Liposomes as carriers for colchicine-derived prodrugs: Vascular disrupting nanomedicines with tailorable drug release kinetics

Article abstract of DOI:10.1016/j.ejps.2011.08.027

Newly formed tumor vasculature has proven to be an effective target for tumor therapy. A strategy to attack this angiogenic tumor vasculature is to initiate local blood vessel congestion and consequently induce massive tumor cell necrosis. Vascular disrupting agents (VDAs) typically bind to tubulin and consequently disrupt microtubule dynamics. Colchicine and its derivatives (colchicinoids) are very potent tubulin binding compounds but have a narrow therapeutic index, which may be improved by employing a liposomal targeting strategy. However, as a result of their physicochemical properties, colchicinoids are problematic to retain in liposomes, as they are released relatively rapidly upon encapsulation. To overcome this limitation, two hydrolyzable PEGylated derivatives of colchicine were developed for encapsulation into the aqueous core of long-circulating liposomes: a moderately rapid hydrolyzing PEGylated colchicinoid containing a glycolic acid linker (prodrug I), and a slower hydrolyzing PEGylated colchicinoid with a lactic acid linker (prodrug II). Hydrolysis studies at 37 °C and pH 7.4 showed that prodrug I possessed relatively rapid conversion characteristics (t_1/2 = 5.4 h) whereas prodrug II hydrolyzed much slower (t_1/2 = 217 h). Upon encapsulation into liposomes, colchicine was released rapidly, whereas both PEGylated colchicine derivatives were efficiently retained and appeared to be released only after cleavage of the PEG-linker. This study therefore demonstrates that, in contrast to colchicine, these novel PEGylated colchicine-derived prodrugs are retained within the aqueous interior after encapsulation into liposomes, and that the release of the active parent can be controlled by using different biodegradable linkers.

Full text of DOI:10.1016/j.ejps.2011.08.027

European Journal of Pharmaceutical Sciences 45 (2012) 429–435
Contents lists available at SciVerse ScienceDirect
European Journal of Pharmaceutical Sciences
journal homepage: www.elsevier.com/locate/ejps
Liposomes as carriers for colchicine-derived prodrugs: Vascular disrupting
nanomedicines with tailorable drug release kinetics
Bart J. Crielaard 1, Steffen van der Wal 1, Huong Thu Le, Aloïs T.L. Bode, Twan Lammers, Wim E. Hennink,
⇑
Raymond M. Schiffelers, Marcel H.A.M. Fens, Gert Storm
Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS), Utrecht University, Utrecht, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 1 September 2011
Newly formed tumor vasculature has proven to be an effective target for tumor therapy. A strategy to
attack this angiogenic tumor vasculature is to initiate local blood vessel congestion and consequently
induce massive tumor cell necrosis. Vascular disrupting agents (VDAs) typically bind to tubulin and con-
sequently disrupt microtubule dynamics. Colchicine and its derivatives (colchicinoids) are very potent
tubulin binding compounds but have a narrow therapeutic index, which may be improved by employing
a liposomal targeting strategy. However, as a result of their physicochemical properties, colchicinoids are
problematic to retain in liposomes, as they are released relatively rapidly upon encapsulation. To over-
come this limitation, two hydrolyzable PEGylated derivatives of colchicine were developed for encapsu-
lation into the aqueous core of long-circulating liposomes: a moderately rapid hydrolyzing PEGylated
colchicinoid containing a glycolic acid linker (prodrug I), and a slower hydrolyzing PEGylated colchicinoid
with a lactic acid linker (prodrug II). Hydrolysis studies at 37 °C and pH 7.4 showed that prodrug I pos-
sessed relatively rapid conversion characteristics (t1/2 = 5.4 h) whereas prodrug II hydrolyzed much
slower (t1/2 = 217 h). Upon encapsulation into liposomes, colchicine was released rapidly, whereas both
PEGylated colchicine derivatives were efficiently retained and appeared to be released only after cleavage
of the PEG-linker. This study therefore demonstrates that, in contrast to colchicine, these novel PEGylated
colchicine-derived prodrugs are retained within the aqueous interior after encapsulation into liposomes,
and that the release of the active parent can be controlled by using different biodegradable linkers.
Ó 2011 Elsevier B.V. All rights reserved.
Keywords:
Liposomes
Colchicine
Colchicinoids
Cancer
Prodrugs
Vascular disrupting agents
1. Introduction
vascular disrupting activity by binding irreversibly to tubulin at
its colchicine domain, thereby inhibiting tubulin dynamics and
Newly formed angiogenic endothelium has become an impor-
tant target for the design of anticancer agents (Carmeliet and Jain,
2000). As tumor vasculature develops relatively fast during early
tumor growth, blood vessels appear immature, disorganized and
imperfect, which makes them a vulnerable target for cancer ther-
apy (Baluk et al., 2005; Thorpe, 2004). Vascular disrupting agents
(VDAs) initiate local disruption of tumor endothelium by interfer-
ing with the immature vasculature, causing site-specific blood ves-
sel congestion. As a consequence, tumor cells are deprived from
nutrients and oxygen, leading to rapid and massive tumor cell
necrosis (Kanthou and Tozer, 2007; Lippert, 2007; Tozer et al.,
2005). Tubulin binding agents (TBAs), such as colchicine and the
structurally and pharmacologically similar colchicinoids, display
microtubule formation (Bhattacharyya et al., 2008; Ravelli et al.,
2004). Because endothelial cells in the tumor vasculature rely more
on tubulin than actin to maintain their cell shape, the binding of
TBAs to tubulin leads to rapid rounding of these cells, resulting
in loss of vascular integrity and, ultimately, to hemostasis (Jordan
and Wilson, 2004; Thorpe, 2004). TBAs are considered to be useful
in cancer therapy not only because of their ability to disrupt exist-
ing angiogenic tumor vasculature, but also because of their capa-
bilities to induce mitotic arrest of tumor cells and inhibition of
angiogenesis (Jordan and Wilson, 2004; Lippert, 2007; Pasquier
et al., 2006; Pasquier and Kavallaris, 2008).
