0090-9556/08/3608-1570–1577$20.00
DRUG METABOLISM AND DISPOSITION
Copyright © 2008 by The American Society for Pharmacology and Experimental Therapeutics
DMD 36:1570–1577, 2008
Vol. 36, No. 8
20735/3363173
Printed in U.S.A.
Squalenoylation Favorably Modifies the in Vivo Pharmacokinetics
and Biodistribution of Gemcitabine in Mice
L. Harivardhan Reddy, Hania Khoury, Angelo Paci, Alain Deroussent, Humberto Ferreira,
Catherine Dubernet, Xavier Declèves, Madeleine Besnard, Helène Chacun,
Sinda Lepêtre-Mouelhi, Didier Desmaële, Bernard Rousseau, Christelle Laugier,
Jean-Christophe Cintrat, Gilles Vassal, and Patrick Couvreur
Received January 31, 2008; accepted May 9, 2008
ABSTRACT:
Gemcitabine (2ⴕ,2ⴕ-difluorodeoxyribofuranosylcytosine; dFdC) is
an anticancer nucleoside analog active against wide variety of
solid tumors. However, this compound is rapidly inactivated by
enzymatic deamination and can also induce drug resistance. To
overcome the above drawbacks, we recently designed a new
squalenoyl nanomedicine of dFdC [4-(N)-trisnorsqualenoyl-gemcitabine (SQdFdC)] by covalently coupling gemcitabine with the
1,1ⴕ,2-trisnorsqualenic acid; the resultant nanomedicine displayed
impressively greater anticancer activity compared with the parent
drug in an experimental murine model. In the present study, we
report that SQdFdC nanoassemblies triggered controlled and prolonged release of dFdC and displayed considerably greater t1/2
(⬃3.9-fold), mean residence time (⬃7.5-fold) compared with the
dFdC administered as a free drug in mice. It was also observed that
the linkage of gemcitabine to the 1,1ⴕ,2-trisnorsqualenic acid noticeably delayed the metabolism of dFdC into its inactive difluorodeoxyuridine (dFdU) metabolite, compared with dFdC. Additionally,
the elimination of SQdFdC nanoassemblies was considerably
lower compared with free dFdC, as indicated by lower radioactivity
found in urine and kidneys, in accordance with the plasmatic concentrations of dFdU. SQdFdC nanoassemblies also underwent
considerably higher distribution to the organs of the reticuloendothelial system, such as spleen and liver (p < 0.05), both after singleor multiple-dose administration schedule. Herein, this paper brings
comprehensive pharmacokinetic and biodistribution insights that
may explain the previously observed greater efficacy of SQdFdC
nanoassemblies against experimental leukemia.
Gemcitabine (2⬘,2⬘-difluorodeoxyribofuranosylcytosine; dFdC) is a
nucleoside analog active against a wide variety of cancers including
nonsmall cell lung and pancreatic cancers. Marketed as Gemzar, it
acts as an antimetabolite, inhibiting ribonucleotide reductase and
DNA synthesis and inducing apoptosis (Cappella et al., 2001). How-
ever, this compound has a narrow therapeutic index due to rapid
enzymatic deamination by deoxycytidine deaminase into its corresponding inactive uracil derivative [difluorodeoxyuridine (dFdU)]
(Abbruzzese et al., 1991; Moog et al., 2002; Reid et al., 2004).
Therefore, the administration of high doses of drug is needed to
achieve the desired therapeutic response but simultaneously, this leads
to various adverse effects mainly myelosuppression (Abbruzzese et
al., 1991). Another limitation associated with the use of gemcitabine
is the induction of anticancer drug resistance due to the downregulation of deoxycytidine kinase enzyme responsible for the primary phosphorylation of gemcitabine (Ruiz van Haperen et al., 1994;
Bergman et al., 2002). To address the problem of deamination, which
is the key of the above issues, various attempts have been made to
This study was supported by the Agence Nationale de la Recherche (Grant
SYLIANU), by the Centre National de la Recherche Scientifique (Grant Ingénieur
de Valorisation), by the Université Paris-Sud (postdoctoral fellowship to L.H.R.),
and by the Institut Gustave Roussy (postdoctoral fellowship to H.K.).
L.H.R. and H.K. contributed equally to this work.
Article, publication date, and citation information can be found at
http://dmd.aspetjournals.org.
doi:10.1124/dmd.108.020735.
ABBREVIATIONS: dFdC, gemcitabine, 2⬘,2⬘-difluorodeoxyribofuranosylcytosine; dFdU, difluorodeoxyuridine; SQdFdC, squalenoyl gemcitabine,
4-(N)-trisnorsqualenoyl-gemcitabine; LC, liquid chromatography; MS, mass spectrometry; HPLC, high-performance liquid chromatography; THU,
tetrahydrouridine; AUC, area(s) under the plasma concentration-time curve; Cl, total body clearance; Vss, volume of distribution; dFdCNA, dFdC
released from SQdFdC nanoassemblies; MRT, mean residence time.
1570
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
Université Paris XI, Faculté de Pharmacie, Unité Mixte de Recherche Centre National de la Recherche Scientifique 8612,
Institut Fédératif de Recherche 141, Châtenay-Malabry cedex, France (L.H.R., H.F., C.D., M.B., H.C., P.C.); Unité Propre de
Recherche et de l’Enseignement Supérieur, Equipe d’Accueil 3535-IFR 54 Pharmacology and New Cancer Treatments,
Institute Gustave Roussy and Paris XI University, Villejuif, France (H.K., A.P., G.V.); Mass Spectrometry Platform, Institut
Fédératif de Recherche 54, Institute Gustave Roussy, Villejuif, France (A.D.); Université Paris V, Faculté de Pharmacie, Unité
Mixte de Recherche Centre National de la Recherche Scientifique 7157, Paris, France (X.D.); Université Paris XI, Faculté de
Pharmacie, Unité Mixte de Recherche Centre National de la Recherche Scientifique 8076 Biocis, Institut Fédératif de
Recherche 141, Châtenay-Malabry cedex, France (S.L.-M., D.D.); and iBiTecS, Service de Chimie Bioorganique et de
Marquage, Commissariat à l’Energie Atomique, Gif sur Yvette, France (B.R., C.L., J.-C.C.)
IN VIVO FATE OF SQUALENOYL GEMCITABINE NANOMEDICINE
1571
netics and biodistribution in vivo are key issues that may represent a
possible explanation.
Thus, in this communication, the pharmacokinetics of the SQdFdC
nanoassemblies were assessed after i.v. administration in healthy
mice, compared with dFdC, by simultaneous quantification of the
drug concentrations in plasma using liquid chromatography (LC)tandem mass spectrometry. The in vivo metabolic kinetics of SQdFdC
versus dFdC conversion into the inactive metabolite, dFdU, was also
investigated, using an additional LC-MS method. Finally, the organ
distribution of SQdFdC nanoassemblies was estimated and compared
with that of dFdC by using the corresponding tritiated compounds.
