial adhe-sionmolecule 1)monoclonal antibodywaspurchased fromBD PharmMingen, and Cy3-conjugated goat anti-rat IgGsecondary antibody was purchased from Jackson Immu-noResearch Laboratories, Inc. Pantoprazole (Nycomed)was purchased from the hospital pharmacy as a lyophilizedpowder and dissolved in 0.9% saline. Radiolabeleddoxorubicin was obtained from Amersham Life Sciences.Lansoprazole was purchased from Sigma and dissolvedin ethanol.

Cell linesThe parental mouse mammary sarcoma EMT-6 was pro-

vided originally by Dr. Peter Twentyman, Cambridge, Unit-ed Kingdom. The human breast cancer cell line MCF-7 andthe human vulvar epidermoid carcinoma cell line A431were obtained from the American Type Culture Collection.

EMT-6 and MCF-7 cells were maintained as monolayersin a-MEM media, supplemented with 10% FBS (Hyclone).

A431 cells were maintained in Dulbecco’s Modified EagleMedium supplemented with 10% FBS. All media wereobtained from the hospital media facility. Cells weregrown in a humidified atmosphere of 95% air and 5%CO2 at 37�C. Routine tests to exclude mycoplasma wereperformed.

Tumors were generated by subcutaneous injection of 1to 5 � 106 exponentially growing cells into the left andright flank regions of female athymic nude mice (MCF7and A431) or syngeneic Balb/C mice (EMT-6). Estrogenpellets were implanted into the nude mice one day beforeinjection of MCF-7 tumor cells. All procedures werecarried out following approval of the Institutional AnimalCare Committee.

Measurement of endosomal pHTo measure endosomal pH, EMT6 cells (106/mL) were

treated with varying concentrations of lansoprazole, pan-toprazole, or tamoxifen (included because of priorreports of effects to raise endosomal pH; ref. 25), andMCF7 cells were treated with varying concentration ofpantoprazole in the presence or absence of doxorubicin(1.8 mmol/L). They were incubated for 3 hours withdextran–fluorescein–tetramethylrhodamine 10,000 MW,anionic (FITC/TMR-dextran; Molecular Probes, Inc.),which is taken up into endosomes, followed by exposureto media for 2 hours. Fluorescence was measured using aCoulter Epics Elite flow cytometer (Beckman Coulter)equipped with an argon laser emitting at 488 nm. Theargon laser was used to excite FITC and TMR with emis-sion evaluated at 525 nm (pH-dependent) and 575 nm(pH-independent), respectively. Calibration of fluores-cence measurements was performed using the ionophorenigericin (Sigma) in buffers of known pH (26).

Doxorubicin uptake and distribution in cellsDoxorubicin distribution in cells was evaluated by fluo-

rescence microscopy. Cells attached to a chambered coverglass were pretreated with pantoprazole (1 mmol/L) for 2hours and then incubated in media containing 2 mg/mLdoxorubicin for 1 hour. The drugs were washed out and thefluorescence signal was recorded with excitation at 514 nmand emission at 488 nm using a Zeiss Axiovert 200 Mfluorescence inverted microscope and the fluorescence sig-nals were captured with a Roper Scientific CoolSnap HQCCD camera. To visualize endosomes, the cells wereexposed to the pH-sensitive endosomal dye, LysoSensorYellow/Blue DND-160 (Molecular Probe, Inc.) at a concen-tration of 5 mmol/L for 15 minutes. The fluorescent signalwas measured with excitation at 360 nm and emission at420 nm.

To determine net uptake of doxorubicin in EMT-6, MCF-7, and A431 cell lines, 106 cells were treated in vials witheither saline or pantoprazole (1 mmol/L). After 2 hours,doxorubicin (1.8 mmol/L) was added to each vial andincubated for 1 hour and then washed twice with PBS.Mean doxorubicin fluorescence was measured using theBecton Dickinson FACScan and Cell Quest software. The

Translational RelevanceThe tumor microenvironment is likely to play an

important role in drug resistance. Our group and othershave shown that limited drug distributionmay influencethe therapeutic efficacy of anticancer drugs. This articleexamines the use of the proton pump inhibitor panto-prazole in combination with doxorubicin in in vitro andin vivo model systems. Pantoprazole was found toincrease doxorubicin uptake and penetration throughtissue in in vitro studies. In solid tumors, pantoprazolewas found to improve doxorubicin distribution andcombined treatment increased tumor growth delay inmice. This study suggests that pantoprazole mayimprove the therapeutic efficacy of doxorubicin by alter-ing intracellular and extracellular drug distribution.

Patel et al.

Clin Cancer Res; 2013 Clinical Cancer ResearchOF2




FL1 filter (530 nm) was used to detect doxorubicinfluorescence.

