Piperazinylalkyl prodrugs of naproxen improve in vitro skin permeation

Article abstract of DOI:10.1016/S0928-0987(00)00090-7

Novel morpholinyl (4a) and piperazinylalkyl (4b-e) esters were synthesized and evaluated in vitro for their properties as bioreversible topically administered dermal prodrugs of naproxen. These ionizable prodrugs exhibited various aqueous solubilities and lipophilicities, depending on the pH of medium. As indicated by octanol-buffer partition coefficients (logP(app)) at pH 7.4, all of the prodrugs were significantly more lipophilic (logP(app)=0.7-3.9) than naproxen (logP(app)=0.3). Furthermore, the most aqueous of the soluble prodrugs (4b-d) were only 2-3-fold less soluble in an aqueous buffer of pH 7.4 (~30-50 mM) than was naproxen (~100 mM). At a pH of 5.0, prodrugs showed a generally higher aqueous solubility and similar logP(app) values, compared to naproxen. The chemical and enzymatic hydrolysis of prodrugs at 37°C was investigated in aqueous buffer solutions (pH 5.0 and 7.4) and in 80% human serum (pH 7.4), respectively. The prodrugs showed moderate chemical stability (t(1/2)=15-150 days at pH 5.0), and they were hydrolyzed enzymatically to naproxen, with half-lives ranging from 0.4 to 77 min. In permeation studies using post-mortem human skin in vitro, the flux of naproxen was 6.5 and 1.6 nmol/cm².h in a saturated aqueous buffer vehicle of pH 7.4 and 5.0, respectively. Among the prodrugs, two piperazinyl derivatives (4c and 4d) resulted in a 9- and 4-fold enhancement of permeation, respectively, when compared to naproxen itself at pH 7.4. 4c also resulted in a significantly (4-fold) better permeation than naproxen at pH 5.0. In conclusion, piperazinyl esters improved skin permeation of naproxen and are promising prodrugs of naproxen for topical drug delivery. Copyright (C) 2000 Elsevier Science B.V.

Full text of DOI:10.1016/S0928-0987(00)00090-7

European Journal of Pharmaceutical Sciences 11 (2000) 157–163
www.elsevier.nl/locate/ejps
Piperazinylalkyl prodrugs of naproxen improve in vitro skin permeation
a
a
b
a
Jarkko Rautioa , , Tapio Nevalainen , Hannu Taipale , Jouko Vepsalainen , Jukka Gynther ,
*
¨ ¨
a
a
¨
Krista Laine , Tomi Jarvinen
aDepartment of Pharmaceutical Chemistry, University of Kuopio, P.O. Box 1627, FIN-70211 Kuopio, Finland
bDepartment of Chemistry, University of Kuopio, P.O. Box 1627, FIN-70211 Kuopio, Finland
Received 13 January 2000; received in revised form 22 March 2000; accepted 24 March 2000
Abstract
Novel morpholinyl (4a) and piperazinylalkyl (4b-e) esters were synthesized and evaluated in vitro for their properties as bioreversible
topically administered dermal prodrugs of naproxen. These ionizable prodrugs exhibited various aqueous solubilities and lipophilicities,
depending on the pH of medium. As indicated by octanol–buffer partition coefficients (log Papp) at pH 7.4, all of the prodrugs were
significantly more lipophilic (log Papp50.7–3.9) than naproxen (log Papp50.3). Furthermore, the most aqueous of the soluble prodrugs
(4b–d) were only 2–3-fold less soluble in an aqueous buffer of pH 7.4 (|30–50 mM) than was naproxen (|100 mM). At a pH of 5.0,
prodrugs showed a generally higher aqueous solubility and similar log Papp values, compared to naproxen. The chemical and enzymatic
hydrolysis of prodrugs at 378C was investigated in aqueous buffer solutions (pH 5.0 and 7.4) and in 80% human serum (pH 7.4),
respectively. The prodrugs showed moderate chemical stability (t1 / 2515–150 days at pH 5.0), and they were hydrolyzed enzymatically to
naproxen, with half-lives ranging from 0.4 to 77 min. In permeation studies using post-mortem human skin in vitro, the flux of naproxen
was 6.5 and 1.6 nmol/cm2?h in a saturated aqueous buffer vehicle of pH 7.4 and 5.0, respectively. Among the prodrugs, two piperazinyl
derivatives (4c and 4d) resulted in a 9- and 4-fold enhancement of permeation, respectively, when compared to naproxen itself at pH 7.4.
4c also resulted in a significantly (4-fold) better permeation than naproxen at pH 5.0. In conclusion, piperazinyl esters improved skin
permeation of naproxen and are promising prodrugs of naproxen for topical drug delivery.
reserved.
2000 Elsevier Science B.V. All rights
Keywords: Naproxen; Prodrug; Solubility; Lipophilicity; Hydrolysis kinetics; Skin permeation
1. Introduction
of the parent drug and thereby enhancing its skin permea-
tion, efficacy and therapeutic value (Chan and Li Wan Po,
1989; Sloan, 1989, 1992).
