Biological and metabolic study of naproxen-propyphenazone mutual prodrug

Article abstract of DOI:10.1016/S0928-0987(02)00159-8

Naproxen-propyphenazone (NAP-PP) esters were synthesized as prodrugs with the aim of improving the therapeutic index through prevention of gastrointestinal irritation and bleeding. The structures of the synthesized NAP-PP hybrid esters were confirmed by I

Full text of DOI:10.1016/S0928-0987(02)00159-8

European Journal of Pharmaceutical Sciences 17 (2002) 121–130
www.elsevier.nl/locate/ejps
Biological and metabolic study of naproxen–propyphenazone mutual
prodrug
a
b
Mahmoud Shehaa , , Alaa Khedr , Hosny Elsherief
*
aFaculty of Pharmacy, Assiut University, Assiut 71526, Egypt
bPharmaceutical Chemistry Department, Faculty of Pharmacy, Al-Azhar University 71524, Assiut, Egypt
Received 6 May 2002; received in revised form 29 July 2002; accepted 8 August 2002
Abstract
Naproxen–propyphenazone (NAP-PP) esters were synthesized as prodrugs with the aim of improving the therapeutic index through
prevention of gastrointestinal irritation and bleeding. The structures of the synthesized NAP-PP hybrid esters were confirmed by IR and
1H NMR spectroscopy and their purity was established by elemental analysis, HPLC and TLC. The release of NAP as well as PP
derivatives, from the ester prodrugs was studied. A validated analytical HPLC method for the estimation of the NAP, and the prodrugs
was developed. Also the enzymatic hydrolysis products of the ester were identified by GC–MS and in conjugation with HPLC. The
kinetics of ester hydrolysis was studied in two different non-enzymatic buffer solutions, at pH 1.2, and 7.4 as well as in liver
homogenates. Study of analgesic and anti-inflammatory properties in comparison with the reference compounds has shown that both
analgesic and anti-inflammatory activities were present at the same doses of the investigated compounds. The ester III was found to be
less irritating to gastric mucosal membrane than the parent drugs. These results suggest that the synthesized prodrugs are characterized by
better therapeutic index than the parent drugs.
2002 Elsevier Science B.V. All rights reserved.
Keywords: Naproxen prodrug; Propyphenazone; Prodrugs; Ulcerogenicity; NSAIDs; Chemical and enzymatic stability
1. Introduction
of naproxen with different alkyl esters and thioesters lead
to prodrugs with retained anti-inflammatory activity but
exhibited greatly reduced gastrointestinal erosive prop-
erties and significantly reduced analgesic potency (Venuti
et al., 1989). But esterification with ethylpiperazine
showed that the analgesic activity was preserved whereas
anti-inflammatory activity was generally reduced (Calvi et
al., 1985). Propyphenazone (PP) is a non-acidic pyrazole
drug and has good analgesic and antipyretic activity with
no anti-inflammatory activity (Burne, 1986). The rational
of this work was to couple naproxen with propyphenazone
to achieve many advantages related to synergistic analgesic
effect with reduced GI irritation. The metabolism of
propyphenazone has been reported to proceed via the
formation of 3-hydroxymethyl-propyphenazone (HMP),
which is pharmacologically active as the parent drug
(Goromaru et al., 1984; Neugebauer et al., 1997). Cou-
pling of both compounds as a hybrid drug or through a
spacer as a mutual prodrug results in a potent analgesic
anti-inflammatory compound with reduction of the main
adverse local effects related to the activity of NSAID. The
preparation of naproxen esters/amides with PP derivatives
For many years, several attempts have been made to
develop bioreversible derivatives or prodrugs of non-ster-
oidal anti-inflammatory drugs (NSAID) containing car-
boxylic acid function in order to depress upper gastroin-
testinal (GI) irritation and bleeding (Laporte et al., 1991;
Wynne et al., 1998). Esterification of the carboxylic acid
moiety of NSAIDS would suppress gastro-toxicity without
adversely affecting their anti-inflammatory activity. In
addition the biotransformation of the prodrugs to the parent
compounds at its target site or sites of activity may be used
to achieve rate and time controlled drug delivery of the
active entities (Bonina et al., 2001; Rautio et al., 2000,
Thorsteinsson et al., 1999). Naproxen (NAP), a potent
NSAID with moderate analgesic activity associated with
GI irritation was selected as a model drug for carboxylic
acid derivatization (Shanbhag et al., 1992). Esterification
*
Corresponding author. Tel.: 120-12-397-1632; fax: 120-88-332-
776.
E-mail address: sheha@acc.aun.eun.eg (M. Sheha).
0928-0987/02/$ – see front matter
PII: S0928-0987(02)00159-8
2002 Elsevier Science B.V. All rights reserved.

122
M. Sheha et al. / European Journal of Pharmaceutical Sciences 17 (2002) 121–130
is described. The availability of NAP and PP derivatives
after hydrolysis is studied by chemical and by enzymatic
means.