Materials and Methods
synthesize lipidic prodrugs (Myhren et al., 2000; Castelli et al., 2006).
One of the synthetic options used is to link the 4-amino site of dFdC
to a fatty acid derivative, thereby protecting it from deamination by
cytidine deaminase (Castelli et al., 2006). This strategy of linking a
long chain of fatty acids also aims to improve the lipophilicity of
dFdC and, hence, increase its interactions with the lipidic membranes
and its cellular uptake (Castelli et al., 2006). However, such compounds were poorly water soluble, which rendered their i.v. administration challenging.
In this context, we recently discovered that by coupling nucleoside
analogs to squalene (a natural lipid, precursor of the cholesterol
biosynthesis), the resulting compounds self-organized in water as
nanoassemblies with an average diameter of 100 to 200 nm, whatever
the nature of the nucleoside analog used and the location of the
linkage. “Squalenoylation” is a new promising concept in drug delivery (Couvreur et al., 2006). In the case of gemcitabine, the covalent
coupling with squalene was performed on the 4-amino moiety of the
molecule (Fig. 1). The obtained nanoassemblies [4-(N)-trisnorsqualenoyl-gemcitabine (SQdFdC)] had an average diameter of 130 nm
when dispersed in water (Couvreur et al., 2006). In vitro, SQdFdC
nanoassemblies exhibited a greater cytotoxicity on sensitive and resistant cancer murine and human cell lines than gemcitabine
(Couvreur et al., 2006; Reddy et al., 2007). In vivo, SQdFdC nanoassemblies exhibited impressively greater anticancer activity against
experimental murine leukemia, producing also higher rates of apoptosis in cancer cells as compared with free dFdC (Reddy et al., 2007).
However, the mechanism behind the observed anticancer efficacy of
SQdFdC nanoassemblies is not yet understood. Altered pharmacoki-
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
FIG. 1. Chemical structure of gemcitabine (A) and SQdFdC (B).
Drugs and Reagents. dFdC was purchased from Sequoia (Pangbourne,
UK) with 98% minimum purity, and [5⬘-3H]-Gemcitabine hydrochloride was
obtained from Moravek Biochemicals (Brea, CA). Deoxycytidine, used as
internal standard in the pharmacokinetic analysis, was obtained from SigmaAldrich (St. Louis, MO) with 99% purity. dFdU was synthesized by Jena
Bioscience (Jena, Germany) with 95% minimum HPLC purity. Reagents
and HPLC-grade solvents were provided by Thermo Electron Corporation
(Waltham, MA), formic acid by Merck (Darmstadt, Germany). Ultrapure water
was prepared using a Milli-Q system (Millipore Corporation, Billerica, MA).
Drug-free heparinized human plasma was obtained from EFS (Rungis,
France). Tetrahydrouridine (THU) was provided by Calbiochem (San Diego,
CA). Dextrose was purchased from Sigma-Aldrich. Soluene, Ultima Gold, and
Hionic-Fluor were purchased from PerkinElmer Life and Analytical Sciences
(Waltham, MA).
Synthesis of SQdFdC and [5ⴕ-3H]SQdFdC. The synthesis of SQdFdC was
performed as previously described (Couvreur et al., 2006). Practically, triethylamine (0.15 g, 1.4 mmol) was added, under nitrogen, to a stirred solution of
trisnorsqualenoic acid (0.50 g, 1.2 mmol) in anhydrous THF (4 ml). The
mixture was cooled to 0°C, and ethyl chloroformate (0.135 g, 1.2 mmol) in
anhydrous THF (3 ml) was added drop-wise. The mixture was stirred at 0°C
for 15 min, and a solution of gemcitabine hydrochloride (0.37 g, 1.2 mmol) and
triethylamine (0.15 g, 1.4 mmol) in anhydrous DMF (5 ml) was added
drop-wise to the reaction at the same temperature. The reaction was stirred for
72 h at room temperature, and the reaction mixture was concentrated in vacuo.
Aqueous sodium hydrogen carbonate was added, and the mixture was extracted with ethyl acetate (3 ⫻ 50 ml). The combined extracts were washed
with water, dried on MgSO4, and concentrated. The crude product was purified
by chromatography on silica gel and eluting with CH2Cl2/MeOH/Et3N, 100:
2:1, then with CH2Cl2/MeOH/Et3N, 100:5:1, to give pure 4-N-squalenoylgemcitabine (0.46 g, 57%) as an amorphous white solid. [␣]D ⫽ 3.1 (c ⫽ 0.95,
CH2Cl2); IR (neat, cm⫺1) 3500 –3150, 2950, 2921, 2856, 1709, 1656, 1635,
1557, 1490, 1435, 1384, 1319, 1275, 1197, 1130, 1071; 1H NMR (300 MHz,
CDCl3) ␦␮ 9.15 (broad s, 1 H, NHCO), 8.16 (d, 1 H, J ⫽ 7.5 Hz, H6), 7.47
(d, 1 H, J ⫽ 7.5 Hz, H5), 6.18 (t, 1 H, J ⫽ 7.0 Hz, H1⬘), 5.22–5.15 (m, 5 H,
HC ⫽ C(CH3)), 4.49 (m, 1 H, H3⬘), 4.86 – 4.09 (m, 3 H, H4⬘ and H5⬘), 2.55
(m, 2 H), 2.38 –2.28 (m, 2 H), 2.13–1.91 (m, 16 H, CH2), 1.69 –1.55 (m, 18 H,
C ⫽ C(CH3)); 13C NMR (75 MHz, CDCl3) ␦␮ 173.7 (CONH), 163.0 (HNC ⫽
N), 155.8 (NCON), 145.4 (CH, C6), 135.1 (C), 134.9 (2 C), 132.7 (C), 131.2
(C), 125.7 (CH), 124.4 (2 CH), 124.2 (2 CH), 122.4 (CF2), 97.7 (CH, C5), 81.7
(2 CH), 69.2 (CH), 59.9 (CH2), 39.7 (2 CH2), 39.5 (CH2), 36.5 (CH2), 34.3
(CH2), 28.3 (2 CH2), 26.8 (CH2), 26.7 (CH2), 26.6 (CH2), 25.6 (CH3), 17.6
(CH3), 16.0 (3 CH3), 15.8 (CH3); MS (-ESI) m/z ⫽ 644 ([M-H]-, 100%); Anal.
Calcd for C36H53N3O5: C, 66.95, H, 8.27, N, 6.51. Found: C, 66.76, H, 8.40,
N, 6.39.