Colchicine is currently used clinically in low doses for the
treatment of acute gout (Keith and Gilliland, 2007; Petersel and
Schlesinger, 2007), familial Mediterranean fever (FMF) (Cerquaglia
et al., 2005), and dermatologic (Bibas et al., 2005) and auto-
inflammatory diseases (Terkeltaub, 2009). However, the utility of
colchicine for cancer therapy is currently limited, as only doses
close to the maximal tolerated dose (MTD) can induce reduction
in tumor blood perfusion leading to a high risk for toxicity (Baguley
et al., 1991; Nihei et al., 1999).
⇑
Corresponding author. Address: Department of Pharmaceutics, Utrecht Institute
for Pharmaceutical Sciences (UIPS), Utrecht University, P.O. Box 80082, 3508 TB
Utrecht, The Netherlands. Tel.: +31 30 2537388; fax: +31 30 2517839.
E-mail address: G.Storm@uu.nl (G. Storm).
1
These authors contributed equally to this work.
0928-0987/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.ejps.2011.08.027

430
B.J. Crielaard et al. / European Journal of Pharmaceutical Sciences 45 (2012) 429–435
A potential strategy for improving the therapeutic index of col-
kinetics (Fig. 1). The hydrolysis rate is controlled by using a hydro-
lysable PEG-linker based on a primary alcohol, which is more
prone hydrolysis, or a linker based on a secondary alcohol, which
is more resistant to hydrolysis. Since colchicinoids are able to pass
the lipid bilayer, the release rate of colchicinoid from the liposome
is directly related to the cleavage rate of the colchicinoid prodrug.
Obviously, the release kinetics of colchicinoids from the liposomes,
once they have accumulated in the tumor, is a critical determinant
for therapeutic activity. Therefore, a tailorable drug delivery sys-
tem for colchicine derivates may prove of great value for improving
the therapeutic index of colchicinoids in the therapy of solid
tumors.
chicine, i.e. limiting the side effects and improving the efficacy,
may be the employment of a colloidal drug delivery system, such
as long-circulating liposomes (Fenske and Cullis, 2008; Lammers
et al., 2008). Advantages associated with the use of long-circulating
liposomes include their prolonged circulation kinetics, their pas-
sive targeting properties, as well as the possibility to encapsulate
both hydrophobic and hydrophilic drugs (Torchilin, 2005). How-
ever, it remains a challenge to formulate long-circulating lipo-
somes that allow the retention of drugs with moderate
lipophilicity, such as colchicinoids, while enabling appropriate re-
lease kinetics once accumulated in the target site (Coimbra et al.,
2011). Colchicinoids localize mainly in the lipid bilayer and leak
out readily, which makes it troublesome to control their retention
and release rate (Kulkarni et al., 1997; Mons et al., 2000). A poten-
tial way to overcome this challenge is to make use of prodrugs,
which are (inactive) bioreversible derivatives of the active drug
but often possess different physicochemical properties than the
parent compound (Coimbra et al., 2011).
After enzymatic and/or chemical conversion of the prodrug, the
parent molecule with its original pharmacological activity is recov-
ered (Rautio et al., 2008). If the prodrug is designed to be very
hydrophilic, it may be encapsulated solely in the aqueous core of
liposomes, thereby limiting its diffusion over the lipid bilayer. Only
upon conversion of the prodrug in the aqueous interior, the parent
drug molecule, in this case the colchicinoid, may be released from
the liposome by its ability to diffuse over the lipid bilayer. A useful
method for developing hydrophilic prodrugs of small molecular
weight drugs is the conjugation with poly(ethylene glycol) (PEG).
PEG has high aqueous solubility, it is non-toxic, non-immunogenic
and non-antigenic (Bell et al., 2007; Parveen and Sahoo, 2006). In
addition, by utilizing a PEG-linker that is susceptible to hydrolysis,
PEGylated prodrugs that can be cleaved in the liposomal interior
are created. The molecular structure of the linker can influence
the rate of hydrolysis, and therefore, by employing different link-
ers, the rate of conversion to the active drug can be controlled
(Roberts et al., 2002).
2. Material and methods
2.1. Materials
1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy
(polyethylene glycol)-2000] (DSPE-PEG2000) and dipalmitoylphos-
phatidylcholine (DPPC) were provided by Lipoid (Germany). All
other chemicals were ordered from Sigma–Aldrich (Germany)
and used without further purification. Phosphate Buffered Saline
(PBS) with pH 7.4 was purchased from B. Braun (The Netherlands).
Solvents were obtained from Biosolve (The Netherlands) and used
without distillation. Acetonitrile and dichloromethane were stored
on molecular sieves prior to use (3 and 4 Å, respectively). All reac-
tions were performed without direct lighting and with flasks cov-
ered in aluminum foil to prevent degradation of colchicine. Flash
chromatography was performed using silica gel of 0.035–
0.070 mm 60 Å mesh (Acros Organics, Belgium). TLC analysis was
performed using plastic backed silica 60 F254 plates (Merck,
Germany) and NMR spectra were recorded on a Varian Gemini
300 MHz spectrometer. ESI mass analysis was performed on a
Shimadzu LCMS-QP8000 single quadrupole spectrometer in
positive ionization mode.
