Synthesis, in vitro and in vivo characterization of novel ethyl dioxy phosphate prodrug of propofol

Article abstract of DOI:10.1016/j.ejps.2008.02.121

A novel ethyl dioxy phosphate prodrug of propofol (3) was synthesized and characterized in vitro and in vivo as safer alternative for phosphonooxymethyl prodrugs. The synthesis of 3 was achieved via vinyl and 1-chloroethyl ether intermediates, followed by

Full text of DOI:10.1016/j.ejps.2008.02.121

european journal of pharmaceutical sciences 3 4 ( 2 0 0 8 ) 110–117
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/ejps
Synthesis, in vitro and in vivo characterization of novel ethyl
dioxy phosphate prodrug of propofol
Hanna Kumpulainena,∗, Tomi Ja¨rvinena, Anne Mannilaa, Jukka Leppa¨nena,
Tapio Nevalainena, Antti Ma¨ntyla¨ a, Jouko Vepsa¨la¨inenb, Jarkko Rautioa
a
Department of Pharmaceutical Chemistry, University of Kuopio, PO Box 1627, FI-70211 Kuopio, Finland
Department of Chemistry, University of Kuopio, PO Box 1627, FI-70211 Kuopio, Finland
b
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel ethyl dioxy phosphate prodrug of propofol (3) was synthesized and characterized
in vitro and in vivo as safer alternative for phosphonooxymethyl prodrugs. The synthesis
of 3 was achieved via vinyl and 1-chloroethyl ether intermediates, followed by addition of
phosphate group. Aqueous solubility and chemical stability of 3 was determined in buffer
solutions and the bioconversion of 3 to propofol was determined in vitro and in vivo. The
results show that 3 greatly enhanced the aqueous solubility of propofol (solubility over
10 mg/mL) and the stability in buffer solution (t1/2 = 5.2 0.2 days at pH 7.4, r.t.) was suffi-
cient for i.v. administration. The enzymatic hydrolysis of 3 to propofol was extremely rapid
in vitro (t1/2 = 21 3 s) and 3 was readily converted to propofol in vivo in rats. During biocon-
version, 3 releases acetaldehyde, a less toxic compound than the formaldehyde released
from the phosphonooxymethyl prodrug of propofol (Aquavan), currently undergoing clin-
ical trials. The maximum plasma concentration of propofol, 3.0 0.2 ␮g/mL, was reached
within 2.1 0.8 min after the i.v. administration of 3. The present study indicates that ethyl
dioxy phosphate represents a potentially useful water-soluble prodrug structure suitable for
i.v. administration.
Received 9 January 2008
Received in revised form
6 February 2008
Accepted 25 February 2008
Published on line 2 March 2008
Keywords:
Prodrug
Ethyl dioxy phosphate
Acetaldehyde
Phosphatase
Formaldehyde
© 2008 Elsevier B.V. All rights reserved.
1.
Introduction
the parent drug, either directly or via an oxymethyl spacer
group. Oxymethyl spacers are generally used to increase
Solubility limitations of drug molecules are an emerging
challenge for the pharmaceutical industry with an objec-
tive to find pharmacologically more potent drug molecules.
Since sufficient aqueous solubility is crucial in both i.v. and
non-parenteral administration and because the possibilities
to formulate the poorly water-soluble drug molecules are
limited, prodrug technology has become a popular way to
enhance the aqueous solubility of drug molecules.
the space around the enzymatically cleavable bond (phos-
phates, esters, carbonates) (Safadi et al., 1993; Varia et al.,
1984) and they undergo rapid spontaneous chemical hydroly-
sis after enzymatic hydrolysis of the promoiety. One possible
drawback of the oxymethyl linker structure is the systemic
liberation of the toxic compound, formaldehyde, in the body,
which in addition to its toxicity, may alter homeostasis within
cells (such as induction of enzymes, metabolic switching and
cell proliferation) (Heck et al., 1990). An alternative strat-
egy to achieve sufficiently fast enzymatic hydrolysis rate
and more favourable safety profile of prodrugs is to use an
Phosphates are probably the most extensively used water-
soluble prodrugs for i.v. administration. Phosphates have been
used as promoieties in hydroxyl or amine functionalities of
∗
Corresponding author. Tel.: +358 44 303 7577; fax: +358 17 162 456.
E-mail address: Hanna.Kumpulainen@uku.fi (H. Kumpulainen).
0928-0987/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.ejps.2008.02.121

european journal of pharmaceutical sciences 3 4 ( 2 0 0 8 ) 110–117
111
Fig. 1 – The hydrolysis of the ethyl dioxy phosphate
prodrug of propofol to the parent drug, promoiety and
acetaldehyde.
ethyl dioxy-spacer structure that liberates the much less toxic
compound, acetaldehyde, from the prodrug structure (Fig. 1).
Although acetaldehyde is not fully harmless, as it can cause
gastrointestinal tract cancer after long-term chronic alcohol
consumption (Seitz et al., 2005), it is substantially less toxic
than formaldehyde.