Doxorubicin penetration in MCCsDrug penetration was evaluated in an in vitro tumor-like

environment using MCCs (27, 28). Approximately 2 � 105

cells were seeded on collagen-coated microporous Teflonmembranes (Millipore) and given 4 to 6 hours to attach.Experiments using MCCs were carried out using EMT-6 andMCF-7 cells. A431-derived MCCs were not used becausecells did not growwell in these conditions. Themembraneswere immersed in a-MEM media in a large vessel withcontinuous stirring and allowed to incubate at 37�C forapproximately 6 to 8days; this led to the formationofMCCscontaining approximately 5 � 106 cells. The MCCs wereincubated in either media alone or media containing 1mmol/L pantoprazole for 2 hours. The penetration experi-ments were performed at 37�C in an atmosphere of 95% airand 5% CO2. To evaluate doxorubicin penetration, solu-tions containing 10 mmol/L radiolabeled doxorubicin wereprepared in 2 � a-MEM (without FBS) and mixed in a 1:1ratio with 1% agar solution; the agar is added to preventconvection.A volumeof 0.5mLof thismixturewas added toone side of the MCC. The membranes were then floated inglass polyshell vials containing 18 mL of a-MEM media. Acell-freemembrane insertwas included in all experiments asa control. The drug, after penetrating through the MCC,appeared in the compartment below and 150 mL sampleswere taken from this compartment over time and assessedby liquid scintillation counting. 3H-sucrose was used asan internal standard at a concentration of 2 mmol/L toensure minimal inter-experimental variations of the MCCthickness.Fluorescent micrographs of MCCs were generated after

exposure to doxorubicin, which is fluorescent. After 1 hourof incubation, MCCs were removed and frozen, 10 mmsections were imaged using an Olympus Upright BX50microscope with a 100 W HBO mercury light sourceequipped with 530 to 560 nm excitation and 573 to 647nm emission filter sets. Images were pseudo-colored usingImage Pro Plus software.

Toxicity studies in miceMice were treated with either PBS containing CaCl2 and

MgCl2 alone, doxorubicin alone (8 mg/kg), or pantopra-zole at each of the following doses: 100, 150, 200, 250, 300mg/kg, alone and 2 hours before doxorubicin. The bodyweight of the mice was recorded every other day.

Plasma concentrations of pantoprazole in miceAt various times after mice were treated with pantopra-

zole (200 mg/kg), blood was collected using cardiac punc-ture in heparin-coated tubes on ice, after sample collectionthe mice were killed. Plasma was isolated and frozen bycentrifugation. Pantoprazole concentration was deter-mined using high-performance liquid chromatography(HPLC) using a protocol similar to that described by Peresand colleagues (29). Shimadzu SIL-20-ACauto-injector and

Shimadzulc-20AD pumps were used for HPLC-MS/MSanalysis and Applied Biosystem MDS Sciex (Concord)API3200 tandem mass spectrometry equipped with a Tur-boIonSpray interface was used subsequently. The data wereprocessed using Analyst 1.4.2 software (Applied BiosystemMDS Sciex).

Doxorubicin distribution in solid tumorsMice bearing EMT-6 or MCF-7 subcutaneous tumors in

both flank regions were divided randomly into groups of 5and were treated when themean tumor diameter was in therange of 8 to 12 mm. Animals were treated with PBScontaining CaCl2 and MgCl2, doxorubicin alone, or pan-toprazole before doxorubicin. Doxorubicinwas given intra-venously at a dose of 25 mg/kg to facilitate detection andquantification of drug auto-fluorescence. Pantoprazole wasadministered intraperitoneally 2 hours before doxorubicintreatment at a dose of 200 mg/kg. To detect hypoxia, EF5was injected intraperitoneally approximately 2 hours beforekilling the mice (0.2 mL of a 10 mmol/L stock per mouse),and to detect functional blood vessels the perfusionmarkerDiOC7 (1 mg/kg) was injected intravenously 1 minutebefore killing the mice. Mice were killed 10 minutes afterdoxorubicin injection and the tumors were excised. Previ-ous studies in our laboratory have showed that doxorubicinis maximally distributed in solid tumors between 10 min-utes and 3 hours after administration (6). Also, removingthe tumor 10 minutes after doxorubicin administrationensures an adequate concentration of pantoprazole in theplasma. The tissues were embedded immediately in OCTcompound, frozen in liquid nitrogen, and stored at�70�C.Cryostat sections 10-mm-thick were cut at 3 levels approx-imately 100 mm apart from each tumor, mounted on glassslides.

Doxorubicin fluorescence was quantified with an Olym-pus Upright BX50 microscope with a 100 W HBO mercurylight source equippedwith 530 to 560 nm excitation/573 to647 nm emission filter sets. Tissue sections were imagedwith a Photometrics CoolSNAP HQ2 (monochrome forfluorescence imaging) camera and tiled using a motorizedstage so that the distribution of doxorubicin was obtainedfor the entire tissue section. All images were captured in 8-bit signal depth and subsequently pseudo-colored.

Tumor sections were first imaged for doxorubicin and theperfusion marker DiOC7 using a TRITC and FITC filter set,respectively. Sections were then stained for blood vesselsusing antibodies specific for the endothelial cell markerCD31 [rat anti-CD31 primary antibody (1:100); BD Bio-sciences; and C73-conjugated goat anti-rat IgG secondaryantibody (1:400)], hypoxic regions were identified using aCy5-conjugated mouse anti-EF5 antibody (1:50). Tumorsections were imaged for CD31using theCy3 (530–560 nmexcitation/573–647 nm emission) filter set, and EF5 usingthe Cy5 far-red filter set. Image analysis was performed aspreviously described in our laboratory (30), where doxo-rubicin 8-bit grayscale images were overlaid with binaryCD31 images. Nonfunctional vessels from the CD31 imagewere determined by comparing the image with a binarized

Improved Therapeutic Effect of Doxorubicin with Pantoprazole

www.aacrjournals.org Clin Cancer Res; 2013 OF3




DiOC7 image and these vessels were removed. The overlaidimages were then run through a customized algorithm togenerate drug-intensity distributions in relation to distancefrom the nearest blood vessel.

