Optimizing delivery of flurbiprofen to the colon using a targeted prodrug approach

Article abstract of DOI:10.1211/jpp.60.5.0006

The carboxylic group responsible for the gastric side-effects of the propionic acid derivative, flurbiprofen, was masked temporarily to overcome these side-effects and to accomplish colon-specific delivery of the drug. An amide prodrug (FLU-GLY) was synthesized by coupling flurbiprofen with L-glycine. Confirmation and characterization of the structure of the synthesized prodrug included elemental analysis, Fourier transform (FT)-IR, FT-NMR, mass (FAB) spectroscopy, and determinations of R_f, R_t and R _M values, respectively. Aqueous solubility and lipophilicity (log P) value were determined at pH 1.2, 4.0, 6.8 and 7.4. In-vitro reversion of FLU-GLY to flurbiprofen was measured at different pHs and in a simulated colonic environment. Acute toxicity and ulceration potential were evaluated in-vivo in albino rats. Pre-formulation studies showed increased hydrophilicity but a non-significant increase in lipophilicity of the prodrug. In-vitro reversion studies suggested that the prodrug remained intact until colonic pH was attained, when the colonic microfloral enzymes (amidase) hydrolysed the FLU-GLY amide linkage, releasing the free drug. In-vivo evaluation indicated that the prodrug was much less toxic and had less ulcerogenic activity than the parent drug. Selective delivery of drugs to the colon can be useful in terms of reducing the dose administered and reducing undesirable side-effects.

Full text of DOI:10.1211/jpp.60.5.0006

JPP 2008, 60: 607–613
2008 The Authors
©
Received October 26, 2007
Accepted February 13, 2008
DOI 10.1211/jpp.60.5.0006
ISSN 0022-3573
Optimizing delivery of flurbiprofen to the colon using
a targeted prodrug approach
Anil K. Philip, Rajesh K. Dubey and Kamla Pathak
Abstract
The carboxylic group responsible for the gastric side-effects of the propionic acid derivative, flurbi-
profen, was masked temporarily to overcome these side-effects and to accomplish colon-specific
delivery of the drug. An amide prodrug (FLU-GLY) was synthesized by coupling flurbiprofen with L-
glycine. Confirmation and characterization of the structure of the synthesized prodrug included ele-
mental analysis, Fourier transform (FT)-IR, FT-NMR, mass (FAB) spectroscopy, and determinations of
R , R and R values, respectively. Aqueous solubility and lipophilicity (log P) value were determined
f
t
M
at pH 1.2, 4.0, 6.8 and 7.4. In-vitro reversion of FLU-GLY to flurbiprofen was measured at different
pHs and in a simulated colonic environment. Acute toxicity and ulceration potential were evaluated
in-vivo in albino rats. Pre-formulation studies showed increased hydrophilicity but a non-significant
increase in lipophilicity of the prodrug. In-vitro reversion studies suggested that the prodrug
remained intact until colonic pH was attained, when the colonic microfloral enzymes (amidase)
hydrolysed the FLU-GLY amide linkage, releasing the free drug. In-vivo evaluation indicated that
the prodrug was much less toxic and had less ulcerogenic activity than the parent drug. Selective
delivery of drugs to the colon can be useful in terms of reducing the dose administered and reduc-
ing undesirable side-effects.
Introduction
Inflammatory bowel disease (IBD), comprising Crohn’s disease and ulcerative colitis, is
characterized by relapsing and remitting episodes of active inflammation and chronic
mucosal injury (Hugot et al 1999). The pathogenesis of IBD remains unclear. Numerous
studies on the aetiology of IBD have shown that the diseases are the result of an exaggerated
or insufficiently suppressed immune response, which leads to the beginning and persistence
of these diseases (Helper & Rex 2001).
Currently, no drugs are available, and IBD treatment relies heavily on non-steroidal
anti-inflammatory drugs (NSAIDs), glucocorticosteroids and immunomodulators. The
primary goal of drug therapy for IBD is to reduce inflammation in the colon, requiring
frequent intake of high doses of NSAIDs, which may lead to gastric ulceration, bleeding
and other gastric complications (Dhaneshwar et al 2007). The study of Cioli et al (1979)
suggested that the direct contact of tissue with NSAIDs plays an important role in the
production of gastrointestinal (GI) tract lesions. The study also confirmed that gastric
side-effects were due to the presence of a carboxylic group in the parent drug. NSAIDs
containing carboxylic groups are poorly absorbed from the GI tract because of unfavourable
physiochemical properties. Colon-specific drug delivery (CSDD) has evolved out of the
need to overcome this barrier of the GI tract, as an ideal delivery system for the topical
treatment of diseases of the colon, namely IBD, colorectal cancer and amoebiosis. To
achieve colonic delivery, a drug needs to be chemically and biochemically stable; it
should not be absorbed in the upper intestine; and should be released abruptly into the
proximal colon, which is considered to be the optimum site for CSDD (Chourasia & Jain
2003).
Department of Pharmaceutics,
Rajiv Academy for Pharmacy,
Mathura-286001, Uttar Pradesh,
India
Anil K. Philip, Rajesh K. Dubey,
Kamla Pathak
Correspondence: Anil K. Philip,
Assistant Professor, Department
of Pharmaceutics, Rajiv Academy
for Pharmacy, Mathura-286001,
Uttar Pradesh, India. E-mail:
anilphilip@sancharnet.in,
The prodrug approach is one of several promising tools for targeting drugs to the colon.
CSDD through colon-specific prodrug activation may be accomplished by exploiting
the high activity of certain enzymes at the target site compared with non-target tissues for
conversion of the prodrug to active drug.
philipanil23@yahoo.co.in
6
07

