ORIGINAL PAPER

PEGylated liposomal Gemcitabine: insights into a potentialbreast cancer therapeutic

Anne-Laure Papa & Almas Sidiqui & Sri Uma Aisvarya Balasubramanian & Sasmit Sarangi &Matthew Luchette & Shiladitya Sengupta & Rania Harfouche

Accepted: 12 August 2013 /Published online: 1 October 2013# International Society for Cellular Oncology 2013

AbstractPurpose Nanoencapsulation of chemotherapeutics is anestablished method to target breast tumors and has beenshown to enhance the efficacy of therapy in various animalmodels. During the past two decades, the nucleoside analogGemcitabine has been under investigation to treat both recal-citrant and localized breast cancer, often in combination withother chemotherapeutics. In this study, we investigatedthe chemotherapeutic efficacy of a novel Gemcitabine-encapsulated liposome previously formulated by our group,GemPo, on both sensitive (4T1) and recalcitrant (MDA-MB-231) breast cancer cell lines.Methods Gemcitabine free drug and liposomal Gemcitabinewere compared both in vitro and in vivo using breast cancermodels.Results We demonstrated that GemPo differently hindered thegrowth, survival and migration of breast cancer cells,

according to their drug sensitivities. Specifically, whereasGemPo was a more potent cytotoxic and apoptotic agent insensitive breast cancer cells, it more potently inhibited cellmigration in the resistant cell line. However, GemPo still actedas a more potent inhibitor of migration, in comparison withfree Gemcitabine, irrespective of cell sensitivity. Administra-tion of GemPo in a 4T1-bearing mouse model inhibited tumorgrowth while increasing mice survival, as compared with freeGemcitabine and a vehicle control. Interestingly, the inclusionof a mitotic inhibitor, Paclitaxel, synergized only with freeGemcitabine in this model, yet was as effective as GemPoalone. However, inclusion of Paclitaxel with GemPo signifi-cantly improved mouse survival.Conclusions Our study is the first to demonstrate the pleio-tropic effects of Gemcitabine and Gemcitabine-loadednanoparticles in breast cancer, and opens the door for a noveltreatment for breast cancer patients.

Keywords PEGylated liposomes . Gemcitabine . Paclitaxel .

Breast cancer

1 Introduction

Breast cancer is the second most common cause of death fromcancer and the most frequently diagnosed solid organ cancerin women in United States [1]. Recent evidence indicates thatmetastatic breast cancer has increased among women underthe age of 40 over the last three decades (increase of 2.07 %per year) [2]. These younger patients often have a poorerprognosis and respond poorly to standard therapy, thus requir-ing aggressive chemotherapeutic regimens, which negativelyhinder the quality of life. Hence, there is an urgent need toenhance the efficacy of chemotherapeutic treatments while

Electronic supplementary material The online version of this article(doi:10.1007/s13402-013-0146-4) contains supplementary material,which is available to authorized users.

A.<L. Papa :A. Sidiqui : S. U. A. Balasubramanian : S. Sarangi :M. Luchette : S. Sengupta :R. HarfoucheBWH-HST Center for Biomedical Engineering,Harvard Medical School, 65 Landsdowne Street,Cambridge, MA 02139, USA

A.<L. Papa :A. Sidiqui : S. U. A. Balasubramanian : S. Sarangi :M. Luchette : S. Sengupta (*) :R. Harfouche (*)Harvard-MIT Division of Health Sciences and Technology,Harvard Medical School, 65 Landsdowne Street,Cambridge, MA 02139, USAe-mail: shiladit@MIT.EDUe-mail: rharfo@gmail.com

Cell Oncol. (2013) 36:449–457DOI 10.1007/s13402-013-0146-4

minimizing toxicity, so as to improve metastatic breast cancermanagement and patient survival.

Over the last decades the use of nanoparticles has improveddrug half‐life in plasma and has enhanced drug delivery at thetumor site by passive targeting (Enhanced Permeability andRetention effect, EPR), due to the “leakiness” of the vasculaturesurrounding the tumor [3, 4]. This strategy decreases adverseeffects because the drug is mostly delivered at the tumor site,allowing a decrease in the effective administered dose. Further-more, the engineering of stealth nanoparticles by PEGylationprotects them from degradation by the reticular endothelialsystem [5] so as to improve drugs’ plasma circulation time.Several drug-loaded nanoformulations are currently under clin-ical investigation or have been approved by the Food and DrugAdministration (FDA) for oncology [5, 6], including the lipo-some Doxil in 1995 [7] and, more recently, Marqibo [8].

Gemcitabine is a nucleoside analog that inhibits DNAsynthesis, thus leading to apoptosis. It is the first line treatmentagainst pancreatic cancer [9, 10]. This drug has also demon-strated additional efficacy in cancer treatments with differentdrug combinations. For example, Gemcitabine in combinationwith Cisplatin is the first line chemotherapy in the treatment ofadvanced stage non-small cell lung cancer [11]. Gemcitabinein combination with carboplatin has been shown to be effec-tive for the treatment of advanced ovarian cancer patients whohave relapsed after platinum-based therapy [12]. Additionally,the clinical relevance of a combined Gemcitabine/Paclitaxeltreatment has been demonstrated in breast cancer patients whowere first treated with anthracyclines [13–15]. However,these new regimens come with a major drawback: increasedside effects [12]. The encapsulation of one or both drugscould considerably decrease these effects and improve pa-tients’ quality of life. We have recently shown that a novelGemcitabine-encapsulated stealth liposome, GemPo, robustlyinhibited growth while increasing apoptosis in a human resis-tant pancreatic cancer cell line [16].