Tumor vasculature was quantified using Media Cyber-netics Image Pro PLUS Software. The total number of bloodvessels was measured by setting a threshold for CD31positive pixel intensity and minimum blood vessel area(62 mm2) and counting the number of objects within thesesettings. Objects above this threshold range but below theminimum area were removed as artifacts (31). The tumorarea was recorded and areas of necrosis and artifact wereexcluded. The mean number of total blood vessels pertumor area was determined. The number of functionalblood vessels per tumor area was calculated in a similarmanner using DiOC7 positive pixels.

Growth delay studiesTwo perpendicular diameters of tumors growing in the

flanks of mice were measured with a caliper, and treatmentbegan once tumors reached a diameter of 5 to 8mm. Tumorvolume was estimated using the formula: 0.5(ab2), where ais the longest diameter and b is the shortest diameter.

To determine the effects of pantoprazole, mice bearingeither MCF-7 or A431 tumors were divided into 4 groups of4 to 5mice each and treated with either saline, doxorubicinalone (8 mg/kg i.v.), pantoprazole alone (200 mg/kg, i.p.),or pantoprazole 2 hours before doxorubicin (200 mg/kg þ8 mg/kg). For evaluation of growth delay followingmultiple doses of drugs, mice were treated with eitherdoxorubicin alone (6 mg/kg i.v.), pantoprazole alone(200 mg/kg i.p.), or pantoprazole combined with doxoru-bicin (200 mg/kg þ 6 mg/kg), once a week for 3 weeks.Every 2 to 3 days, the tumor volume and body weight weremeasured.Measurementswere taken until tumors reached amaximum diameter of 1.2 cm or began to ulcerate, whenmice were killed humanely. All mice were ear tagged andrandomized to avoid bias with measurements. Growthdelay experiments were also done in EMT-6 tumors butbecause of the fast-growing nature of these tumors, ulcera-tions and overgrown tumors prevented data from beingcollected beyond 8 days.

Statistical analysisA one-way ANOVA, followed by a post hoc t test were

performed todetermine statistical differences between treat-ment groups. For drug uptake using flow cytometry andquantification of functional tumor vasculature and hypox-ia, t tests were performed to determine significant differ-ences between treatment groups. P < 0.05 was used toindicate statistical significance.

ResultsPPIs increase endosomal pH in tumor cells

Pantoprazole, lansoprazole, or tamoxifen led to concen-tration-dependent increases in endosomal pH in EMT-6cells; increases in endosomal pH were observed with con-centrations of tamoxifen above 10 mmol/L, and with lan-

soprazole and pantoprazole at concentrations above 200mmol/L (Fig. 1A). Similar effects were observed followingexposure ofMCF-7 cells to pantoprazole; doxorubicin alonehad a small effect to raise endosomal pH and enhancedthe concentration-dependent effects of pantoprazole toincrease endosomal pH (Fig. 1B).

Pantoprazole pretreatment influences doxorubicinuptake in tumor cells in culture

Photomicrographs of fluorescence in MCF-7 and EMT-6(not shown) cells exposed to doxorubicin alone show thedrug to be present in the nucleus and in punctuate compart-ments within the cytoplasm; and localized staining ofdoxorubicin with lysosensor gives evidence of endosomalsequestration (Fig. 2A). Treatment with pantoprazolereduces the amount of doxorubicin fluorescence in thecytoplasm whereas it is retained within the nucleus (Fig.2A, panel iii).

Doxorubicin uptake within cells pretreated with eithersaline or pantoprazole 1mmol/L) wasmeasured using flowcytometry. In EMT-6 cells, doxorubicin fluorescence signif-icantly increased with pantoprazole pretreatment (P <0.05). In contrast, MCF-7 and A431 (not shown) cellsshowed a decrease in doxorubicin fluorescence by 24%

Concentration (mmol/L)

Concentration (mmol/L)

10,0001,000100101

En

do

som

al p

H

4.0

4.5

5.0

5.5

6.0

6.5

7.0A

B

10,0001,000100101

En

do

som

al p

H

4.5

5.0

5.5

6.0

6.5

7.0

Figure 1. Endosomal pH measurements in cancer cells. A, EMT-6 cellsexposed to varying concentrations of lansoprazole (&), pantoprazole(^), and tamoxifen (~). B, MCF7 cells treated with varying doses ofpantoprazole in the presence (&) or absence (^) of doxorubicin. Cellswere exposed to the pH sensitive dye lysosensor and fluorescence wasmeasured using flow cytometry (data represent mean � SE; n ¼ 3).

Patel et al.





(P < 0.05) and 36% (P < 0.05), respectively, with panto-prazole pretreatment (Fig. 2B).When cellswere treatedwitha lower concentration of pantoprazole (100 mmol/L), sim-ilar changes in distribution were observed, but these resultswere not significant (data not shown).

