ArticlePDF Available

Liposomal encapsulation of the natural flavonoid fisetin improves bioavailability and antitumor efficacy

Authors:

Abstract and Figures

The natural flavonoid fisetin (3,3',4',7-tetrahydroxyflavone) has shown antiangiogenic and anticancer properties. Because of fisetin limited water solubility, we designed a liposomal formulation and evaluated its biological properties in vitro and in Lewis lung carcinoma (LLC) bearing mice. A liposomal formulation was developed with DOPC and DODA-PEG2000, possessing a diameter in the nanometer range (173.5±2.4nm), a high homogeneity (polydispersity index 0.181±0.016) and high fisetin encapsulation (58%). Liposomal fisetin incubated with LLC cells were internalized, induced a typical fisetin morphological effect and increased the sub-G1 cell distribution. In vivo, liposomal fisetin allowed a 47-fold increase in relative bioavailability compared to free fisetin. The effect of liposomal fisetin on LLC tumor growth in mice at low dose (21mg/kg) allowed a higher tumor growth delay (3.3 days) compared to free fisetin at the same dose (1.6 day). Optimization of liposomal fisetin therapy was attempted by co-treatment with cyclophosphamide which led to a significant improvement in tumor growth delay (7.2 days) compared to cyclophosphamide with control liposomes (4.2 days). In conclusion, fisetin liposomes markedly improved fisetin bioavailability and anticancer efficacy in mice and this formulation could facilitate the administration of this flavonoid in the clinical setting.
Content may be subject to copyright.
This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
International
Journal
of
Pharmaceutics
444 (2013) 146–
154
Contents
lists
available
at
SciVerse
ScienceDirect
International
Journal
of
Pharmaceutics
journa
l
h
omepa
g
e:
www.elsevier.com/locate/ijpharm
Pharmaceutical
nanotechnology
Liposomal
encapsulation
of
the
natural
flavonoid
fisetin
improves
bioavailability
and
antitumor
efficacy
Johanne
Seguin, Laura
Brullé,
Renaud
Boyer,
Yen
Mei
Lu,
Miriam
Ramos
Romano,
Yasmine
S.
Touil,
Daniel
Scherman,
Michel
Bessodes,
Nathalie
Mignet,
Guy
G.
Chabot
Paris
Descartes
University,
Faculty
of
Pharmacy,
INSERM
U1022,
CNRS
UMR8151,
Sorbonne
Paris
Cité,
Chimie
ParisTech,
Chemical,
Genetic
&
Imaging
Pharmacology
Laboratory,
F-75006
Paris,
France
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Received
12
October
2012
Received
in
revised
form
22
January
2013
Accepted
24
January
2013
Available online 1 February 2013
Keywords:
Fisetin
Liposomes
Pharmacokinetics
Lewis
lung
carcinoma
Cyclophosphamide
Anticancer
effect
a
b
s
t
r
a
c
t
The
natural
flavonoid
fisetin
(3,3,4,7-tetrahydroxyflavone)
has
shown
antiangiogenic
and
anticancer
properties.
Because
of
fisetin
limited
water
solubility,
we
designed
a
liposomal
formulation
and
evaluated
its
biological
properties
in
vitro
and
in
Lewis
lung
carcinoma
(LLC)
bearing
mice.
A
liposomal
formulation
was
developed
with
DOPC
and
DODA-PEG2000,
possessing
a
diameter
in
the
nanometer
range
(173.5
±
2.4
nm),
a
high
homogeneity
(polydispersity
index
0.181
±
0.016)
and
high
fisetin
encapsulation
(58%).
Liposomal
fisetin
incubated
with
LLC
cells
were
internalized,
induced
a
typical
fisetin
morphological
effect
and
increased
the
sub-G1
cell
distribution.
In
vivo,
liposomal
fisetin
allowed
a
47-fold
increase
in
relative
bioavailability
compared
to
free
fisetin.
The
effect
of
liposomal
fisetin
on
LLC
tumor
growth
in
mice
at
low
dose
(21
mg/kg)
allowed
a
higher
tumor
growth
delay
(3.3
days)
compared
to
free
fisetin
at
the
same
dose
(1.6
day).
Optimization
of
lipo-
somal
fisetin
therapy
was
attempted
by
co-treatment
with
cyclophosphamide
which
led
to
a
significant
improvement
in
tumor
growth
delay
(7.2
days)
compared
to
cyclophosphamide
with
control
liposomes
(4.2
days).
In
conclusion,
fisetin
liposomes
markedly
improved
fisetin
bioavailability
and
anticancer
efficacy
in
mice
and
this
formulation
could
facilitate
the
administration
of
this
flavonoid
in
the
clinical
setting.
© 2013 Elsevier B.V. All rights reserved.
1.
Introduction
Numerous
plant-derived
compounds
have
been
linked
to
the
chemoprevention
and
treatment
of
cancer,
including
the
flavonoids
which
occupy
a
central
place
due
to
their
widespread
abundance
in
human
food
and
because
they
are
relatively
non-
toxic
(Gupta
et
al.,
2010;
Havsteen,
2002;
Lopez-Lazaro,
2002;
Middleton
et
al.,
2000;
Surh,
2003).
As
part
of
a
research
pro-
gram
aimed
at
finding
new
antiangiogenic
agents
in
the
flavonoid
family,
we
have
recently
identified
the
natural
flavonoid
fisetin
Abbreviations:
DODA,
dioctadecyldimethylammonium
chloride;
DOPC,
1,2-
dioleoyl-sn-glycero-3-phosphocholine;
DOPE-rhodamine,
1,2-dioleoyl-sn-glycero-
3-phosphoethanolamine-rhodamine;
DSPE-PEG(2000),
1,2-distearoyl-sn-glycero-
3-phosphoethanolamine-N-[amino(polyethylene
glycol)-2000];
EPR,
enhanced
permeability
and
retention
effect;
GLY,
glycine;
i.p.,
intraperitoneal;
i.v.,
intra-
venous;
LLC,
Lewis
lung
carcinoma;
P90G,
Phospholipon®90
G;
PDI,
polydispersity
index;
PEG2000,
poly(ethylene
glycol)45;
SDS,
sodium
dodecyl
sulfate.
Corresponding
author
at:
Faculty
of
Pharmacy,
Paris
Descartes
University,
Chemical,
Genetic
&
Imaging
Pharmacology
Laboratory
(INSERM
U1022
CNRS
UMR
8151),
4
avenue
de
l’Observatoire,
F-75006
Paris,
France.