For [5⬘-3H]dFdC, 3.5 mCi of 3H-gemcitabine ([5⬘-3H]dFdC) was mixed
with cold dFdC hydrochloride (10 mg) to obtain 10 mg of tritiated dFdC
hydrochloride (3.5 mCi, 0.033 mmol). The synthesis of tritiated 4-(N)trisnorsqualenoyl-[5⬘-3H]gemcitabine was then performed according to the
experimental procedure described before to provide pure 3 H-4-(N)trisnorsqualenoyl gemcitabine ([5⬘-3H]SQdFdC) (0.017 g, 84%, 84.2 mCi/
mmol). The specific activity was adjusted to 2.89 mCi/mmol by mixing this
material together with cold SQdFdC (343 mg) to obtain finally 360 mg of
[5⬘-3H]SQdFdC. [5⬘-3H]SQdFdC was stored at ⫺20°C as a solution of 6
mg/ml in acetone/ethanol (1:1).
1572
REDDY ET AL.
MA). SQdFdC and dFdC were detected in the positive ion mode using tandem
mass spectrometry using multiple reaction monitoring transitions, m/z 646.5 3
112 and m/z 264 3 112, respectively. The quantification process was performed over the 10 to 10,000 ng/ml concentration range using MassLynx
software (Waters).
Method 2: Quantification of dFdU in plasma. Chromatographic conditions
for the quantification of dFdU were adapted from Xu et al. (2004) using an
Uptisphere C18 column, 3 ␮m, 2-mm i.d. ⫻ 100-mm length (Interchim) at a
flow rate of 0.2 ml/min. Gradient elution started with the initial step for 2 min
of 100% of solvent A [ammonium formate (5 mM, pH 7.5)/methanol, 98:2
(v/v)], followed by 8-min linear gradient up to 100% solvent B [methanol/
ammonium formate (5 mM, pH 7.5), 98:2 (v/v)] and held for 0.1 min, then
back again to 100% solvent A. The column was then equilibrated for 7 min at
the initial conditions before the next injection. Prior to each sample injection,
the autosampler was washed with 20 ␮l of solvent methanol/ammonium
formate [5 mM, pH 7.5, 50:50 (v/v)] and then with 30 ␮l with ammonium
formate (5 mM, pH 7.5). The total time of analysis per each sample was
17 min.
Detection was performed with a Quattro-LCZ triple quadrupole mass spectrometer equipped with an orthogonal electrospray source (Waters). dFdU was
detected in the positive ion mode using single-ion recording mode of the [M ⫹
H]⫹ ion at m/z 265.0 and 228.0 for dFdU and the internal standard, respectively. The quantification process was performed over the 10 to 10,000 ng/ml
concentration range using MassLynx software (Waters).
Pharmacokinetic analysis. Pharmacokinetic calculations were performed by
noncompartmental analysis using WinNonlin, version 3.2 (Pharsight, Mountain View, CA). The areas under the plasma concentration-time curve (AUC;
hours per nanogram per milliliter) starting from the first to the last sampling
time were calculated using the trapezoidal rule (linear up-log down) with
extrapolation to infinity using the ratio Cn/Ke, where Cn was the last measurable concentration. The extrapolated AUC before the first time point and after
the last time point did not exceed 15% of the total AUC0-⬁. The elimination t1/2
was determined using the three last time points of the kinetics. In the SQdFdC
nanoassembly-treated mice, total body clearance (Cl) and volume of distribution (Vss) of SQdFdC were calculated by using a dose of 36.6 mg/kg (the molar
equivalent of 15 mg/kg dFdC). The ratio (r) was defined as the ratio of
plasmatic concentrations of dFdU to those of dFdC.
Organ Distribution Studies of 3H-SQdFdC Nanoassemblies Compared
with 3H-dFdC. Tritiated (3H-) versions of dFdC or SQdFdC nanoassemblies
were used to study the organ distribution after i.v. administration in healthy
DBA/2 mice. The mice were divided into various groups of three in each and
injected i.v. with 3H-dFdC and 3H-SQdFdC nanoassemblies at doses equivalent to 10 mg/kg. The radioactivity at this dose corresponded to 0.9 ␮Ci/mouse
for 3H-dFdC and 3H-SQdFdC nanoassemblies, and the final volume of injections remained 100 ␮l for both drugs. For single-dose biodistribution studies,
at 5 min (0.083 h) and 30 min (0.5 h) and 1, 2, 6, and 24 h after i.v.
administration, blood was collected from the retro-orbital plexus into the
heparinized tubes from the respective injected groups, and the mice were then
sacrificed. The pharmacokinetic parameters were calculated by WinNonlin,
version 3.2 (Pharsight, Mountain View, CA) as described before. For multipledose biodistribution studies after the two-dose schedule (injections on days 0
and 4) and the four-dose schedule (injections on days 0, 4, 8, and 13), blood
was collected after 4 and 2 days of completion of the schedules, respectively.
The blood was centrifuged at 3000 rpm for 5 min at 4°C, and the supernatant
plasma was isolated. To 100 ␮l of plasma, 10 ml of Ultima Gold scintillant was
added, and the mixture was vortexed vigorously for 1 min. The vials were kept
aside for 2 h to allow the foam to dissipate and then counted in a beta counter.
The organs, including the liver, lung, spleen, kidney, muscle, stomach, intestine, and brain, were dissected and blotted using tissue paper to remove
adherent blood and any fatty matter. Fecal samples were processed in a similar
manner. Approximately 50 to 100 mg of the tissues was weighed and placed
in the scintillation counting vials. One milliliter of soluene was added into the
vials containing tissue and kept in an incubator overnight at 60°C to allow
complete dissolution. Later, 10 ml of Hionic-Fluor was added to the vials,
which were vortexed vigorously for 1 min. The vials were kept aside for 1 h
to allow the foam to dissipate. Separately, the urine samples were mixed with
10 ml of Ultima Gold, vortexed for 1 min, and set aside for 1 h for the foam
to dissipate. The radioactivity present in the samples and standards was
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
Preparation and Characterization of 4-(N)-Trisnorsqualenoylgemcitabine Nanoassemblies. Cold or 3H-SQdFdC nanoassemblies were prepared by
the nanoprecipitation technique. Briefly, SQdFdC was dissolved in ethanol (10
mg/ml) and added drop-wise under magnetic stirring (500 rpm) into a 5%
(w/v) aqueous dextrose solution. Precipitation of the nanoassemblies occurred
spontaneously. Ethanol was completely evaporated using a Rotavapor at 37°C
to obtain an aqueous suspension of pure SQdFdC nanoassemblies (5 mg/ml).
The mean particle size of the nanoassemblies was determined at 20°C by
quasielastic light scattering using a nanosizer (Coulter N4MD; Beckman
Coulter, Inc., Fullerton, CA). The selected angle was 90°, and the measurement
was made after dilution of the suspension of nanoassemblies in water.
Animal Experiments. The animal experiments were carried out according
to the principles of laboratory animal care and legislation in force in France.