Doxorubicin penetration in MCCs pretreated withpantoprazoleWhen MCCs were pretreated with 1 mmol/L pantopra-

zole, there was a greater than 2-fold increase in doxorubicinpenetration for those grown from EMT-6 cells (P < 0.05;Fig. 3A) and �1.3-fold increase through those grown fromMCF-7 cells (P < 0.02; Fig. 3B). Internal standards indicatedby 3H-sucrose penetration were used to control for varia-tions in thickness of the cell cultures, and showed <10%variation.Photomicrographs of MCCs derived from MCF-7 cells

indicate an increase in doxorubicin fluorescence in cellsmore distal from the source of drug that are pretreated withpantoprazole compared with control MCCs at 2 hours afterthe start of doxorubicin exposure (Fig. 3C and D). Similarchanges in fluorescence were observed in MCCs derivedfrom EMT-6 cells (photomicrographs not shown).

Plasma concentration of pantoprazole and toxicity inmice

After intraperitoneal injections of 200 mg/kg pantopra-zole in mice, the peak plasma concentration was �300mmol/L within the first hour after administration, and�150 mmol/L after 2 hours. After 5 hours, the concentrationin the plasma decreased to less than 1% of the maximumconcentration in the blood and by 24 hours less than 0.01mmol/L of pantoprazole was detectable (Fig. 4).

Toxicity studies in miceMice treated with saline alone showed a gradual increase

in body weight over more than 20 days. Mice treated withdoxorubicin alone (8 mg/kg i.v.) or pantoprazole alone(up to 300 mg/kg) showed minimal increase in bodyweight. Combined treatment led to no change in bodyweight at pantoprazole doses of 100 or 150 mg/kg, but at200 mg/kg mice showed a temporary decrease in bodyweight (�15%) within the first 5 to 8 days after treatmentfollowed by rapid recovery to their original body weight.Mice treated with 250 or 300 mg/kg of pantoprazole andthen doxorubicin showed continual loss in body weightbeyond 20 days (data not shown). The maximum tolerated

0

20

40

60

80

100

120

140

160

180

A431MCF-7EMT-6

Mea

n D

OX

flu

ore

scen

ce

Cell line

DOX

PTP + DOX

Ai ii iii

B

Figure 2. Doxorubicin uptake intumor cells in culture. Fluorescentmicrographs show alterations inintracellular doxorubicindistribution in cancer cells in culturewith pantoprazole pretreatment.A, MCF-7 cells were treated with(i) doxorubicin, (ii) lysosensor, or(iii) pantoprazole beforedoxorubicin. B, doxorubicinfluorescence in EMT-6,MCF-7, andA431 cells pretreated with eithersaline or 1 mmol/L pantoprazolewasmeasuredusing flowcytometry(data represent mean � SEM;n ¼ 3).






dose was determined to be 200mg/kgwhen combinedwithdoxorubicin.

Effect of pantoprazole on distribution of doxorubicinin tumors

Photomicrographs taken from MCF-7 tumors show sub-stantial increases in doxorubicin fluorescence in tumors inmice pretreated with pantoprazole, compared with micetreated with doxorubicin alone (Fig. 5A and B).

Doxorubicin distribution was quantified in EMT-6 andMCF-7 tumors in mice at 10 minutes after injection. Doxo-rubicin fluorescence intensity was determined in relation todistance to the nearest functional blood vessel in areas ofinterest with similar numbers of functional vessels. In alltumors, there were steep gradients of decreasing doxorubi-cin fluorescence in relation to distance from the nearestfunctional blood vessel in the section. In MCF-7 tumors,there was a shallower gradient of decrease in doxorubicindistribution in the tumors pretreated with pantoprazolecompared with the tumors treated with doxorubicin alone(P < 0.05; Fig. 5C). However, in EMT-6 tumors there was no

0

0.05

0.1

0.15

0.2

0.25

0.3

86420

Rel

ativ

e p

enet

rati

on

of

do

xoru

bic

in

Time (h)

DOX

PTP + DOX

A

E

F

C

B

Figure 3. The penetration ofdoxorubicin in MCCs derived fromMCF-7 cells as a function of time.MCCs were pretreated with eithersaline or pantoprazole (1 mmol/L)followed by 14C-doxorubicin (datarepresent mean � SEM; n ¼ 4; A).Photomicrographs of fluorescentdoxorubicin in MCCs derived fromMCF-7 cells. MCCs werepretreated with either saline (B) orpantoprazole before doxorubicintreatment (C; scale bars represent100 mm).

0.001

0.01

0.1

1

10

100

1,000

2520151050

Pan

top

razo

le c

on

cen

trat

ion

(µ

mo

l/L)

Time (h)

Figure 4. Plasma concentration of pantoprazole at various times afteradministration of 200 mg/kg i.p. to Balb/C nude mice. Plasma levels ofpantoprazole were determined by a validated high-performance liquidchromatography method (data represent mean � SEM; n ¼ 4).

Patel et al.





significant difference in drug distribution between treat-ment groups (data not shown).

Pantoprazole pretreatment increases growth delay inMCF-7 tumorsDoxorubicin alone led to growth delay of MCF-7 tumors

(P < 0.05), and there was no effect of pantoprazole alone(Fig. 6A). Mice treated with pantoprazole before a singledose of doxorubicin significantly increased tumor growthdelay compared with the other treatment groups (P < 0.05),with minimal tumor growth out to 42 days; because ofulceration, mice were killed thereafter. Studies were alsocarried out in a second human xenograft model that wasderived from the epidermoid carcinoma cell line A431: thistumor was resistant to doxorubicin but pretreatmentwith pantoprazole led to a small increase in growth delaycomparedwith groups treatedwith the control (P¼0.06) ordoxorubicin alone (P ¼ 0.05; Fig. 6B). There was no sub-stantial loss in body weight after treatment (data notshown).