Tel.:
+33
1
53
73
95
71;
fax:
+33
1
43
26
69
18.
E-mail
address:
guy.chabot@parisdescartes.fr
(G.G.
Chabot).
(3,3,4,7-tetrahydroxyflavone)
as
an
interesting
lead
that
can
stabi-
lize
endothelial
cells
in
vitro
at
non-cytotoxic
concentrations
(Touil
et
al.,
2009).
Fisetin
is
mainly
found
in
fruits,
vegetables,
nuts
and
wine
(Arai
et
al.,
2000;
Kimira
et
al.,
1998)
and
displays
a
variety
of
biological
effects
including
antioxidant
and
anti-inflammatory
(Park
et
al.,
2007;
Woodman
and
Chan,
2004).
Several
studies
have
shown
in
vitro
cytotoxic
and
apoptotic
properties
of
fisetin
(Jang
et
al.,
2012;
Lee
et
al.,
2009;
Suh
et
al.,
2009;
Syed
et
al.,
2011;
Yang
et
al.,
2012;
Ying
et
al.,
2012).
Antiangiogenic
activity
of
fisetin
has
also
been
documented
in
vitro
and
in
vivo
in
mice
(Bhat
et
al.,
2012;
Fotsis
et
al.,
1997;
Touil
et
al.,
2011b).
Fisetin
mechanism
of
action
appears
to
be
complex
and
involves
the
inhibition
of
several
molecular
targets
and
pathways,
includ-
ing
cyclin-dependent
kinases
(Lu
et
al.,
2005a,b;
Sung
et
al.,
2007),
DNA
topoisomerases
I
and
II
(Constantinou
et
al.,
1995;
Olaharski
et
al.,
2005),
urokinase
(Jankun
et
al.,
2006),
actin
polymerization
(Böhl
et
al.,
2007),
androgen
receptor
signaling
(Khan
et
al.,
2008),
activation
of
p53
and
inhibition
of
NFkB
pathways
(Li
et
al.,
2011).
Fisetin
has
recently
been
shown
to
possess
interesting
in
vivo
anticancer
activity
in
mice
bearing
mouse
lung
carcinoma
(Touil
et
al.,
2011b),
human
melanoma
(Syed
et
al.,
2011)
and
human
prostate
tumors
(Khan
et
al.,
2008).
Because
fisetin
possesses
a
low
water
solubility
(Guzzo
et
al.,
2006)
and
poor
bioavailability
0378-5173/$
see
front
matter ©
2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ijpharm.2013.01.050
Author's personal copy
J.
Seguin
et
al.
/
International
Journal
of
Pharmaceutics
444 (2013) 146–
154 147
(Ragelle
et
al.,
2012),
its
in
vivo
administration
remains
a
chal-
lenge.
We
therefore
undertook
to
formulate
fisetin
into
liposomes
to
hopefully
improve
its
bioavailability
and
increase
its
antitumoral
activity.
Liposomes
are
artificial
vesicles
composed
of
lipidic
amphiphiles,
usually
phospholipids,
which
organize
themselves
in
water
to
form
an
aqueous
core
surrounded
by
a
lipidic
bilayer.
This
structure
allows
liposomes
to
transport
both
hydrophilic
and
lipophilic
compounds,
and
this
pharmaceutical
preparation
is
now
used
in
the
clinical
setting
as
drug
carrier
of
several
drug
classes
including
antibiotics,
antifungals
and
anticancer
agents
(Allen
and
Cullis,
2004;
Langer,
1998).
Concerning
the
anticancer
agents,
liposomes
have
been
shown
to
increase
drug
accumulation
in
tumors
(Gabizon,
1992).
This
tumor
retention
effect
is
apparently
due
to
the
liposomal
drug
extravasation
through
the
tumor
porous
capillary
endothelium,
which
is
attributed
to
the
enhanced
per-
meability
and
retention
effect
(EPR)
due
to
hypervascularization,
defective
vascular
architecture,
impaired
lymphatic
drainage
and
increased
production
of
permeability
mediators
in
tumors
(Maeda
et
al.,
2000;
Yuan
et
al.,
1995).
Several
liposomal
forms
of
anthra-
cyclines
are
currently
being
employed
in
the
clinic
and
these
formulations
have
contributed
to
significantly
reduce
toxicity
while
maintaining
their
anticancer
activity
in
breast
(Batist
et
al.,
2001;
O’Brien
et
al.,
2004)
and
soft
tissue
carcinoma
(Siehl
et
al.,
2005).
Hydrophobic
compounds
like
flavonoids
are
not
frequently
for-
mulated
as
liposomes,
because
a
rapid
exchange
of
the
compounds
may
occur
between
the
liposomal
membrane
and
the
cellular
mem-
brane
(Fahr
et
al.,
2006).
However,
this
pharmaceutical
formulation
has
recently
been
shown
to
improve
the
solubility
of
the
flavonoid
quercetin
and
to
increase
its
blood
residence
time
and
anticancer
activity
in
mice
(Yuan
et
al.,
2006).
We
have
recently
developed
a
liposomal
formulation
of
fisetin
that
was
shown
to
retain
its
biological
activities
in
vitro
(Mignet
et
al.,
2012).
The
purpose
of
the
present
study
was
to
determine
the
bioavailability
of
fisetin
administered
as
a
liposomal
formula-
tion,
and
to
evaluate
its
anticancer
effects
in
tumor
bearing
mice.
We
also
studied
the
combination
of
the
liposomal
fisetin
with
the
cytotoxic
drug
cyclophosphamide.
Our
results
demonstrate
that
the
liposomal
formulation
can
markedly
increase
fisetin
bioavail-
ability
and
improve
its
anticancer
activity
in
Lewis
lung
carcinoma
bearing
mice.
2.
Materials
and
methods
2.1.
Chemicals
Fisetin
(98%
purity)
was
purchased
from
Shanghai
FWD
Chemicals
Limited
(Shanghai,
China).
Cholesterol,
DMSO,
morin,
cyclophosphamide,
Hepes
and
phosphate
buffers
were
purchased
from
Sigma–Aldrich
(Saint
Quentin
Fallavier,
France).
DOPE-
rhodamine
was
obtained
from
Avanti
Polar
Lipids,
Inc.
(Coger,
Paris,
France)
and
P90G
was
from
Lipoid
(Steinhausen,
Switzerland).
Dichloromethane,
triethylamine,
chloroform,
methanol,
absolute
ethanol
and
silica
gel
were
provided
by
Carlo
Erba
Reactif,
SDS
(Peypin,
France).