DBA/2 mice (5– 6 weeks old) weighing about 15 to 18 g were used for the
studies and were purchased from Janvier (Le-Genest-Saint-Isle, France). Mice
were provided with standard mouse food and water ad libitum. For pharmacokinetic and biodistribution studies, mice were divided into two main groups
for treatment with dFdC or with SQdFdC nanoassemblies, with subgroups of
three to five mice for each time point. The control group contained the mice
that were not treated with any of the above drugs. The pharmacokinetic and
biodistribution studies were performed separately with different animals.
Pharmacokinetic Studies. Drug administration and sampling. SQdFdC
nanoassemblies or free dFdC were injected at equimolar doses of 15 mg of
dFdC/kg. dFdC was prepared in 0.9% NaCl, and SQdFdC nanoassemblies
(36.6 mg/kg SQdFdC) were prepared in 5% dextrose (see Preparation and
Characterization of 4-(N)-Trisnorsqualenoylgemcitabine Nanoassemblies).
Both compounds were administered by the retro-orbital route. Blood samples
were collected into tubes containing 8 ␮l of THU (10 mg/ml in water) from
retro-orbital plexus (from the side other than that administered) after 1, 5, 15,
and 30 min and 1, 2, and 4 h of administration (n ⫽ 4 or 5). As for the
long-time pharmacokinetic study, blood samples were collected from the
retro-orbital plexus after 2, 4, 8, 24, and 48 h (n ⫽ 4). Without delay, plasma
was separated by centrifugation at 3000g for 10 min at 4°C. Extraction and
quantification of SQdFdC and dFdC in the plasma samples were performed
using a validated high-performance liquid chromatography tandem mass spectrometry (Khoury et al., 2007), and dFdU, the metabolite of dFdC, was assayed
by a modified LC-MS method described by Xu et al. (2004).
High-performance liquid chromatography and tandem mass spectrometry.
Practically, SQdFdC, dFdC, and dFdU plasmatic concentrations were determined as follows. Briefly, to 100 ␮l of THU-pretreated mice plasma, 20 ␮l of
10 ␮g/ml deoxycytidine (in methanol) was added. The mixture was first
vortexed for 10 s, then 1 ml of acetonitrile/methanol [9:1 (v/v)] was added. The
samples were vortexed for 10 s and then centrifuged at 15,800g (Eppendorf
centrifuge 5415R; Eppendorf North America, New York, NY) for 20 min at
4°C. The supernatant was transferred to a conical polypropylene tube and
evaporated to dryness under a nitrogen flow, at room temperature. The residues
were then stored at ⫺20°C until analysis. For SQdFdC/dFdC quantitation, the
residues were dissolved in 200 ␮l of a mixture of methanol/water containing
0.1% formic acid [95:5 (v/v)], and for the quantification of dFdU, the residues
were resuspended in ammonium formate (5 mM, pH 7.5). Twenty microliters
of the extracted samples was injected into the chromatographic system, and the
analyses were performed using an 1100 series HPLC system (Agilent Technologies, Palo Alto, CA) including an autosampler and a binary pump.
Method 1: Quantification of dFdC and SQdFdC in plasma. Chromatographic conditions for the quantification of dFdC and SQdFdC were similar to
those described earlier (Khoury et al., 2007) using an Uptisphere C18 column,
3 ␮m, 2-mm i.d. ⫻ 50-mm length (Interchim, Montluçon, France) at a flow
rate of 0.2 ml/min. Gradient elution started with the initial step for 1 min of
100% of solvent A [methanol/water with 0.1% formic acid, 10:90 (v/v)],
followed by 1-min linear gradient up to 100% solvent B [methanol/water with
0.2% formic acid, 95:5 (v/v)] and held for 6 min, then back again to 100%
solvent A. The column was then equilibrated for 6 min at the initial conditions
before the next injection. Prior to each sample injection, the autosampler
needle was washed twice with 30 ␮l of solvent methanol/water with 0.1%
formic acid [95:5 (v/v)]. The total time of analysis per each sample was
14 min.
Detection was performed with a Quattro-LCZ triple quadrupole mass spectrometer equipped with an orthogonal electrospray source (Waters, Milford,
IN VIVO FATE OF SQUALENOYL GEMCITABINE NANOMEDICINE
1573
FIG. 2. Comparative plasma concentration versus
time curves of dFdC (‚; 15 mg/kg), SQdFdC (⽧)
(15 mg/kg mol Eq of dFdC), and dFdCNA (f) (i.e.,
the dFdC released from SQdFdC nanoassemblies)
in mice. The values are the mean ⫾ S.D. of four
mice.
Results
Pharmacokinetic Study. SQdFdC, dFdC, and dFdC released from
SQdFdC nanoassemblies in mice plasma. The pharmacokinetic analysis after i.v. administration of SQdFdC nanoassemblies was performed compared with free dFdC. Additionally, the plasmatic appearance of dFdC released from SQdFdC nanoassemblies, i.e., dFdC
released from SQdFdC nanoassemblies (dFdCNA), in the blood stream
was also quantified. The plasma elimination curves of dFdC and
SQdFdC nanoassemblies are depicted in Fig. 2.
The plasma concentration-time curve of dFdC, in mice treated with
15 mg/kg free dFdC, was characterized by non-monoexponential
decay, suggesting that dFdC pharmacokinetics followed at least a
two-compartment model (Fig. 2, open triangles), as already described
in mice and human plasma (Fogli et al., 2002; Immordino et al., 2004;
Wang et al., 2005). dFdC rapidly disappeared from plasma with a
short terminal t1/2 of 1.6 h and Cl of 2770 ml/h/kg (Table 1).
The plasma drug time course in the group of SQdFdC nanoassembly-treated mice seemed to follow a similar two-exponential decay
(Fig. 2, diamonds) with a t1/2 of 8.6 h and Cl of 831 ml/h/kg. dFdCNA
presented a higher terminal elimination t1/2 (6.3 h) as compared with
that when administered as free drug (1.6 h), but a similar total plasma
clearance (2481 ml/h/kg) (Table 1). dFdCNA total plasma clearance
was calculated because a complete cleavage of the SQdFdC into
dFdCNA was observed. Indeed, the AUC0-⬁ of dFdC and dFdCNA
were comparable. Additionally, the MRT of dFdCNA was 7.5-fold
higher, and its Vss increased 6.8-fold (1074 and 7344 ml/kg for dFdC
and dFdCNA, respectively).
dFdU in mice plasma. The plasmatic concentrations of dFdU were
measured after i.v. administration of either free dFdC or SQdFdC
nanoassemblies. Figure 3 illustrates the plasmatic concentration-time
curves of dFdU in the dFdC-treated group (Fig. 3, open triangles) and
in the SQdFdC nanoassembly-treated group (Fig. 3, diamonds).
In the dFdC-treated group, the highest plasmatic concentrations of
dFdU (Cmax, 10,375 ng/ml) was measured at 0.5 h postinjection,
indicating the rapid deamination of dFdC (Table 1). On the other
hand, in the SQdFdC nanoassembly-treated group, a comparable
value of Cmax (10,756 ng/ml) was observed at 2 h postinjection.