Mice treated with multiple doses of pantoprazole anddoxorubicin (once aweek for 3weeks), showed even greatergrowth delay of MCF-7 xenografts compared with thesingle-dose combination (Fig. 6C). In A431 xenograftshowever, multiple doses of pantoprazole and doxorubicindid not lead to significant growth delay (Fig. 6D).

DiscussionThe acidic microenvironment in solid tumors may

cause resistance to some anticancer drugs. The pH gradientbetween the cytoplasm and intracellular organelles maybe modified in cancer cells and there may be sequestrationof basic drugs in acidic organelles (16). PPIs increase pHin acidic vesicles such as endosomes (22), thereby inhibit-ing the accumulation of basic cytotoxic agents in acidicorganelles (32) and they may lead to apoptosis of cancercells (33).

In this study, we investigated the effect of pantoprazoleon endosomal pH of cancer cells. We showed that panto-prazole increased endosomal pH at concentrations of 200

0

2

4

6

8

10

12

14

140120100806040200

DO

X f

luo

resc

ence

inte

nsi

ty (

i.u.)

Distance (mm)

DOX

PTP+DOX

BA

C

Figure 5. The distribution ofdoxorubicin fluorescence intensityin relation to the nearest functionalblood vessel in MCF7 xenografts.Mice were treated with eitherdoxorubicin alone (10 minutes) (A)or pantoprazole (2 hours) anddoxorubicin (B) before tumorexcision. Photomicrographs showblood vessels (red), functionalblood vessels (yellow), hypoxicregions (green), and doxorubicin(blue; scale bar represents 100mm). Tumors were imaged andquantified using a customizedalgorithm to show doxorubicinfluorescence intensity in relation todistance from the nearestfunctional blood vessel (C; datarepresent mean � SEM; n ¼ 6).






mmol/L or higher, and the effect of pantoprazole to raiseendosomal pH was enhanced in the presence of (basic)doxorubicin. In addition to pantoprazole, lansoprazole,another type of PPI, and tamoxifen also showed an increasein endosomal pH likely caused by blocking acidification inthese intracellular compartments.

We observed increased uptake of doxorubicin in EMT-6cells with pantoprazole pretreatment in contrast todecreased uptake with the pretreatment in MCF-7 andA431 cells. It is likely that EMT-6 cells have slightly higherlevels of P-glycoprotein (PgP) compared with MCF-7 cells(34, 35). Similar patterns of doxorubicin uptake wereseen in EMT-6 cells and PgP overexpressing AR1 and NCIcells (Supplementary A). This would explain the lowuptake in EMT-6 cells treated with doxorubicin alonecompared with that in MCF-7 and A431 cells. It has beenshown that PPIs, including pantoprazole, inhibit PgP(36), and this may contribute to the increase in druguptake after pantoprazole pretreatment of EMT6 cells.There is evidence that pantoprazole may enhance uptakeof PgP substrates across the blood–brain barrier thatexpresses PgP: for example, there was increased penetra-tion of imatinib in the brain with pantoprazole pretreat-ment (37). We did not measure doxorubicin concentra-tion in mouse brain with and without co-administration

of pantoprazole, but we did not observe neurotoxicity atthe doses used in our experiments.

PPIs have been shown to increase the sensitivity ofresistant tumor cells to the cytotoxic effects of several drugs,including cisplatin, 5-flurouracil, and vinblastine, andincrease the cytotoxicity of these agents in cells (22–24).In the MCF-7 cells, we observed consistently decreaseddoxorubicin uptake with pantoprazole pretreatment. Inthese cells, the nucleus may be saturated with drug andexcess concentrations are likely stored in acidic vesicles,whereas in PPI-treated cells this sequestration will bereduced leading to a decreased net uptake of drug. Withdecreased uptake of drug, there is more drug available todistribute to cells distal from blood vessels. Our in vitro datashow an increase in doxorubicin penetration in pantopra-zole-treated MCCs.

If pantoprazole decreases net drug uptake in cells close toblood vessels, there is likely to be increased doxorubicindistribution to more distant cells and the gradient ofdecreasing doxorubicin intensity will be significantly shal-lower. In MCF-7 xenografts, we observed a significantincrease in doxorubicin distribution in relation to thenearest blood vessel following treatment with pantopra-zole, but this effect was not observed in EMT-6 tumors. Thismay again be attributed to slightly higher levels of PgP in

10

100

1,000

3020100

Days

ControlPTP (×3)DOX (×3)DOX+PTP (×3)

10

100

1,000

50403020100

Days

ControlPTP (×3)DOX (×3)DOX+PTP (×3)

10

100

1,000

50403020100

Mea

n t

um

or

volu

me

(mm

3 )

Mea

n t

um

or

volu

me

(mm

3 )

Mea

n t

um

or

volu

me

(mm

3 )

Mea

n t

um

or

volu

me

(mm

3 )Days

ControlDOXPTPDOX+PTP

10

100

1,000

3020100

Days

ControlDOXPTPDOX+PTP

BA

C D

MCF-7 A431

Figure 6. Effect of pantoprazole anddoxorubicin treatment on tumorgrowth inmice. Mice bearingMCF-7 or A431 xenografts were treatedonce (A and B) or once a week for 3weeks (C and D) with saline,doxorubicin alone, pantoprazolealone, or pantoprazolepretreatment before doxorubicin.Tumor volume in mice wasmeasured every 2 to 4 days (datarepresent mean � SEM; n ¼ 8).