The present study is the first to investigate the effects ofGemPo versus Gemcitabine in vitro and in vivo, using resis-tant and sensitive breast cancer cells. Cell viability, apoptosisand migration studies highlighted the greater effect of thenanoformulation as opposed to the free drug. Additionally,in vivo studies demonstrated that Paclitaxel inclusion differ-entially modulated the tumorigenic effects of Gemcitabine andGemPo.

2 Materials and methods

2.1 Materials

All the solvents were purchased from Sigma-Aldrich (St-Louis, MO) and Fisher Scientific (Pittsburgh, PA). L-α-Phosphatidylcholine (PC), cholesterol and Sephadex G-25

were obtained from Sigma-Aldrich, whereas 1,2-distearoyl-sn -glycero-3-phosphoethanolamine-N-[amino(polyethyleneglycol)-2000] (DSPE-PEG2,000) was fromAvanti Polar Lipids(Alabaster, AL). Gemcitabine hydrochloride was purchasedfrom Tocris (Ellisville, MO). Paclitaxel was purchased fromLC Laboratories. Annexin V-Alexa Fluor 488 and propidiumiodide were obtained from Invitrogen (Carlsbad, CA). TheCellTiter 96 AQueous One Solution Cell Proliferation (MTS)Assay was obtained from Promega Corporation (Madison,WI). UV-visible spectra were measured using a ShimadzuUV-2450 UV-visible spectrophotometer. A Malvern ZetasizerNano (Westborough, MA) was used to measure the size dis-tribution and zeta potential of the nanoparticles.

2.2 Synthesis of Gemcitabine-encapsulated nanoliposomes(GemPo)

Liposomes were synthesized using a lipid film hydration tech-nique as previously described [16]. Briefly, phosphatidylcho-line, cholesterol and DSPE-PEG2000 (10:5:1 mass ratio) weredissolved and mixed in dichloromethane. For GemPo, 4 mg ofGemcitabine dissolved in methanol was added. Solvent wasthen evaporated leading to lipid film formation. Lipid filmswere then hydrated for 1 h30 in phosphate buffered saline andsubsequently subjected to 25 extrusion cycles with a 0.2 μmpore size polycarbonate membrane. Samples were then puri-fied on a Sephadex G‐25 collumn. Gemcitabine loading inGemPo was assessed by UV-visible spectroscopy at 268 nm.DMSO was used as solvent to disintegrate liposomes.

Alternative methods are described in the literature, in whichweak base drugs are actively entrapped in liposomes with anammonium ion gradient used to actively load the drug andimprove its stability [17].

2.3 Transmission Electron Microscopy (TEM)

A drop of sample was deposited onto a carbon membranesupported by a copper grid. After drying, a drop of 2 % uranylacetate solution was added to enhance the contrast of thesample. A JEOL 1200-EX operating at 80 kV was used forthe observations.

2.4 Zeta potential of GemPo and liposomes

The zeta potentials of GemPo and liposomes were measuredat different pH values in NaCl 10−2 mol.L−1 using a MalvernZetasizer Nano.

2.5 Cell culture

MDA-MB-231 and 4T1 breast cancer cells were obtained fromthe American Type Tissue Culture Collection (Rockville,MD). MDA-MB-231 was maintained in DMEM medium

450 A.-L. Papa et al.

and 4T1 in RPMImedium, both supplemented with 10% FBSand 1 % antibiotic/antimycotic (all from Invitrogen). Cellswere grown in 100mm dishes and replated at 5,000 cells.cm−2.Cells were washed extensively with serum-deprived medium(1% FBS) prior to drug addition. The conditions used through-out in vitro experiments consisted of free Gemcitabine, emptyliposomes and GemPo. Solvent-treated cells served as internalcontrols.

2.6 Cell viability assay

Cancer cells in 96-well plates were incubated with drugs for48 and 72 h. Cell viability was quantified using 3-(4,5-Di-methylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) from the CellTiter 96AQueous One Solution kit. Briefly, the cells were washedwith PBS, incubated with 0.3 mg.mL−1 of MTS in basalmedium without phenol red for 2 h at 37 °C. Absorbancevalues were then measured at 490 nm in a plate reader(Versamax, Molecular Devices, Sunnyvale, CA). Absorbancevalues were ploted after background subtraction and normal-ized to solvent-treated cells.

2.7 Apoptosis study

Cells grown in 6-well plates were treated with drugs(Gemcitabine, 1 μM) for 48 and 72 h. Cells were first incu-bated with 5 μL AnnexinV-Alexa Fluor 488 in binding buffer(10 mM HEPES, 140 mM NaCl, 2.5 mM CaCl2, pH 7.4) for15 min, then rinsed and counterstained with propidium iodide(PI). Samples were analyzed using an Accuri C6 flowcytometer (BD Accuri Cytometers, Ann Arbor, MI).AnnexinV-Alexa Fluor 488, propidium iodide or both wereomitted for the negative controls and for cell population gating.