Dermal (i.e., topical) drug delivery has been gaining
increasing popularity since it offers some advantages over
more conventional treatments. Topical therapy can, for
example, deliver therapeutic levels of drug to tissue close
to the site of application in a safer and more effective
manner than those obtainable with oral delivery (Guy and
Maibach, 1983; McNeill et al., 1992; Suh et al., 1997;
Mikulak et al., 1998). However, the dermal drug transport
is greatly limited by the unsuitable physicochemical prop-
erties of most drugs and the efficient barrier function of the
skin, and is frequently insufficient for medical uses. Thus,
many attempts of improving topical absorption have been
performed, such as the prodrug approach with the aim of
changing pharmaceutical and/or pharmacokinetic character
Naproxen is a potent nonsteroidal anti-inflammatory
drug (NSAID) that is widely used for the treatment of
rheumatic diseases and related painful conditions (Todd
and Clissold, 1990). Unfortunately, following topical
administration in man naproxen shows bioavailability of
only 1–2% (Yano et al., 1986; van den Ouweland et al.,
1989; Singh and Roberts, 1994; Suh et al., 1997) which
may be attributed to ionizable carboxylic acid group.
Therefore, various prodrugs have been studied in attempt
to increase the dermal permeation of naproxen by tran-
siently masking its ionized group (Mueller, 1990; Kasting
et al., 1992; Bonina et al., 1993; Weber et al., 1994).
In our previous studies, we synthesized and evaluated a
series of acyloxyalkyl and aminoacyloxyalkyl esters of
naproxen, which readily hydrolyzed to naproxen in vitro
and possessed highly variable aqueous solubilities and
lipophilicities (Rautio et al., 1998, 1999). The highly
*Corresponding author. Tel.: 1358-17-163-445; fax.: 1358-17-162-
456.
E-mail address: jarkko.rautio@uku.fi (J. Rautio)
0928-0987/00/$ – see front matter
PII: S0928-0987(00)00090-7
2000 Elsevier Science B.V. All rights reserved.

158
J. Rautio et al. / European Journal of Pharmaceutical Sciences 11 (2000) 157–163
lipophilic acyloxyalkyl esters did not enhance dermal
permeation of naproxen, probably due to their very low
aqueous solubilities (Rautio et al., 1998). The amino-
acyloxyalkyl esters (amino acids as a promoiety) of
naproxen had combination of adequate aqueous solubility
and lipophilicity and resulted in a 3-fold increase in the in
vitro skin permeation of naproxen (Rautio et al., 1999).
Based on these studies, and the recent work of others
(Milosovich et al., 1989; Tammara et al., 1993, 1994; Pop
et al., 1996), we synthesized and evaluated a series of
morpholinyl and piperazinyl esters as potential prodrugs of
naproxen for topical drug delivery.
CH2OCH2), 2.23 (4H, m, CH2NCH2), 2.21 (2H, m,
CH2N), 1.58 (2H, m, OCH3CH2), 1.57 (3H, d, CCH3,
J57.2 Hz), 1.39 (2H, m, CH2CH2N), HRMS: m/z
371.2145, calcd for C22H29NO4 371.2096.
2.2.2. 2-(1-Piperazinyl)ethyl-2-(6-methoxy-2-
naphthyl)propanoate (4b)
A
solution of 2-bromoethyl 2-(6-methoxy-2-naph-
thyl)propanoate (2.0 g, 8.7 mmol) and piperazine (1.87 g,
21.7 mmol) in dry toluene (50 ml) was refluxed for 24 h.
The solution was washed with water and dried with
MgSO4, and evaporated to yield an oil. The residue was
purified by flash chromatography on silica gel, eluting with
MeOH to yield 4b as an oil (0.96 g, 32%): 1H NMR
(CDCl3, 400 MHz) d 7.72–7.10 (6H, m, aromatic), 4.20
(2H, dt, OCH2CH2, J55.8 Hz, J53.0 Hz), 3.91 (3H, s,
CH3O) 3.86 (1H, q, J57.1 Hz, CHMe), 2.72 (4H, t,
CH2NCH2, J54.9 Hz), 2.54 (2H, t, CH2N, J55.8 Hz),
2.31 (4H, m, CH2NCH2), 1.58 (3H, d, J57.0 Hz, CH3C),
HRMS: m/z 342.2020, calcd for C20H26N2O3 342.1943.
2. Materials and methods
2.1. Equipment
1H and 13C NMR spectra were recorded on a Bruker
AM 400 WB operating at 400.1 MHz. NMR data are
reported in parts per million (d ) and are referenced to TMS
as the internal standard. The splitting pattern abbreviations
are as follows: s, singlet; d, doublet; t, triplet; q, quartet; p,
pentet; m, multiplet; dt, double triplet; bs, broad singlet;
bm, broad multiplet. Electron impact (EI) mass spectra of
the prodrugs were determined by a VG 70-250SE magnetic
sector mass spectrometer (VG Analytical, Manchester,
UK). Flash chromatography was accomplished over silica
gel (30–60 mm, J.T. Baker 7024-02). Thin-layer chroma-
tography (TLC) analyses of reactions were run on alu-
minium foil plates coated with silica gel 60 F254 (Merck).