DODA-GLY-PEG2000
has
been
synthesized
as
pre-
viously
described
(Mignet
et
al.,
2012).
All
other
chemicals
were
of
pharmaceutical
grade
or
of
the
highest
analytical
purity
available.
2.2.
Liposomal
fisetin
preparation
The
liposomal
fisetin
was
basically
prepared
according
to
our
previously
developed
optimal
preparation
(Mignet
et
al.,
2012),
by
replacing
P90G
by
DOPC.
Briefly,
DOPC,
cholesterol,
DODA-
PEG2000
and
fisetin
were
mixed
in
chloroform:
ethanol
(10:1,
v/v),
using
the
following
weight
ratios
79.6;
4;
13.2
and
3.2.
The
solution
was
evaporated
under
vacuum
at
37 C
during
5
h.
Film
hydration
was
then
performed
by
gently
mixing
the
film
overnight
in
Hepes
buffer
10
mM
and
glucose
5%
to
reach
a
final
lipid
con-
centration
of
100
mM.
Lipid
extrusion
was
carried
out
on
a
Lipex
extruder
(Biomembranes,
Inc.)
under
nitrogen
pressure
in
order
to
concentrate
and
homogenize
the
liposomal
preparation.
The
sam-
ple
was
successively
passed
through
several
filters
with
porosity
of
0.8
m
and
0.4
m.
Liposomal
purification
was
accomplished
on
a
Sephadex
G-25
column
conditioned
with
Hepes
buffer
glu-
cose
(10
mM,
5%,
respectively
pH
7.4)
that
allowed
to
eliminate
free
fisetin.
The
mean
liposome
diameter
and
the
polydispersity
index
(PDI)
were
determined
by
diffusion
light
scattering
(Malvern
Zeta
Sizer
NanoSeries).
The
percent
fisetin
encapsulation
was
deter-
mined
spectrophotometrically
at
320
nm
using
a
standard
curve
ranging
from
10
g/mL
to
100
g/mL
and
was
calculated
based
on
the
amount
initially
incorporated
in
the
preparation.
2.3.
Effect
of
encapsulated
fisetin
on
Lewis
lung
carcinoma
cells
(LLC)
Lewis
lung
carcinoma
cells
were
plated
on
a
24
well
plate
(150,000
cells/mL)
and
incubated
overnight
at
37 C
in
5%
CO2
humidified
atmosphere.
The
DMEM
medium
was
discarded
and
liposomal
fisetin
(200
L,
4.4
g/mL
with
1%
of
DOPE-rhodamine)
or
free
fisetin
at
4.4
g/mL,
were
added
for
2
h
at
37 C
onto
the
cells.
After
incubation,
the
cells
were
fixed
(4%
paraformaldehyde),
permeabilized,
and
saturated
with
a
3%
BSA
solution
containing
1%
Triton
X100
(1
h
at
room
temperature).
Cells
were
processed
for
indirect
immunofluorescence
as
follows:
cells
were
incubated
1
h
at
37 C
with
the
mouse
anti--tubulin
monoclonal
antibody
(Sigma)
at
1/2000
dilution
and
further
incubated
with
an
anti-mouse
fluo-
rescein
isothiocyanate
(FITC)
secondary
antibody
(Sigma)
at
1/400
dilution
for
45
min
in
the
dark
at
room
temperature.
Photographs
were
taken
on
a
Zeiss
fluorescence
microscope
and
a
Zeiss
LSM-510
confocal
microscope
(FITC:
excitation
488
nm,
emission
530
nm;
Carl
Zeiss
France,
Le
Pecq,
France).
2.4.
Quantitative
assessment
of
apoptosis
and
analysis
of
cell
cycle
The
tumor
cell
(LLC)
were
plated
onto
6
well
plate
(200,000
cells/mL)
and
incubated
overnight
at
37 C
in
5%
CO2
humidified
atmosphere.
Cells
were
incubated
thereafter
with
empty
liposomes
(0.24
mg/mL
of
lipids),
liposomal
fisetin
(200
L,
11
g/mL)
or
free
fisetin
at
11
g/mL
for
24
h
at
37 C.
After
this
second
incubation
the
supernatant
was
discarded
and
cells
were
trypsinized
and
put
into
15
mL
tubes
and
centrifuged
for
5
min
at
290
×
g.
The
cells
were
washed
tree
times
with
phosphate
buffer,
permeabilized
with
80%
ethanol
during
30
min
at
20 C,
and
incu-
bated
with
a
staining
solution
30
min
in
the
dark.
The
staining
solution
was
composed
of
50
g/mL
of
propidium
iodide,
50
g/mL
of
RNase,
0.1%
of
Triton
X100
and
1
mg/mL
of
sodium
citrate.
After
staining,
the
cells
were
analyzed
on
a
Coulter
Epics
cytometer
(ex
488
nm,
em 625
nm).
The
rate
of
cell
in
phase
Sub-G1,
G1,
S
and
G2/M
were
determined
using
the
WinMDI
software.
The
cells
in
Sub-G1
phase
were
considered
as
apoptotic
cells.
2.5.
Fisetin
pharmacokinetics
in
mice
2.5.1.
Mice
and
treatments
Female
8
weeks
old
C57BL/6J
mice
(body
weight
18–22
g),
were
purchased
from
Janvier
(Le
Genest-St-Isle,
France).
After
an
overnight
fasting
period,
mice
were
administered
the
vari-
ous
treatments
as
described
hereafter.
For
the
intravenous
(i.v.)
administration
into
the
tail
vein,
21
mice
received
the
free
fisetin
Author's personal copy
148 J.
Seguin
et
al.
/
International
Journal
of
Pharmaceutics
444 (2013) 146–
154
formulated
in
20%
DMSO,
20%
PEG
200
and
60%
saline
at
a
final
concentration
of
1.3
mg/mL
(hereafter
referred
as
“free
fisetin”).
The
total
volume
injected
i.v.
was
200
L
for
a
20
g
mouse,
which
corresponds
to
a
volume
of
DMSO
of
40
L.
It
should
be
noted
that
an
undiluted
DMSO
volume
of
50
L
can
be
administered
safely
i.v.
to
mice
without
toxicity
(Willson
et
al.,
1965),
and
that
in
our
stud-
ies,
the
final
DMSO
dose
per
mouse
corresponds
to
40
L
(for
a
20
g
mouse),
which
was
further
diluted
in
saline
and
injected
slowly
over
1
min.