Herein, the tmax of dFdU resulting from the administration of SQdFdC
nanoassemblies (at equivalent dose than dFdC) was markedly delayed
(Table 1; Fig. 3). Other relevant pharmacokinetic parameters were
listed in Table 1. Surprisingly, the AUC of dFdU in the SQdFdCtreated group was 2-fold higher (112,701 versus 54,243 ng/h/ml).
Organ Distribution. Tissue distribution of radioactivity was determined in mice after single or multiple (two doses on days 0 and 4
and four doses on days 0, 4, 8, and 13) i.v. dosing of either 3H-dFdC
or 3H-SQdFdC nanoassemblies. Of all the organs evaluated up to 24 h
after single-dose injection of 3H-dFdC, higher distribution occurred in
spleen (Fig. 4). At 1 h postinjection, the splenic accumulation of
3
H-dFdC radioactivity was the highest (37.4% injected dose/g), followed by a constant decrease with time. The overall distribution of
3
H-dFdC radioactivity to intestine was relatively higher than the other
organs evaluated except the spleen (Table 2; Fig. 4). The accumulation of 3H-dFdC radioactivity in the liver slightly increased at 30 min
postinjection (9.3% injected dose/g) but later showed a constant
decrease with time.
The distribution of the radioactivity resulting from 3H-SQdFdC
nanoassemblies to organs such as heart, kidney, muscle, stomach, and
intestine was lower compared with the radioactivity resulting from
3
H-dFdC at all the time points studied (Table 2). The low concentrations in the kidneys suggest slower excretion of the 3H-SQdFdC
nanoassemblies compared with the 3H-dFdC. Interestingly, the distribution and retention of 3H-SQdFdC nanoassembly radioactivity in the
organs of the reticuloendothelial system were significantly higher than
that of 3H-dFdC radioactivity at all the time points studied up to 6 h
postinjection (Fig. 4). In liver, accumulation of the 3H-SQdFdC nanoassembly radioactivity increased up to 1 h postinjection, followed by
a constant decrease in the later time points. In spleen, highest concentrations of the 3H-SQdFdC nanoassembly radioactivity were found
up to 6 h postinjection before the decrease, and 7.35% injected dose/g
still remaining at 24 h postinjection ( p ⬍ 0.05). 3H-SQdFdC nanoassembly radioactivity exhibited significantly higher accumulation in
lung as compared with the 3H-dFdC radioactivity, up to 6 h postinjection ( p ⬍ 0.05) (Table 2). Additionally, calculation of pharmacokinetic parameters in organs revealed a higher Cmax and AUC0-⬁
values in the liver, lung, and spleen for the radioactivity of 3H-
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
measured using a beta counter (Beckmann, LS 6000TA; Beckman Coulter).
The results obtained for the biodistribution samples were converted and
interpreted as percentage of injected dose per gram of tissue or milliliter of
plasma or urine. The statistical comparisons were made using Student’s t test
using GraphPad software (GraphPad Software Inc., (San Diego, CA).
1574
REDDY ET AL.
TABLE 1
Pharmacokinetic parameters of SQdFdC, dFdC or dFdCNA, and dFdU in mice treated i.v. with either SQdFdC nanoassemblies (15 mg/kg mol Eq of dFdC)
or dFdC (15 mg/kg) in DBA/2 female mice
SQdFdC-Treated Mice (15 mg/kg Equivalent in Gemcitabine, n ⫽ 4)
dFdCcon-Treated Mice (15 mg/kg, n ⫽ 4)
Parameter
t1/2 (h)
C0 (ng/ml)
Cmax (ng/ml)
Tmax (h)
Y0-⬁ (ng/h/ml)
Cl (ml/h/kg)
Vss (ml/kg)
MRT (h)
SQdFdC
dFdCNA
8.6
120,122
6.3
44,178
831a
1740a
2.1
6627
0.017
6043
2481b
7344b
3.0
dFdU
10.5
10,756
2.0
112,701
15.3
dFdC
1.6
38,355
5413
2770
1074
0.4
dFdU
8.5
10,375
0.5
54,243
8.7
t1/2 (hours), terminal elimination half-life; C0 (nanograms per milliliter), extrapolated plasma concentration at t ⫽ 0; Cmax (nanograms per milliliter), maximum observed plasma concentration;
Tmax (hours), time to reach the Cmax; Cl (milliliters per hour per kilogram), total body plasma clearance; Vss, volume of distribution at steady-state (milliliters per kilogram); MRT, mean residence
time.
a
Cl and Vss values of SQdFdC were calculated considering a dose of 36.6 mg/kg SQdFdC.
b
Cl and Vss of dFdCNA were calculated using a dose of 15 mg/kg dFdC and i.v. route.
FIG. 4. Distribution of radioactivity in the liver and spleen after administration of
a single dose (10 mg/kg) of either 3H-dFdC or 3H-SQdFdC nanoassemblies. liver
3
H-dFdC (u), spleen 3H-dFdC (`), liver 3H-SQdFdC nanoassemblies (t), spleen
3
H-SQdFdC nanoassemblies (2). The values are the mean ⫾ S.D. of three mice. ⴱ,
p ⬍ 0.05 compared with that of 3H-dFdC.
SQdFdC nanoassemblies compared with that of the 3H-dFdC counterpart (Table 3).
The urinary excretion of 3H-dFdC radioactivity (in terms of 3HdFdC or 3H-dFdU) was biphasic, with rapid initial phase followed by
a slow excretion (Fig. 5). 3H-SQdFdC nanoassembly radioactivity (in
terms of 3H-SQdFdC, 3H-dFdC, or 3H-dFdU) exhibited considerably
lower excretion rate ( p ⬍ 0.05) as compared with 3H-dFdC radioactivity. Excretion of the 3H-dFdC radioactivity was rapid, with maximum amounts excreted at 1 h postinjection (99,255 Bq/ml urine),
reaching a minimum (5935 Bq/ml urine) at 24 h postinjection. The
3
H-SQdFdC nanoassembly radioactivity showed a similar trend, with
maximum excretion occurring at 1 h postinjection (39,717 Bq/ml
urine). However, the excreted amounts were considerably lower ( p ⬍
0.05 up to 6 h postinjection) than those observed with 3H-dFdC.
3
H-dFdC radioactivity underwent a rapid distribution to the brain,
initially with peak concentration appearing at 1 h postinjection (3.27%
injected dose/g), followed by a decrease later (0.61% injected dose/g
at 24 h postinjection) (Table 2). The overall brain distribution of
3
H-SQdFdC nanoassemblies was lower than that of 3H-dFdC as
observed from the AUC values (3H-dFdC, 47.86% radioactivity/h/ml;
3
H-SQdFdC nanoassemblies, 31.57% radioactivity/h/ml). Other pharmacokinetic parameters are presented in Table 3.