2.2. Synthesis of N-Boc-colchicine (2)
In the current study, two PEGylated colchicinoid prodrugs were
synthesized and encapsulated into long-circulating liposomes. The
presented strategy allows retention of the colchicinoids – in their
prodrug form – in the aqueous core of the liposomes, and at the
same time enables release of the cleaved prodrug from the lipo-
some, at a rate that can be tailored by adjustment of the hydrolysis
Colchicine (1, 4.9 g, 12.3 mmol) and DMAP (1.5 g, 12.5 mmol)
were dissolved in acetonitrile (50 mL). Subsequently Boc2O
(17.4 g, 79.7 mmol), dissolved in acetonitrile (50 mL), and dry
Et3N (3 mL, 21.5 mmol) were added. The reaction mixture was re-
fluxed for 1.5 h after which additional Boc2O was added (1.6 g,
Fig. 1. Schematic outline of the liposomal colchicine-derived prodrug strategy. PEGylated colchicine-derived prodrugs with different hydrolysis kinetics are synthesized by
using different biodegradable PEG-linkers (A). These colchicinoid prodrugs are encapsulated into long-circulating liposomes (B). Upon hydrolysis of prodrug I or II in the
aqueous core of the liposomes, the pharmacologically active parent is released from the liposomes with rate kI and kII (kI > kII), respectively (C).

B.J. Crielaard et al. / European Journal of Pharmaceutical Sciences 45 (2012) 429–435
431
7.3 mmol) and stirred at RT for an additional 2 h. The reaction mix-
ture was concentrated in vacuo and re-dissolved in chloroform. The
chloroform was extracted three times with saturated aqueous cit-
ric acid (pH 2), and the aqueous fractions extracted again with
chloroform. The combined organic extracts were washed with
brine and dried over MgSO4. After flash chromatography in ethyl-
acetate/acetone 4:1, 2 was obtained as a brownish yellow solid
(3.52 g, 57%).
1H NMR (300 MHz, CDCl3) d 1.96–2.58 (4 ꢀ m, 5H), 3.64 (s, 3H),
3.90, 3.93, 3.98 (3 ꢀ s, 9H), 4.04–4.20 (m, 2H), 4.63–4.75 (m, 1H),
6.54 (s, 1H), 6.86 (d, 1H, J = 10.7 Hz), 7.33 (d, 1H, J = 10.7 Hz),
7.60 (s + broad s, 2H) ppm.
13C NMR (300 MHz, CDCl3) d 30.1, 37.0, 52.0, 56.3, 56.5, 61.6,
61.7, 62.5, 107.6, 113.0, 125.9, 131.3, 134.4, 135.7, 137.1, 141.9,
151.4, 152.0, 153.8, 164.2, 172.8, 179.9 ppm.
2.6. Synthesis of mPEG-acetyl-N-(2-hydroxyacetyl)-deacetylcolchicine
ester (prodrug I, 6)
TLC : Rf ¼ 0:35 in EtOAc=acetone 4 : 1
1H NMR (300 MHz, CDCl3) d 1.58 (s, 9H), 1.90–2.16 (m, 2H), 2.28
(s, 3H), 2.44–2.70 (m, 2H), 3.65 (s, 3H), 3.89, 3.93, 3.97 (3 ꢀ s,
3 ꢀ 3H), 5.14 (dd, 1H), 6.52 (s, 1H), 6.77 (d, 1H, J = 10.7 Hz), 7.21
(d, 1H, J = 10.7 Hz), 7.60 (s, 1H) ppm.
N-(2-Hydroxyacetyl)-deacetylcolchicine (5, 20 mg, 0.048 mmol),
DMAP (15 mg, 0.12 mmol) and mPEG5000-acetic acid (210 mg,
0.041 mmol) were dissolved in 5 mL CH2Cl2. DIC (15 lL,
0.095 mmol) was added and the reaction mixture was stirred for
16 h at room temp. Next, the reaction mixture was concentrated
in vacuo and dissolved in ammonium acetate buffer (3 mL, pH
5.5, 20 mM). This was filtered and separated on a PD-10 column
(GE Healthcare, Belgium) equilibrated with the same buffer. After
two gel filtrations, no low molecular weight starting material
could be detected on TLC (CHCl3/MeOH/AcOH, 90:10:0.1). The
PEG-conjugate was precipitated in diethyl ether, collected by
centrifugation, and after drying in vacuo, prodrug I (6) was
obtained as a slightly yellow solid (120 mg, 52%, composed of
a mixture of unreacted mPEG-acetic acid and 6). The amount
of colchicine derivate conjugated to PEG was determined with
an UPLC system (Waters) equipped with an Acquity BEH C18
2.3. Synthesis of N-Boc-deacetylcolchicine (3)
N-Boc-colchicine (2, 3.52 g, 7.0 mmol) was dissolved in metha-
nol (40 mL). Sodium methoxide in methanol (9 mL 25% w/v,
ꢁ40 mmol) was added at 4 °C and stirred for 45 min. The reaction
was stopped by addition of saturated NH4Cl solution (40 mL).
Excess brine was added and the now cloudy suspension was
extracted with Et2O. After drying over MgSO4 and concentration
in vacuo a light brown solid was obtained.
The progress of the reaction was followed by NMR (Rf was also
0.35 in EtOAc: acetone 4:1), which showed complete removal of
the acetyl group. The product was deemed pure enough and was
used directly without further purification.
1.7
l
m column (2.1 ꢀ 50 mm) (Waters) and photodiode array
1H NMR (300 MHz, CDCl3) d 1.37 (s, 9H), 1.63–1.70 (m, 1H),
2.22–2.51 ppm (m, 3H), 3.65 (s, 3H), 3.89, 3.93, 3.99 (3 ꢀ s, 9H),
4.36–4.41 (m, 1H), 4.95 (d, 1H, J = 7.4 Hz), 6.52 (s, 1H), 6.81 (d,
1H, J = 10.7 Hz), 7.26 (d, 1H, J = 10.7 Hz), 7.51 (s, 1H) ppm.
detector (Acquity PDA, Waters) set at 350 nm, using a gradient
mobile phase from 5% to 95% acetonitrile in water with
trifluoroacetic acid (TFA) as modifier.