Ulm, Germany). Naproxen was obtained as a gift from
Misr Company for Pharmaceuticals (Cairo, Egypt), and
propyphenazone was obtained from Kahira Company for
Pharmaceuticals (Cairo, Egypt). Dimethylformamide
(DMF) was purchased from El-Naser Chemical Industry
Company (Cairo, Egypt), dried over phosphorus pentoxide
and successively distilled under reduced pressure before
use. All other chemicals and solvents were of analytical
2. Material and methods
2.1. Materials
grade.
The
starting
compound
3-bromomethyl-
Melting points were recorded (uncorrected) on a melting
point/MS (Stuart Scientific, UK). Infrared spectra were
recorded with a Shimadzu IR-470 spectrometer (Shimadzu,
Tokyo, Japan) and UV spectra on a Shimadzu UV-120-02
spectrophotometer (Shimadzu, Kyoto, Japan). Thin layer
chromatography (TLC) was carried out on precoated silica
gel plates (Merck, 0.25 MM, 60 F254). Column chroma-
tography was carried out on Merck silica gel 60 (Particle
size 0.063–0.200 mm). 1H NMR spectra were recorded on
a Jeol JNM-EX 270 MHz spectrometer (Jeol, Tokyo,
Japan); all chemical shifts are given in d ppm, relative to
tetramethylsilane (TMS) as internal standard. Elemental
analyses were performed at the Microanalysis Center,
Chemistry Department, Faculty of Science, Assiut Uni-
versity.
The HPLC system consisted of a Knauer model 64,
solvent delivery module (Knauer, Germany), Knauer vari-
able wavelength UV detector, 20-ml sample loop, and a
Shimadzu CR-6A Chromatopac integrator (Shimadzu,
Tokyo, Japan). The column used was Du-Pont Zorbax C8
(250 mm34.6 mm i.d., 6 mm), heated at 40 8C (Du-Pont
column heater, Du-Pont, Delawara, USA). The effluent
was monitored at 284 nm at a flow rate of 1 ml/min. Two
mobile phases were used. Mobile A, consisting of
methanol/acetonitrile/0.05 M sodium acetate (30:15:55)
adjusted to pH 5.0 with glacial acetic acid was used for the
estimation of HMP and NAP; Mobile B, consisting of
acetonitrile/isopropanol/0.05 M sodium acetate (45:5:50),
adjusted to pH 5.0 with glacial acetic acid was used for the
estimation of ester. The GC–MS apparatus used was a
Finnigan/MAT 1020B (Finnigan/MAT, Bremen, Ger-
many); data system: INCOSI. Capillary column: DB-5
(J&W Scientific, Rancho, Cordova, USA), fused silica 30
m30.25 mm i.d., 0.25 mm film thickness. Carrier gas:
helium, pressure 26 p.s.i., flow rate 60 cm/s. Gerstel-Split
injector, split 1:20 direct coupling. Source temperature
180 8C, ionization energy 70 eV, and multiplier voltage
2000 V. Scanning from 35 to 400 amu/s. Column port
temperature 260 8C, detector (FID), temperature 320 8C,
column oven temperature programmed from 150 8C (1
min) to 280 8C (5 min) to 300 8C (7 min) were used.
b-Glucouronidase-arylsulfatase (from Helix Pomatia,
stabilized aqueous solution, b-Glucouronidase 40 U/ml;
arylsulfatase 20 U/ml) was purchased from Merck (Darm-
stadt, Germany). N-Methyl-N-trimethylsilyltrifluro-acet-
amide (MSTFA), trimethylchlorosilane (TMSCI), and
NADP were purchased from Fluka (Feinchemikalien, Neu-
propyphenazone was prepared in a pure crystalline form
according to the reported method (Lucius, 1907).
2.2. Chemistry
2.2.1. Preparation of 3-aminomethyl-4-isopropyl-2-
methyl-1-phenyl-3-pyrazoline-5-one (AMP)
A solution of 3-bromomethylpropyphenazone (3 g, 10
mmol) in chloroform (100 ml) and hexamethylenetet-
ramine (1.7 g, 12 mmol) was stirred at 50 8C for 3 h. The
solvent was evaporated under vacuum to dryness. The
yellowish crystalline adduct obtained was digested with 10
N HCl for about 2 h and evaporated under vacuum. The
residue was stirred with 20 ml of water and cooled in a
refrigerator for about 3 h and filtered. The product obtained
was recrystallized from ethanol to give 1.9 g (65% yield)
of pure 3-aminomethyl-propyphenazone hydrochloride
1
(AMP.HCl), m.p. 164–5 8C, H NMR (CDCl3): 1.12 (d,
6H), 2.5 (s, 3H), 2.62 (m, 1H), 3.25 (m, 2H), 3.45–3.6 (m,
2H, exchangeable), 6.85–7.22 (m, 5 H). Elemental analy-
sis: calculated (C: 59.67, H: 7.15, N: 14.91) found (C:
58.95, H: 7.25, N: 15.08).