The organ distribution studies were then performed using similar
i.v. administration schedules as described by Reddy et al. (2007) on
leukemia-bearing mice (i.e., two doses on days 0 and 4, followed by
tissue sampling on day 8, or four doses on days 0, 4, 8, and 13,
followed by tissue sampling on day 15). In the two-dose injection
schedule, the distribution of 3H-dFdC radioactivity and 3H-SQdFdC
nanoassembly radioactivity into organs such as heart, lung, kidney,
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
FIG. 3. Appearance of the deaminated metabolite dFdU as indicated by the plasma
concentration-time curves after treatment
with dFdC (‚) (15 mg/kg) or SQdFdC
nanoassemblies (⽧); 15 mg/kg mol Eq of
dFdC). The values are the mean ⫾ S.D. of
four mice.
1575
IN VIVO FATE OF SQUALENOYL GEMCITABINE NANOMEDICINE
TABLE 2
Organ distribution of radioactivity following single-dose i.v. administration (10 mg/kg) in mice of either 3H-dFdC or 3H-SQdFdC nanoassemblies
The values are the mean ⫾ S.D. of three mice.
Time (h)
Organ
Heart
Lung
Kidney
Brain
Muscle
Stomach
Intestine
Feces
0.083
0.5
1
2
6
24
dFdC
SQdFdC
dFdC
SQdFdC
dFdC
SQdFdC
dFdC
SQdFdC
dFdC
SQdFdC
dFdC
SQdFdC
7.67 ⫾ 0.79
8.33 ⫾ 0.45
8.63 ⫾ 0.68
1.38 ⫾ 0.42
8.30 ⫾ 0.20
5.34 ⫾ 0.69
8.45 ⫾ 2.30
1.36
8.73 ⫾ 5.73
34.59 ⫾ 6.20*
3.56 ⫾ 1.05
0.41 ⫾ 0.05
1.66 ⫾ 0.25
2.33 ⫾ 0.58
2.00 ⫾ 0.25
0.45
5.32 ⫾ 1.50
7.97 ⫾ 0.41
6.01 ⫾ 0.76
3.24 ⫾ 0.75
5.33 ⫾ 0.30
5.57 ⫾ 0.99
10.40 ⫾ 1.10
1.32
6.42 ⫾ 5.54
29.09 ⫾ 6.07*
5.46 ⫾ 1.54
0.82 ⫾ 0.07
2.80 ⫾ 0.22
4.64 ⫾ 0.21
5.88 ⫾ 3.59
0.90
6.20 ⫾ 2.43
5.60 ⫾ 0.67
8.05 ⫾ 1.14
3.27 ⫾ 0.24
6.20 ⫾ 0.34
7.30 ⫾ 0.78
11.77 ⫾ 0.13
1.50
5.93 ⫾ 2.47
29.46 ⫾ 4.73*
3.51 ⫾ 0.83
1.22 ⫾ 0.19
3.31 ⫾ 0.24
3.79 ⫾ 0.97
7.10 ⫾ 1.74
0.81
4.38 ⫾ 0.73
4.90 ⫾ 0.23
7.04 ⫾ 1.18
2.92 ⫾ 0.09
4.72 ⫾ 0.22
4.88 ⫾ 0.53
8.45 ⫾ 1.99
2.50
4.01 ⫾ 1.09
17.59 ⫾ 9.25*
3.31 ⫾ 0.84
1.78 ⫾ 0.10
3.07 ⫾ 0.53
2.94 ⫾ 0.84
7.15 ⫾ 0.97
6.54
3.9 ⫾ 1.52
3.61 ⫾ 0.68
5.9 ⫾ 1.63
2.33 ⫾ 0.05
3.52 ⫾ 0.13
2.80 ⫾ 0.10
5.63 ⫾ 1.01
17.91
3.08 ⫾ 0.62
16.89 ⫾ 8.4*
4.12 ⫾ 1.80
1.83 ⫾ 0.19
3.06 ⫾ 0.43
2.62 ⫾ 0.36
3.89 ⫾ 0.88
13.94
0.79 ⫾ 0.62
1.10 ⫾ 0.15
0.72 ⫾ 0.32
0.61 ⫾ 0.07
0.97 ⫾ 0.10
0.99 ⫾ 0.20
1.06 ⫾ 0.30
8.64
0.67 ⫾ 0.14
1.09 ⫾ 0.43
0.79 ⫾ 0.23
0.41 ⫾ 0.06
0.61 ⫾ 0.14
0.64 ⫾ 0.22
0.70 ⫾ 0.13
2.09
* p ⬍ 0.05 compared with that of 3H-dFdC.
TABLE 3
Organ
Brain
Heart
Intestine
Kidney
Liver
Lung
Muscle
Spleen
Stomach
Compound
dFdC
SQdFdC
dFdC
SQdFdC
dFdC
SQdFdC
dFdC
SQdFdC
dFdC
SQdFdC
dFdC
SQdFdC
dFdC
SQdFdC
dFdC
SQdFdC
dFdC
SQdFdC
t1/2
9.62
8.36
8.53
8.40
7.48
7.28
6.43
9.44
11.63
10.63
10.28
5.17
9.67
7.73
7.21
5.82
12.01
9.65
Tmax
Cmax
AUC0-⬁
h
%/ml
%/h/ml
1
6
0
0
1
2
0
1
1
1
0
0
0
1
1
6
1
1
3.27
1.83
7.67
8.73
11.77
7.15
8.63
5.47
9.36
29.67
8.33
34.59
8.31
3.31
37.45
62.64
7.30
4.65
47.86
31.57
72.56
62.53
108.39
74.44
90.93
69.63
67.57
138.88
83.87
234.82
77.19
52.16
458.65
845.91
75.02
52.51
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
Organ pharmacokinetic parameters in mice following single-dose i.v.
administration (10 mg/kg) of either 3H-dFdC or 3H-SQdFdC nanoassemblies
FIG. 6. A, organ distribution of radioactivity after the completion of two i.v. doses
(10 mg/kg on days 0 and 4) of either 3H-dFdC (s) or 3H-SQdFdC nanoassemblies
(`). Sampling of organs was performed on day 8. The values are the mean ⫾ S.D.
of three mice. ⴱ, p ⬍ 0.05 compared with that of 3H-dFdC. B, organ distribution of
radioactivity after the completion of four i.v. doses of 3H-dFdC (s) or 3H-SQdFdC
nanoassemblies (`) (10 mg/kg on days 0, 4, 8, and 13). Sampling of organs was
performed on day 15. The values are the mean ⫾ S.D. of three mice. ⴱ, p ⬍ 0.05
compared with that of 3H-dFdC.
FIG. 5. Urinary excretion profiles of radioactivity after single-dose i.v. administration of 10 mg/kg of either 3H-dFdC ‚ or 3H-SQdFdC nanoassemblies (⽧). The
values are the mean ⫾ S.D. of three mice.
muscle, stomach, intestine, and brain was not significantly different.