2.7. Synthesis of N-(2-hydroxypropionyl)-deacetylcolchicine (7)
2.4. Synthesis of N-deacetylcolchicine (4)
N-Deacetylcolchicine (4, 15 mg, 0.04 mmol), N-hydroxy-
succinimide (7 mg, 0.06 mmol), DL-lactic acid (5.4 mg, 0.06 mmol)
and 10 lL of Et3N were dissolved in CH2Cl2 (3 mL). DCC (12.4 mg,
0.06 mmol) in CH2Cl2 (3 mL) was subsequently added and stirred
for 16 h. After concentration in vacuo the product was purified
using flash chromatography in CH2Cl2/MeOH 9:1. Further purifica-
tion with flash chromatography using chloroform/MeOH/Et3N
91:9:0.2 (Rf = 0.4) yielded 7 as a yellow solid (14 mg, 82%).
N-Boc-deacetylcolchicine (3) was dissolved in CH2Cl2 (40 mL)
after which TFA (9 mL) was added. The reaction mixture was stir-
red for 2 h after which a saturated sodium carbonate solution (pH
approximately 10) was carefully added to quench the reaction. This
was extracted twice with CH2Cl2 and once with EtOAc. After pool-
ing the organic fractions, drying over MgSO4 and concentration in
vacuo, a thick brown oil was obtained. After flash chromatography
with CH2Cl2/MeOH/Et3N (90:10:0.1 Rf 0.38) and concentration, 4
was obtained as a slightly yellow solid (1.98 g, 79% in two steps).
ESI ꢂ MS½M þ Naꢃþ ¼ 380:05; calculated 380:15
½M þ Na þ CH3CNꢃþ ¼ 420:90; calculated 421:17
ESI ꢂ MS½Mꢃþ ¼ 428:10; calculated 428:46
1H NMR (300 MHz, CDCl3) d = 1.35 (d, 3H), 1.88–2.59 (4 ꢀ m,
5H), 3.62 (s, 3H), 3.89, 3.93, 3.98 (3 ꢀ s, 9H), 4.22 (q, 1H), 4.64
(m, 1H), 6.54 (s, 1H), 6.87 (d, 1H, J = 11.0), 7.32 (d, 1H, J = 10.7),
7.45 (s, 2 ꢀ 1H) ppm.
1H NMR (300 MHz, CDCl3) d 1.56–1.62 (m, 1H + broad S, ꢂNH2,
disappeared on D2O addition), 2.26–2.47 (m, 3H), 3.64 (s, 3H),
3.64–3.72 (m, 1H), 3.89 (s, 6H), 3.98 (s, 3H), 6.52 (s, 1H), 6.78 (d,
1H, J = 10.7 Hz), 7.17 (d, 1H, J = 10.7 Hz), 7.73 (s, 1H) ppm.
13C NMR (300 MHz, CDCl3) d 20.8, 30.1, 36.8, 51.8, 56.3, 56.5,
61.4, 61.7, 68.4, 107.6, 113.1, 125.8, 131.2, 134.5, 135.7, 137.0,
141.8, 151.4, 151.9, 153.7, 164.1, 175.6, 179.8 ppm.
2.8. Synthesis of mPEG-propionyl-N-(2-hydroxypropionyl-)-deacetyl
colchicine ester (prodrug II, 8)
2.5. Synthesis of N-(2-hydroxyacetyl)-deacetylcolchicine (5)
N-Deacetylcolchicine (4, 250 mg 0.67 mmol), glycolic acid (50 mg,
0.66 mmol), Et3N (0.3 mL, 2 mmol) and N-hydroxysuccinimide
N-(2-Hydroxypropionyl)-deacetylcolchicine (7, 14 mg, 0.033 mmol),
DMAP (15 mg, 0.12 mmol) and mPEG5000-propionic acid (210 mg,
0.041 mmol) were dissolved in CH2Cl2 (5 mL). DIC (15 lL,
(60 mg, 0.5 mmol) were dissolved in CH2Cl2 (10 mL). DIC (150 lL,
0.95 mmol) was added and stirred for 16 h. After concentration in va-
cuo, flash column chromatography was performed using CH2Cl2/
MeOH (90:10). After filtration in water followed by lyophilization, 5
was obtained as a slightly yellow solid (75 mg, 27%).