2.8 Wound healing assay

Confluent cell monolayers were maintained in a 24-wellplate and mechanically wounded with a sterile 10 μL pi-pette tip. Cells were then washed extensively in basal mediato remove debris and incubated with 1 μM drug or vehiclein serum-deprived media over 12 h. Wound healing wasvisualized at the start‐ (t 0) and end-points of the experi-ment, using inverted bright field microscopy. Images werequantified using Image J software and reported as follows:% wound healing=[100− (wound area at t 12 hrs/woundarea at t0)]×100.

2.9 Syngeneic breast cancer mouse model

1×106 4T1 cells were inoculated subcutaneously in the flanksof 6-week-old BALB/c female mice (weighing 18 g, CharlesRiver Laboratories, Wilmington, MA). Mice were split into

six groups with five animals per group. Drug therapy wasstarted 10 days following the implantation of tumor cells,when tumors reached a volume of 120 mm3. Drug therapyconsisted of the administration of 200 μL GemPo orGemcitabine injections in the tail vein on alternate days(7 mg.kg−1 per injection). In parallel, intraperitoneal injectionof Paclitaxel (5 mg.kg−1 per injection) was administered to therelevant groups. Mice were sacrificed when the tumorsreached a volume of 750 mm3 or when the skin started toulcerate. Tumor volumes and body weights were monitoredevery alternate day. All animal procedures were approved bythe Harvard institutional IUCAC committee.

2.10 Statistical analysis

All results were expressed as mean ± SEM of at leastquintuplate samples. Statistical comparisons were obtainedusing the GraphPad Prism software using one-way ANOVA,followed by the Newman-Keuls test. Probability (p) valuesless than 0.05 were considered significant.

3 Results

3.1 Physico-chemical properties of GemPo nanoparticles

Gemcitabine was entrapped into cholesterol-based stealth li-posomes as depicted in Fig. 1a, using lipid film hydration. Wetermed this nanoformulation GemPo, which yielded aGemcitabine loading of 49.5 μg.mg−1 and an encapsulationefficiency of 35 % [16]. TEM images (Fig. 1b) showed thatspherical nanoparticles were obtained. The zeta potential ofthe liposomes and GemPo were measured at different pH(Fig. 1c). The two samples were positively charged at pHvalues lower than their isoelectric point (IEP=pH 6.4 and6.0 for GemPo and liposomes, respectively) while negativelycharged at higher pH values.

3.2 More cytotoxicity of GemPo as comparedwith Gemcitabine

To assess whether Gemcitabine activity is potentiated in thelipoformulation, the cytotoxicity profiles of GemPo versus freedrug were evaluated in breast cancer cells using the MTS assay.MDA-MB-231 (Fig. 2a–b) and 4T1 (Fig. 2c–d) cell lines wereincubated with drugs for 48 and 72 h. At 48 h inMDA-MB-231cells, GemPo was significantly more potent than Gemcitabine,whose effect was closer to the liposome control at up to 1 μM(Fig. 2a). At 72 h, GemPo was significantly more potent atlower doses (<0.5 μM) (Fig. 2b). At 48 h in 4T1 cells, bothdrugs followed a parallel sigmoidal distribution, with GemPoslightly but significantly more potent at up to 1μMdosage, afterwhich dose both drugs caused 100 % cytotoxicity (Fig. 2c). At

A Gemcitabine nanoparticle to potentially treat breast cancer 451

72 h, the density of cells was decimated (Fig. 2d). Representa-tive IC50 values at 48 h were only obtained for the 4T1 cell line:0.4 μM for GemPo against 1 μM for free Gemcitabine. Theresistant nature of the MDA‐MB‐231 cells was manifested by adrug plateau at high doses, explaining the lack of IC50 extrapo-lation. At 48 h, free liposomeswere not cytotoxic, irrespective of

the cell line, as expected. However, a slight proliferation effectresulted from liposomes’ incubation with the MDA‐MB‐231cell line, as previously observed in resistant cells by our group[16]. A slight proliferation effect of Gemcitabine was alsoobserved at low doses (Fig. 2a–b) for the resistant cell line whileit was never the case for GemPo.

Fig. 1 a Schematicrepresentation of GemPo(liposomal Gemcitabine) self-assembly by evaporation-hydration technique. bTransmission ElectronMicroscopy image of GemPo andc size distribution obtained byDynamic Light Scattering ofGemPo after 25 cycles ofextrusion through a 0.2 μm poresize polycarbonate membrane, dZeta potential depending on thepH of both GemPo (blackdiamond) and empty liposome(black square) nanoparticles inNaCl 10−2 mol.L−1

48 h

rs

72 h

rs

4T1

0.01 0.10 1 10 1000

25

50

75

100

125

150

Dose (µM)

Cel

l Pro

lifer

atio

n (%

of

cont

rol)

GemGemPoLipo

C

MDA-MB-231

0.001 0.010 0.100 1 10 1000

25

50

75

100

125

150

Dose (µM)