However, liver and spleen accumulation of 3H-SQdFdC nanoassembly radioactivity was again significantly higher ( p ⬍ 0.05) compared
with the 3H-dFdC (Fig. 6A). After the four-dose injection schedule,
the accumulation of 3H-SQdFdC nanoassembly radioactivity in spleen
(12.7% injected dose/g) was dramatically higher than that of 3H-dFdC
radioactivity (2% injected dose/g) ( p ⬍ 0.05), whereas the distribution in the other organs was similar (Fig. 6B).
Discussion
After i.v. administration, dFdC rapidly disappeared from plasma,
with a terminal t1/2 of 1.6 h, and was rapidly eliminated with short
1576
REDDY ET AL.
FIG. 7. Ratio (r) of the plasmatic concentrations of dFdU to those of dFdC in mice
treated with dFdC (s) (15 mg/kg) or SQdFdC nanoassemblies (`) (15 mg/kg mol
Eq of dFdC).
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
MRT as described previously (Immordino et al., 2004). Nevertheless,
the possibility of a third t1/2 and, thus, a longer terminal t1/2 cannot be
excluded. However, this latter phase was not observed under our
experimental conditions because, at 24 h postinjection, the dFdC
plasmatic concentrations were exceedingly below the detection limit
of our analytical method. Interestingly, the dFdCNA released from
SQdFdC nanoassemblies displayed a ⬃3.9-fold higher t1/2 and ⬃7.5fold higher MRT, showing that SQdFdC nanoassemblies considerably
modified the pharmacokinetics of dFdC with prolonged blood retention and an increased volume of distribution. According to the pharmacokinetic results, the cleavage of SQdFdC into the pharmacologically active dFdCNA seems to occur rapidly, which is in accordance
with the previous in vitro experiments performed using cathepsins B
and D enzymes (Couvreur et al., 2006). Interestingly, because cathepsin B is overexpressed in cancer cells (Kos et al., 2005), part of the
SQdFdC prodrug may be specifically activated in those target cells.
However, in the blood circulation, other amidases are likely implicated in the cleavage of SQdFdC into dFdCNA. On the other hand, the
deamination by deoxycytidine deaminase was significantly delayed.
This may be explained by the association of SQdFdC with plasma
lipoproteins, mainly very low-density lipoprotein, as described previously with squalene (Tilvis et al., 1982; Strandberg et al., 1990),
leading to a delayed accessibility of the released dFdC to deamination.
This is further supported by calculation of the ratio (r) of the plasma
concentrations of dFdU to those of dFdC in dFdC- and SQdFdCtreated groups (Fig. 7). This ratio (r) is constantly lower in the
SQdFdC nanoassembly-treated group than in the dFdC-treated one, at
least up to 8 h postinjection (Fig. 7). The increased t1/2, tmax, AUC,
and the altered ratio of dFdU to dFdC (r) are a consequence of the
increase in the effective volume of distribution after SQdFdC
administration.
In clinical practice, dFdC is administered at doses as high as 1250
mg/m2, contributing to the manifestation of adverse effects. The data
we report here suggest that squalenoylation of dFdC offers various
advantages over free dFdC, such as prolonged t1/2, increased MRT,
and volume of distribution. Although an increased AUC of dFdU was
calculated, a comparable plasma clearance was observed.
The organ distribution patterns of the radioactivity after single-dose
i.v. administration of either 3H-dFdC or 3H-SQdFdC nanoassemblies
were distinct. Primarily, the lower radioactivity concentrations of
3
H-SQdFdC nanoassemblies in kidney and urine suggest a slow
elimination and, hence, longer body retention compared with the
3
H-dFdC radioactivity, which is also supported by the plasma pharmacokinetic data. The significantly higher accumulation of 3H-SQd-
FdC nanoassembly radioactivity in the reticuloendothelial organs such
as the spleen, liver, and lung may possibly be due to the uptake of the
gemcitabine nanomedicine by macrophages, probably due to the hydrophobic surface and colloidal nature of this formulation. This biodistribution pattern suggests that a significant amount of SQdFdC was
directly captured by these tissues as nanoassemblies, before cleavage.
Otherwise, a quite similar organ/tissue distribution as for 3H-dFdC
should have been observed. Although SQdFdC appears to be rapidly
cleaved into dFdCNA as stated by the pharmacokinetic parameters, the
squalenoylation seems to modify the behavior of dFdC through an
altered biodistribution pattern of SQdFdC, explaining also the delayed
dFdU formation.
The organ distribution evaluation performed after the two-dose
schedule (days 0 and 4, Fig. 5) and four-dose schedule (days 0, 4, 8,
and 13, Fig. 6) revealed again significantly higher accumulation of
3
H-SQdFdC nanoassembly radioactivity in the liver and spleen ( p ⬍
0.05), compared with 3H-dFdC radioactivity. As already discussed
before, this distribution pattern may give increased activity against
cancer metastasis because the lymphoid organs, such as the spleen and
liver, are the major organs of metastasis in the case of cancers such as
leukemia.
On the other hand, such distribution should have implications on
the toxicity of this nanomedicine. However, our recent report (Reddy
et al., 2008) revealed no hematologic or hepatic toxicity in mice with
either dFdC or SQdFdC nanoassembly treatment, even at maximal
tolerated doses as assessed from the blood and biochemical parameter
analysis, indicating the safety of these doses. On the contrary, at lethal
doses, SQdFdC nanoassembly treatment led before death to lymphopenia and severe thrombocytopenia, whereas dFdC treatment led
to lymphopenia and granulocytopenia. Additionally, both of these
drugs showed a decrease in spleen weights, whereas the decrease was
higher in the SQdFdC nanoassembly-treated group. Interestingly,
even at the lethal dose and although the observed liver concentration
of SQdFdC, no hepatotoxicity was detected as assessed from the liver
enzyme levels (aspartate aminotransferase, alanine aminotransferase,
and alkaline phosphatase) (Reddy et al., 2008). Regarding the activity,
the enhanced lung accumulation of SQdFdC as observed in this study
could be of interest for the treatment of lung-associated cancers. The
lower tissue distribution to non-reticuloendothelial organs observed
here would allow the anticipation that the SQdFdC nanoassemblies
would display a lower activity in an s.c. grafted tumor. In fact, on the
contrary, SQdFdC nanoassemblies led to a greater tumor control and
survival of the mice compared with dFdC in an s.c. grafted P388
tumor (Couvreur et al., 2008). In summary, the greater efficacy of
SQdFdC nanoassemblies observed both in an i.v. leukemia metastatic
model and an s.c. solid tumor model coupled to the pharmacokinetic
and organ distribution may be explained by: i) a greater distribution to
reticuloendothelial organs, which favored improved treatment of metastatic cancer; and ii) triggered controlled and prolonged release of
dFdC by SQdFdC nanoassemblies in the vascular compartment, leading to an increased blood t1/2, which, in turn, improve exposure of the
tumor site to the drug.