0.095 mmol) was added and stirred for 16 h. The reaction mixture
was concentrated in vacuo, dissolved in ammonium acetate buffer
(pH 5.5, 20 mM), filtered and finally separated on a PD-10 column
(GE Healthcare, Belgium) equilibrated with the same buffer. No
low molecular weight starting material could be detected by TLC
(CHCl3/MeOH/AcOH, 90:10:0.1). After lyophilization, prodrug II
TLC : Rf ¼ 0:40 in CHCl3=MeOH=AcOH 90 : 10 : 0:1
ESI ꢂ MS½M þ Naꢃþ ¼ 438:20; calculated 438:15
(8) was obtained as
a stringy offwhite solid (153 mg, 67%,

432
B.J. Crielaard et al. / European Journal of Pharmaceutical Sciences 45 (2012) 429–435
O
O
MeO
MeO
MeO
MeO
MeO
MeO
O
O
O
A
NH
O
N
NH
B
a
b
O
MeO
MeO
MeO
C
O
O
OMe
OMe
OMe
1
e
3
2
colchicine
N-Boc-colchicine
N-Boc-deacetylcolchicine
c
I
MeO
MeO
NH2
MeO
O
e
OMe
d
4
N-deacetylcolchicine
HO
HO
MeO
MeO
O
O
NH
O
NH
O
II
MeO
MeO
MeO
MeO
OMe
OMe
7
5
N-(2-hydroxyacetyl)-deacetylcolchicine/ colchifoline
N-(2-hydroxypropionyl)-deacetylcolchicine
O
O
n
g
f
O
O
n
O
O
O
O
O
O
MeO
MeO
MeO
MeO
NH
NH
III
MeO
MeO
O
O
OMe
Prodrug I
6
8
OMe
Prodrug II
Fig. 2. Synthesis of PEGylated colchicine-derived prodrugs. Prodrug I and II are synthesized in three phases. In phase I, colchicine is deacetyled in three steps. First, the
acetamide is Bocylated; second, the acetyl group is removed using methoxide; third, the Boc group is removed using TFA, yielding intermediate 4. In phase II, the deacetyl-
colchicine is reacted with glycolic acid or lactic acid by carbodiimide coupling to give the hydroxyl-functionalized colchicine derivatives 5 and 7, respectively. In phase III, the
hydroxyl group is conjugated via a hydrolysable ester bond to PEG bearing a carboxyl moiety (indicated by an arrow).Reagents: (a) Boc2O, DMAP, Et3N, Acetonitrile; (b)
NaOMe, MeOH; (c) TFA, DCM; (d) glycolic acid, DIC, N-hydroxysuccinimide, Et3N, CH2Cl2; (e) lactic acid, DCC, N-hydroxysuccinimide, Et3N, CH2Cl2; (f) mPEG5000-CH2COOH,
DIC, DMAP, CH2Cl2; (g) mPEG5000-(CH2)2COOH, DIC, DMAP, CH2Cl2.
composed of a mixture of unreacted mPEG-propionic acid and 8).
The amount of colchicine derivate conjugated to PEG was deter-
mined by UPLC, as described in Section 2.6.
2.10. Encapsulation in liposomes
DPPC, DPSE-PEG2000 and cholesterol were dissolved in ethanol
in a molar ratio of 1.85:0.15:1. Lipid films were formed by rotary
evaporation, which were dried further under a nitrogen flow for
30 min. The lipid films were hydrated with solutions of colchicine
(40 mg/mL), prodrug I (90 mg/mL) or prodrug II (60 mg/mL) in PBS
to form liposomes. The size and polydispersity of the liposomes
were decreased by repeated extrusions using an extruder (LIPEX,
Northern Lipids Inc, Canada) equipped with polycarbonate filters
with pore sizes of 200 nm and 100 nm (Whatman, USA). Colchicine
and prodrugs not encapsulated into the liposomes was removed by
ultracentrifugation (Beckman Coulter, The Netherlands) for 45 min
at 250,000 g. After resuspending the liposome pellet in PBS, the
particle size and size distribution were measured by dynamic light
2.9. In vitro degradation kinetics
The hydrolysis kinetics of the prodrugs were studied at 4 °C
and 37 °C in phosphate buffer (20 mM, pH 7.4) for both prodrug
I (6) and prodrug II (8), and at 37 °C in sodium hydrogen carbon-
ate buffer (20 mM, pH 9.0) for prodrug II (8). Samples were ta-
ken at several time points during 72 h, and stored at ꢂ20 °C
before analysis. The concentration of both PEGylated and
hydrolyzed colchicine derivatives was determined by UPLC, as
described in Section 2.6.

B.J. Crielaard et al. / European Journal of Pharmaceutical Sciences 45 (2012) 429–435
433
scattering using a Malvern ALV CGS-3 (Malvern Instruments, UK).
The concentration of colchicine and colchicine-derived prodrugs
in the liposome dispersions was determined by UPLC, as described
in Section 2.6.
2.11. In vitro release kinetics
The in vitro release kinetics of colchicine and colchicine-derived
prodrugs from long-circulating liposomes were determined by
dialysis of the liposome formulations against PBS at 4 °C and
37 °C. 1 mL of liposome dispersion was added to a Slide-A-Lyzer
Cassette (Thermo Scientific, USA) with a molecular weight cut-off
of 10 kDa, which was dialyzed against 200 mL of PBS. Samples
were taken from the dialysate at several time points during 72 h,
and stored at ꢂ20 °C before measurement by UPLC, as described
in Section 2.6.
2.12. Data analysis
The hydrolysis and liposomal release data were analysed using
Graphpad Prism v5.03 (Graphpad Software Inc, USA). LogP calcula-
tions were performed with MarvinSketch 5.3.2 (ChemAxon Ltd,
Hungary).
3. Results and discussion
3.1. Synthesis of colchicine-derived prodrugs
Two novel types of PEGylated colchicine-derived prodrugs were
synthesized following the scheme depicted in Fig. 2. Since colchi-
cine lacks functional groups suitable for conjugating PEG, deriva-
tives were synthesized and modified at the acetamide moiety at
the B-ring of colchicine, which is not part of the pharmacophore
and not essential for its tubulin binding activity (Bhattacharyya
et al., 2008; Nguyen et al., 2005). Earlier studies have shown that
modification at this site indeed does not result in a loss of activity,
provided that the modification does not cause too much steric hin-
drance (Bagnato et al., 2004; Bombuwala et al., 2006; Lagnoux
et al., 2005). The N-acetyl moiety of colchicine was substituted
with an N-2-hydroxyacetyl- or N-2-hydroxypropionyl moiety,
which are both suitable for esterification (Fig. 2). These
colchicinoids (5 and 7) were synthesized in two steps, monitored
by ESI-MS and NMR. First, the acetamide moiety of colchicine (1)
was protected (2), deacetylated (3) and deprotected (4), with a
final yield of 40%, using a strategy based on earlier reported
methodology (Fig. 2, phase I) (Bagnato et al., 2004; Lagnoux
et al., 2005). Then, the obtained deacetylcolchicine was re-acylated
with glycolic acid (5, yield 27%) and lactic acid (7, yield 82%),
respectively, allowing for conjugation with an acid-functionalized
methoxy PEG (Fig. 2, phase II). The glycolic acid derivative, also
known as colchifoline, has similar tubulin binding properties and
toxicity as its parent compound colchicine (Brossi et al., 1983).