Patel et al.





EMT-6 cells. Previous studies in our laboratory have shownthat PgP inhibitors may decrease doxorubicin distributionin solid tumors derived from PgP-expressing cells, mostlikely because they increase drug uptake in cells close toblood vessels, such that there is less drug available topenetrate to distal cells (30, 38).Causes of drug resistance in tumors are both multifacto-

rial and interconnected such that overcoming one mecha-nismmay not improve drug resistance. An effective strategyis likely to influence several processes that contribute toresistance to give an overall therapeutic benefit. The effect ofpantoprazole in MCF-7 cells and xenografts may improvetherapeutic efficacy by influencing the net effect of differentmechanisms of resistance.Our HPLC data show that pantoprazole concentrations

decrease exponentially after 2 to 3 hours. By administeringpantoprazole 2 hours before doxorubicin, we hypothesizedthat pantoprazole would be in sufficient concentration toincrease pH within intracellular organelles and inhibitdoxorubicin sequestration into these organelles. However,our HPLC data indicatemaximum levels of pantoprazole inthe plasma of mice that are lower than the concentration ofpantoprazole required to raise endosome pH in the in vitroexperiments. High concentrations of pantoprazole mayaccount for increases in doxorubicin penetration in MCCs,but in drug-sensitive xenografts, lower in vivo concentra-tions of pantoprazole in combinationwith doxorubicin canstill alter drug distribution. After 2 hours, plasma concen-trations of pantoprazole are around 200 mmol/L. In druguptake experiments using flow cytometry, we observed thatpantoprazole alters doxorubicin uptake into cells at 1mmol/L and at 100 mmol/L, albeit to a lesser extent. Thefinding that lower doses of pantoprazole are effective in vivomight be because of the role of the microenvironment andits effects on ion gradients between intracellular compart-ments and between intracellular and extracellular pH. Theextracellular pH is known to be low in tumor regions distantfrom functional blood vessels, an effect that will inhibitcellular uptake of basic drugs (11); the pH gradient betweenthe cytoplasm and the extracellular fluid may be inhibitedby PPIs, leading to improved uptake of basic drugs such asdoxorubicin, although this effect is likely to have limitedtherapeutic benefit if the concentration of doxorubicin insuch regions is low.In addition to contributing to drug sequestration, acid-

ic endosomes have been shown to be important in theprocess of autophagy (39). Autophagy may be a survivalmechanism of nutrient-deprived cells (40, 41). Our lab-oratory has preliminary data to indicate that pantopra-zole is an inhibitor of autophagy because of its role inaltering the pH of acidic compartments; this may be anadditional mechanism through which PPIs may effecttherapeutic efficacy (42).The strategies of altering drug sequestration to improve

drug distribution in tumors, and of inhibiting autophagymay overcomedrug resistance.Our studies of tumor growthdelay in mice bearing MCF-7 xenografts indicate a thera-peutic advantage of using pantoprazole in combination

with doxorubicin. Multiple-dose treatments showed evengreater effects to delay growth of MCF-7 tumors. Theseexperiments were repeated in another xenograft modelusing A431 tumors. A431 xenografts were resistant to doxo-rubicin and pantoprazole pretreatment had only minimaleffects to improve growth delay. Preliminary studies in ourlaboratory indicate that docetaxel has much greater effectsto delay growth of this tumor and using biomarkers todemonstrate the distribution of (nonfluorescent) docetaxel(43), we have shown recently that pantoprazole pretreat-ment enhances bothdrugdistribution andantitumor effectsconsiderably (44). These studies indicate that pantoprazolein combination with some chemotherapeutic agents mayhave important antitumor effects.

ConclusionsOur findings suggest that pantoprazole increases endo-

somal pH in tumor cells andmay increase nuclear uptake ofdoxorubicin within these cells. Pantoprazole increases tis-sue penetration of doxorubicin in MCCs and improvesdoxorubicin distribution from blood vessels in solidtumors. Pantoprazole given before doxorubicin increasesgrowth delay when given as single ormultiple doses tomicebearing MCF7 xenografts.

Our data, and those of others (22–24), suggest thatpretreatment with pantoprazolemay be an effective strategyto improve the therapeutic efficacy of chemotherapy insome solid tumors. Increased drug sequestration and lim-ited drug distribution may have important roles in multi-factorial drug resistance and must be considered whendeveloping novel therapeutics treatment strategies.

The preclinical results described in this report led tofunding by the Komen Foundation of a phase I trial ofescalating doses of pantoprazole given before doxorubicin;this trial has shown that high-dose pantoprazole (240 mg)can be given intravenously before standard dose doxorubi-cin without apparent increase in toxicity, and that highserum levels of pantoprazole (�100 mmol/L) can beachieved in patients. Although the main goal of a phase Itrial is to evaluate tolerance and toxicity in patients whohave exhausted standard therapy, responses were seen inthat trial. A phase II trial using pantoprazole with chemo-therapy has been initiated.

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

Authors' ContributionsConception and design: K.J. Patel, I.F. TannockDevelopment of methodology: K.J. Patel, I.F. TannockAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): K.J. Patel, C. LeeAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): K.J. Patel, C. Lee, Q. Tan, I.F. TannockWriting, review, and/or revision of the manuscript: K.J. Patel, I.F.TannockStudy supervision: I.F. Tannock

AcknowledgmentsThe authors acknowledge Dr. E. Chen for providing support with the

HPLC analysis.