2.2.2. Preparation of glycinyl-3-
hydroxymethylpropyphenazone (Gly-HMP)
To an ice-cold solution of Boc-glycine (1.75 g, 10
mmol) in 30 ml dichloromethane were added HMP (2.46
g, 10 mmol), dimethylaminopyridine (50 mg), and DCC
(2.27 g). The reaction mixture was stirred at 4 8C for 1 h,
and kept overnight at room temperature. The precipitated
DCU was separated by filtration and the filtrate was
washed with cooled 0.05 N HCl and followed by saturated
solution of NaHCO3 and finally with brine, dried over
anhydrous Na2SO4, and evaporated under reduced pres-
sure. The product was then treated with 10 ml of 6 N HCl
in dioxane and stirred for 1 h followed by evaporation
under vacuum at room temperature. The resulting product
was triturated with 100 ml of dry ether and left for 2 h in
refrigerator. The precipitated product was filtered and
vacuum dried to yield crystals of Gly-HMP (2.8 g, 82.6%),
1
m.p. 135–7 8C, H NMR (DMSO 6d) 1.1, 1.2 (6H, d, 9
Hz); 2.45 (3H,s); 2.5–2.65 (1H, m); 3.0–3.2 (m, 2H,
exchangeable); 3.53–3.67 (m, 2H); 4.8 (2H, s); 6.7–7.2 (5
H, m). Elemental analysis: calculated (C: 56.55, H: 6.53,
N: 12.37); found (C: 55.32, H: 6.52, N: 12.41).

M. Sheha et al. / European Journal of Pharmaceutical Sciences 17 (2002) 121–130
123
2.2.3. Preparation of 3-[(hydroxyacetamido)methyl]-4-
isopropyl-2-methyl-1-phenyl-3-pyrazolin-5-one (AcAMP)
To an ice-cold solution of O-tritylglycolic acid (3.5 g,
11 mmol) and AMP.HCl (2.81 g, 10 mmol) in dimethyl-
formamide (20 ml), triethylamine (5 ml), and BOP (5 g)
were added. The reaction mixture was stirred at 4 8C for 1
h, and 3 h at room temperature. The reaction mixture was
evaporated under vacuum, and the residue dissolved in
ethyl acetate, washed with cooled 0.01 N HCl (50 ml) and
followed by saturated solution of NaHCO3 and finally with
brine. The solution was dried over anhydrous Na2SO4, and
evaporated under reduced pressure; 10 ml of 6 N HCl in
dioxane were added to the residue, stirred for 1 h and
evaporated under vacuum at room temperature. The res-
idue was then chromatographed on a silica column eluted
with hexane/ethyl acetate (3:1) to obtain pure AcAMP.
The vacuum dried crystals (2.2 g, 73%) has m.p. 127–
3.7 (s, 3H), 3.9, 4.0 (d, 1H); 4.1 (s, 2H); 4.7 (s, 2H); 6.7 (s,
1H); 6.8–7.6 (m, 11H). Elemental analysis:
C30H33N3O5.H2O (FW 533.62); calculated C 67.52%, H
6.56% and N 7.88%; found C 67.9%, H 6.6% and N
7.63%.
2.2.6. Preparation of naproxen-
hydroxyacetamidopropyphenazone (Nap-AcAMP; III)
To a solution of thionyl chloride (0.8 g, 6.7 mmol) in
dichloromethane (15 ml) was added dimethylformamide
(0.5 g, 6.8 mmol) in dichloromethane (3 ml). After 10 min
at room temperature, the above solution was allowed to
react with a solution of naproxen (1.15 g, 5 mmol)
dissolved in dichloromethane (15 ml) by reflux in the dark
for about 1 h. To the cold stirred reaction mixture, a
solution of AcAMP (1.5 g, 4.95 mmol) in dichloromethane
(10 ml) was added drop-wise. The reaction mixture was
stirred for 1 h at room temperature and then evaporated
under vacuum. The residue was purified by column
chromatography (hexane/ethyl acetate; 2:1) to yield 1.8 g
1
8 8C, H NMR (CDCl3) 1.1, 1.2 (d, 6H); 2.4 (s, 3H);
2.8–3.1 (m, 1H); 3.6–3.9 (m, 2H); 4.8–5.1 (m, 3H); 6.7 (s,
1H); 6.9–7.3 (m, 5 H). Elemental analysis: C16H21N3O3
(FW 303.36); calculated C 63.35%, H 6.98% and N
12.85%; found C 63.0%, H 7.1% and N 13.52%.
1
(70%) of pure Nap-AcAMP. H NMR (DMSO.d6): 1.1,
1.2 (d, 6H); 1.6 (d, 3H); 2.6 (s, 3H); 2.6–2.8 (m, 1H); 3.7
(m, 1H); 3.9 (s, 3H); 4.1 (m, 2H); 5.2 (s, 2H); 6.7 (s, 1H);
6.9–7.6 (m, 11H). Elemental analysis: C30H33N3O5.1/
2H2O (FW 524.6) calculated C 68.68%, H 6.49% and N
8.0%; found C 68.0%, H 6.62% and N 7.83%.