We
did
not
encounter
any
acute
mortality
using
this
formulation
in
our
studies.
A
second
group
of
21
mice
received
the
i.v.
fisetin
liposomes
sterilized
by
filtration
through
0.22
m
filters
at
13
mg/kg.
Mice
were
sacrificed
at
5,
10,
15,
30,
60,
120,
240
min,
and
the
blood
was
obtained
by
cardiac
puncture
onto
heparinized
syringes,
centrifuged
(10,000
×
g,
10
min)
and
the
harvested
plasma
was
kept
frozen
at
20 C
until
HPLC
analysis.
For
the
intraperitoneal
(i.p.)
administration,
the
free
fisetin
(pre-
pared
as
described
above
for
the
i.v.
route)
was
injected
at
the
maximum
tolerated
dose
by
this
route
(223
mg/kg).
The
liposo-
mal
fisetin
dose
was
21
mg/kg
corresponding
to
an
injected
volume
of
260–500
L
depending
of
the
percent
encapsulated
fisetin
of
a
given
preparation.
Mice
were
sacrificed
at
15,
30,
60,
90,
120,
180
and
240
min,
the
blood
was
obtained
by
cardiac
puncture
onto
hep-
arinized
syringes,
centrifuged
(10,000
×
g,
10
min),
and
the
plasma
was
kept
frozen
at
20 C
until
HPLC
analysis.
All
animal
experi-
ments
have
been
carried
out
in
accordance
with
institutional
and
French
regulations
concerning
the
protection
of
animals,
and
with
the
European
Commission
regulations.
2.5.2.
Determination
of
fisetin
plasma
concentration
Fisetin
concentration
in
plasma
was
determined
by
HPLC
as
followed:
to
100
L
of
plasma
was
added
60
L
of
a
methano-
lic
solution
of
morin
(2,3,4,5,7-pentahydroxyflavone)
used
as
an
internal
standard
(0.5
mg/mL),
and
200
L
of
cold
acidified
methanol
(methanol:perchloric
acid
70%,
200:1,
v/v)
to
precipi-
tate
proteins.
After
vortexing
for
5
min,
the
samples
were
kept
on
ice
for
15
min,
and
centrifuged
at
10,000
×
g
at
4C.
The
super-
natant
(100
L)
was
injected
onto
a
reversed-phase
HPLC
system
(Shimadzu
CLASS-VP®,
version
5.3),
equipped
with
an
octadecylsi-
lane
column
(Beckman
Ultrasphere
ODS,
5
m;
4.6
mm
×
250
mm)
thermostated
at
20 C,
and
a
UV
detector
set
at
360
nm.
The
mobile
phase
was
composed
of
25%
acetonitrile
and
75%
acidi-
fied
water
(2%,
v/v
glacial
acetic
acid),
at
a
flow
rate
of
1
mL/min.
In
these
conditions
the
retention
times
were
9.6
and
15
min,
for
fisetin
and
morin,
respectively.
The
ratio
of
the
area
of
the
fisetin
peak
divided
by
the
internal
standard
peak
area
was
reported
to
a
calibration
curve
to
determine
the
concentration
of
fisetin.
Calibration
curves
were
linear
with
correlation
coef-
ficients
near
unity.
The
quantification
limit
of
the
system
was
0.1
g/mL.
2.5.3.
Pharmacokinetic
parameters
determination
The
following
non
compartmental
pharmacokinetic
parame-
ters
were
calculated
using
standard
methods
(Gibaldi
and
Perrier,
1982):
maximum
concentration
(Cmax)
extrapolated
to
time
zero
for
the
i.v.
route;
area
under
the
plasma
concentration
versus
time
curve
from
time
zero
to
the
time
of
the
last
measurable
concentration
(AUC0t)
calculated
by
the
trapezoidal
method;
ter-
minal
half-life
=
ln
2/Kel,
where
Kel is
the
terminal
elimination
rate
constant.
The
mean
residence
time
(MRT)
was
calculated
as
the
AUMC/AUC,
where
AUMC
is
the
area
under
the
first
moment
curve.
Clearance
was
calculated
as
the
dose/AUC,
and
the
volume
of
distri-
bution
Vss
as
the
CL
×
MRT.
The
mean
absorption
time
(MAT)
after
i.p.
administration
was
calculated
as
the
MRTi.p.
minus
MRTi.v.
(Gibaldi
and
Perrier,
1982).
Relative
bioavailability
(Frel)
compar-
ing
the
free
fisetin
and
its
liposomal
formulation
for
the
same
route
of
administration
was
determined
by
the
following
formula:
Frel =
(AUlipo ×
dosefree)/(AUCfree ×
doselipo).
2.6.
Fisetin
tissues
distribution
Female
8
weeks
old
C57BL/6J
mice
were
implanted
subcuta-
neously
in
both
flanks
with
Lewis
lung
carcinoma
tumor
fragments
(20–30
mm3).
Fifteen
days
after
tumor
implantation,
liposomal
or
free
fisetin
was
administered
i.v.
at
the
dose
of
13
mg/kg
in
a
total
volume
of
300
L.
Mice
were
sacrificed
15
min
after
fisetin
injection.
Blood
was
collected
by
cardiac
puncture
onto
hep-
arinized
syringes.
Tumor,
liver,
lungs,
kidneys
and
spleen
were
also
harvested.
Blood
and
tissues
were
kept
frozen
at
20 C
until
deter-
mination
of
fisetin
concentrations
by
HPLC,
as
described
above.
2.7.
Antitumor
activity
in
mice
2.7.1.
Fisetin
treatment
Female
8
weeks
old
C57BL/6J
mice
(body
weight
18–22
g)
(Jan-
vier,
Le
Genest-St-Isle,
France)
were
used
for
antitumor
evaluation.
Lewis
lung
tumor
fragments
(20–30
mm3)
were
implanted
subcu-
taneously
(s.c.)
bilaterally
into
mouse
flanks
using
a
12
gauge
trocar.
Four
days
after
tumor
implantation,
mice
received
the
following
i.p.
treatments
(5
mice
per
group
from
day
4
to
8
and
from
day
11
to
14):
5
free
fisetin
control
mice
received
solvent
(20%
DMSO,
20%
PEG
200,
60%
saline),
5
liposome
control
mice
received
an
empty
liposomal
formulation,
5
mice
received
the
free
fisetin
prepara-
tion
at
21
mg/kg
and
5
mice
received
liposomal
fisetin
at
21
mg/kg.