Grant SupportThisworkwas supported by a grant from theCanadian Institute forHealth

Research.The costs of publication of this article were defrayed in part by the

payment of page charges. This article must therefore be hereby marked

advertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received January 18, 2013; revised September 12, 2013; accepted October3, 2013; published OnlineFirst October 18, 2013.

References1. Di Paolo A, Bocci G. Drug distribution in tumors: mechanisms, role in

drug resistance, and methods for modification. Curr Oncol Rep2007;9:109–14.

2. Minchinton AI, Tannock IF. Drug penetration in solid tumours. Nat Rev2006;6:583–92.

3. Eikenberry SE. A tumor cord model for doxorubicin delivery and doseoptimization in solid tumors. Theor Biol Med Model 2009;6:16.

4. Kyle AH, Huxham LA, Yeoman DM, Minchinton AI. Limited tissuepenetration of taxanes: a mechanism for resistance in solid tumors.Clin Cancer Res 2007;13:2804–10.

5. Lankelma J, Dekker H, Luque FR, Luykx S, Hoekman K, van der Valk P,et al. Doxorubicin gradients in human breast cancer. Clin Cancer Res1999;5:1703–7.

6. Primeau AJ, Rendon A, Hedley D, Lilge L, Tannock IF. The distributionof the anticancer drug Doxorubicin in relation to blood vessels in solidtumors. Clin Cancer Res 2005;11:8782–8.

7. Tunggal JK, Cowan DS, Shaikh H, Tannock IF. Penetration ofanticancer drugs through solid tissue: a factor that limits theeffectiveness of chemotherapy for solid tumors. Clin Cancer Res1999;5:1583–6.

8. Jain RK. Delivery of molecular medicine to solid tumors. Science1996;271:1079–80.

9. Jain RK. Barriers to drug delivery in solid tumors. Sci Am 1994;271:58–65.

10. Simon S, Roy D, Schindler M. Intracellular pH and the control ofmultidrug resistance. Proc Natl Acad Sci U S A 1994;91:1128–32.

11. Tannock IF, Rotin D. Acid pH in tumors and its potential for therapeuticexploitation. Cancer Res 1989;49:4373–84.

12. Rotin D, Robinson B, Tannock IF. Influence of hypoxia and an acidicenvironment on the metabolism and viability of cultured cells:potential implications for cell death in tumors. Cancer Res 1986;46:2821–6.

13. Helmlinger G, Sckell A, DellianM, Forbes NS, Jain RK. Acid productionin glycolysis-impaired tumors provides new insights into tumormetab-olism. Clin Cancer Res 2002;8:1284–91.

14. Gillies RJ, Liu Z, Bhujwalla Z. 31P-MRSmeasurements of extracellularpH of tumors using 3-aminopropylphosphonate. Am J Physiol1994;267:C195–203.

15. McCoyCL,ParkinsCS,ChaplinDJ,Griffiths JR,RodriguesLM,StubbsM. The effect of blood flow modification on intra- and extracellular pHmeasured by 31P magnetic resonance spectroscopy in murinetumours. Br J Cancer 1995;72:905–11.

16. Raghunand N, He X, van Sluis R, Mahoney B, Baggett B, Taylor CW,et al. Enhancement of chemotherapy bymanipulation of tumour pH. BrJ Cancer 1999;80:1005–11.

17. Ouar Z, Lacave R, Bens M, Vandewalle A. Mechanisms of alteredsequestration and efflux of chemotherapeutic drugs by multidrug-resistant cells. Cell Biol Toxicol 1999;15:91–100.

18. Nishi T, Forgac M. The vacuolar (Hþ)-ATPases—nature's most ver-satile proton pumps. Nat Rev Mol Cell Biol 2002;3:94–103.

19. Sabolic I, Brown D, Verbavatz JM, Kleinman J. H(þ)-ATPases of renalcortical and medullary endosomes are differentially sensitive to Sch-28080 and omeprazole. Am J Physiol 1994;266:F868–77.

20. Larsson H, Mattson H, Sundell G, Carlsson E. Animal pharmacody-namics of omeprazole: a survey of its pharmacological properties invivo. Scand J Gastroenterol Suppl 1985;108:23–35.

21. Lee CM, Tannock IF. Inhibition of endosomal sequestration of basicanticancer drugs: influence on cytotoxicity and tissue penetration. Br JCancer 2006;94:863–9.

22. Luciani F, Spada M, De Milito A, Molinari A, Rivoltini L, Montinaro A,et al. Effect of proton pump inhibitor pretreatment on resistance of

solid tumors to cytotoxic drugs. J Natl Cancer Inst 2004;96:1702–13.

23. DeMilito A, Iessi E, LogozziM, Lozupone F, SpadaM,MarinoML, et al.Proton pump inhibitors induce apoptosis of human B-cell tumorsthrough a caspase-independentmechanism involving reactive oxygenspecies. Cancer Res 2007;67:5408–17.

24. De Milito A, Canese R, Marino ML, Borghi M, Iero M, Villa A, et al. pH-dependent antitumor activity of proton pump inhibitors against humanmelanoma is mediated by inhibition of tumor acidity. Int J Cancer2010;127:207–19.