2.2.4. Preparation of naproxen-3-
hydroxymethylpropyphenazone ester (NAP-HMP; I)
A solution of CDI (0.5 g, 3.08 mmol) in anhydrous
DMF (2.5 ml) was added drop-wise to a cold solution of
naproxen (0.5 g, 2.17 mmol) dissolved in anhydrous DMF
(3.5 ml). The mixture was kept cold for 30 min and a
solution of HMP (0.52 g, 2.17 mmol) in dry DMF (10 ml)
was added drop-wise. The reaction mixture was stirred at
room temperature overnight. The reaction mixture was
evaporated under vacuum, and the residue purified by
column chromatography to yield 0.75 g (75.5%) of pure
ester I [m.p. 143–145 8C, IR (cm21): 1773 (strong band),
1685 (strong band); 1H NMR (DMSO.d6): 1.1, 1.2 (6H, d,
9 Hz); 1.45, 1.55 (3H, d, 9 Hz); 2.6 (3H, s); 2.65, 2.70,
2.75, 2.80 (1H, q, 5 Hz); 3.7 (3H, s), 4.0–4.2 (1H, m); 4.9
(2H, s); 7.1–7.6 (11H, m)]. Elemental analysis:
C28H30N2O4.3/2 H2O (FW 485.55) calculated; C 69.26%,
H 6.85% and N 5.77%; found; C 69.25%, H 6.90% and N
5.63%.
2.3. In vitro hydrolysis of naproxen prodrugs
2.3.1. Chemical hydrolysis study
The hydrolysis study was carried out using a Hanson
SR6 dissolution apparatus (Hanson Research, CA, USA).
The test was conducted at 3760.5 8C using apparatus II
and pedal speed rotation of 50 rpm. Two hydrolysis media
were used: 250 ml of non-enzymatic simulated gastric fluid
(SGF) buffer solution of pH 1.2 and an isotonic phosphate
buffer, pH 7.4. An accurate weight of 25.0 mg of the
prodrug was used for the study. From this matrix, aliquots
of 3.0 ml were withdrawn at 0.25, 0.5, 1, 2, 4, 6, 8, 10, 12,
14, 16, 18, 20, 24, and 30 h and were immediately replaced
with 3.0 ml of fresh hydrolysis media equilibrated at
3760.5 8C. The samples were stored at 220 8C pending
analysis. A developed HPLC assay was used for the
determination of prodrug in these samples. All dissolution
experiments were carried out in triplicate and the average
values were reported after a correction was applied for the
dilution effect.
2.2.5. Preparation of naproxen-glycine-3-
hydroxypropyphenazone (Nap-Gly-HMP; II)
A solution of CDI (1 g, 6.16 mmol) in anhydrous DMF
(5 ml) was added drop-wise, at 4 8C, to a solution of
naproxen (1 g, 4.34 mmol) dissolved in anhydrous DMF
(10 ml). To the cold mixture, a solution of Gly-HMP (1.47
g, 4.34 mmol) in dry DMF (20 ml) was added drop-wise.
The reaction mixture was stirred for 10 min at 4 8C, for 3 h
at room temperature and then evaporated under vacuum.
The residue was purified by column chromatography
(hexane/ethyl acetate; 2:1) to yield 1.9 g (85%) of pure
compound II with m.p. 115 8C. 1H NMR (DMSO.d6): 1.1,
1.2 (d, 6H); 1.4, 1.5 (d, 3H); 2.5 (s, 3H); 2.4–2.6 (m, 1H);
2.3.2. In vitro enzymatic hydrolysis study
NADPH was regenerated in the liver homogenate of
mice, using 30 g liver homogenate in 108 ml isotonic
potassium dihydrogen phosphate (pH 7.4). The percentage
recovery of the prodrugs (I, II, and III) were estimated
from 1 ml liver homogenate, spiked with 0.50 mg of the
prodrug, followed by extraction as described below.
The rest of the liver homogenate–NADPH (100 ml) was

124
M. Sheha et al. / European Journal of Pharmaceutical Sciences 17 (2002) 121–130
mixed with 50 mg prodrug. The hydrolysis was carried out
using Vankel dissolution apparatus (Vankel VK 7000, NJ,
USA). The experiment was conducted at 3760.5 8C, and
pedal rotation speed of 50 rpm. Aliquots of 1-ml sample
withdrawn at 1, 20, 40, 60, 80, 100, 120, 140, 160, 180,
200, 220, 240, 260 and 280 min, were transferred to 5-ml
amber colored, capped reaction vials and immediately
replaced with 1 ml of fresh hydrolysis media equilibrated
at 3760.5 8C. Then, 1 ml 10% TFA was added to
deactivate the enzyme in the samples, and it was kept in a
refrigerator at 220 8C until the next step. The deactivated
samples taken from liver homogenate treated with prodrug
ester I (2 ml) were divided to two equal parts. The first
part (1 ml), was transferred to a 5-ml centrifuge tube,
mixed with 0.4 ml water, 1.1 ml acetonitrile, vortexed for
5 min, and centrifuged for 5 min at 5000 rpm. From the
clear supernatant, 20 ml was injected for HPLC analysis of
the prodrug content using method B. The second part (1
ml), was transferred to a capped reaction tube, and mixed
with 0.35 ml acetate buffer pH 5.5, followed by 50 ml
p-glucuronidase-arylsulfatase solution. This solution was
incubated at 40 8C in a thermostatically controlled water
bath for 24 h. This sample was extracted with 1.1 ml
acetonitrile as described previously. From the clear super-
natant, a 20-ml aliquot was analyzed by HPLC using
method A, for estimation of NAP. All sample extracts were
developed three times in parallel with calibration curve
solutions. The recovered amount was calculated from the
calibration curves of the investigated compounds, using
five levels to obtain a calibration line. The calibration
solutions were prepared in methanol/acetonitrile (1:1), in
the range of 0.01–0.25 mmol, of each. The estimated
concentrations were calculated for the final dilution, from
the corresponding regression line, and corrected for the
extraction factor.