The
volumes
of
injection
were
comprised
between
200
and
400
L,
depending
on
the
preparation
used.
Tumor
growth
was
assessed
by
caliper
bi-dimensional
measurements
(in
mm)
and
the
tumor
volume
(mm3)
was
calculated
according
to
the
following
formula:
width2×
length/2.
The
weight,
behavior
and
external
aspect
of
the
mice
were
controlled
every
day
in
order
to
detect
any
suffering.
The
choice
of
the
treatment
protocol
was
based
on
previous
stud-
ies
with
free
fisetin,
where
this
drug
administration
schedule
was
found
to
be
optimal
and
non
toxic
(Touil
et
al.,
2011b).
The
break
on
days
9
and
10
allows
for
mouse
recovery
from
treatment
toxicity.
2.7.2.
Combined
treatments:
cyclophosphamide
and
fisetin
Three
groups
of
four
mice
were
implanted
s.c.
bilaterally
with
LLC
tumor
fragments,
and
four
days
later
the
mice
received
the
following
i.p.
treatments
(4
mice
per
group)
for
12
consecutive
days
(days
4–15):
4
control
mice
received
an
empty
liposomal
preparation
i.p.;
4
mice
received
the
empty
liposomal
prepara-
tion
i.p.
and
cyclophosphamide
at
30
mg/kg
subcutaneously
(s.c.);
4
mice
received
the
i.p.
fisetin
liposomes
corresponding
to
35
mg/kg
of
fisetin
and
cyclophosphamide
s.c.
at
30
mg/kg.
Tumor
volumes
were
assessed
as
described
above.
2.8.
Data
analysis
The
tumor
growth
delay
(treated–control,
or
T–C)
in
days
was
determined
when
the
tumor
volume
reached
500
mm3.
Comparison
between
tumor
volumes
were
assessed
using
the
Mann–Whitney
test,
and
differences
between
area
under
the
plasma
concentrations
versus
time
curves
(AUCs)
for
i.p.
admin-
istration
of
free
and
liposomal
fisetin
were
assessed
using
the
Student’s
t-test.
3.
Results
3.1.
Characteristics
of
the
liposomal
fisetin
preparation
The
liposomal
fisetin
was
formulated
according
to
our
pre-
viously
described
preparation
(Mignet
et
al.,
2012),
with
minor
Author's personal copy
J.
Seguin
et
al.
/
International
Journal
of
Pharmaceutics
444 (2013) 146–
154 149
Fig.
1.
Effects
of
free
and
liposomal
fisetin
on
Lewis
lung
carcinoma
cells
(LLC).
Morphological
effects:
Lewis
lung
carcinoma
cells
(LLC)
were
treated
for
2
h
with
empty
liposomes
(A),
free
fisetin
at
a
concentration
of
4.4
g/mL
(B),
and
with
liposomal
fisetin
at
4.4
g/mL
(C).
The
green
color
shows
the
tubulin
network
and
the
blue
represents
the
cell
nuclei
stained
with
DAPI.
Magnification
400×,
scale
bar
4
m.
Cell
cycle
effects:
LLC
cells
were
exposed
for
24
h
to
empty
liposomes
(D),
to
free
fisetin
at
a
concentration
of
11
g/mL
(E),
and
to
liposomal
fisetin
at
11
g/mL
(F).
(For
interpretation
of
the
references
to
color
in
figure
legend,
the
reader
is
referred
to
the
web
version
of
the
article.)
modifications,
as
described
in
Section
2.
The
resulting
liposomal
preparation
presented
a
fisetin
encapsulation
of
58%,
a
mean
lipo-
some
diameter
of
173.5
±
2.4
nm
and
a
good
homogeneity
with
a
polydispersity
index
of
0.181
±
0.016.
3.2.
Effects
of
liposomal
fisetin
on
Lewis
lung
carcinoma
cells
(LLC)
Before
evaluating
the
liposomal
fisetin
formulation
in
vivo,
we
first
verified
its
morphological
and
cytotoxicity
effects
on
Lewis
lung
carcinoma
cells
(LLC),
which
will
be
grafted
as
solid
tumors
in
mice
in
this
study.
LLC
cells
exposed
to
empty
liposomes
did
not
show
any
noticeable
morphological
change
(Fig.
1A)
and
were
actually
identical
to
control
cells
unexposed
to
empty
liposomes
(not
shown).
Fig.
1B
shows
that
free
fisetin
at
a
concentration
of
4.4
g/mL
presented
a
typical
change
in
the
morphology
of
LLC
cells
characterized
by
elongated
membrane
pseudopods
after
a
2
h
incu-
bation
period.
Liposomal
fisetin
(4.4
g/mL)
also
presented
similar
morphological
alterations
(Fig.
1C),
compared
to
free
fisetin
indi-
cating
that
the
liposomal
formulation
could
indeed
deliver
fisetin.
Concerning
the
cytotoxic
effects,
there
was
no
difference
between
the
free
fisetin
and
the
liposomal
fisetin,
as
the
IC50s
were
15.5
and
15.0
g/mL,
respectively,
for
a
24
h
exposure
time
(Mignet
et
al.,
2012).
The
quantification
of
LLC
cells
in
the
sub-G1
phase
by
flow
cytometry
indicated
that
only
5%
of
cells
were
in
apoptosis
when
exposed
to
empty
liposomes
for
24
h
(Fig.
1D).
When
exposed
to
free
fisetin
at
11.1
g/mL
during
24
h
(Fig.
1E),
35%
of
cells
were
in
apoptosis,
whereas
a
similar
concentration
and
duration
of
treat-
ment
with
liposomal
fisetin
showed
that
the
percent
apoptotic
cells
was
increased
to
46%
(Fig.
1F).
Taken
collectively,
these
data
clearly
indicate
that
liposomal
fisetin
is
indeed
biologically
active
on
LLC
cells,
as
shown
by
its
mor-
phological
effects
similar
to
free
fisetin,
and
also
by
its
induction
of
apoptosis
after
a
24
h
exposure
time.
3.3.
Liposomal
fisetin
pharmacokinetics
and
bioavailability
We
next
examined
the
influence
of
the
liposomal
formulation
on
fisetin
bioavailability
in
mice.
To
do
so,
we
compared
the
phar-
macokinetics
of
free
and
liposomal
fisetin
in
mice
administered
via
the
intravenous
(i.v.)
or
the
intraperitoneal
(i.p.)
route.
Fig.