Naproxen ((1)-2-(6-methoxy-2-naphthyl)-propionic acid)
was kindly donated by Orion Pharma (Espoo, Finland). All
other reagents were obtained from commercial suppliers
and were used without further purification.
2.2.3. 2-(4-Methyl-1-piperazinyl)ethyl-2-(6-methoxy-2-
naphthyl)propanoate (4c)
A
solution of 2-bromoethyl 2-(6-methoxy-2-naph-
thyl)propanoate (2.0 g, 8.7 mmol) and methylpiperazine
(2.18 g, 21.7 mmol) in dry toluene (50 ml) was refluxed
for 20 h. The solution was washed with water and dried
with MgSO4. The solvent was evaporated, and the residue
was purified by flash chromatography on silica gel, eluting
with EtOAc–MeOH (1:1) to afford 4c as an oil (1.65 g,
1
53%): H NMR (CDCl3, 400 MHz) d 7.72–7.10 (6H, m,
aromatic), 4.20 (2H, dt, OCH2CH2, J55.7 Hz, J51.6 Hz),
3.91 (3H, s, CH3O) 3.86 (1H, q, J57.2 Hz, CHMe), 2.55
(2H, t, CH2N, J55.8 Hz 2.38 (4H, bm, CH2NCH2,), 2.28
(4H, bm, CH2NCH2), 2.20 (3H, s, NCH3), 1.58 (3H, d,
J57.2 Hz, CH3C), HRMS: m/z 356.2162, calcd for
C21H28N2O3 356.2100.
2.2. Synthesis of prodrugs
2.2.1. 4-(4-Morpholinyl)butyl-2-(6-methoxy-2-
naphthyl)propanoate (4a)
2.2.4. 4-(4-Methyl-1-piperazinyl)butyl-2-(6-methoxy-2-
naphthyl)propanoate (4d)
2-(6-Methoxy-2-naphthyl)propionic acid chloride (0.31
g, 1.2 mmol) was dissolved in tetrahydrofuran (30 ml) and
treated with 2-morpholinyl-4-butanol (0.37 g, 2.3 mmol) in
a dropwise fashion, and the mixture was stirred for 2 days
at room temperature. The precipitated 2-morpholinyl-4-
butanol hydrochloride was filtered off and the filtrate was
evaporated. The oily product was dissolved in dichlorome-
thane (50 ml) and washed with 5% sodium bicarbonate
(3350 ml) and water (2350 ml). The organic layer was
dried over anhydrous calcium sulfate (CaSO4) and evapo-
rated. The resulting residue was purified by flash silica gel
column chromatography, eluting with 5% MeOH in
4d was prepared as described for 4a from 2-(6-methoxy-
2-naphthyl)propionic acid chloride (0.32 g, 1.3 mmol) and
4-(4-methyl-1-piperazinyl)-butanol (0.386 g, 2.2 mmol).
Flash chromatography (10% MeOH in CH2Cl2) proceeded
1
4d as a viscous oil (0.16 g, 32%); H NMR (CDCl3, 400
MHz) d 7.71–7.10 (6H, m, aromatic), 4.08 (2H, dt,
OCH2CH2, J55.9 Hz, J53.6 Hz), 3.91 (3H, s, CH3O),
3.84 (1H, q, J57.2 Hz, CHMe), 2.48–2.24 (8H, bm,
CH2NCH2), 2.28 (3H, s, NCH3), 2.25 (2H, m, CH2N),
1.57 (3H, d, J56.9 Hz, CH3C), 1.56 (2H, m, CH2CH2O),
1.42 (2H, m, CH2CH2N), HRMS: m/z 384.2473, calcd for
C23H32N2O3 384.2413.
1
CH2Cl2 affording 4a as a white solid (0.25 g, 55%): H
NMR (CDCl3, 400 MHz) d 7.71–7.10 (6H, m, aromatic),
4.09 (2H, dt, OCH2CH2, J56.5 Hz, J54.4 Hz), 3.91 (3H,
s, CH3O), 3.83 (1H, q, J57.2 Hz, CHMe), 3.61 (4H, m,
2.2.5. 6-(4-Methyl-1-piperazinyl)hexyl-2-(6-methoxy-2-
naphthyl)propanoate (4e)
To a solution of 1 (0.78 g, 3.4 mmol), 6-(4-methyl-1-

J. Rautio et al. / European Journal of Pharmaceutical Sciences 11 (2000) 157–163
159
piperazinyl)-hexanol (0.64 g, 3.2 mmol), and 4-dimethyl-
aminopyridine (DMAP) (40 mg, 0.2 mmol) in dry CH2Cl2
(20 ml), was added dicyclohexylcarbodiimide (DCC) (1.30
g, 6.3 mmol) and the mixture was stirred for 3 days at
room temperature. The precipitated dicyclohexylurea
(DCU) was filtered off and the filtrate was evaporated. The
resulting residue was purified by flash chromatography on
silica gel eluting with 3% MeOH in CH2Cl2 affording 4e
as a white solid (0.31 g, 23%): 1H NMR (CDCl3, 400
MHz) d 7.72–7.10 (6H, m, aromatic), 3.91 (3H, s, CH3O),
3.86 (1H, q, J57.2 Hz, CHMe), 3.20 (2H, m, OCH2CH2),
1.91 (2H, m, CH2N), 1.80–1.50 (8H, bm, CH2NCH2),
1.68 (3H, s, NCH3), 1.58 (3H, d, J57.3 Hz, CH3C),
1.37–1.15 (8H, m, (CH2)4), HRMS: the molecular peak of
4e was not obtained.