NAP and HMP, separately. Male Swiss mice (23–26 g)
and male Sprague–Dawley rats (150–170 g) were used.
The animals were starved for about 15 h before test
compound administration. The tested compounds were
prepared for oral administration in aqueous 0.5% carboxy-
methylcellulose (CMC) solution. The properties of the test
compounds were compared with those of HMP and NAP.
Statistical analysis was performed with P#0.05 as the
level of significance.
2.5.1. Analgesic activity
The analgesic activity was evaluated using a reported
hot plate method (Jansen and Jageneau, 1957; Galina et al.,
1983). this depends upon observing the normal response to
a pain stimulus in untreated animals and comparing it with
the response to the same stimulus after the administration
of a drug at definite time intervals. The relative mouse
response to the plate temperature was a convenient meth-
od, to compare the relative drug activity.
Five groups of six mice were orally treated with a
solution of prodrugs (I or II or III), naproxen or HMP (in
equivalent doses, 21.7 mmol/kg). The mice were placed
individually on a preheated hot plate (5461 8C) and the
time (s) which elapsed before the mouse reacted was
recorded. The reaction time (s) was determined at 1, 2, 3,
4, 6, 7 and 8 h. The differences in latencies, with respect to
the control group (six mice), were used as an indication of
analgesia.
2.5.2. Anti-inflammatory activity (Winter et al., 1962)
Six groups of six rats were used. Each group of animals
was orally dosed with the test compound (NAP, MHP,
Prodrug I, Prodrug II, or Prodrug III; 21.7 mmol/kg) or
drug-free vehicle (CMC). After 1 h, 0.1 ml of a 0.1%
carrageenan solution was injected into the sub-plantar
tissue of the right paw. Swelling (volume) of the injected
paw of each animal was measured at 2, 3, 4, 5, 6 and 8 h
by a mercury plethysmometer.
2.4. Identification of the prodrug metabolites
After 280 min enzymatic hydrolysis of prodrug ester I, 1
ml of the sample was mixed with 2 ml phosphate buffer
pH 6.0, saturated with about 1 g sodium chloride; and
extracted by vortex with 5 ml dichloromethane. From the
clear supernatant, 3 ml were withdrawn, transferred to a
2-ml capped reaction vial, and dried under a stream of
nitrogen gas. The residue was dissolved in 0.3 ml chloro-
form and 50 ml MSTFA containing 10% TMSCI. The
mixture was heated at about 80 8C for 30 min. The sample
was then cooled to room temperature, and 1 ml was
developed for GC–MS analysis. A derivative standard
solution mixture of the HMP, and NAP was silylated and
developed in parallel to the sample extract.
2.5.3. Ulcerogenicity study
A Jeol, JSM-5400LV scanning electron microscope
(Electron Microscope Unit, Assiut University) was used
for observing mucosal injury from a scanning micrograph
of stomach specimens. Four groups of four mice each were
fasted for 12 h prior to the administration of the drug. The
first group was administered a daily oral dose (21.7 mmol/
kg) as a 1 ml suspension of prodrug III in 0.1% CMC
solution, for four successive days. In a similar manner, the
second and third groups received equivalent doses of NAP
and HMP, separately. The fourth group received an equiva-
lent amount of the vehicle (CMC) and was considered as
control group. Food was withdrawn from all groups until
24 h after the last dose. The mice were then sacrificed, so
that the stomach could be removed, opened along the
greater curvature, investigated for appearance of lesions
and prepared for scanning in the electron microscope. The
2.5. In-vivo evaluation
The prodrugs prepared were screened for analgesic and
anti-inflammatory activity compared with that exerted by

M. Sheha et al. / European Journal of Pharmaceutical Sciences 17 (2002) 121–130
125
specimens were fixed by soaking in glutaraldehyde solu-
tion (5% in cacodylate buffer, pH 7.2) for 24 h followed
by three washings each for 20 min with cacodylate buffer.
They were then treated with osmium tetraoxide (1%
solution) for 2 h and washed with cacodylate buffer as
described above. Then the specimens were subjected to
dehydration by ethanolic solutions of 30, 50 and 70%, for
30 min with each, followed by 90% ethanol for 1 h and
finally in absolute ethanol for 2 days. After discharge of
the alcohol, the specimens were soaked in amyl acetate
solution for 2 days, dried under reduced pressure, and
mounted on holders coated for scanning in a scanning
electron microscope (SEM).