The (acyloxy)alkyl group has been widely used as link-
ers between an ester, a carbonate or a carbamate promoiety
and parent drugs containing a carboxylic acid functional-
ity (Beaumont et al., 2003), in fact many such prodrugs are
even marketed (e.g. candesartan cilexitil (Easthope and Jarvis,
2002), cefuroxime axetil (Scott et al., 2001), bacmecillinam
(Josefsson et al., 1982)). By comparison, no ethyl dioxy linked
phosphate prodrugs have been described in the literature. One
possible reason is the challenge involved in the synthesis of
these compounds, in fact there are no techniques for the syn-
thesis of ethyl dioxy phosphates described in the literature
(SciFinder Scholar, Beilstein Crossfire). This represented our
starting point; we were interested in determining whether
an ethyl dioxy linked phosphate could be synthesized and
whether the ethyl dioxy spacer could serve as a safe and
hydrolysable linker between the parent drug and the phos-
phate promoiety.
For this study, we choose an anesthetic drug propofol as
our model drug. Propofol (Fig. 2) is a widely used anesthetic
with a rapid onset, a short duration of action and minimal
side effects (Baker and Naguib, 2005). Because of the high
lipophilicity of propofol, it needs to be administered in oil-in-
water emulsion. However, the extremely low water-solubility,
high lipophilicity, inherent emulsion instability, pain on injec-
tion and hyperlipidemia on prolonged administration (Baker
and Naguib, 2005) are major drawbacks, and thus have lead to
the development of more water-soluble prodrugs. At least two
phosphate prodrugs of propofol have been described in the lit-
erature (Fig. 2), one with the phosphate group attached directly
to the hydroxyl functionality of propofol (Propofol Phosphate,
PP)(Banaszczyk et al., 2002) and another with an oxymethyl
spacer between the parent drug and the phosphate promoiety
(5, GPI 15715, Aquavan®) (Gibiansky et al., 2005; Schywalsky et
al., 2003; Struys et al., 2005). Both prodrugs enhance the water-
solubility of propofol by many-fold, as would be predicted
from their structures. The enzymatic hydrolysis of GPI 15715
(Schywalsky et al., 2003) is faster than that of PP (Banaszczyk et
al., 2002), but on the other hand, PP does not liberate the toxic
formaldehyde as a breakdown product to the body. We hoped
Fig. 2 – The structures of propofol and its phosphate-,
phosphonooxymethyl- (5, GPI 15715, Aquavan®) and ethyl
dioxy phosphate- (3) prodrugs.
to combine the properties of these two prodrugs, and there-
fore, we synthesized an ethyl dioxy linked phosphate prodrug
3 of propofol. Furthermore, we have extensively characterized
the ethyl dioxy structure as a novel water-soluble phosphate
prodrug structure in vitro and in vivo in rats.
2.
Materials and methods
All materials, reagents and enzymes were obtained from com-
mercial suppliers and were used without further purifications.
2.1.
Synthesis of compounds
2.1.1. General synthetic procedures
All the described reactions were monitored by thin-layer chro-
matography using aluminum sheets precoated with Merck
silica gel 60 F254. Samples were visualized by UV-light. Col-
umn chromatography was executed on Merck silica gel 60
F254 (0.063–0.200 mm mesh). 1H, 13C and 31P NMR-spectra
were recorded on a Bruker Avance DRX500 spectrometer
(Bruker, Rheinstetter, Germany) operating at 500 MHz, 125.76
and 202.46 MHz at 25 ◦C, respectively. TMS for CDCl3-samples
and TSP for D2O samples were used as an internal reference.
The splitting pattern abbreviations are as follows: s = singlet,
d = doublet, t = triplet, q = quartet, m = multiplet, sep = septet.
Electrospray ionization mass spectra were acquired by an LCQ
quadrupole ion trap mass spectrometer with an electronspray
ionization source (Finnigan MAT, San Jose, CA). Elemental
analyses were carried out on a ThermoQuest CE Instruments
EA 1110-CHNS-O elemental analyzer.

112
european journal of pharmaceutical sciences 3 4 ( 2 0 0 8 ) 110–117
2.1.2. 2,6-Diisopropyl-2-vinyloxy-benzene (1)
2.1.6. Phosphoric acid mono-(2,6-diisopropyl-
Anhydrous copper(II)acetate (3.67 g, 20.19 mmol) was added
to a solution of 2,6-diisopropyl phenol (3 g, 16.83 mmol) in
dry acetonitrile. The reaction flask was rinsed with O2.
Tetravinyltin (3.74 g, 16.48 mmol) was added to the reaction
mixture via a syringe and the reaction was stirred for 1 h at
room temperature. O2-bubbling was replaced by a O2-balloon
and reaction was stirred at 60 ◦C for 16 h. The resulting brown
mixture was poured into a 25% NH4OAc-solution (30 mL)
and stirred for 15 min. The reaction was extracted with Et2O
(150 mL). The organic layer was washed with saturated brine
(2 × 40 mL), dried over Na2SO4 and evaporated to dryness.