Cel

l Pro

lifer

atio

n (%

of

cont

rol)

GemGemPoLipo

A

4T1

0.05 0.50 1.00 5.00 10.00 20.000

50

100

150

200

250

Dose (µM)

Cel

l Pro

lifer

atio

n (%

con

trol

)

D

MDA-MB-231

0.05 0.50 1.00 5.00 10.00 20.000

50

100

150

200

250

Cel

l Pro

lifer

atio

n (%

con

trol

)

B

Dose (µM)

Fig. 2 MTS (3-(4,5-Dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium)viability assays of (a–b) MDA-MB-231 and (c–d) 4T1 cellsusing increasing doses ofliposomes, Gemcitabine andGemPo (liposomal Gemcitabine)at (a–c) 48 h and (b–d) 72 h.4,000 cells/well were seeded in 96well plates before being serumstarved overnight after a growthperiod of 24 h. Drugs versuscontrol were then incubated foreither 48 or 72 h

452 A.-L. Papa et al.

3.3 More apoptosis by GemPo as comparedwith Gemcitabine

As a nucleoside analog, Gemcitabine replaces one of the basesduring DNA replication, thus inhibiting replication and lead-ing to DNA fragmentation and cell death [18, 19]. To deter-mine whether there is a difference in the induction of apopto-sis between GemPo and Gemcitabine, cancer cells were incu-bated with these drugs for 48 h (Fig. S1) and 72 h (Fig. 3) andthe apoptotic profiles between both nanoformulations wereinvestigated using the Annexin V-propidium iodide (PI) co-staining assay and analyzed by flow cytometry, which depictsearly apoptotic cells as Annexin V+/PI− and late apoptotic cellsas Annexin V+/PI+. In both cell lines, empty liposomes did notinduce an increase in the apoptotic cell percentage as comparedwith control cells (Fig. 3a, b), whereas Gemcitabine exertedapoptosis in both cell lines. This effect was more explicit in the4T1 model. Most importantly, the percentage of late apoptoticcells was significantly higher for GemPo than the freeGemcitabine: GemPo 19.5 % and 54.8 %, Gemcitabine17.4 % and 39.2 %, respectively for MDA-MB-231 and 4T1.The same trend was seen at 48 h.

3.4 GemPo reverses the pro-migratory effect of Gemcitabinein MDA-MB-231 cells

To ascertain whether GemPo affects other mechanisms asidefrom cell death, its effect on migration, an indicator of metas-tasis, was assessed by means of the wound-healing assay.Cells were incubated with 1 μM of GemPo or Gemcitabinefor 12 h. This dose (1 μM) represents IC50 at 48 h (experi-mental time was reduced to 12 h in order to observe migration,and not proliferation, effects). Free Gemcitabine induced sig-nificant cell migration in the resistant cell line (Fig. 4a), andthis effect was reversed to below control levels by GemPo(Fig. 4, Histograms). Figure 4a shows a marked migrationinhibition by GemPo as opposed to the free drug and thecontrol in MDA-MB-231 cells, with 43.1 % of cell migrationinhibited in comparison with the control. In 4T1 cells(Fig. 4b), Gemcitabine had no significant migratory effect,whereas GemPo administration inhibited migration by 19.5 %in comparison with the control group.

3.5 In vivo assessment of GemPo in breast cancer ‐ drugcombination study

The effects of GemPo versus free Gemcitabine were evaluatedin vivo using a 4T1 syngeneic breast cancer model. Treatmentsconsisted of GemPo, Gemcitabine or PBS control, alone or incombination with Paclitaxel. GemPo treatment resulted in asignificantly decreased tumor burden (Fig. 5a) and increase inanimal survival, an indicator of systemic toxicity (Fig. 5b) , ascompared with free Gemcitabine, demonstrating that an

enhanced permeability and retention (EPR) effect wasachieved. As for Paclitaxel, GemPo was not able to stabilizetumor growth by itself, but showed an additive effect whencombined with free Gemcitabine in terms of both tumor burden(Fig. 5a, at day 16: 1.66 fold change for Paclitaxel and 0.69 forGemcitabine/Paclitaxel) and mice survival (Fig. 5b). Interest-ingly, tumor burden was not significantly different betweenGemPo and the GemPo/Paclitaxel combination (Fig. 5a, atday 16: 0.65 fold change for GemPo and 0.59 for GemPo/Paclitaxel). In addition, both GemPo and Gemcitabine/Paclitaxel followed the same trend in decreasing tumorburden. According to IACUC standards, mice wereeuthanized as soon as any early skin ulceration wassuspected, due to the growth of the tumor. Interestingly,21 days after tumor implantation, the Paclitaxel group hadbeen fully euthanized (5 mice), followed by the control (4mice), Gemcitabine (3 mice), GemPo (2 mice) andGemcitabine/Paclitaxel (1 mouse) groups (Fig. 5b). At day20, GemPo/Paclitaxel was the only group in which none ofthe mice had been euthanized. In addition, no significantchange in body weight was observed among the differenttreatment groups, demonstrating that the liposomal deliverysystem did not induce systemic toxicity (Fig. 5c).