Purospher RP-C18 column (12534 mm, 5 mm) was used.
A mobile phase mixture of acetonitrile and a 0.02 M
phosphate buffer solution of pH 5.0–5.5 (depending on
compound) at a flow rate of 1.2 ml/min were used.
2.4. Aqueous solubility
The aqueous solubility of naproxen and its prodrugs was
determined in phosphate buffer (0.16 M) at both pH 5.0
and 7.4 at room temperature. Excess amounts of each
compound were added to 1–1.5 ml of buffer, the mixtures
were stirred for either 60 min (pH 5.0) or for 30 min (pH
7.4), centrifuged at 14 000 rpm, and filtered (Millipore
0.45 mm). The concentrations of the compounds in their
saturated solutions were analyzed by the HPLC. The pH of
the mixtures were held constant throughout the experi-
ment.
2.2.6. 4-(4-Methyl-1-piperazinyl)-butanol
Bromobutyl acetate (4.0 g, 20.5 mmol) in 5 ml of
benzene was added dropwise to
a
solution of
2.5. Apparent partition coefficients
methylpiperazine (4.1 g, 41 mmol) in 5 ml of benzene and
the solution was refluxed for 2 h. After cooling,
methylpiperazine hydrobromide was filtered off and the
filtrate was evaporated. Flash chromatography (MeOH)
proceeded 4-(4-methyl-1-piperazinyl)butyl acetate (2.3 g,
10.7 mmol) as a viscous oil, which was refluxed with 70
ml 2 M NaOH for 24 h. To a cooled solution, 5 g NaCl
was added, and a solution was extracted (1:2 ethanol/
chloroform, 3360 ml), dried, and evaporated to yield
4-(1-hydroxybutyl)piperazine (1.6 g, 87%).
The apparent partition coefficients (log Papp) of naprox-
en and its prodrugs were determined at room temperature
between 1-octanol and phosphate buffer (0.16 M) at both
pH 5.0 and 7.4. Before use, the 1-octanol was saturated
with phosphate buffer for 24 h by stirring vigorously. A
known concentration of compound in phosphate buffer was
shaken for either 30 min (pH 5.0) or 15 min (pH 7.4), with
a suitable volume of 1-octanol. After shaking, the phases
were separated by centrifugation at 14 000 rpm for 4 min.
The concentrations of the compounds in the buffer phase
before and after partitioning were determined by the
HPLC.
This method was used to prepare 2-morpholinyl-4-
butanol and 6-(4-methyl-1-piperazinyl)-hexanol.
2.2.7. 2-Bromoethyl 2-(6-methoxy-2-naphthyl)propanoate
(3)
2.6. Hydrolysis in aqueous solution
A mixture of 1 (5.00 g, 21.4 mmol), 2-bromoethanol
(3.25 g, 26.0 mmol), DMAP (0.27 g, 2.2 mmol), and DCC
(2.48 g, 26 mmol) in dry CH2Cl2 (80 ml) was was stirred
for 26 h at room temperature. The precipitated DCU was
filtered off and the filtrate was evaporated. The crude
material was purified by flash chromatography on silica gel
eluting with methanol–ethyl acetate 1:1 affording 3 as a
The rates of chemical hydrolysis of prodrugs were
studied in aqueous phosphate buffer solutions of pH 7.4
and pH 5.0 (0.16 M, ionic strength 0.5) at 378C. An
appropriate amount of prodrug was dissolved in 10 ml of
preheated buffer and the solutions were placed in a
thermostatically controlled water bath at 378C. At appro-
priate intervals, samples were taken and analyzed for
remaining prodrug by the HPLC. Pseudo-first-order half-
time (t1 / 2) for the hydrolysis of prodrug was calculated
from the slope of the linear portion of the plotted logarithm
of remaining prodrugs versus time.
1
white solid (3.93 g, 73%): H NMR (CDCl3, 400 MHz) d
7.72–7.10 (6H, m, aromatic), 4.38 (2H, m, OCH2CH2),
3.91 (3H, s, CH3O), 3.90 (1H, q, J57.0 Hz, CHMe), 3.45
(2H, dt, J56.1 Hz, J51.4 Hz, CH2Br), 1.60 (3H, d, J57.0
Hz, CH3C).