Flash chromatography on SiO2 (2% ethyl acetate in petrol
ether) gave 1 as a colorless oil (1.34 g, 6.56 mmol, 39%). 1H
NMR (CDCl3) ı 7.18–7.10 (3H, m), 6.60 (1H, dd, J = 14.0 Hz and
6.5 Hz), 4.15–4.08 (2H, m), 3.18 (2H, sep, J = 6.9 Hz), 1.19 (12 H, d,
J = 6.9 Hz).
phenoxymethyl) ester disodium salt (5)
To a solution of triethylamine (13.03 mL, 9.46 g, 93.5 mmol) and
85% phosphoric acid (5.86 mL, 9.87 g, 101 mmol) was added 4
(3.2 g, 14.1 mmol) and the reaction mixture was heated at 64 ◦C
for 2.5 h. The solvents were evaporated, 100 mL of water was
added to the residue and the pH of the mixture was adjusted
to 1.5 with 6 M HCl. The aqueous solution was extracted with
diethyl ether (3 × 50 mL), washed with brine and evaporated
to dryness. 50 mL of water was added to the residue and pH
was adjusted to 9 with saturated NaOH-solution. The mixture
was washed with toluene (2 × 30 mL) and concentrated under
reduced pressure to half of the volume. 100 mL of isopropanol
was added, the mixture was frozen, allowed to warm to 0 ◦C,
filtered and dried in a vacuum oven to obtain 5 as a white
solid (0.87 g, 3.0 mmol, 21%) 1H NMR (D2O) ı 7.32–7.24 (3H, m),
5.27 (2H, d, J = 7.5 Hz), 3.46 (2H, m), 1.22 (12 H, d, J = 6.9 Hz). 13
C
NMR (D2O) 152.96, 145.90, 128.63, 127.03, 96.11, 28.97, 25.95.
31P NMR (D2O) ı 2.02. ESI-MS: 287.1 (M−2Na + 1). Anal. Calcd
for C13H19Na2O5P·0.2H2O: C, 46.49; H, 5.82. Found: C, 46.12; H,
5.82.
2.1.3. 2-(1-Chloro-ethoxy)-1,3-diisopropyl-benzene (2)
Ethyl acetate was saturated with dry HCl-gas by bubbling
gaseous HCl through the solution for 10 min.
2 (1.04 g,
5.1 mmol) was dissolved in 10 mL of HCl-saturated ethyl
acetate. The mixture was heated in a microwave-oven at
100 ◦C at 5 bar for 10 min and evaporated to dryness. The
next reaction was continued immediately without any further
purification.
2.2.
HPLC analysis
HPLC analysis of 3, 5 and propofol was performed using an
Agilent Technologies 1100 series gradient RP-HPLC system
with UV detection (220 nm) and for rat whole blood sam-
ples the fluorescence detection (Ex = 276 nm, Em = 310). The
HPLC system consisted of Agilent 1100 Series Binary Pump,
1100 Series Autosampler, 1100 Series Micro Vacuum Degasser,
1100 Series Thermostated Column Compartment, 1100 Series
Fluorescence Detector, HP 1050 Variable Wavelenght Detec-
tor, 1100 Series Control Module and a Zorbax Eclipse XDB-C8
(4.6 mm × 150 mm, 5 ␮m) analytical column (Agilent Technolo-
gies Inc., Little Falls, Wilmington, DE). Loop injection volume
was 20 ␮L. The isocratic elution was performed by using a
mobile phase consisting of 90% (v/v) acetonitrile and 10 mM
tetrabutylammonium dihydrogen phosphate at a ratio of 65:35
at a flow rate of 1.0 mL/min at 30 ◦C.
2.1.4. Phosphoric acid mono-[1-(2,6-diisopropyl-phenoxy)-
ethyl] ester disodium salt (3)
To a solution of tetrabutylammonium phosphate (0.4 M in
acetonitrile, 30 mL, 12 mmol) was added 2 (5.1 mmol, theoret-
ical maximum yield) in triethylamine (5 mL, 36 mmol) under
argon. The reaction mixture was stirred for 18 h and the sol-
vents were evaporated. Thirty milliliters of water was added
and the residue was extracted with diethyl ether (3 × 50 mL).
Solvents were evaporated, 10 mL of acetonitrile was added
and the pH of the mixture was adjusted to 11 with saturated
NaOH. After stirring for 10 min, the solvents were evaporated,
the residue was purified with preparative HPLC on a reversed
˚
phase Kromasil 100 A (C8) column by gradient elution using
water and acetonitrile (40–80% ACN) as eluent and lyophilized
to obtain 3 as a white solid (494 mg, 1,43 mmol, 28%). 1H NMR
(D2O) ı 7.29–7.22 (3H, M), 5.54 (1H, q, J = 5.1 Hz), 3.42 (2H, m), 1.66
(3H, d, J = 5.1 Hz), 1.21 (12 H, d, J = 6.9 Hz). 13C NMR (D2O) ı 152.30,
146.45, 128.57, 127.08, 103.04, 29.28, 26.03 (d, J = 15.2 Hz), 24.97.
31P NMR (D2O) ı 0.03. ESI-MS: 301.3 (M−2Na + 1). Anal. Calcd
for C14H21Na2O5P·0.7H2O: C, 46.85; H, 6.29. Found: C, 46.79; H,
6.29.
2.3.