4 Discussion

Liposomes with a size distribution centered at 149.5±1.7 nmwere synthesized, as previously reported [16], a size at which theEPR effect is well-known to be achieved in vivo [20].Gemcitabine release from liposomes has previously been studiedby our group [16] and others [21, 22]: it follows a burst releaseprofile within a few hours. In the current study we additionallydemonstrated that the zeta potential of GemPo and liposomeswere positive below their IEP value and negative above it. Thisresult suggests that while circulating in the blood stream orbefore cell internalization, the particles are slightly negativelycharged, whereas they would become neutral or slightly posi-tively charged in the tumor microenvironment. Indeed, the ex-tracellular pH (pHe) of a malignant tumor is acidic (pHe≈6.5–6.9) in comparison with the pHe of healthy tissue [23, 24]. Theneutral [25] or positive [26, 27] surface charges would thenfacilitate cell uptake, as previously established [25–27].

Using both resistant (MDA-MD-231) and sensitive (4T1)breast cancer cell lines, we showed that GemPo was signifi-cantly more cytotoxic than free Gemcitabine at all doses tested.Moreover, the gap between both drugs was more visible in theresistant cell line, most likely due to the higher drug-sensitivityin 4T1. GemPo and Gemcitabine were effective in inducingapoptosis in both cell lines, but more potently in 4T1, asexpected from the cytotoxicity study. Interestingly, these resultshave shown that GemPo treatment provoked more apoptosisthan Gemcitabine in both models as well. This finding is in

A Gemcitabine nanoparticle to potentially treat breast cancer 453

agreement with prior literature where liposomal Gemcitabinetreatment led to a higher level of apoptosis than free drug inmultiple myeloma cancer cells [28]. The improved cytotoxicand apoptotic effects obtained by the nanoparticle formulationcould be explained by (i) a better intracellular trafficking, whichis characteristic of PEGylated nanoparticles [29], and (ii) abetter internalization by both membrane fusion [30, 31] andendocytosis of liposomes [30–32], as opposed to the free drug.In turn, this increases the concentration of drug per surface areaat the membrane level.

The effect of liposomal Gemcitabine on cell migration hasbeen assessed by the wound-healing assay. Gemcitabine actu-ally increased migration in the resistant cell line only,

consistent with a previous report in pancreatic cancer [33]and in agreement with our MTS data at low doses. Interesting-ly, GemPo completely reversed this effect. Although the un-derlying mechanisms are not the scope of this study, previousreports have shown that this effect is achieved by liposomesbypassing MDR-mediated resistance [34, 35]. This resultcould be potentially significant for the treatment of metastaticcancer, as this pro-migratory response can be avoided withGemPo. In the sensitive cell line, however, Gemcitabine hadno significant effect, whereas GemPo could still inhibit migra-tion. To the best of our knowledge, this is the first time thatGemcitabine-loaded liposomes were observed to reverseGemcitabine migratory effects in breast cancer cell lines.

AEmpty LiposomesControl

Gemcitabine GemPo

Contro

lLip

oGem

GemPo

0

10

20

30

40

LR: early apoptoticUR: apoptoticPI positive (UR+UL)

% o

f M

DA

-MB

-231

&

&

BEmpty LiposomesControl

Gemcitabine GemPo

Contro

lLip

oGem

GemPo

0

10

20

30

40

% o

f 4T

1

Fig. 3 Apoptosis study (Annexin-FITC/PI co-staining) performed on aMDA-MB-231 and b 4T1 cells at 72 h. Cells grown in 6-well plates weretreated with both free and liposomal Gemcitabine (GemPo) at a dose of1 μM for 48 and 72 h. Cells were first incubated with 5 μL AnnexinV-Alexa Fluor 488 in binding buffer (10 mM HEPES, 140 mM NaCl,2.5 mM CaCl2, pH 7.4) for 15 min, then rinsed and counterstained with

propidium iodide (PI). Representative images are shown as well ashistograms presenting the percentages of viable cells (LL, Lower Left),early apoptotic cells (LR, Lower Right), late apoptotic cells (UR, UpperRight) and necrotic cells (UL, Upper Left). Significances (values less than#p <0.05, ##p<0.01, ###p <0.001) were calculated and compared to (#)vehicle and (&) Gemcitabine

454 A.-L. Papa et al.

Control Gemcitabine GemPo

MDA-MB-231

4T1

A

B

Fig. 4 Wound healing assay of a MDA-MB-231 and b 4T1 cell lines at12 h. Confluent cell monolayers were maintained in a 24-well plate andmechanically scratched with a sterile 10 μL pipette tip. Cells were thenwashed extensively in basal media to remove debris and incubated with1 μM of drug (Gemcitabine and GemPo: liposomal Gemcitabine) or

vehicle in serum-deprived media over 12 h. Representative pictures areshown on left, with histograms on right. Significances (values less than#p <0.05, ##p<0.01, ###p <0.001) were calculated and compared to (#)vehicle and (&) Gemcitabine

10 12 14 160.0

0.4

0.8

1.2

1.6

2.0

2.4

2.8

Vehicle

GemGemPo

Paclitaxel

Gem + PaclitaxelGemPo+ Paclitaxel

Days (following tumor implantation)