This study has demonstrated that squalenoylation of the anticancer
nucleoside analog, gemcitabine, dramatically modified the pharmacokinetics, metabolism, and biodistribution of this compound. Our findings indicate that the squalenoyl nanomedicine of gemcitabine acted
as a reservoir of gemcitabine due to its modified distribution compared with free dFdC, then releasing the active molecule, thereby
minimizing the rapid exposure of the drug to deamination. Apart from
the favorable pharmacokinetics, selective accumulation of the
squalenoyl gemcitabine nanoassemblies into the organs of reticuloendothelial system provides the basis to explain the previously observed
IN VIVO FATE OF SQUALENOYL GEMCITABINE NANOMEDICINE
superior anticancer activity of gemcitabine nanomedicine in comparison with gemcitabine free. Thus, squalenoylation of gemcitabine
offers several pharmacological benefits and, hence, is expected to
provide a positive therapeutic advantage in cancer therapy.
References
Moog R, Burger AM, Brandl M, Schüler J, Schubert R, Unger C, Fiebig HH, and Massing U
(2002) Change in pharmacokinetic and pharmacodynamic behavior of gemcitabine in human
tumor xenografts upon entrapment in vesicular phospholipid gels. Cancer Chemother Pharmacol 49:356 –366.
Myhren F, Borretzen B, Dalen A, and Sandvold ML (2000), inventors; Clavis Pharma, assignee.
Oral dosage forms of gemcitabine derivatives. European patent EP0986570. 2007 Nov 28.
Reddy LH, Dubernet C, Lepêtre-Mouelhi S, Desmaële D, and Couvreur P (2007) A new
nanomedicine of gemcitabine displays enhanced anticancer activity in sensitive and resistant
leukemia types. J Control Rel 124:20 –27.
Reddy LH, Marque PE, Dubernet C, Lepêtre Mouelhi S, Desmaële D, and Couvreur P (2008)
Preclinical toxicology (sub-acute and acute) and efficacy of a new squalenoyl gemcitabine
anticancer nanomedicine. J Pharmacol Exp Ther 325:484 – 490.
Reid JM, Qu W, Safgren SL, Ames MM, Krailo MD, Seibel NL, Kuttesch J, and Holcenberg J
(2004) Phase I trial and pharmacokinetics of gemcitabine in children with advanced solid
tumors. J Clin Oncol 22:2445–2451.
Ruiz van Haperen VW, Veerman G, Eriksson S, Boven E, Stegmann AP, Hermsen M, Vermorken JB, Pinedo HM, and Peters GJ (1994) Development and molecular characterization of
a 2⬘,2⬘-difluorodeoxycytidine-resistant variant of the human ovarian carcinoma cell line
A2780. Cancer Res 54:4138 – 4143.
Strandberg TE, Tilvis RS, and Miettinen TA (1990) Metabolic variables of cholesterol during
squalene feeding in humans: comparison with cholestyramine treatment. J Lipid Res 31:1637–
1643.
Tilvis R, Kovanen PT, and Miettinen TA (1982) Metabolism of squalene in human fat cells.
Demonstration of two-pool system. J Biol Chem 257:10300 –10305.
Wang LR, Huang MZ, Xu N, Shentu JZ, Liu J, and Cai J (2005) Pharmacokinetics of
gemcitabine in chinese patients with non-small-cell lung cancer. J Zhejiang Univ Sci B
6:446 – 450.
Xu Y, Keith B, and Grem JL (2004) Measurement of the anticancer agent gemcitabine and its
deaminated metabolite at low concentrations in human plasma by liquid chromatography-mass
spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci 802:263–270.
Address correspondence to: Patrick Couvreur, Unité Mixte de Recherche
Centre National de la Recherche Scientifique 8612, Physico-chimie, Pharmacotechnie, et Biopharmacie, Université Paris-Sud XI, Faculté de Pharmacie, 5 rue
Jean-Baptiste Clément, 92296 Châtenay-Malabry, France. E-mail: patrick.
[email protected]
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
Abbruzzese JL, Grunewald R, Weeks EA, Gravel D, Adams T, Nowak B, Mineishi S, Tarassoff
P, Satterlee W, Raber MN, et al. (1991) A phase I clinical, plasma, and cellular pharmacology
study of gemcitabine. J Clin Oncol 9:491– 498.
Bergman AM, Pinedo HM, Peters GJ (2002) Determinants of resistance to 2⬘,2⬘difluorodeoxycytidine (gemcitabine). Drug Resist Updat 5:19 –33.
Cappella P, Tomasoni D, Faretta M, Lupi M, Montalenti F, Viale F, Banzato F, D’Incalci M, and
Ubezio P (2001) Cell cycle effects of gemcitabine. Int J Cancer 93:401– 408.
Castelli F, Sarpietro MG, Ceruti M, Rocco F, and Cattel L (2006) Characterization of lipophilic
gemcitabine prodrug-liposomal membrane interaction by differential scanning calorimetry.
Mol Pharm 3:737–744.
Couvreur P, Reddy LH, Mangenot S, Poupaert JH, Desmaële D, Lepêtre-Mouelhi S, Pili B,
Bourgaux C, Amenitsch H, and Ollivon M (2008) Discovery of new hexagonal supramolecular
nanostructures formed by squalenoylation of an anticancer nucleoside analogue. Small 4:247–
253.
Couvreur P, Stella B, Reddy LH, Hillaireau H, Dubernet C, Desmaële D, Lepêtre-Mouelhi S,
Rocco F, Dereuddre-Bosquet N, Clayette P, et al. (2006) Squalenoyl nanomedicines as
potential therapeutics. Nano Lett 6:2544 –2548.
Fogli S, Danesi R, Gennari A, Donati S, Conte PF, and Del Tacca M (2002) Gemcitabine,
epirubicin and paclitaxel: pharmacokinetic and pharmacodynamic interactions in advanced
breast cancer. Ann Oncol 13:919 –927.
Immordino ML, Brusa P, Rocco F, Arpicco S, Ceruti M, and Cattel L (2004) Preparation,
characterization, cytotoxicity and pharmacokinetics of liposomes containing lipophilic gemcitabine prodrugs. J Control Release 100:331–346.
Khoury H, Deroussent A, Reddy LH, Couvreur P, Vassal G, and Paci A (2007) Simultaneous
determination of gemcitabine and gemcitabine-squalene by liquid chromatography-tandem
mass spectrometry in human plasma. J Chromatogr B Analyt Technol Biomed Life Sci
858:71–78.
Kos J, Sekirnik A, Premzl A, Zavasnik-Bergant V, Langerholc T, Turk B, Werle B, Golouh R,
Repnik U, Jeras M, et al. (2005) Carboxypeptidases cathepsins X and B display distinct protein
profile in human cells and tissues. Exp Cell Res 306:103–113.
1577