25. Altan N, Chen Y, Schindler M, Simon SM. Tamoxifen inhibits acidifi-cation in cells independent of the estrogen receptor. Proc Natl AcadSci U S A 1999;96:4432–7.

26. Mayer LD, Bally MB, Cullis PR. Uptake of adriamycin into largeunilamellar vesicles in response to a pH gradient. Biochim BiophysActa 1986;857:123–6.

27. Cowan DS, Hicks KO, Wilson WR. Multicellular membranes as an invitromodel for extravascular diffusion in tumours.Br JCancer 1996;27:S28–31.

28. Hicks KO, Ohms SJ, van Zijl PL, Denny WA, Hunter PJ, WilsonWR. Anexperimental and mathematical model for the extravascular transportof a DNA intercalator in tumours. Br J Cancer 1997;76:894–903.

29. Peres O, Oliveira CH, Barrientos-Astigarraga RE, Rezende VM,Mendes GD, de Nucci G. Determination of pantoprazole in humanplasma by LC-MS-MS using lansoprazole as internal standard.Arzneimittelforschung 2004;54:314–9.

30. Patel KJ, Tannock IF. The influence of P-glycoprotein expression andits inhibitors on the distribution of doxorubicin in breast tumors. BMCCancer 2009;9:356.

31. Less JR, Skalak TC, Sevick EM, Jain RK. Microvascular architecture ina mammary carcinoma: branching patterns and vessel dimensions.Cancer Res 1991;51:265–73.

32. Ouar Z, Bens M, Vignes C, Paulais M, Pringel C, Fleury J, et al.Inhibitors of vacuolar Hþ-ATPase impair the preferential accumu-lation of daunomycin in lysosomes and reverse the resistance toanthracyclines in drug-resistant renal epithelial cells. Biochem J2003;370:185–93.

33. Yeo M, Kim DK, Kim YB, Oh TY, Lee JE, Cho SW, et al. Selectiveinduction of apoptosis with proton pump inhibitor in gastric cancercells. Clin Cancer Res 2004;10:8687–96.

34. Twentyman PR, Wright KA, Fox NE. Characterisation of a mousetumour cell line with in vitro derived resistance to verapamil. Br JCancer 1990;61:279–84.

35. Chen YN, Mickley LA, Schwartz AM, Acton EM, Hwang JL, Fojo AT.Characterization of adriamycin-resistant human breast cancer cellswhich display overexpression of a novel resistance-relatedmembraneprotein. J Biol Chem 1990;265:10073–80.

36. Pauli-Magnus C, Rekersbrink S, Klotz U, Fromm MF. Interactionof omeprazole, lansoprazole and pantoprazole with P-glycoprotein.Naunyn Schmiedeberg's Arch Pharmacol 2001;364:551–7.

37. Breedveld P, Pluim D, Cipriani G,Wielinga P, van Tellingen O, SchinkelAH, et al. The effect of Bcrp1 (Abcg2) on the in vivo pharmacokineticsand brain penetration of imatinib mesylate (Gleevec): implications forthe use of breast cancer resistance protein and P-glycoprotein inhi-bitors to enable the brain penetration of imatinib in patients. CancerRes 2005;65:2577–82.

38. Tunggal JK, Melo T, Ballinger JR, Tannock IF. The influence of expres-sion of P-glycoprotein on the penetration of anticancer drugs throughmulticellular layers. Int J Cancer 2000;86:101–7.

39. Bampton ET, Goemans CG, Niranjan D, Mizushima N, TolkovskyAM. The dynamics of autophagy visualized in live cells: from

Patel et al.





autophagosome formation to fusion with endo/lysosomes. Autop-hagy 2005;1:23–36.

40. Apel A,Herr I, SchwarzH,RodemannHP,Mayer A.Blocked autophagysensitizes resistant carcinoma cells to radiation therapy. Cancer Res2008;68:1485–94.

41. Abedin MJ, Wang D, McDonnell MA, Lehmann U, Kelekar A. Autop-hagy delays apoptotic death in breast cancer cells following DNAdamage. Cell Death Differ 2007;14:500–10.

42. MarinoML, Fais S, Djavaheri-MergnyM, Villa A, Meschini S, LozuponeF, et al. Proton pump inhibition induces autophagy as a survival

mechanism following oxidative stress in human melanoma cells. CellDeath Dis 2010;1:e87.

43. Saggar JK, Fung AS, Patel KJ, Tannock IF. Use of molecularbiomarkers to quantify the spatial distribution of effects of antican-cer drugs in solid tumors. Mol Cancer Ther 2013;12:542–52.

44. Tan Q, Tannock IF. Inhibition of autophagy as a potential mechanismby which pantoprazole, a proton pump inhibitor (PPI), enhances thetherapeutic effect of docetaxel chemotherapy for solid tumors. Postersession presented at: 103rd Annual Meeting of the American Associ-ation for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL.






Published OnlineFirst October 18, 2013.Clin Cancer Res Krupa J. Patel, Carol Lee, Qian Tan, et al. to Improve the Therapy of Solid Tumors

StrategyDistribution and Activity of Doxorubicin: A Potential Use of the Proton Pump Inhibitor Pantoprazole to Modify the

Updated version

10.1158/1078-0432.CCR-13-0128doi:

Access the most recent version of this article at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/early/2013/11/22/1078-0432.CCR-13-0128To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-13-0128

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/early/2013/11/22/1078-0432.CCR-13-0128


use of the proton pump inhibitor pantoprazole to modify the

Documents