Fol

d ch

ange

in t

umor

vol

ume

A

10 12 14 160.90

0.95

1.00

1.05

1.10

Vehicle

GemGemPo

Paclitaxel

Gem + PaclitaxelGemPo + Paclitaxel

Days (following tumor implantation)

Bod

y w

eigh

t (g

)

C

Vehicl

e

Paclit

axel

Gem

GemPo

Gem+Pac

litax

el

GemPo+

Paclit

axel

0

1

2

3

4

5

Num

ber

of e

utha

nize

d m

ice

at d

ay 2

1

B

Fig. 5 In vivo efficacy of GemPo in a subcutaneous 4T1 syngeneicmouse model. a Fold change in tumor volume is presented over time.The first double injection (Gemcitabine or GemPo IV: intravenously, andPaclitaxel IP: intraperitoneally) was performed 10 days after tumor im-plantation (1×106 4T1 cells were inoculated subcutaneously in the flanksof 6-week-old BALB/c female mice, 5 mice per group). Double injectionswere administered every alternate day (arrows ). b Effect of drug

treatment on survival as a measure of systemic toxicity. c Change inbody weight of treated mice over the course of the experiment. Doubleinjections were administered every alternate day (arrows). The firstdouble injection (Gemcitabine or GemPo IV: intravenously, and Pacli-taxel IP: intraperitoneally) was performed 10 days after tumor implanta-tion (1×106 4T1 cells were inoculated subcutaneously in the flanks of 6-week-old BALB/c female mice, 5 mice per group)

A Gemcitabine nanoparticle to potentially treat breast cancer 455

Metastatic breast cancer patients treated with Gemcitabineand Paclitaxel in combination versus Paclitaxel alone havebeen investigated in a phase III clinical trial in 2008 [15]. Theoutcome of the trial was that this combination was mostefficient for the treatment of advanced breast cancer. Howev-er, this combination significantly increased the incidence ofsystemic side effects [15]. In the present study, the rationalefor a GemPo and Paclitaxel combination was thatnanoencapsulation would reduce Gemcitabine-associated sys-temic toxicity, due to passive homing to the tumor site by theEPR effect. It is also important to mention that Gemcitabine,in combination with Paclitaxel, is an important second linechoice for treatment of metastatic breast cancer patients whodo not respond to or cannot tolerate anthracyclines [13–15].Our results showed that GemPo in combination with Pacli-taxel was as efficient as the Gemcitabine/Paclitaxel co-treatment in reducing tumor size burden. However, the lipo-somal Gemcitabine/Paclitaxel combination was the most ef-fective doublet chemotherapy for enhancing animal survival,which is indicative of lessened systemic side effects.

Chemotherapeutic regimens consisting of a liposome and afree drug have been studied in the past and have shownequivalent responses in comparison to more toxic chemother-apy [36]. Decreasing toxicity of chemotherapy is extremelyimportant in combination therapy as toxicity might be additiveand could necessitate discontinuation of therapy. The majordose limiting toxicity of gemcitabine is myelosuppression[37], which is difficult to reproduce in mice at therapeuticdosages. This could be because, unlike in the clinic, longcourses of chemotherapy are not practically achievable inanimal studies. The average weight lost by the mice duringthe study has been used as a surrogate marker of toxicity [38],however our data indicated no significant weight differencesbetween the various treatment groups.

5 Conclusions

Gemcitabine-loaded liposomes have been evaluated in com-parison with free Gemcitabine in resistant and sensitive breastcancer models. It has been demonstrated that GemPo is signif-icantly more effective than Gemcitabine in inhibiting cell via-bility and promoting apoptosis, regardless of breast cancersensitivity, both in vitro and in vivo. Interestingly, whileGemcitabine showed a pro-migratory effect on the resistant cellline only, GemPo was able to significantly reverse this effect.This result highlights previously unknown pleiotropic effects ofthe GemPo nanoformulation. In vivo experiments indicatedthat liposomal Gemcitabine was as effective as aGemcitabine/Paclitaxel combination in reducing tumor vol-umes, whereas it decreased systemic side effects when com-bined with Paclitaxel, highlighting the clinical potential ofGemPo. Significant systemic toxicity was not observed by the

combination of GemPo and Paclitaxel, as was evident by thelack of significant weight loss in the treated mice. These resultshighlight that GemPo could replace free Gemcitabine treatmentfor breast cancer patients, alone or with adjuvant chemotherapysuch as Paclitaxel, a regimen with already established clinicalrelevance. Nanoparticle albumin bound (nab)‐Paclitaxel hasalso emerged as an important drug to reduce the toxicity ofPaclitaxel allowing the administration of larger doses and facil-itating its use in a first line setting [39, 40]. A combination ofnab‐Paclitaxel andGemPomay also hold promise for the futureand could potentially reduce the increased adverse effects ofthis doublet chemotherapy regimen while improving the pa-tients’ prognosis and quality of life.