2.3. HPLC analysis
2.7. Hydrolysis in human serum
The analytical high-pressure liquid chromatography
(HPLC) system consisted of a Merck Hitachi L-6200A
intelligent pump, Hewlett Packard HP1046A program-
mable fluorescence detector (excitation, 226 nm; emission,
368 nm), a Merck Hitachi D-6000A interface module, a
Merck Hitachi AS-2000 autosampler, and a Merck LaCh-
rom column oven L-7350. For all sample separations a
The rates of enzymatic hydrolysis for naproxen prodrugs
were studied in human serum at 378C (Institute of Public
Health, University of Kuopio) which was diluted to 80%
with 0.16 M phosphate buffer, pH 7.4. The reactions were
initiated by dissolving an appropriate amount of prodrug in
phosphate buffer, and preheated human serum was added.
The solutions were kept in a water bath at 378C, and 0.5

160
J. Rautio et al. / European Journal of Pharmaceutical Sciences 11 (2000) 157–163
ml of serum/buffer mixture were withdrawn and added to
1.0 ml of ethanol to precipitate protein from the serum.
After immediate mixing and centrifugation for 10 min at
14 000 rpm, the supernatant was analyzed for remaining
prodrug and released naproxen by the HPLC. Pseudo-first-
order half-times (t1 / 2) for the hydrolysis of prodrug was
calculated from the slope of the linear portion of the
plotted logarithm of remaining prodrugs against time.
where Kp is the permeability coefficient, Jss is the steady-
state skin flux and C is the saturation solubility of the
compound in the used vehicle (phosphate buffer of pH 5.0
or 7.4).
2.9. Statistical analysis
A one-factor analysis of variance (ANOVA factorial)
was used to test the statistical significance of differences
between naproxen and prodrugs. Significance in the differ-
ences in the means was tested using Fisher’s protected
least significance difference (PLSD) at 95% confidence.
2.8. In vitro skin permeation study
Samples of human skin were obtained from the abdomi-
nal region of adult human cadavers from the Kuopio
University Hospital (Kuopio, Finland). The epidermis was
isolated from the underlying dermis by heat separation at
608C in distilled water for 2 min, after which the skin
specimens were dried and frozen prior to use. The permea-
tion studies were carried out using the Franz-type diffusion
cell (PermeGear, Riegel, PA, USA) a previously described
(Rautio et al., 1998). Skin specimens were rehydrated
before being mounted in the diffusion cell. The receptor
medium (0.05 M isotonic phosphate buffer solution of pH
7.4) was stirred and kept at 378C throughout the study. The
compounds were applied as suspensions in 0.05 M phos-
phate buffer of both pH 5.0 and 7.4. At specified time
intervals 250-ml aliquots were withdrawn from the receptor
compartment and replaced with fresh buffer. The drug
concentrations were assayed by HPLC. The steady-state
flux for naproxen and its prodrug (4a–e) was determined
by plotting the cumulative amount (in nmol) of the parent
drug and intact prodrug as measured in the receptor phase
against time, and dividing the slope of the steady-state
position by the surface area of the diffusion cell (0.71
cm2). The permeability coefficients of naproxen and
selected prodrugs were calculated using Eq. (1):
3. Results and discussion
3.1. Synthesis of morpholinyl and piperazinylalkyl esters
of naproxen
The synthesis of morpholinyl (4a) and piperazinylalkyl
(4b–e) prodrugs of naproxen is illustrated in Scheme 1.
The esters 4a and 4d were prepared from naproxen acid
chloride (2) (Franssen et al., 1992) according to the
procedure reported by Tammara et al. (1993). The esters
4b and 4c were prepared from naproxen 2-bromoethyl
ester (3) and, the appropriate piperazine, and 4e was
prepared by coupling naproxen (1) with 4-(4-methyl-1-
piperazinyl)-hexanol
using
dicyclohexylcarbodiimide
(DCC). The purities of the prodrug substances were
determined by HPLC and NMR and were .98% (mol%)
for each prodrug.
3.2. Aqueous solubility and lipophilicity
Physicochemical parameters, such as aqueous solubility
and lipophilicity, have been shown to influence membrane
flux, therapeutic activity, and pharmacokinetic profiles of
Kp 5 Jss /C
(1)
1.

J. Rautio et al. / European Journal of Pharmaceutical Sciences 11 (2000) 157–163
161
Table 2
medicines (Goosen et al., 1998). The aqueous solubilities
and lipophilicities (evaluated by drug partitioning between
1-octanol and phosphate buffer) of naproxen and its
prodrugs at pH 7.4 and 5.0 are shown in Table 1.
Except for the highly lipophilic prodrug 4e, the prodrugs
derived from cyclic amines showed moderate solubility
both in neutral (pH 7.4) and in acidic (pH 5.0) aqueous
solutions. Their aqueous solubilities decreased at pH 7.4
but increased markedly at pH 5.0 compared to those of
naproxen. All of the prodrugs were also more lipophilic
than naproxen at pH 7.4, as illustrated by their log Papp
values, but they also maintained a lipophilicity comparable
to naproxen at pH 5.0. Lipophilicity is very important for
dermal permeation because the stratum corneum, the major
barrier to drug permeation, is lipid in nature and generally
favors permeation by lipophilic drugs. However, it has also
been reported that an effective dermal prodrug should
possess not only high lipophilicity, but also adequate
aqueous solubility (Chang et al., 1997; Kerr et al., 1998;
Taylor and Sloan, 1998; Rautio et al., 1999). From the
aspect of solubility properties, prodrugs 4c and 4d seem
the most promising naproxen prodrugs for topical drug
delivery.