2
depicts
the
fisetin
plasma
concentrations
following
the
i.v.
and
i.p.
admin-
istration
of
either
the
free
fisetin
or
its
liposomal
preparation.
The
pharmacokinetic
parameter
values
are
presented
in
Table
1.
Fig.
2.
Pharmacokinetics
of
free
fisetin
and
liposomal
fisetin.
Mice
received
free
fisetin
administered
either
i.v.
at
13
mg/kg
(X
symbol),
or
i.p.
at
223
mg/kg
(solid
circles).
Liposomal
fisetin
was
administered
i.v.
at
13
mg/kg
(open
triangles)
or
i.p.
at
21
mg/kg
(open
circles).
Fisetin
plasma
concentrations
were
determined
by
HPLC,
as
described
in
Section
2.
Mean
of
at
least
3
mice
per
time
point
±
SEM.
Author's personal copy
150 J.
Seguin
et
al.
/
International
Journal
of
Pharmaceutics
444 (2013) 146–
154
Table
1
Fisetin
pharmacokinetic
parameters
after
intravenous
or
intraperitoneal
administration
of
free
fisetin
and
liposomal
fisetin
in
mice.a
Parameter
Intravenous
Intraperitoneal
Free
fisetin
Liposomal
fisetin
Free
fisetin
Liposomal
fisetin
Dose
(mg/kg)
13
13
223
21
Cmax (g/mL) 6.0
10.0
2.5
6.75
Elimination
constant
(Kel)
(h1)
1.136
0.18
0.17
0.0125
Terminal
half-life
(h) 0.61
3.8
4.1
55.4
AUC0
4
h(g
h/mL)
1.12
1.84
2.27
10.06*
AUMC
(g
h2/mL)
1.09
1.92
3.87
19.48
Mean
residence
time
(MRT)
(h)
0.97
1.04
1.71
1.93
Mean
absorption
time
(MAT)
(h) – 0.74
1.48
Clearance
(CL)
(L/kg/h) 11.64
7.06
98.2
2.09
Volume
of
distribution
(Vss)
(L/kg)
11.33
7.35
167
4.04
Relative
bioavailabilityb(Frel)
1
47
aPharmacokinetic
parameters
are
defined
and
calculated
as
described
in
Section
2.
bFrel =
(AUlipo ×
dosefree)/(AUCfree ×
doselipo).
*Statistically
different
from
the
AUC
of
free
i.p.
administration
of
fisetin
at
the
P
<
0.05
level.
Concerning
the
i.v.
administration
of
free
fisetin
and
the
fisetin
liposomal
formulation,
some
differences
were
observed
in
terms
of
maximum
concentrations
(6
versus
10
g/mL,
respec-
tively),
terminal
half-lives
(0.61
versus
3.8
h,
respectively),
and
clearance
(11.64
versus
7.06
L/kg/h).
The
comparison
of
free
versus
liposomal
fisetin
administered
at
the
same
dose
level
of
13
mg/kg
disclosed
that
the
liposomal
formulation
offered
a
modest
advantage
in
systemic
exposure
as
expressed
by
its
1.6-fold
increase
in
area
under
the
plasma
concentra-
tion
versus
time
curve
(AUC)
for
the
liposomal
formulation
versus
the
free
fisetin
(AUC
of
1.12
versus
1.84
g
h/mL,
respec-
tively).
Considering
that
the
i.v.
administration
was
relatively
toxic
to
mice
and
did
not
offer
a
clear
advantage
for
the
liposomal
form
over
the
free
fisetin,
we
next
explored
its
i.p.
administra-
tion.
After
the
i.p.
administration
of
liposomal
fisetin
at
21
mg/kg
and
the
i.p.
injection
of
free
fisetin
at
223
mg/kg,
which
is
the
active
dose
in
LLC
tumor
bearing
mice,
it
could
be
observed
that
liposomal
fisetin
yielded
higher
fisetin
plasma
concentra-
tions,
although
the
dose
was
10
times
lower
than
that
of
the
free
fisetin
dose
(Fig.
2).
Table
1
compares
the
pharmacokinetic
param-
eter
values
for
the
free
and
the
liposomal
fisetin
injected
via
the
i.p.
route.
The
liposomal
fisetin
allowed
increasing
significantly
the
plasma
concentrations
of
fisetin
throughout
the
examina-
tion
period
which
translated
into
a
significant
4.4-fold
increase
in
AUC
(P
<
0.05),
with
a
liposomal
fisetin
dose
10
times
lower
than
that
of
the
free
fisetin.
Taking
into
account
the
adminis-
tered
dose,
the
calculated
relative
bioavailability
(Frel)
was
found
to
be
47-fold
higher
for
the
liposomal
fisetin
compared
to
the
free
fisetin.
3.4.
Organ
distribution
of
liposomal
fisetin
We
next
assayed
the
fisetin
concentrations
in
the
major
organs
at
15
min
after
the
i.v.
administration
of
a
dose
of
13
mg/kg
of
free
fisetin
or
of
its
liposomal
formulation.
Fig.
3A
shows
that
fisetin
blood
concentrations
were
2-fold
higher
in
mice
receiving
liposo-
mal
fisetin
compared
to
free
fisetin.
The
liver
also
presented
a
5-fold
increase
in
fisetin
concentration
after
liposomal
fisetin
adminis-
tration
compared
to
the
free
fisetin
(Fig.
3B).
The
other
sampled
organs
(lungs,
kidneys,
spleens)
and
the
LLC
tumors
did
not
show
significant
differences
in
fisetin
exposure
between
the
liposomal
formulation
versus
the
free
fisetin.
Similar
organ
distribution
results
were
observed
at
2
h
post
i.p.
administration
of
liposomal
fisetin
(data
not
shown).
3.5.
Liposomal
fisetin
antitumoral
activity
The
effect
of
liposomal
fisetin
on
Lewis
lung
tumor
growth
in
mice
is
presented
in
Fig.
4A.
It
can
be
observed
that
a
low
free
fisetin
dose
of
21
mg/kg
can
nonetheless
elicit
a
tumor
growth
delay
(T–C)
of
1.6
day
at
a
tumor
volume
of
500
mm3(Table
2).
When
the
same
fisetin
dose
was
administered
as
a
liposomal
formulation,
it
could
increase
the
tumor
growth
delay
to
3.3
days,
which
is
a
sig-
nificant
effect
in
this
rapidly
growing
and
highly
invasive
tumor.