Acknowledgments This work was supported by a Department of De-fense BCRP Era of Hope Scholar award [W81XWH-07-1-0482] and aMary Kay Ash Charitable Foundation Grant to Prof. S. Sengupta. R.Harfouche is supported by a CIHR Fellowship. A.L. Papa is grateful forthe support extended by the Carnot Foundation. We acknowledge LouiseTrakimas, Maria Ericsson and Elizabeth Benecchi from the Departmentof Cell Biology at Harvard Medical School for their advice and technicalsupport during TEM imaging.

Conflict of interest The authors declare no conflict of interest.

References

1. American Cancer Society, Cancer Facts & Figures 2013 (AmericanCancer Society, Atlanta, 2013)

2. R.H. Johnson, F.L. Chien, A. Bleyer, Incidence of breast cancer withdistant involvement among women in the United States, 1976 to2009. JAMA: J. Am. Med. Assoc. 309 , 800–805 (2013)

3. R. Harfouche, S. Basu, S. Soni et al., Nanoparticle-mediated targetingof phosphatidylinositol-3-kinase signaling inhibits angiogenesis. An-giogenesis 12 , 325–338 (2009)

4. A. Paraskar, S. Soni, B. Roy et al., Rationally designed oxaliplatin-nanoparticle for enhanced antitumor efficacy. Nanotechnology 23 ,075103 (2012)

5. M. Wang, M. Thanou, Targeting nanoparticles to cancer. Pharmacol.Res. Off. J. Ital. Pharmacol. Soc. 62 , 90–99 (2010)

6. B.Y. Kim, J.T. Rutka, W.C. Chan, Nanomedicine. N. Engl. J. Med.363 , 2434–2443 (2010)

7. Y.C. Barenholz, D. Peer, Liposomes, lipid biophysics, andsphingolipid research: from basic to translation research. Chem. Phys.Lipids 165 , 363–364 (2012)

8. J.A. Silverman, S.R. Deitcher, Marqibo((R)) (vincristine sulfate lipo-some injection) improves the pharmacokinetics and pharmacody-namics of vincristine. Cancer Chemother. Pharmacol. 71 , 555–564(2013)

9. H.A. Burris 3rd, M.J. Moore, J. Andersen et al., Improvements insurvival and clinical benefit with gemcitabine as first-line therapy forpatients with advanced pancreas cancer: a randomized trial. J. Clin.Oncol. Off. J. Am. Soc. Clin. Oncol. 15 , 2403–2413 (1997)

10. J.L. Van Laethem, C. Verslype, J.L. Iovanna et al., New strategies anddesigns in pancreatic cancer research: consensus guidelines reportfrom a European expert panel. Ann. Oncol. Off. J. Eur. Soc. Med.Oncol. / ESMO 23, 570–576 (2012)

11. G.V. Scagliotti, P. Parikh, J. von Pawel et al., Phase III study com-paring cisplatin plus gemcitabine with cisplatin plus pemetrexed in

456 A.-L. Papa et al.

chemotherapy-naive patients with advanced-stage non-small-cell lungcancer. J. Clin. Oncol. Off. J. Am. Soc. Clin. Oncol. 26 , 3543–3551(2008)

12. J. Pfisterer, M. Plante, I. Vergote et al., Gemcitabine plus carboplatincompared with carboplatin in patients with platinum-sensitive recur-rent ovarian cancer: an intergroup trial of the AGO-OVAR, the NCICCTG, and the EORTC GCG. J. Clin. Oncol. Off. J. Am. Soc. Clin.Oncol. 24, 4699–4707 (2006)

13. A.M. Murad, R.C. Guimaraes, B.C. Aragao et al., Phase II trial of theuse of paclitaxel and gemcitabine as a salvage treatment in metastaticbreast cancer. Am. J. Clin. Oncol. 24 , 264–268 (2001)

14. P. Vici, A. Fabi, G. Metro et al., Phase I-II trial of prolongedgemcitabine infusion plus paclitaxel as a biweekly schedule foradvanced breast cancer patients pretreated with anthracyclines. Can-cer Chemother. Pharmacol. 67, 687–693 (2011)

15. K.S. Albain, S.M. Nag, G. Calderillo-Ruiz et al., Gemcitabine plusPaclitaxel versus Paclitaxel monotherapy in patients with metastaticbreast cancer and prior anthracycline treatment. J. Clin. Oncol. Off. J.Am. Soc. Clin. Oncol. 26 , 3950–3957 (2008)

16. A.L. Papa, S. Basu, P. Sengupta et al., Mechanistic studies ofGemcitabine-loaded nanoplatforms in resistant pancreatic cancercells. BMC Cancer 12, 419 (2012)

17. G. Haran, R. Cohen, L.K. Bar et al., Transmembrane ammonium sulfategradients in liposomes produce efficient and stable entrapment of am-phipathic weak bases. Biochim. Biophys. Acta 1151, 201–215 (1993)

18. K. Pliarchopoulou, D. Pectasides, Pancreatic cancer: current andfuture treatment strategies. Cancer Treat. Rev. 35 , 431–436 (2009)

19. M.R. Kano, Y. Bae, C. Iwata et al., Improvement of cancer-targetingtherapy, using nanocarriers for intractable solid tumors by inhibitionof TGF-beta signaling. Proc. Natl. Acad. Sci. U. S. A. 104 , 3460–3465 (2007)