Aqueous solubility
The aqueous solubilities of 3 and propofol were determined
at room temperature in Tris–HCl-buffer (50 mM) at pH 7.4. The
pH of the mixtures was held constant during the study. Six
milligrams of 3 or excess amounts of propofol were added
to 0.5 mL of buffer, the mixtures were stirred for 1 h, filtered
(0.45 ␮m Millipore) and analyzed by HPLC.
2.1.5. 2-Chloromethoxy-1,3-diisopropyl-benzene (4)
Sodium hydroxide pellets (1.2 g, 30.0 mmol) and bro-
mochloromethane (28.7 mL, 57.1 g, 441 mmol) were added to
a solution of 2,6-diisopropyl phenol (2.7 g, 15.2 mmol) in dry
tetrahydrofuran under nitrogen. The reaction mixture was
heated at 64 ◦C for 2 h, allowed to cool to room temperature
and filtered. The filtrate was evaporated to dryness to obtain
4 as a yellowish oil (3.2 g, 14.1 mmol, 93%). 4 was used in
the next reaction without any further purification. 1H NMR
(CDCl3) ı 7.20–7.09 (3H, m), 5.75 (2H, s), 3.35 (2H, m), 1.22 (12
H, d, J = 6.9 Hz).
2.4.
Hydrolysis in aqueous solutions
The rate of the chemical hydrolysis of 3 and 5 was deter-
mined in borate buffer (180 mM) and in Tris buffer (50 mM)
at pH 7.4 at 37 ◦C and at r.t. An appropriate amount of 3 or
5 (concentration approx. 50 ␮M) was dissolved in preheated
buffer solution or in buffer solution at room temperature and
the solutions were placed in a thermostatically controlled
water bath at 37 ◦C or to magnetic stirrer at r.t. Samples were
taken at appropriate intervals and analyzed by HPLC. Pseudo-
first-order half-lives (t1/2) for the hydrolysis of prodrugs were

european journal of pharmaceutical sciences 3 4 ( 2 0 0 8 ) 110–117
113
calculated from the slope of the linear portion of the plotted
logarithm of the remaining prodrug versus time.
0.9% NaCl). For the determination of propofol, venous blood
samples of 200 ␮L were collected immediately before and 1, 2,
4, 6, 10 and 30 min after the i.v. bolus administration, mixed
with 20 ␮L of 3% EDTA in 0.7% NaCl-solution and frozen imme-
diately. The drawn blood volume was replaced with 0.9% NaCl.
2.5.
Enzymatic hydrolysis in alkaline phosphatase
solution
2.7.
Rat blood sample preparations
Hydrolysis of 3 in alkaline phosphatase solution (type VII-S
from bovine intestinal mucosa, Sigma–Aldrich) was deter-
mined at 37 ◦C. 4 ␮L (81.8 units) of alkaline phosphatase was
added to a solution of 3 in 50 mM Tris–HCl buffer (pH 7.4,
final concentration 50 ␮M) at 37 ◦C. In blank solutions, alka-
line phosphatase was replaced with the same volume of water
to ensure that the hydrolysis was only enzymatic. At prede-
termined time intervals, 200 ␮L samples were removed and
200 ␮L of ice-cold acetonitrile was added to each sample to
stop the enzymatic hydrolysis. The samples were kept on ice,
centrifuged for 10 min at 14,000 rpm, and the supernatant was
analyzed by HPLC. Pseudo-first-order half-lives (t1/2) for the
hydrolysis of 3 were calculated from the slope of the linear por-
tion of the plotted logarithm of the remaining prodrug versus
time.
For the assay of propofol in rat whole blood samples, the
red blood cells were lysed with a total of three freeze/thaw
cycles. 10 ␮L of 90 ␮g/mL thymol as internal standard was
added to 100 ␮L of lysed whole blood, then 250 ␮L of ace-
tonitrile was added to precipitate proteins and the samples
were centrifuged for 3 min at 12,000 rpm. The supernatant
was injected directly into the HPLC column. The quantity of
propofol was determined using external, spiked standards
of lysed whole blood and the concentration was estimated
based on peak-area ratio of propofol to the internal stan-
dard (thymol). All concentration curves and pharmacokinetic
parameters (Tmax, Cmax) after i.v. administration for each ani-
mal were calculated by standard pharmacokinetic methods
using WinNonlin Nonlinear Estimation Program, Professional
version 5.0.1 (Pharsight Corporation).
2.6.
Animal treatments
Adult male Wistar rats weighing 250 5 g were purchased
from the National Laboratory Animal Centre (Kuopio, Finland).
Rats were housed on a 12-h light/12-h dark cycle. All experi-
ments were carried out during the light phase. The rats had
free access to tap water and food pellets. All procedures were
reviewed and approved by the Animal Ethics Committee at the
University of Kuopio. Rats were anesthetized with ketamine
(90 mg/kg, i.p.) and xylazine (8 mg/kg, i.p.). The right jugular
vein was exposed aseptically as described by Waynforth and
Flecnell, (Waynforth and Fleckell, 1994) and cannulated with a
PE-50 catheter filled with 100 IE/mL heparin. The catheter was
exteriorised through a small incision made in the neck. The
rats were allowed to recover until the following day.