Although the synthesis of colchifoline has been described in the
literature (Iorio et al., 1981), a more straightforward method for
the synthesis of both colchifoline and the related compound 7 is
presented here. By using of N,N0-dicyclohexylcarbodiimide (DCC)
instead of acid chlorides in the modification of colchicine, a higher
yield of colchicine derivatives could be achieved (Sharma et al.,
1984).
Fig. 3. Hydrolysis of PEGylated colchicine-derived prodrugs. The hydrolysis in
phosphate buffer of pH 7.4 at 4 °C (d) and 37 °C (j) was measured for both prodrug
I (A) and prodrug II (B). For prodrug I, the calculated hydrolysis half-life (t1/2) at 4 °C
was 14 d (zero-order kinetics, r2 0.64) and at 37 °C 5.4 h (first-order kinetics, r2
1.00). For prodrug II in phosphate buffer of pH 7.4 at 4 °C (d), the calculated
hydrolysis half-life (t1/2) was 500 d (zero-order kinetics, r2 0.69) and at pH7.4 at
37 °C (j) it was 9 d (zero-order kinetics, r2 0.87). In order to demonstrate the
liability of prodrug II to hydrolysis, the hydrolysis was also determined in carbonate
buffer of pH 9 at 37 °C (e), resulting in a calculated hydrolysis half-life of 7.1 h
(first-order kinetics, r2 = 0.99).
quantified by UPLC, which was 34% (w/w) for prodrug I and 74%
(w/w) for prodrug II.
3.2. In vitro hydrolysis kinetics
It was anticipated that the synthesized PEG-colchicinoids are
hydrolyzed at different rates under physiological conditions. The
main difference between prodrug I and prodrug II is that prodrug
I is an ester of a primary alcohol, which is more prone to hydrolysis
than esters of secondary alcohols, such as prodrug II. Additionally,
the closer proximity of the ether moiety to the carboxylic end
group of the mPEG in prodrug I further influences the degradation
rate due to electronic destabilization of the ester bond (Roberts
et al., 2002). Incubation of prodrug I in 20 mM phosphate buffer
with pH 7.4 at 37 °C resulted in first-order hydrolysis kinetics with
a calculated half-life (t1/2) of 5.4 h and degradation of approxi-
mately 90% of the total amount after 24 h incubation (Fig. 3A).
Furthermore, the hydrolysis in PBS at 4 °C appeared to be slow
(zero-order kinetics, calculated hydrolysis t1/2 = 14 d). Prodrug II
was found to be more resistant to hydrolysis at pH 7.4, with a
calculated half-life of approximately 9 d at 37 °C and a virtually
absent degradation at 4 °C (0.4% after 72 h) (Fig. 3B). When the
pH was increased to pH 9, to assess the susceptibility of prodrug
II to hydrolysis, the calculated half-life at 37 °C dropped to 7.1 h.
Finally, the conjugation of PEG5000 by esterification was per-
formed using methoxy PEG-acetic acid to obtain prodrug I (6)
and methoxy PEG-propionic acid to obtain prodrug II (8) (Fig. 2,
phase III). After conjugation, the product consisted of a mixture
of unreacted PEG and colchicine-derived prodrug. The amount of
colchicinoid prodrug present in each synthesized product was

434
B.J. Crielaard et al. / European Journal of Pharmaceutical Sciences 45 (2012) 429–435
Fig. 4. Release of colchicine and colchicine derivatives from long-circulating liposomes. Colchicine and colchicine-derived prodrugs were encapsulated into long-circulating
liposomes. Upon dialysis against PBS at 4 °C (d) and 37 °C (j), unmodified colchicine leaked out very rapidly (A). The release kinetics of colchicinoid from liposomes
encapsulating prodrug I (B) was more controlled, with slow release at 4 °C (zero-order kinetics, r2 0.96, release t1/2 = 13 d) and faster release at 37 °C (first-order kinetics, r2
1.00, release t1/2 = 8.2 h). The colchicinoid release from liposomes encapsulating prodrug II (C) was much slower: at 4 °C the release was negligible (zero-order kinetics, r2
0.73, release t1/2 = 181 d), while at 37 °C the release was higher (zero-order kinetics, r2 0.94, release t1/2 = 10 d), but still slow as compared to prodrug I.
3.3. Liposomal encapsulation and release
each prodrug. At 37 °C, long-circulating liposomes loaded with
prodrug I showed a rapid release of colchicinoid (first-order kinet-
ics, release t1/2 = 8.2 h) (Fig. 4B), whereas liposomes containing
prodrug II released the colchicinoid much slower (zero-order
kinetics, release t1/2 = 10 d) (Fig. 4C). At 4 °C, the release of colchi-
cinoid from liposomes loaded with prodrug I was slow (zero-order
kinetics, release t1/2 = 13 d h), whereas almost no release was ob-
served from liposomes loaded with prodrug II (zero-order kinetics,
release t1/2 = 181 d). These data demonstrate that the release of
colchicinoid is mainly determined by the hydrolysis rate of the
prodrug, as both prodrugs have similar physicochemical character-
istics as such. The partitioning coefficients of the colchicinoids
used for the synthesis of prodrug I and II are similar to that of col-
chicine (calculated logP: 0.6 and 1.2 respectively), which indicates
that these colchicinoids, like colchicine, will be rapidly released
from the long-circulating liposomes. Both prodrugs, however, are
retained within the aqueous core of the liposomes, and only after
cleavage of the PEG chain by hydrolysis of the ester bond, the
formed colchicinoid can diffuse over the lipid bilayer. These results
demonstrate that by adjusting the PEG-linker, and thus modifying
the hydrolysis rate of the colchicine-derived prodrugs, the release
rate of colchicinoid from liposomes can be tailored. After i.v.