20. M.Gaumet, A.Vargas, R. Gurny et al., Nanoparticles for drug delivery:the need for precision in reporting particle size parameters. Eur. J.Pharm. Biopharm. Off. J. Arbeitsgemeinschaft fur PharmazeutischeVerfahrenstechnik e.V 69 , 1–9 (2008)

21. T. Lammers, V. Subr, K. Ulbrich et al., Simultaneous delivery ofdoxorubicin and gemcitabine to tumors in vivo using prototypicpolymeric drug carriers. Biomaterials 30, 3466–3475 (2009)

22. D. Cosco, A. Bulotta,M. Ventura et al., In vivo activity of gemcitabine-loaded PEGylated small unilamellar liposomes against pancreatic can-cer. Cancer Chemother. Pharmacol. 64, 1009–1020 (2009)

23. R.J. Gillies, Z. Liu, Z. Bhujwalla, 31P-MRS measurements of extra-cellular pH of tumors using 3-aminopropylphosphonate. Am. J.Physiol. 267 , C195–C203 (1994)

24. V. Estrella, T. Chen, M. Lloyd et al., Acidity generated by the tumormicroenvironment drives local invasion. Cancer Res. 73 , 1524–1535(2013)

25. R.R. Arvizo, O.R. Miranda, D.F. Moyano et al., Modulating phar-macokinetics, tumor uptake and biodistribution by engineerednanoparticles. PloS One 6 , e24374 (2011)

26. S. Lerch, M. Dass, A. Musyanovych, et al., Polymeric nanoparticlesof different sizes overcome the cell membrane barrier. Eur. J. Pharm.Biopharm. Off. J. Arbeitsgemeinschaft fur PharmazeutischeVerfahrenstechnik e.V, (2013)

27. R. Luo, B. Neu, S.S. Venkatraman, Surface functionalization ofnanoparticles to control cell interactions and drug release. Small 8 ,2585–2594 (2012)

28. C. Celia, N. Malara, R. Terracciano et al., Liposomal delivery improvesthe growth-inhibitory and apoptotic activity of low doses of gemcitabinein multiple myeloma cancer cells. Nanomed. Nanotechnol. Biol. Med.4, 155–166 (2008)

29. J. Suh, K.L. Choy, S.K. Lai et al., PEGylation of nanoparticlesimproves their cytoplasmic transport. Int. J. Nanomedicine 2 , 735–741 (2007)

30. M.R. Almofti, H. Harashima, Y. Shinohara et al., Cationic liposome-mediated gene delivery: biophysical study and mechanism of inter-nalization. Arch. Biochem. Biophys. 410 , 246–253 (2003)

31. M. Kapoor, D.J. Burgess, Cellular uptake mechanisms of novelanionic siRNA lipoplexes. Pharm. Res. 30, 1161–1175 (2013)

32. J. Rejman, V. Oberle, I.S. Zuhorn et al., Size-dependent internaliza-tion of particles via the pathways of clathrin- and caveolae-mediatedendocytosis. Biochem. J. 377 , 159–169 (2004)

33. K. Quint, M. Tonigold, P. Di Fazio et al., Pancreatic cancer cellssurviving gemcitabine treatment express markers of stem cell differ-entiation and epithelial-mesenchymal transition. Int. J. Oncol. 41 ,2093–2102 (2012)

34. F.M. Kievit, F.Y.Wang, C. Fang et al., Doxorubicin loaded iron oxidenanoparticles overcome multidrug resistance in cancer in vitro. J.Control. Release Off. J. Control. Release Soc. 152 , 76–83 (2011)

35. X. Dong, R.J. Mumper, Nanomedicinal strategies to treat multidrug-resistant tumors: current progress. Nanomedicine (Lond) 5 , 597–615(2010)

36. V. Adamo, V. Lorusso, R. Rossello et al., Pegylated liposomal doxo-rubicin and gemcitabine in the front-line treatment of recurrent/metastatic breast cancer: a multicentre phase II study. Br. J. Cancer98 , 1916–1921 (2008)

37. A.M. Storniolo, S.R. Allerheiligen, H.L. Pearce, Preclinical, pharma-cologic, and phase I studies of gemcitabine. Semin. Oncol. 24, S7-2–S7-7 (1997)

38. R. Graeser, C. Bornmann, N. Esser et al., Antimetastatic effects ofliposomal gemcitabine and empty liposomes in an orthotopic mousemodel of pancreatic cancer. Pancreas 38, 330–337 (2009)

39. W.J. Gradishar, S. Tjulandin, N. Davidson et al., Phase III trial ofnanoparticle albumin-bound paclitaxel compared with polyethylatedcastor oil-based paclitaxel in women with breast cancer. J. Clin.Oncol. Off. J. Am. Soc. Clin. Oncol. 23, 7794–7803 (2005)

40. W.J. Gradishar, D. Krasnojon, S. Cheporov et al., Significantly longerprogression-free survival with nab-paclitaxel compared withdocetaxel as first-line therapy for metastatic breast cancer. J. Clin.Oncol. Off. J. Am. Soc. Clin. Oncol. 27, 3611–3619 (2009)

A Gemcitabine nanoparticle to potentially treat breast cancer 457