3.
Results
3.1.
Synthesis
The phosphonooxymethyl derivative of propofol (5, GPI 15715)
was synthesized as previously described (Bonneville et al.,
2003) from propofol with bromochloromethane and phospho-
ric acid (Fig. 3) so that chemical stability studies could be
undertaken. The synthesis of ethyl dioxy linked prodrug 3
was more challenging. Our first attempt to develop a suit-
able synthesis route for ethyl dioxy linked prodrugs involved
the method for the synthesis of 1-chloroethyl phosphates
(Kumpulainen et al., 2005, 2006). It was thought that 1-
chloroethyl phosphates might possibly be used as starting
materials for the synthesis of ethyl dioxy linked phosphate
The drug solutions were administered to the conscious rats
as 30-s i.v. injections (including the rinsing of the catheter with
Fig. 3 – Synthesis of the ethyl dioxy phosphate prodrug 3 and phosphonooxymethyl prodrug 5, GPI 15715, of propofol.

114
european journal of pharmaceutical sciences 3 4 ( 2 0 0 8 ) 110–117
Table 1 – The aqueous solubility of 3 and propofol in buffer solution, half-lives of 3 and 5 in buffer solutions and half-life
of 3 in alkaline phosphatase solution and, all at pH 7.4 (mean S.D.; n = at least 3)
Compound
Solubility
(mg mL−1); Tris
buffer r.t.c
Chemical
stability, t1/2; Tris
buffer 37 ◦C
Chemical
Chemical
Enzymatic hydrolysis,
t1/2; alkaline
stability, t1/2
;
stability, t1/2;
borate buffer 37 ◦C
borate buffer r.t.c
phosphatase, 37 ◦C
3
5
>10
–
0.13 0.002
25.8 3.3 h
21.5 0.8 h
Stableb
–
5.2 0.2 days
21 3 s
a
a
a
a
–
–
a
–
a
a
a
Propofol
–
–
–
a
Not determined.
No degradation was observed after three weeks.
Room temperature.
b
c
prodrugs. However, probably because of the steric hindrance
and the presence of a phenolic hydroxyl group which is a
good leaving group, the 1-chloroethyl phosphates did not react
with propofol. The second attempt was to use the same syn-
thesis route to synthesize 1-chloroethyl ether of propofol via
vinyl ether, which could be further derivatized to phosphate.
We found that the only way to make vinyl ether of propo-
fol was the use of tetravinyl tin and cuprous acetate in an
oxygen atmosphere (Blouin and Frenette, 2001). The vinyl
ether underwent fast microwave-assisted conversion to 1-
chloroethyl ether in the presence of HCl-gas at 100 ◦C within
10 min.
Since the ethyl dioxy structure was found to be chemically
acid-labile and since many protecting groups of phosphates
require acidic conditions for their removal, the synthesis of
an ethyl dioxy phosphate was carried out without any pro-
tecting groups being present. The synthesis was achieved by
using tetrabutylammonium phosphate in acetonitrile with an
excess of triethylamine under dry and basic conditions in
organic solvent. The sodium salt 3 was obtained with the
treatment of the product with sodium hydroxide and the final
product was purified with preparative HPLC. Since most of the
marketed (acyloxy) alkyl prodrugs are racemic mixtures, the
enantiomers were not separated, but the racemic mixture was
used in all assays which will be described.
the drug were to be used for i.v. administration, although the
chemical degradation was significantly faster than that of 5
(no degradation was observed after three weeks). There were
no significant differences between the stabilities of 3 at 37 ◦C
in different buffer solutions (Tris- and borate buffer).
3.4.
Enzymatic hydrolysis in alkaline phosphatase
solution
The enzymatic hydrolysis of 3 was determined in alkaline
phosphatase solution at pH 7.4. Only the half-life of 3 but not
the exact amount of formed propofol was determined due to
the low solubility of propofol. The enzymatic hydrolysis of 3 to
propofol was extremely rapid with a half-life of approximately
20 s (Table 1).
3.5.
Bioconversion of 3 to propofol in vivo
The bioconversion of 3 to propofol was demonstrated in vivo
in rats after a single i.v. bolus administration. One group of
three adult male Wistar rats received a single bolus dose
of 10 mg/kg of propofol (0.056 mmol/kg) and another group
of three rats received 19.4 mg/kg of 3 (0.056 mmol/kg). The
plasma concentration of propofol was determined from whole
blood samples using thymol as an internal standard with
the method described previously in the literature (Yeganeh
and Ramzan, 1997). Only the amount of propofol, not the
amount of prodrug, was determined and compared between
the groups. The concentration curves of propofol after admin-
istration of 3 and propofol are presented in Fig. 4, in which each
line shows the values of individual rats. After administration
of prodrug or propofol, the pharmacokinetics revealed a fast
initial decline and a slower elimination phase and these were
best described by a two-compartmental model. The levels of
propofol in whole blood (␮g/mL) and the pharmacokinetic
parameters (Tmax, Cmax) were determined using the above
mentioned simulation and are shown in Table 2. The phar-
macokinetic parameters of GPI 15715 and propofol phosphate
are gathered from the publications of (Schywalsky et al., 2003)
and (Banaszczyk et al., 2002), respectively.