NAP and prodrugs was developed. The developed chro-
matographic methods were validated for selectivity, preci-
sion, accuracy, suitability, linearity and range. Also, the
applied chromatographic procedures were validated to
separate selectively, the ester, HMP and NAP from the
biogenic materials of the hydrolyzate extract. GC–MS was
used for the identification of enzymatic hydrolysis products
of the prodrug (I). The hydrolyzate after 280 min contains
naproxen and HMP in addition to O-desmethyl-naproxen
ester with HMP and O-desmethyl-naproxen. These data
prove the low activity of this sterically hindered ester
prodrug so metabolism mainly occurs at the terminal
demethylation rather than by ester hydrolysis.
The kinetics of ester hydrolysis was studied in two
different buffer solutions (pH 1.2 and 7.4), without en-
zymes. Also, the effect of activated esterase enzymes
present in liver homogenates was investigated. The data
for chemical hydrolysis, presented in Fig. 1, fit pseudo-
first-order kinetics (correlation coefficient .0.98) and the
rate constants computed at pH 1.2 were 0.7531022 (h21),
1.0231022 (h21) and 1.1531022 (h21) for prodrugs I, II
and III, respectively. These values correspond to half
times in vitro of about 91, 67, and 60 h for Prodrugs I, II
and III, respectively. The data for chemical hydrolysis,
presented in Fig. 2, fit pseudo-first-order kinetics (correla-
tion coefficient .0.99) and the rate constants computed at
pH 7.4 were 3.5431022 (h21), 5.4831022 (h21) and
6.9431022 (h21) for prodrugs I, II and III, respectively.
These values correspond to half times in vitro of about 19,
12, and 10 h for Prodrugs I, II and III, respectively. The
results were the average of three runs. As expected, the
rate of prodrug hydrolysis was found to be highly pH-
dependent as shown in Figs. 1 and 2. While the prodrug
was hydrolyzed at higher pH (7.4), it showed insignificant
hydrolysis in the acidic medium (pH 1.2).
3. Results and discussion
3.1. Chemistry
Bromopropyphenazone (BMP) was used for preparation
of hydroxymethyl-propyphenazone (HMP) by reflux with
water, or 3-aminomethylpropyphenazone (AMP) by re-
action with hexamine followed by acid treatment. Direct
coupling of naproxen with HMP was carried out by three
different reaction conditions as shown in Scheme 1. The
carbonyldiimidazole (CDI) method gave better yield of the
corresponding ester with minimal reaction by-products. In
order to avoid or minimize the steric hindrance effect on
the hydrolysis of naproxen-HMP ester prodrug (I), a
spacer was introduced. The spacers chosen were glycine
and glycolic acid. N-Protected glycine coupled with HMP
by DCC followed by N-deprotection and coupling with
naproxen using CDI yielded naproxen amide prodrug (II).
Coupling of O-tritylglycolic acid with aminomethyl-
propyphenazone (AMP) was carried out using strong
coupling reagent BOP under very mild conditions. The
O-trityl group was removed using mild acidic conditions to
yield the amide AcAMP which reacted with naproxen acid
chloride to yield naproxen ester prodrug (III). The struc-
tures and purity of the prodrugs formed were confirmed by
instrumental methods including UV, 1H NMR, and elemen-
tal analysis.
The hydrolysis of the ester was tested in the rat liver
homogenate, to investigate the effect of activated esterase
enzyme on the drug. Fig. 3 shows that the metabolic rate
of the ester is linear. The rate constants computed were
0.831022 (h21), 1.0431022 (h21) and 1.2731022 (h21
)
for prodrugs I, II and III, respectively. These values
correspond to half times in vitro liver enzyme of about 85,
66 and 54 min for prodrugs I, II and III, respectively.
Moreover, after a 240-min incubation period of the ester in
liver homogenate, less than 10.5% of intact prodrug was
recovered.
Sample stability in the liver homogenate was tested by
incubating the prodrug, HMP, and NAP at 220 8C for 24
h, followed by re-estimating the content by applying
HPLC. The recovered amounts of the prodrug (I), prodrug
(II), prodrug (III), and NAP were found to be 10062,
9762, 9761, and 8562.3%, respectively. The extracted
samples have shown non-significant difference, and typical
chromatograms compared with those extracted at zero time
were observed.
3.2. In vitro hydrolysis of ester
Since the carboxyl group of naproxen is essential for the
therapeutic action, the prodrugs of a prolonged action were
designed in a form which the biologically active moiety
can be released in its original state with time. Therefore,
the release of NAP from NAP–propyphenazone prodrugs
was studied in vitro in order to evaluate the possible time
span in which the drug could be available from different
prodrugs. The prodrugs that quantitatively dissociate and
release naproxen were selected for in vivo study. A
validated analytical HPLC method for the estimation of the

126
M. Sheha et al. / European Journal of Pharmaceutical Sciences 17 (2002) 121–130
Scheme 1. General steps for synthesis of the targeted prodrugs (I–III).