Hydrolysis rates of naproxen prodrugs in phosphate buffer solutions (pH
7.4 and 5.0), and 80% human serum (pH 7.4) at 378C
Compound
t1 / 2 (days)
phosphate
buffer pH 7.4
t1 / 2 (days)
phosphate
buffer pH 5.0
t1 / 2 (min)
80% human
serum
4a
4b
4c
4d
4e
10.4
2.4
1.1
14.2
29.9
157.0
14.8
27.8
26.4
2463
0.460.06
461
3863
7762
a
–
a Not determined due to high chemical stability.
faster than in aqueous solutions having half-lives ranging
from 0.4 to 77 min, which indicates enzymatic hydrolysis.
3.4. In vitro skin permeation study
In skin permeation studies, excised post-mortem human
skin was used in a Franz-type diffusion cell to examine the
permeation characteristics of naproxen and the prodrugs
4a–e. Prodrug suspensions were applied as phosphate
buffers to maintain constant diffusion and maximum flux,
and each compound was applied in the isotonic phosphate
buffer (0.05 M) at both pH 7.4 and 5.0
3.3. Chemical and enzymatic hydrolyses
The diffusion experiments showed that both naproxen
and its prodrugs are able to permeate human skin in vitro.
Fig. 2 represents the cumulative amounts of permeated
naproxen (in nmol) or intact prodrug measured in the
receptor compartment, divided by the surface area of the
diffusion cell and plotted against the time of sampling. The
steady-state flux (Jss) was calculated from the slope of the
linear portions of these plots and are presented in Table 3.
The permeability coefficients (Kp) for the steady-state
delivery were obtained by dividing the steady-state flux by
Chemical hydrolysis of prodrugs (4a–e) in aqueous
buffer solutions, and enzymatic hydrolysis of the com-
pounds in 80% human serum was investigated. Human
serum or plasma has commonly been employed for testing
the susceptibility of the skin esterases capable of hydrolyz-
ing the ester bond of prodrugs (Bundgaard et al., 1989;
Kasting et al., 1992; Roy and Manoukian, 1994). The
kinetics of the chemical and enzymatic hydrolyses of each
prodrug followed pseudo-first-order kinetics over several
half-lives (Table 2).
Prodrugs showed a substantially higher chemical stabili-
ty in an aqueous buffer solution of pH 5.0 than pH 7.4,
half-lives ranging from 15 to 157 days and from 1 to 30
days, respectively. In human serum, each prodrug hydro-
lyzed quantitatively to naproxen (Fig. 1) significantly
Table 1
Aqueous solubility (mean6S.D.; n52–4) and apparent partition coeffi-
cient (log Papp, mean6S.D.; n52–3) of naproxen and prodrugs
a
Compound
Aqueous solubility (mM)
Log Papp
pH 7.4
pH 5.0
pH 7.4
pH 5.0
Naproxen
101.961.3
0.260.0
30.367.1
32.661.8
51.862.6
0.00660.001
0.460.0
0.3060.03
2.6060.10
0.7460.02
2.2960.02
2.4460.09
3.9260.07
2.3860.02
0.9060.08
0.2060.12
0.3760.03
1.1360.09
3.8460.05
4a
4b
4c
4d
4e
17.760.5
18.262.6
61.066.8
54.661.0
0.00660.001
a Papp is the apparent partition coefficient between 1-octanol and
phosphate buffer (0.16 M) at room temperature.
Fig. 1. Time courses for naproxen prodrug (4e) (d) and naproxen (s)
during hydrolysis of the prodrug in 80% human serum (pH 7.4) at 378C.

162
J. Rautio et al. / European Journal of Pharmaceutical Sciences 11 (2000) 157–163
It should be noted that at pH 7.4, where the highest flux
values of both naproxen and prodrugs were achieved, a
significantly lower concentration of prodrugs than naprox-
en was used in the donor phase. Permeability coefficient
(Kp) is independent of donor concentration, unlike flux,
and serves as an useful parameter to evaluate the permea-
tion potential of a compound. The Kp value of all prodrugs
was enhanced by several orders of magnitude compared to
naproxen at pH 7.4. However, at pH 5.0, only prodrug 4e
exhibited a higher Kp value than naproxen itself, despite its
flux value being the lowest among the prodrugs. Because
permeability of the compounds from aqueous solutions
increases as the solubility of the compounds in these
solutions decreases, the high Kp value of 4e can be
explained by its very poor aqueous solubility and de-
creased concentration in the applied vehicle, compared to
the other compounds. Therefore, the flux, which measures
the mass of material transported through the skin, is more a
therapeutically relevant parameter than permeability coeffi-
cient.
Fig. 2. Permeation profiles (mean6S.E.M.; n53–12) for naproxen (h)
and prodrugs 4c (d), and 4d (m) through excised human skin in 0.05 M
phosphate buffer (pH 7.4), and for naproxen (s) in 0.05 M phosphate
buffer (pH 5.0) vehicle (in all cases except 4c the error bars are smaller
than the symbols). The data represents the sum of naproxen and its
prodrug.
the solubilities of the compounds in the applied corre-
sponding vehicle (Table 3).