administration, targeted nanomedicines such as liposomes need
time to localize passively in the target tissue, and during this per-
iod the drug should not be released. On the other hand, there
should be efficient drug release after localization in the target area
3.3.1. Liposomes containing colchicine
Colchicine was encapsulated in long-circulating liposomes with
a mean particle size of around 100 nm and low polydispersity,
using the well-known lipid film hydration method. As expected,
due to its physicochemical properties, colchicine was released
readily (Fig. 4A). This has been observed before (Kulkarni et al.,
1997). Colchicine has moderate water solubility with a partition
coefficient (logP) of around 1.0 (Quinn et al., 1981; Zamora et al.,
1988), and localizes in the lipid bilayer (Mons et al., 2000). The ra-
pid release, i.e. 100% within 24 h, indicates that colchicine will also
leak out of the liposomes when they are circulating in the blood-
stream before they reach the target site, making this formulation
unsuitable for therapeutic applications.
3.3.2. Liposomes containing colchicine-derived prodrugs
By conjugating colchicine to PEG5000 using a biodegradable lin-
ker, prodrugs with large molecular weight and high aqueous solu-
bility (calculated logP: ꢂ4.44 (prodrug I) and ꢂ3.64 (prodrug II))
were obtained. The improved solubility allows encapsulation into
the aqueous core of long-circulating liposomes. Both prodrug I
and II were entrapped into long-circulating liposomes with a mean
diameter of around 100 nm and low polydispersity. At pH 7.4, the
two liposomal prodrug formulations showed a release profile of
colchicinoid that strongly resembled the hydrolysis kinetics of

B.J. Crielaard et al. / European Journal of Pharmaceutical Sciences 45 (2012) 429–435
435
Cerquaglia, C., Diaco, M., Nucera, G., La Regina, M., Montalto, M., Manna, R., 2005.
Pharmacological and clinical basis of treatment of Familial Mediterranean Fever
(FMF) with colchicine or analogues: an update. Curr. Drug Targets Inflamm.
Allergy 4, 117–124.
Coimbra, M., Isacchi, B., van Bloois, L., Torano, J.S., Ket, A., Wu, X., Broere, F.,
Metselaar, J.M., Rijcken, C.J.F., Storm, G., Bilia, R., Schiffelers, R.M., 2011.
Improving solubility and chemical stability of natural compounds for medicinal
use by incorporation into liposomes. Int. J. Pharm. 416, 433–442.
Fenske, D.B., Cullis, P.R., 2008. Liposomal nanomedicines. Expert. Opin. Drug. Deliv.
5, 25–44.
to induce a therapeutic effect. Although the optimal release profile
at the target site needs to be established experimentally, the pre-
sented prodrug strategy shows that premature leakage in the cir-
culation is prevented.
4. Conclusions
This work demonstrates that the release kinetics of colchici-
noids from long-circulating liposomes can be tailored by encapsu-
lating rationally designed PEGylated colchicinoid prodrugs that
hydrolyze in the aqueous interior. Two PEGylated colchicine-de-
rived prodrugs were synthesized by substituting the acyl-moiety
with a hydroxyl containing moiety to allow for esterification with
an acid-functionalized PEG. By employing glycolic or lactic acid to
functionalize colchicine, and by using different PEG acids, the
hydrolysis rate of the prodrug could be controlled. Both designed
prodrugs were retained in liposomes, whereas colchicine leaked
out instantaneously. Indeed, the release kinetics of colchicinoids
from the liposomal prodrug formulations appeared determined
by the hydrolysis rate of the prodrug.
Iorio, M.A., Molinari, M., Brossi, A., 1981. Synthesis of colchifoline from deacetyl-
colchiceine. Can. J. Chem. 59, 283–284.
Jordan, M.A., Wilson, L., 2004. Microtubules as a target for anticancer drugs. Nat.
Rev. Cancer 4, 253–265.
Kanthou, C., Tozer, G.M., 2007. Tumour targeting by microtubule-depolymerizing
vascular disrupting agents. Expert Opin Ther Targets 11, 1443–1457.
Keith, M.P., Gilliland, W.R., 2007. Updates in the management of gout. Am. J. Med.
120, 221–224.
Kulkarni, S.B., Singh, M., Betageri, G.V., 1997. Encapsulation, stability and in-vitro
release characteristics of liposomal formulations of colchicine. J. Pharm.
Pharmacol. 49, 491–495.
Lagnoux, D., Darbre, T., Schmitz, M.L., Reymond, J.L., 2005. Inhibition of mitosis
by glycopeptide dendrimer conjugates of colchicine. Chem. Eur. J. 11,
3941–3950.
Lammers, T., Hennink, W.E., Storm, G., 2008. Tumour-targeted nanomedicines:
principles and practice. Br. J. Cancer 99, 392–397.
Lippert, J.W., 3rd, 2007. Vascular disrupting agents. Biorg. Med. Chem. 15,
605-615.
Mons, S., Veretout, F., Carlier, M.F., Erk, I., Lepault, J., Trudel, E., Salesse, C., Ducray, P.,
Mioskowski, C., Lebeau, L., 2000. The interaction between lipid derivatives of
colchicine and tubulin: Consequences of the interaction of the alkaloid with
lipid membranes. BBA-Biomembranes 1468, 381–395.