M. Sheha et al. / European Journal of Pharmaceutical Sciences 17 (2002) 121–130
127
Fig. 3. Pseudo-first-order-plot of the enzymatic hydrolysis of the naprox-
en prodrugs.
tions of HMP, and NAP. HMP was trimethylsilylated at its
hydroxyl group, and showed characteristic mass fragment
representing: TMS group at m/z573 (47%), abundant
mass fragment of m/z5318 (82%) representing the molar
peak (3-O-TMS) and base peak at 303 representing 3-O-
TMS that loses one methyl group from the TMS part. NAP
was trimethylsilylated at its carboxylic group and shows
characteristic mass fragments of 302.7 (22%), 287.3
(25%), 230.2 (64%) and base peak at 215.3. However, the
mass spectrum of the prodrug (I), was not silylated, but its
desmethyl metabolite was identified. This new metabolite
has shown a TMS derivative at 516 (20%) represented the
molar peak of ester-OTMS (instead of ester-OCH3), 502
(10%) represents N-desmethyl-ester-OTMS, 309 (20%),
230 (55%), and base peak at m/z5215, represent the
demethylsilylated NAP. The desmethyl-ester metabolite
could not be identified in the HPLC chromatogram.
Fig. 1. Pseudo-first-order plot of the chemical hydrolysis profile of the
prodrugs in SGF.
3.3. GC–MS data interpretation
The extracted prodrug I metabolites have shown the
same GC-mass spectrum of the derivatized standard solu-
3.4. In-vivo evaluation
Analgesic, anti-inflammatory and ulcerogenic activities
of the prodrugs were studied in comparison to equivalent
doses (21.7 mmol/kg) of NAP and HMP. The analgesic
activity was investigated in mice according to reported
methods (Jansen and Jageneau, 1957; Galina et al., 1983).
In this procedure, the rodent is placed on a hot plate that is
preheated to a temperature that is presumed to be aversive.
The latency to lick a paw and/or jump is recorded. The
differences between the latencies and those of controls are
used as a measure of the degree of analgesia (Table 1). As
a general pattern, the analgesic activities of the prodrugs
show improvement over time. The maximum analgesic
activity is reached after about 4–6 h for all prodrugs. The
observed latent analgesia may result from prodrug bio-
availability and/or hydrolysis to parent drugs.
Fig. 2. Pseudo-first-order plot of the chemical hydrolysis profile of the
prodrugs in SIF.

128
M. Sheha et al. / European Journal of Pharmaceutical Sciences 17 (2002) 121–130
Table 1
The latencies of pain threshold (seconds, s)in mice treated with the prodrugs I–III, naproxen (NAP) and hydroxymethylpropyphenazone (HMP)
a
Compound
Mean6S.E. of the pain threshold (s)
1 h
85.063.14
137.169.10
30.663.78
35.062.18
51.363.16
2 h
72.563.33
125.568.38
50.364.10
72.564.10
64.165.27
3 h
61.662.52
109.268.09
87.068.59
97.665.70
87.365.20
4 h
50.961.28
84.665.32
98.269.71
107.6611.95
99.167.21
6 h
45.162.01
60.063.10
92.867.26
103.165.72
101.366.25
7 h
8 h
NAP
HMP
Prodrug I
Prodrug II
Prodrug III
36.761.19
56.563.91
95.963.58
87.566.12
97.267.72
30.060.92
45.866.51
71.368.06
58.463.29
81.264.17
a n56 mice; dose 21.7 mmol/kg; the results are significant at P,0.01.
The effectiveness of NSAIDs in acute inflammation can
be estimated by using intraplantar injection of carrageenan
in the rat. Swelling of the injected paw was measured at 2,
3, 4, 5, 6, and 8 h using a mercury plethysmometer. The
pretreatment with oral naproxen (5 mg/kg) resulted in a
significant reduction in swelling that lasted about 5 h
post-carrageenan injection (Fig. 4). The anti-inflammatory
activity of prodrugs significantly improves over time. This
means that the prodrugs per se are devoid of anti-in-
flammatory activity and that the observed latent activity
results from hydrolysis to the parent drugs. An equimolar
dose of prodrug III (11.17 mg/kg) reduced inflammation
in a pattern approximately similar to naproxen. Prodrugs I
and II show moderate anti-inflammatory activity with time
in comparison with prodrug III that may be due to the
slow rate of hydrolysis of prodrugs I and II to release
naproxen.
system, was tested in comparison with parent drugs (NAP
and HMP) following oral administration for 4 days in
mice. As could be seen from the gross observation, linear
or oval-shaped lesions were found mainly in the corpus of
the stomach. The group subjected to NAP prodrug III had
significantly fewer lesions than the NAP group. The
examination of the stomach specimens of the treated
experimental animals under scanning electron microscope
affords a highly precise method for investigation of the
ulcerogenic potential of NSAIDs. Fig. 5 shows scanning
electromicrographs, at a constant magnification power, for
stomach specimens of mice treated with a chronic dose of
prodrug III (Fig. 5A); the parent drugs HMP (Fig. 5B);
NAP (Fig. 5C) and the control group (Fig. 5D) which
receive only the vehicle. As shown in Fig. 5, the NAP-
treated group (Fig. 5C) and to some extent HMP-treated
group were characterized by complete damage of the
mucous layer besides ulceration of the sub-mucosal cells.