3.2.
Aqueous solubility
The aqueous solubility of propofol (130 2 ␮g/mL) was greatly
enhanced by the prodrug 3 since this compound has solubil-
ity over 10 mg/mL (Table 1). The maximum solubility of 3 was
not determined due to the small amount of prodrug available.
However, this limit was chosen since it has been estimated
that in general a drug needs to have a solubility of at least
10 mg/mL to be suitable for i.v. administration (Stella, 2006).
Due to the fast and complete dissolution and limited stability
of 3, the shaking time was reduced to 1 h from the conven-
tional 72 h.
3.3.
Stability in buffer solutions
Prodrug 3 was readily converted to propofol in each
rat after i.v. administration. The maximum plasma con-
centration (Cmax) of propofol, 3.0 0.2 ␮g/mL, was reached
within 2.1 0.8 min (Tmax) after the administration of 3.
Rats that received 3 were uncoordinated and had difficul-
ties in maintaining their balance. In contrast, the dose of
10 mg/kg of propofol was sufficient to evoke sedation which
The chemical hydrolysis rates of 3 and 5 were determined in
180 mM borate buffers at pH 7.4 at 37 ◦C and at room temper-
ature (r.t.). The chemical degradation of both 3 and 5 followed
first-order kinetics and the half-lives are presented in Table 1.
The stability of 3 in buffer solution (t1/2 = 5.2 0.2 days at r.t.
and t1/2 = 25.8 3.3 h at 37 ◦C, pH 7.4) was reasonably good if

european journal of pharmaceutical sciences 3 4 ( 2 0 0 8 ) 110–117
115
Table 2 – Estimated pharmacokinetic parameters (mean S.D.; n = 3) of propofol after single i.v. bolus dose of 3
(19.4 mg/kg, 0.056 mmol/kg) and propofol (10 mg/kg, 0.056 mmol/kg) into rats and referenced pharmacokinetic
parameters of GPI 15715 and Propofol Phosphate (PP) from the literature
Compound
3
Propofol
GPI 15715a
PPb
Dose
Cmax (␮g/mL)
Tmax (min)
19.4 mg/kg (0.056 mmol/kg)
3.0 0.2
2.1 0.8
10 mg/kg (0.056 mmol/kg)
14.3 1.8
40 mg/kg (0.120 mmol/kg)
7.1 1.6
130 mg/kg (0.499 mmol/kg)
9.6 2.1
c
–
3.7 0.2
7.3 3.8
Cmax = maximum concentration of propofol. Tmax = time to maximum concentration of propofol.
a
From Schywalsky et al. (2003).
From Banaszczyk et al. (2002).
After i.v. bolus Tmax = 0 min.
b
c
and more effective, when compared to PP (Banaszczyk et al.,
2002) (Tmax of 7.3 3.8 min and a Cmax of 9.6 2.1 ␮g/mL after
i.v. administration of PP with a dose as great as 130 mg/kg
(0.499 mmol/kg)).
4.
Discussion
Ethyl dioxy phosphates are a promising strategy for the devel-
opment of novel phosphate prodrugs with fast bioconversion
and lack of the formaldehyde release. Ethyl dioxy phosphate of
propofol (3) greatly enhanced the aqueous solubility of propo-
fol, the solubility being over 10 mg/mL and sufficient for i.v.
administration. Moreover, 3 was bioconverted to propofol as
efficiently as the phosphonooxymethyl prodrug of propofol (5,
GPI 15715), the latter being synthesized for the purpose of com-
parison of chemical stability of these two prodrugs. The faster
chemical hydrolysis of 3 compared to that of 5 is probably due
to chemical structure of the ethyl dioxy acetal. It is known
that acetals with a good leaving group (such as phenol) and
a neighbouring carboxylic acid group undergo intramolecu-
lar facilitated hydrolysis (Anderson and Fife, 1973; Fife et al.,
1996). Therefore it can be postulated that the acidic groups of
phosphate facilitate the hydrolysis of 3 by forming an oxocar-
bonium ion, which is internally greatly resonance-stabilized,
leading to the breaking of the C O bond and the release of
propofol (Fig. 5). Nonetheless, the chemical stability of 3 is ade-
quate for i.v. use, though it may require the reconstitution of
the preparation prior to use.
Fig. 4 – Propofol levels after single i.v. bolus dose of (A) 3
(19.4 mg/kg) and (B) propofol (10 mg/kg) in rats (n = 3). Each
line shows the values of one rat.
occurred at approximately 13 s after the i.v. bolus. When
compared to GPI 15715, the maximum concentration of
propofol, 3.0 0.2 ␮g/mL after administration of 3 (19.4 mg/kg,
0.056 mmol/kg), is approximately half of the concentration
achieved after administration of GPI 15715 with over dou-
ble the dose (7.1 1.6 ␮g/mL after 40 mg/kg, 0.120 mmol/kg)
(Schywalsky et al., 2003). The maximum concentration of
propofol was also reached at approximately in the same time
(2.1 0.8 min and 3.7 0.2 min for 3 and GPI 15715, respec-
tively). The bioconversion of 3 to propofol was much faster
The fast enzymatic hydrolysis of 3 suggests that ethyl
dioxy linked prodrugs act in a similar manner as conventional
phosphate and phosphonooxymethyl prodrugs. The differ-
ence between the concentrations of propofol after equimolar
administration of 3 and propofol is similar to the results
reported for GPI 15715 (Schywalsky et al., 2003) and is probably
due to formulation-dependent pharmacokinetics of propofol.