Nguyen, T.L., McGrath, C., Hermone, A.R., Burnett, J.C., Zaharevitz, D.W., Day, B.W.,
Wipf, P., Hamel, E., Gussio, R., 2005. A common pharmacophore for a diverse set
of colchicine site inhibitors using a structure-based approach. J. Med. Chem. 48,
6107–6116.
Nihei, Y., Suzuki, M., Okano, A., Tsuji, T., Akiyama, Y., Tsuruo, T., Saito, S., Hori, K.,
Sato, Y., 1999. Evaluation of antivascular and antimitotic effects of tubulin
binding agents in solid tumor therapy. Jpn. J. Cancer Res. 90, 1387–1395.
Parveen, S., Sahoo, S.K., 2006. Nanomedicine: clinical applications of polyethylene
glycol conjugated proteins and drugs. Clin. Pharmacokinet. 45, 965–988.
Pasquier, E., Honore, S., Braguer, D., 2006. Microtubule-targeting agents in
angiogenesis: where do we stand? Drug Resist. Update 9, 74–86.
Pasquier, E., Kavallaris, M., 2008. Microtubules: a dynamic target in cancer therapy.
IUBMB Life 60, 165–170.
Consequently, by using different biodegradable linkers, the release
of colchicinoids from long-circulating liposomes can be tailored.
Acknowledgements
This work was supported by MediTrans, an Integrated Project
funded by the European Commission under the ‘‘nanotechnologies
and nano-sciences, knowledge-based multifunctional materials
and new production processes and devices’’ (NMP), thematic prior-
ity of the Sixth Framework Program.
References
Bagnato, J.D., Eilers, A.L., Horton, R.A., Grissom, C.B., 2004. Synthesis and
Petersel, D., Schlesinger, N., 2007. Treatment of acute gout in hospitalized patients.
J. Rheumatol. 34, 1566–1568.
characterization of
a cobalamin-colchicine conjugate as a novel tumor-
targeted cytotoxin. J. Org. Chem. 69, 8987–8996.
Quinn, F.R., Neiman, Z., Beisler, J.A., 1981. Toxicity and quantitative structure-
activity relationships of colchicines. J. Med. Chem. 24, 636–639.
Rautio, J., Kumpulainen, H., Heimbach, T., Oliyai, R., Oh, D., Jarvinen, T., Savolainen,
J., 2008. Prodrugs: design and clinical applications. Nat. Rev. Drug Discov. 7,
255–270.
Ravelli, R.B.G., Gigant, B., Curmi, P.A., Jourdain, I., Lachkar, S., Sobel, A., Knossow, M.,
2004. Insight into tubulin regulation from a complex with colchicine and a
stathmin-like domain. Nature 428, 198–202.
Roberts, M.J., Bentley, M.D., Harris, J.M., 2002. Chemistry for peptide and protein
PEGylation. Adv. Drug Deliv. Rev. 54, 459–476.
Sharma, P.N., Brossi, A., Silverton, J.V., Chignell, C.F., 1984. Synthesis and binding to
tubulin of colchicine spin probes. J. Med. Chem. 27, 1729–1733.
Terkeltaub, R.A., 2009. Colchicine update: 2008. Semin. Arthritis Rheum. 38, 411–
419.
Thorpe, P.E., 2004. Vascular targeting agents as cancer therapeutics. Clin. Cancer
Res. 10, 415–427.
Torchilin, V.P., 2005. Recent advances with liposomes as pharmaceutical carriers.
Nat. Rev. Drug Discov. 4, 145–160.
Tozer, G.M., Kanthou, C., Baguley, B.C., 2005. Disrupting tumour blood vessels. Nat.
Rev. Cancer 5, 423–435.
Zamora, J.M., Pearce, H.L., Beck, W.T., 1988. Physical-chemical properties shared by
compounds that modulate multidrug resistance in human leukemic cells. Mol.
Pharmacol. 33, 454–462.
Baguley, B.C., Holdaway, K.M., Thomsen, L.L., Zhuang, L., Zwi, L.J., 1991. Inhibition of
growth of colon 38 adenocarcinoma by vinblastine and colchicine: evidence for
a vascular mechanism. Eur. J. Cancer 27, 482–487.
Baluk, P., Hashizume, H., McDonald, D.M., 2005. Cellular abnormalities of blood
vessels as targets in cancer. Curr. Opin. Genet. Dev. 15, 102–111.
Bell, S.J., Fam, C.M., Chlipala, E.A., Carlson, S.J., Lee, J.I., Rosendahl, M.S., Doherty,
D.H., Cox, G.N., 2007. Enhanced circulating half-life and antitumor activity of a
site-specific pegylated interferon-
299–305.
a protein therapeutic. Bioconjug. Chem. 19,
Bhattacharyya, B., Panda, D., Gupta, S., Banerjee, M., 2008. Anti-mitotic activity of
colchicine and the structural basis for its interaction with tubulin. Med. Res.
Rev. 28, 155–183.
Bibas, R., Gaspar, N.K., Ramos-e-Silva, M., 2005. Colchicine for dermatologic
diseases. J. Drugs Dermatol. 4, 196–204.
Bombuwala, K., Kinstle, T., Popik, V., Uppal, S.O., Olesen, J.B., Vina, J., Heckman, C.A.,
2006. Colchitaxel, a coupled compound made from microtubule inhibitors
colchicine and paclitaxel. Beilstein J. Org. Chem. 2, 13.
Brossi, A., Sharma, P.N., Atwell, L., Jacobson, A.E., Iorio, M.A., Molinari, M., Chignell,
C.F., 1983. Biological effects of modified colchicines. 2. Evaluation of catecholic
colchicines, colchifolines, colchicide, and novel N-acyl- and N-aroyldeacetyl-
colchicines. J. Med. Chem. 26, 1365–1369.
Carmeliet, P., Jain, R.K., 2000. Angiogenesis in cancer and other diseases. Nature 407,
249–257.