These effects are not observed in the prodrug-treated group
The ulcerogenic activity of the prodrug III, as a
representative example for the synthesized oral delivery
Fig. 4. The mitigation of acute inflammation by test compounds (21.7 mmol/kg) in a carrageenan model in rats. The filled boxes represent animals to
whom vehicle was administered, diagonally lined boxes represent naproxen, the horizontally lined boxes represent hydroxymethyl-propyphenazone, the
open boxes represent prodrug I, the large checker board boxes represent prodrug II, and the dashed horizontal lined boxes represent prodrug III.

M. Sheha et al. / European Journal of Pharmaceutical Sciences 17 (2002) 121–130
129
Fig. 5. Scanning electromicrographs of mouse stomach after chronic doses (4 days) of: Prodrug III (A), HMP (B), NAP (C), and the vehicle (D).
pharmacological study of new esters of naproxen with derivatives of
N-oxyethylpiprazine. Farmaco 40, 334–346.
Galina, Z.H., Sutherland, C.J., Amit, Z., 1983. Effects of heat-stress on
behaviour and the pituitary adrenal axis in rats. Pharmacol. Biochem.
(Fig. 5A) or control group (Fig. 5D). The previous
observations afford good evidence for the safety of the
suggested oral delivery system of NSAIDs compared with
the traditional use of the parent drugs.
Behav. 19, 251–256.
Goromaru, T., Matsura, H., Furuta, T., Baba, S., 1984. Identification of
isopropylantipyrine metabolites in rat and man by using stable isotope
tracer techniques. Chem. Pharm. Bull. 32, 3179–3186.
4. Conclusion
Jansen, P.A., Jageneau, A.H., 1957. J. Pharm. Pharmacol. 9, 381.
Laporte, J.R., Carne, X., Vidal, X., Moreno, V., Juan, J., 1991. Upper
gastrointestinal bleeding in relation to previous use of analgesics and
non-steroidal anti-inflammatory drugs. Catalan Countries Study on
Upper Gastrointestinal Bleeding. Lancet 337 (8733), 85–89.
Lucius M. and Brueing in Hoechst, AM. Patentschrift Nr. 206637, Klasse
12p, Gruppe 8, Patentiert im Deutschen Reiche vom 3 October 1907
ab.
The in vitro and in vivo evaluation of the synthesized
naproxen–propyphenazone hybrid drug ester and/or amide
revealed improvement in the therapeutic index of the
parent drugs. The derivatives are characterized by prodrug
profile, adequate chemical stability, and reduced ul-
cerogenic liability. The prodrugs retained the anti-inflam-
matory and analgesic activities of the parent drugs.
Neugebauer, M., Khedr, A., El-Rabbat, N., El-Kommes, M., Saleh, G.,
1997. Stereoselective metabolic study of Famprofazone. Biomed.
Chromatogr. 11, 356–361.
Rautio, J., Nevalainen, T., Taipale, H., Vepsalainen, J., Gynther, J., Laine,
K., Jarvinen, T., 2000. Piperazinylalkyl prodrugs of naproxen improve
in vitro skin permeation. Eur. J. Pharm. Sci. 11, 157–163.
Shanbhag, V.R., Crider, A.M., Gokhale, R., Harpalani, A., Dick, R.M.,
1992. Esters and amide prodrugs of ibuprofen and naproxen: synthesis,
antiinflammatory activity, and gastrointestinal toxicity. J. Pharm. Sci.
81, 149–154.
Thorsteinsson, T., Masson, M., Loftsson, T., Haraldsson, G.G., Stefan-
sson, E., 1999. Diacyl glyceryl ester prodrugs for slow release in the
skin: synthesis and in vitro degradation and absorption studies for
naproxen derivatives. Pharmazie 54, 831–836.
References
Bonina, F.P., Puglia, C., Barbuzzi, T., de Caprariis, P., Palagiano, F.,
Rimoli, M.G., Saija, A., 2001. In vitro and in vivo evaluation of
polyoxyethylene esters as dermal prodrugs of ketoprofen, naproxen
and diclofenac. Eur. J. Pharm. Sci. 14, 123–134.
Burne, K., 1986. Comparative pharmacology of non-opioid analgesics.
Med. Toxicol. Suppl. 1, 1–9.
Calvi, E., Picciola, G., Garenini, G., Gentili, P., 1985. Synthesis and
Venuti, M.C., Young, J.M., Maloney, P.J., Johnson, D., McGreevy, K.,

130
M. Sheha et al. / European Journal of Pharmaceutical Sciences 17 (2002) 121–130
1989. Synthesis and biological evaluation of omega-(N,N,N-tri-
Wynne, H.A., Long, A., Nicholson, E., Ward, A., Keir, D., 1998. Are
altered pharmacokinetics of non-steroidal anti-inflammatory drugs
(NSAIDs) a risk factor for gastrointestinal bleeding? Br. J. Clin.
Pharmacol. 45, 405–408.
alkylammonium)alkyl esters and thioesters of carboxylic acid non-
steroidal antiinflammatory agents. Pharm. Res. 6, 867–873.
Winter, C.A., Risley, E.A., Nuss, G.W., 1962. Carrageenin-induced edema
in hind paw of the rat as an assay for anti-inflammatory drugs. Proc.
Soc. Exp. Biol. Med. 111, 544.