Indeed,
tumor
volumes
of
the
liposomal
fisetin
group
were
statis-
tically
smaller
compared
to
the
empty
liposomes
group
on
days
7,
8
and
11
(cf.
Fig.
4A).
With
regard
to
toxicity,
the
day
that
corresponded
to
the
maxi-
mum
body
weight
loss
(nadir)
in
this
study
was
day
5.
We
observed
a
negligible
body
weight
loss
for
the
empty
liposomes
(0.1
g)
and
a
weight
gain
of
0.3
g
for
the
fisetin
liposomes
on
day
5
(Fig.
4B,
Table
2).
For
the
solvent
used
to
dissolve
the
free
fisetin
and
the
free
fisetin
solution,
they
were
clearly
more
toxic
that
the
lipo-
somal
formulations
as
shown
in
Fig.
4B,
with
a
mean
body
weight
change
of
1.2
g
and
0.9
g,
respectively
(Table
2).
This
last
toxicity
could
probably
be
attributed
to
the
solvent
composition
containing
20%
DMSO,
20%
PEG200
and
60%
saline.
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
SpleenKidney Tumor Liver Lung
Free Fisetin
Fisetin liposome
**
[Fisétine] µg/g
0
1
2
3
4
5
6
*
Blood
[Fisétine] µg/ml
A
B
Fig.
3.
Fisetin
tissues
distribution.
Fisetin
concentration
in
tissues
15
min
after
the
i.v.
administration
of
free
fisetin
(empty
bars)
or
liposomal
fisetin
(solid
bars)
at
a
dose
of
13
mg/kg
in
Lewis
lung
tumor
bearing
mice.
(A)
Blood
concentration
expressed
in
g/mL.
(B)
Fisetin
concentrations
in
tumors
and
the
indicated
organs,
expressed
in
g/g.
Mean
±
SEM
from
3
independent
values.
*P
<
0.05;
**P
<
0.01
(Stu-
dent’s
t-test).
Author's personal copy
J.
Seguin
et
al.
/
International
Journal
of
Pharmaceutics
444 (2013) 146–
154 151
Fig.
4.
Antitumoral
effect
of
liposomal
fisetin.
(A)
Mice
bearing
Lewis
lung
carcinoma
tumors
grafted
on
day
1
were
treated
i.p.
daily
from
day
4
to
day
8
and
from
day
11
to
day
14
with
the
following
treatment:
solvent
of
free
fisetin
containing
20%
DMSO,
20%
PEG200,
60%
saline
(solid
squares);
free
fisetin
at
21
mg/kg
(solid
circles);
empty
liposomes
(empty
squares);
and
liposomal
fisetin
at
21
mg/kg
(empty
circles).
Tumor
volumes
of
the
liposomal
fisetin
group
were
statistically
smaller
compared
to
the
empty
liposomes
group
on
days
7,
8
and
11
(P
<
0.05,
Mann–Whitney
test).
(B)
Mice
body
weight
variation
as
a
function
of
time
(symbols
are
the
same
as
in
Fig.
4A).
3.6.
Antitumoral
activity
of
the
combination
of
liposomal
fisetin
and
cyclophosphamide
Because
additive
effects
of
free
fisetin
combined
with
cyclophos-
phamide
have
previously
been
reported
on
Lewis
lung
tumors
(Touil
et
al.,
2011b),
we
next
attempted
to
optimize
the
lipo-
somal
fisetin
therapy
by
co-treatment
with
cyclophosphamide.
The
dose
of
i.p.
liposomal
fisetin
was
increased
to
35
mg/kg
and
cyclophosphamide
was
administered
s.c.
at
a
relatively
low
dose
of
30
mg/kg.
Fig.
5
shows
that
the
treatment
of
Lewis
lung
car-
cinoma
bearing
mice
led
to
a
tumor
growth
delay
(T–C)
of
4.2
days
for
the
treatment
of
cyclophosphamide
combined
with
empty
liposomes,
whereas
the
combination
of
cyclophosphamide
with
liposomal
fisetin
led
to
a
significantly
improved
anticancer
effect
with
a
T–C
of
7.2
days
(Table
3).
Tumor
volumes
of
the
empty
liposomes
+
cyclophosphamide
group
were
statistically
smaller
compared
to
the
empty
liposomes
group
from
day
4
until
day
15
(P
<
0.01,
Mann–Whitney
test),
whereas
tumor
volumes
of
the
liposomal
fisetin
+
cyclophosphamide
group
were
smaller
than
the
empty
liposomes
+
cyclophosphamide
group
starting
on
day
8
until
day
19
(P
<
0.05,
Mann–Whitney
test).
This
drug
combination
was
not
toxic,
as
shown
by
negligible
body
weight
change
in
the
treated
animals
(cf.
Table
3).
4.
Discussion
The
main
objectives
of
this
study
were
to
evaluate
the
in
vivo
pharmacokinetics
and
antitumor
activity
of
liposomal
fisetin
in
tumor
bearing
mice.
Cellular
results
indicated
that
the
cytotoxic
activity
of
fisetin
was
maintained
on
Lewis
lung
carcinoma
cells
despite
its
encapsulation,
which
confirmed
our
previous
study
where
we
had
shown
that
liposomal
fisetin
was
internalized
and
retained
the
cytotoxic
and
morphological
effect
of
the
free
fisetin
on
endothelial
cells
(Mignet
et
al.,
2012).
Cell
cycle
analysis
yielded
some
information
about
the
rate
of
apoptosis.
Induction
of
apoptosis
by
fisetin
in
various
tumor
cells
has
recently
been
reported
in
the
literature
(Jang
et
al.,
2012;
Khan
et
al.,
2008;
Syed
et
al.,
2011;
Yang
et
al.,
2012;
Ying
et
al.,
2012).
After
exposure
of
the
tumor
cells
(LLC)
to
free
or
liposomal
fisetin,
we
observed
an
increase
of
the
sub-G1
phase,
meaning
that
apopto-
sis
could
either
be
induced
by
free
or
encapsulated
fisetin.
This
is
the
first
time
that
cell
cycle
arrest
is
evidenced
with
an
encapsulated
form
of
fisetin.
Following
these
encouraging
cellular
results,
liposomal
fisetin
was
next
evaluated
for
its
potential
improvement
of
fisetin
phar-
macokinetics,
as
could
be
expected
from
similar
sized
liposomes
(Gabizon
et
al.,
2012).