Fig. 5 – The chemical hydrolysis of the ethyl dioxy phosphate prodrug of propofol (3).

116
european journal of pharmaceutical sciences 3 4 ( 2 0 0 8 ) 110–117
A lipid-free formulation of propofol has a greater volume of
distribution and a greater elimination clearance, but a simi-
lar terminal half-life and larger distribution in the lungs when
compared to the oil-in-water formulation (Dutta and Ebling,
1998b). The slow release of propofol from the lungs leads to
decreased concentrations in the circulation (Dutta and Ebling,
1998a). 3, like GPI 15715 (Schywalsky et al., 2003), probably acts
like the lipid-free formulation, releasing the lower concentra-
tions of propofol to the circulation than is seen after equimolar
administration of propofol. 3 and GPI 15715 (Schywalsky et al.,
2003) have approximately the same Tmax and proportionate
Cmax. We therefore conclude that the bioconversion of 3 to
propofol is as effective as that of GPI 15715 and much faster
and more effective, when compared to PP (Banaszczyk et al.,
2002).
The purpose of this study was to confirm the bioconver-
sion of 3 to propofol in vivo, not to evoke complete sedation
in the rats. The sedated behaviour of rats demonstrated that
propofol had been released from 3, but larger doses would be
needed to achieve complete sedation. This effect is similar
to that reported earlier in case of GPI 15715 (Schywalsky et
al., 2003). Even though the pharmacokinetic parameters are
not truly comparable between different studies due to the
dose-dependent onset of the action and recovery of propo-
fol from prodrugs, this study indicates that 3 is bioconverted
to propofol in a similar manner as the phosphonooxymethyl
prodrug GPI 15715. In this study, no evidence was observed
of any abnormal pharmacokinetics that could be explained
by chirality. However, since enantiomers often have differ-
ent pharmaceutical and pharmacokinetic profiles (Hutt and
Tan, 1996) and also differences in binding to serum albu-
min (Chuang and Otagiri, 2006), in the future it might be
worthwhile to separate the enantiomers and to evaluate their
individual pharmaceutical and pharmacokinetic behaviours.
This is not a unique problem of 3, as most of the (acyloxy) alkyl
prodrugs are marketed as racemic mixtures.
In a conclusion, a novel ethyl dioxy phosphate prodrug of
propofol (3) has been shown to increase the water-solubility of
the poorly water-soluble drug propofol. The chemical stability
of 3 is sufficient for i.v. administration, but is limited due to the
presence of the more labile ethyl dioxy acetal. The enzymatic
release of propofol from 3 is rapid both in vitro with alkaline
phosphatase solution and in vivo in rats. The bioconversion
of 3 to propofol is similar to that of GPI 15715, the phos-
phonooxymethyl prodrug of propofol, and much faster when
compared to propofol phosphate. The results provide clear
evidence that the ethyl dioxy linker is suitable for preparing
phosphate prodrugs designed for i.v. administration by com-
bining the properties of previously described phosphate and
phosphonooxymethyl prodrugs (i.e. efficient bioconversion to
propofol and lack of systemic release of formaldehyde). Fur-
ther studies with different model drugs containing hydroxyl
and amine functionalities are ongoing in order to explore the
overall versatility of this novel phosphate prodrug structure.
Tiina Koivunen, Mrs. Helly Rissanen and Mrs. Anne Riekkinen
for their skillful technical assistance. The Graduate School of
Bioorganic and Medicinal Chemistry, the National Technology
Agency of Finland, the Academy of Finland (grants 105590 JR;
110277 TN) have provided financial support for this study.
r e f e r e n c e s
Anderson, E., Fife, T.H., 1973. Carboxyl group participation in
acetal hydrolysis. Hydrolysis of disalicyl acetals. J. Am. Chem.
Soc. 95, 6437–6441.
Baker, M.T., Naguib, M., 2005. Propofol: the challenges of
formulation. Anesthesiology 103, 860–876.
Banaszczyk, M.G., Carlo, A.T., Millan, V., Lindsey, A., Moss, R.,
Carlo, D.J., Hendler, S.S., 2002. Propofol phosphate, a
water-soluble propofol prodrug: In vivo evaluation. Anesth.
Analg. 95, 1285–1292.
Beaumont, K., Webster, R., Gardner, I., Dack, K., 2003. Design of
ester prodrugs to enhance oral absorption of poorly
permeable compounds: challenges to the discovery scientist.
Curr. Drug Metab. 4, 461–485.
Blouin, M., Frenette, R., 2001. A new method for the preparation
of aryl vinyl ethers. J. Org. Chem. 66, 9043–9045.