In conclusion, the present study shows that the permea-
tion of naproxen through human skin can be significantly
improved by using piperazinylalkyl prodrugs of naproxen.
The good skin permeation observed for these prodrugs can
most likely be attributed to their simultaneously high
aqueous solubility and lipophilicity, the combination of
which is an important for permeation across the skin.
Based on flux and Kp values, the prodrug 4c seems the
most promising prodrug to improve the topical drug
delivery of naproxen.
The steady-state flux for naproxen applied as a suspen-
sion was considerably higher at pH 7.4 (6.560.6 nmol/
cm2?h) than at pH 5.0 (1.660.2 nmol/cm2?h), which is a
result of the higher aqueous solubility (more than 250-fold)
of naproxen at pH 7.4 compared to pH 5.0. The flux values
of each piperazine prodrug from an aqueous vehicle of pH
7.4 were also higher than those from the aqueous vehicle at
pH 5.0. Since the aqueous solubilities of prodrugs were
similar at both pH values, the improved permeation
resulted from their higher partition coefficients at pH 7.4.
At this pH, the prodrugs 4c and 4d exhibited the highest
flux values, which were 9- and 4-fold higher than that of
naproxen itself at pH 7.4, respectively. In addition, at pH
5.0 4c resulted in significantly (4-fold) better permeation
than naproxen at pH 5.0. The only tested morpholinylalkyl
prodrug (4a) resulted in a similar flux at both pH values
which were, however, lower (pH 7.4) or similar (pH 5.0)
than the flux of naproxen. This is most probably due to a
high partition coefficient (log Papp |3.9) associated with
the very poor aqueous solubility of 4a (6 mM).
Acknowledgements
Authors are grateful to thank Mrs. Helly Rissanen, Mrs.
Tarja Ihalainen and Mr. Jukka Knuutinen for expert help
during this study. This work was financially supported by
the Academy of Finland and Technology Development
Centre (Finland).
References
Table 3
Steady-state flux (Jss) and permeability coefficients (Kp) (mean6S.E.M.;
n53–12) for the delivery of total naproxen species through excised
human skin in vitro in isotonic phosphate buffer (0.05 M, pH 7.4 and 5.0)
at 378C
Bonina, F.P., Montenegro, L., Guerrera, F., 1993. Naproxen 1-
alkylazacycloalkan-2-one esters as dermal prodrugs: in vitro evalua-
tion. Int. J. Pharm. 100, 99–105.
Bundgaard, H., Mørk, N., Hoelgaard, A., 1989. Enhanced delivery of
nalidixic acid through human skin via acyloxymethyl ester prodrugs.
Int. J. Pharm. 55, 91–97.
Compound
Jss (nmol/cm2?h)
Kp 3106 (cm/h)
pH 7.4
pH 5.0
pH 7.4
pH 5.0
Chan, S.Y., Li Wan Po, A., 1989. Prodrugs for dermal delivery. Int. J.
Pharm. 55, 1–16.
Chang, S.Y., Park, S.B., Jung, J.H., Shon, S.I., Yoon, H.J., 1997.
Ibuprofen heterocyclic esters as dermal prodrugs in vitro evaluation.
STP Pharm. Sci. 7, 315–319.
Franssen, E.J.F., Koiter, J., Kuipers, C.A.M., Bruins, A.P., Moolenaar, F.,
de Zeeuw, D., Kruizinga, W.H., Kellogg, R.M., Meijer, D.K.F., 1992.
Low molecular weight proteins as carriers for renal drug targeting.
Preparation of drug-protein conjugates and drug-spacer derivatives and
Naproxen
6.560.6
1.560.4a
5.160.4
1.6 60.2
2.360.2
2.560.1
6.461.9a
0.860.2
0.360.1a
0.0660.01
6.5461.62a
0.1760.01a
1.7960.40a
0.5460.03a
59.8865.66a
4.0760.47
0.1360.01a
0.1460.01a
0.1060.03a
0.0160.00a
52.7869.77a
4a
4b
4c
4d
4e
58.5613.0a
27.761.8a
0.460.0a
a P,0.05 compared to naproxen (ANOVA, Fisher’s PLSD test).

J. Rautio et al. / European Journal of Pharmaceutical Sciences 11 (2000) 157–163
163
their catabolism on renal cortex homogenates and lysosomal lysates. J.
Med. Chem. 35, 1246–1259.
as novel naproxen prodrugs for dermal drug delivery. Pharm. Res. 16,
1172–1178.
¨
Goosen, C., du Plessis, J., Muller, D.G., Janse van Rensburg, L.F., 1998.
Roy, S.D., Manoukian, E., 1994. Permeability of ketorolac acid and its
ester analogs (prodrug) through human cadaver skin. J. Pharm. Sci. 83,
1548–1553.
Singh, P., Roberts, M.S., 1994. Skin permeability and local tissue
concentrations of nonsteroidal anti-inflammatory drugs after topical
application. J. Pharmacol. Exp. Ther. 268, 144–151.