The
i.v.
administration
of
the
liposomal
Table
2
Antitumoral
effect
of
free
and
liposomal
fisetin.a
Treatment
Fisetin
dosage
(mg/kg/dose)
Schedule
(day)
Total
fisetin
dose
(mg/kg)
Mean
body
wt
change
(g/mouse)
Median
tumor
volume
(mm3
d11)
%T/C
Time
for
median
tumor
to
reach
500
mg
(day)
T–C
(day)
Solvent
controlb
1.2
(d5)
804.5
9.2
Free
fisetinc21
d4–d8
d11–d14
189
0.9
(d5)
526.0
65.4
10.8
1.6
Empty
liposomed
d4–d8
d11–d14
0.1
(d5)
569.1
70.7
9.8
0.6
Liposomal
fisetine21 d4–d8
d11–d14
189
+0.3
(d5)
317.0
39.4
12.5
3.3
aFour
groups
composed
of
5
female
8
weeks
old
C57BL/6J
mice
were
implanted
s.c.
bilaterally
with
Lewis
lung
tumor
fragments
and
treatments
were
initiated
4
days
later.
The
indicated
treatments
were
administered
from
day
4
to
8
and
from
day
11
to
14.
bControl
mice
received
the
solvent
used
to
dissolve
the
free
fisetin
preparation
composed
of
20%
DMSO,
20%
PEG
200,
60%
saline.
There
were
no
drug
death
recorded
during
this
study.
cFree
fisetin
preparation
at
21
mg/kg
containing
the
solvent
composed
of
20%
DMSO,
20%
PEG
200,
60%
saline.
dEmpty
liposomal
formulation
as
described
in
Section
2.
eFisetin
liposomal
preparation
at
21
mg/kg,
as
described
in
Section
2.
Author's personal copy
152 J.
Seguin
et
al.
/
International
Journal
of
Pharmaceutics
444 (2013) 146–
154
Table
3
Antitumoral
effect
of
liposomal
fisetin
combined
with
cyclophosphamide
(CPA).a
Treatment
Drug
dosage
(mg/kg/dose)
Schedule
(day)
Total
dose
(mg/kg)
Mean
body
wt
change
[g/mouse
(day
of
nadir)]
Median
tumor
volume
(mm3
d11)
%T/C
Time
for
median
tumor
to
reach
500
mg
(day)
T–C
(day)
Controlb
d4–d15
644.9
9.8
Empty
liposome
+
CPAc
d4–d15
+0.3
d4–d8 186.5
39.8
14
4.2
30
330
0.25
d7–d8
Liposomal
fisetin
+
CPAd
35 d4–d15 420
+0.75
d4–d8 109.7
22.1
17
7.2
30 330 0.25
d7–d8
aThree
groups
of
4
mice
bearing
LLC
tumors.
bControl
mice
received
an
empty
liposomal
preparation
i.p.
as
described
in
Section
2.
cMice
received
the
empty
liposomal
preparation
i.p.
and
cyclophosphamide
(CPA)
at
30
mg/kg
s.c.
dMice
received
the
i.p.
fisetin
liposomes
corresponding
to
35
mg/kg
of
fisetin
and
cyclophosphamide
(CPA)
s.c.
at
30
mg/kg.
fisetin
led
to
a
modest
improvement
in
the
fisetin
bioavail-
ability.
This
could
indicate
a
fast
exchange
of
this
lipophilic
compound
with
physiological
membranes,
as
suggested
by
Fahr
et
al.
(2006).
However,
the
fact
that
the
amount
of
fisetin
could
significantly
be
increased
in
the
liver
15
min
post
i.v.
administra-
tion
tend
to
show
that
part
of
the
fisetin
was
still
associated
to
its
carrier.
Concerning
the
i.p.
administration,
the
liposomal
fisetin
led
to
a
significant
improvement
in
bioavailability
compared
to
the
i.p.
administration
of
the
free
fisetin,
with
a
47-fold
increase
in
relative
bioavailability.
This
increased
bioavailability
and
pro-
longation
of
residence
time
has
previously
been
observed
for
quercetin,
a
closely
related
flavonoid
(Yuan
et
al.,
2006),
although
this
report
is
the
first
to
demonstrate
the
pharmacokinetic
advan-
tage
of
administering
fisetin
in
its
liposomal
formulation.
This
enhanced
fisetin
bioavailability
is
probably
resulting
from
the
observed
liver
accumulation
of
liposomal
fisetin
which
could
act
as
a
slow
releasing
reservoir
that
could
prolong
the
residence
time
in
the
blood.
Intraperitoneal
administration
also
favors
the
lymphatic
distribution
of
drugs,
which
is
a
favorable
property,
Fig.
5.
Antitumoral
effect
of
liposomal
fisetin
combined
with
cyclophosphamide.
Mice
bearing
Lewis
lung
carcinoma
tumor
were
treated
with
empty
liposomes
(black
squares),
with
empty
liposomes
and
cyclophosphamide
at
30
mg/kg
(black
triangles),
and
liposomal
fisetin
at
the
dosage
of
35
mg/kg
in
combination
with
cyclophosphamide
30
mg/kg
(white
triangle).
Tumors
were
grafted
on
day
1
and
mice
were
treated
daily
from
day
4
to
day
15.
Tumor
volumes
of
the
empty
lipo-
somes
+
cyclophosphamide
group
were
statistically
smaller
compared
to
the
empty
liposomes
group
starting
on
day
4
until
day
15
(P
<
0.01,
Mann–Whitney
test);
tumor
volumes
of
the
liposomal
fisetin
+
cyclophosphamide
group
were
smaller
than
the
empty
liposomes
+
cyclophosphamide
group
starting
on
day
8
until
day
19
(P
<
0.05,
Mann–Whitney
test).
especially
with
anticancer
drugs
that
must
access
lymph
nodes
which
are
frequently
harboring
metastases
(Nishioka
and
Yoshino,
2001).
Liposomal
forms
of
anthracyclines
were
shown
to
improve
the
antitumor
effect
of
the
drug
via
the
enhanced
permeability
and
retention
effect
(EPR)
(Maeda
et
al.,
2000).
The
abnormalities
of
the
endothelial
barriers
in
the
tumors
would
favor
liposome
extravasa-
tion
in
this
target
tissue
and
ultimately
increase
the
concentration
of
the
drug
in
the
tumor.
In
this
study,
we
could
show
that
the
bioavailability
of
fisetin
was
increased
in
the
blood,
but
we
could
not
observe
any
increase
in
fisetin
concentration
in
the
tumor,
at
least
at
the
time
points
studied.
Another
point
to
consider
here
is
the
fact
that
only
<