Bonneville, G., Walz, A.J., Delahanty, G. (2003). Process for
preparing water-soluble phosphonooxymethyl derivatives of
alcohol and phenol. WO03059255, 24 July, 2003.
Chuang, V.T., Otagiri, M., 2006. Stereoselective binding of human
serum albumin. Chirality 18, 159–166.
Dutta, S., Ebling, W.F., 1998a. Formulation-dependent brain and
lung distribution kinetics of propofol in rats. Anesthesiology
89, 678–685.
Dutta, S., Ebling, W.F., 1998b. Formulation-dependent
pharmacokinetics and pharmacodynamics of propofol in rats.
J. Pharm. Pharmacol. 50, 37–42.
Easthope, S.E., Jarvis, B., 2002. Candesartan cilexetil: an update of
its use in essential hypertension. Drugs 62, 1253–1287.
Fife, T.H., Bembi, R., Natarajan, R., 1996. Neighbouring carboxyl
group participation in the hydrolysis of acetals. Hydrolysis of
o-carboxybenzaldehyde cis-and trans-1,2-cyclohexanediyl
acetals. J. Am. Chem. Soc. 118, 12956–12963.
Gibiansky, E., Struys, M.M., Gibiansky, L., Vanluchene, A.L.,
Vornov, J., Mortier, E.P., Burak, E., Van Bortel, L., 2005. Aquavan
injection, a water-soluble prodrug of propofol, as a bolus
injection: a phase i dose-escalation comparison with diprivan
(part 1): Pharmacokinetics. Anesthesiology 103, 718–729.
Heck, H.D., Casanova, M., Starr, T.B., 1990. Formaldehyde
toxicity–new understanding. Crit. Rev. Toxicol. 20, 397–426.
Hutt, A.J., Tan, S.C., 1996. Drug chirality and its clinical
significance. Drugs 52 (Suppl. 5), 1–12.
Josefsson, K., Bergan, T., Magni, L., Pring, B.G., Westerlund, D.,
1982. Pharmacokinetics of bacmecillinam and pivmecillinam
in volunteers. Eur. J. Clin. Pharmacol. 23, 249–252.
Kumpulainen, H., Jarvinen, T., Saari, R., Lehtonen, M.,
Vepsalainen, J., 2005. An efficient strategy for the synthesis of
1-chloroethyl phosphates and phosphoramidates. J. Org.
Chem. 70, 9056–9058.
Kumpulainen, H., Saari, R., Lehtonen, M., Rautio, J., Ja¨rvinen, T.,
Vepsa¨la¨inen, J., 2006. Convenient microwave-assisted
synthesis of 1-chloroethyl phosphates. Tetrahedron Lett. 47,
2003–2004.
Safadi, M., Oliyai, R., Stella, V.J., 1993. Phosphoryloxymethyl
carbamates and carbonates–novel water-soluble prodrugs for
amines and hindered alcohols. Pharm. Res. 10, 1350–
1355.
Schywalsky, M., Ihmsen, H., Tzabazis, A., Fechner, J., Burak, E.,
Vornov, J., Schwilden, H., 2003. Pharmacokinetics and
Acknowledgements
We thank Prof. Valentino J. Stella for valuable scientific discus-
sion concerning this publication and Mrs. Miia Reponen, Mrs.

european journal of pharmaceutical sciences 3 4 ( 2 0 0 8 ) 110–117
117
pharmacodynamics of the new propofol prodrug gpi 15715 in
a water-soluble prodrug of propofol, as a bolus injection: a
rats. Eur. J. Anaesthesiol. 20, 182–190.
phase i dose-escalation comparison with diprivan (part 2):
pharmacodynamics and safety. Anesthesiology 103,
730–743.
Scott, L.J., Ormrod, D., Goa, K.L., 2001. Cefuroxime axetil: an
updated review of its use in the management of bacterial
infections. Drugs 61, 1455–1500.
Seitz, H.K., Maurer, B., Stickel, F., 2005. Alcohol consumption and
cancer of the gastrointestinal tract. Dig. Dis. 23, 297–303.
Stella, V.J., 2006. Prodrugs strategies for improving drug-like
properties. In: Borchardt, R.T., Kerns, E.H., Hageman, M.J.,
Thakker, D.R., Stevens, J.L. (Eds.), Optimizing the “drug-like”
Properties of Leads in Drug Discovery, first ed. Springer
Science+Business Media, New York, pp. 221–237.
Varia, S.A., Schuller, S., Stella, V.J., 1984. Phenytoin prodrugs iv:
hydrolysis of various 3-(hydroxymethyl)phenytoin esters. J.
Pharm. Sci. 73, 1074–1080.
Waynforth, H.B., Fleckell, P.A., 1994. Experimental and Surgical
Technique in the Rat, second ed. Academic Press limited,
London.
Yeganeh, M.H., Ramzan, I., 1997. Determination of propofol in rat
whole blood and plasma by high-performance liquid
chromatography. J. Chromatogr. B Biomed. Sci. Appl. 691,
478–482.
Struys, M.M., Vanluchene, A.L., Gibiansky, E., Gibiansky, L.,
Vornov, J., Mortier, E.P., Van Bortel, L., 2005. Aquavan injection,