Sloan, K.B., 1989. Prodrugs for dermal delivery. Adv. Drug. Deliv. Rev.
3, 67–101.
Sloan, K.B., 1992. Functional group considerations in the development of
prodrug approaches to solving topical delivery problems. In: Sloan,
K.B. (Ed.), Prodrugs, Topical and Ocular Drug Delivery, Marcel
Dekker, New York, pp. 17–116.
Suh, H., Jun, H.W., Dzimianski, M.T., Lu, G.W., 1997. Pharmacokinetic
and local tissue disposition studies of naproxen following topical and
systemic administration in dogs and rats. Biopharm. Drug. Disp. 18,
623–633.
Tammara, V.K., Narurkar, M.M., Crider, A.M., Khan, M.A., 1993.
Synthesis and evaluation of morpholinoalkyl ester prodrugs of in-
domethacin and naproxen. Pharm. Res. 10, 1191–1199.
Tammara, V.K., Narurkar, M.M., Crider, A.M., Khan, M.A., 1994.
Morpholinoalkyl ester prodrugs of diclofenac: synthesis, in vitro and
in vivo evaluation. J. Pharm. Sci. 83, 644–648.
Taylor, H.E., Sloan, K.B., 1998. 1-Alkylcarbonyloxymethyl prodrugs of
5-fluorouracil (5-FU): synthesis, physicochemical properties, and
topical delivery of 5-FU. J. Pharm. Sci. 87, 15–20.
Todd, P.A., Clissold, S.P., 1990. Naproxen. A reappraisal of its pharma-
cology, and therapeutic use in rheumatic diseases and pain states.
Drugs. 40, 91–137.
van den Ouweland, F.A., Eenhoorn, P.C., Tan, Y., Gribnau, F.W.J., 1989.
Transcutaneous absorption of naproxen gel. Eur. J. Clin. Pharmacol.
36, 209–211.
Correlation between physicochemical characteristics, pharmacokinetic
properties and transdermal absorption of NSAID’s. Int. J. Pharm. 163,
203–209.
Guy, R.H., Maibach, H.I., 1983. Drug delivery to local subcutaneous
structures following topical administration. J. Pharm. Sci. 72, 1375–
1380.
Kasting, G.B., Smith, R.L., Anderson, B.D., 1992. Prodrugs for dermal
delivery: solubility, molecular size, and functional group effects. In:
Sloan, K.B. (Ed.), Prodrugs, Topical and Ocular Drug Delivery,
Marcel Dekker, New York, pp. 117–161.
Kerr, D., Roberts, W., Tebbett, I., Sloan, K.B., 1998. 7-Alkylcarbonylox-
ymethyl prodrugs of theophylline: topical delivery of theophylline. Int.
J. Pharm. 167, 37–48.
McNeill, S.C., Potts, R.O., Francoeur, M.L., 1992. Local enhanced topical
delivery (LETD) of drugs: does it truly exist? Pharm. Res. 9, 1422–
1427.
Mikulak, S.A.,Vangsness, C.T., Nimni, M.E., 1998. Transdermal delivery
and accumulation of indomethacin in subcutaneous tissues in rats. J.
Pharm. Pharmacol. 50, 153–158.
Milosovich, S.M., Hussain, A.A., Hussain, M., Dittert, L., 1989. The
utilization of prodrugs to enhance transdermal absorption of testo-
sterone, deoxycorticosterone and indomethacin. Prog. Clin. Biol. Res.
292, 273–277.
Mueller, L.G., 1990. Novel anti-inflammatory agents, pharmaceutical
compositions and methods for reducing inflammation. US Patent 4,
912, 248.
Pop, E., Soti, F., Brewster, M.E., Barenholz, Y., Korablyov, V., Mech-
oulam, R., Nadler,V., Biegon, A., 1996. Derivatives of dexanabinol. II.
Salts of amino acid esters containing tertiary and quaternary
heterocyclic nitrogen with increased water-solubility. Pharm. Res. 13,
469–475.
¨
Weber, H., Meyer-Trumpener, K., Lippold, B.C., 1994. Ester des
¨ ¨
Rautio, J., Taipale, H., Gynther, J., Vepsalainen, J., Nevalainen, T.,
naproxens als potentielle prodrugs zur hautpenetration, 2. Mitt.:
¨
¨
Jarvinen, T., 1998. In vitro evaluation of acyloxyalkyl esters as dermal
Penetrationseigenschaften an exzidierter mausehaut. Arch. Pharm.
prodrugs of ketoprofen and naproxen. J. Pharm. Sci. 87, 1622–1628.
(Weinheim) 327, 681–686.
¨ ¨
Rautio, J., Nevalainen, T., Taipale, H., Vepsalainen, J., Gynther, J.,
Yano, T., Nakawaga, A., Tsuji, M., Noda, K., 1986. Skin permeability of
various non-steroidal anti-inflammatory drugs in man. Life. Sci. 39,
1043–1050.
¨
Pedersen, T., Jarvinen, T., 1999. Synthesis and in vitro evaluation of
aminoacyloxyalkyl esters of 2-(6-methoxy-2-naphthyl)propionic acid