6
08
Philip et al
The amino acid glycine is useful as a promoiety; it has Qualikems Fine Chemicals Pvt. Ltd (New Delhi, India). Tri-
broad-spectrum anti-inflammatory, cytoprotective and immu- ethylamine and n-butanol were purchased from Qualigens
nomodulatory properties (Habib et al 2006), and would also Fine Chemicals (Mumbai, India). HPLC-grade methanol and
be expected to synergize with flurbiprofen activity. It is not water, sodium sulfate (anhydrous), sodium bicarbonate and
only non-toxic but is also free from side-effects, being a natu- chloroform were purchased from Ranbaxy Fine Chemicals
ral component of our bodies. In order to reduce the gastric Ltd (New Delhi, India). Distilled water was used throughout
side-effects of flurbiprofen, structural modification with an the study. All other materials used were of analytical grade;
amino acid was carried out to mask the carboxylic group tem- those of synthetic grade were purified before use.
porarily. A strategic group attached to mask the carboxylic
group would not only protect the vulnerable group and stabi-
lize the molecule, but will also direct the drug to its target
Animals
site. An increase in hydrophilicity has also been proposed due Albino rats were purchased from the Central Drug Research
to the amino acid’s polar group (Sinha & Kumria 2003).
Institute (Lucknow, Uttar Pradesh, India) and were housed in
To meet these requirements for colon specificity and reduc- the animal house at the Department of Pharmacology, Rajiv
tion of the gastric side-effects of NSAIDs with a carboxylic Academy for Pharmacy, Mathura, Uttar Pradesh, India. The
group, a colon-specific prodrug of flurbiprofen using the amino in-vivo ulcerogenicity and acute toxicity studies were con-
acid glycine as a promoiety was synthesized, characterized and ducted with prior approval from the institutional animal ethi-
evaluated in pre-formulation and in-vitro kinetic studies. Its cal committee (IAEC/04/07/R1).
ulcerogenic activity and acute toxicity compared with those of
flurbiprofen were evaluated in-vivo in albino rats.
Synthesis of the prodrug
The flurbiprofen–glycine prodrug (FLU-GLY) was synthe-
Materials and Methods
sized in a two-step reaction (Figure 1). The first step was syn-
thesis of L-glycine methyl ester (GME) hydrochloride. Freshly
distilled thionylchloride (0.1mol) was slowly added to metha-
Materials
Flurbiprofen was a gift from Elcon Drugs Pvt. Ltd (Gurgaon, nol (300 mL) with cooling, after which 0.2mol L-glycine was
Haryana, India). L-Glycine and HPLC-grade acetonitrile were added. The mixture was refluxed for 10h at 60–70°C with
purchased from Merck (India) (Bombay, India). Thionylchlo- continuous stirring on a magnetic stirrer. The solvent was dis-
ride and N,N-dicyclohexylcarbodiimide were purchased from tilled at 64–65°C and the resulting product was collected and
Spectrochem. Pvt. Ltd (Mumbai, India). Diethylether, metha- triturated with cold diethyl ether (50mL) to remove excess
nol and N,N-dimethylformamide were obtained from dimethyl sulphite. The crude product obtained was recrystallized
Step 1
NH2
O
NH2
6
0–70°C
.
OCH2 HCl
CH2
L-glyine
COOH
ThionylchlorideMethanol
SOCl2
CH2OH
CH2
C
8 h
GME HCl
Step 2
CH2
O
C
C
OH
NH2
O
F
.
CH2
C
OCH2 HCl
GME HCl
Flurbiprofen
N.N-Dimethylformamide
N.N-Dicyclohexylcarbodiimide
Triethylamine
CH2
O
O
C
C
NH
CH2
C
O
CH2
F
flurbiprofen–glycine conjugate
Figure 1 Reaction scheme for the synthesis of flurbiprofen–glycine prodrug. Step 1 was synthesis of L-glycine methyl ester (GME) hydrochloride.
The second step was synthesis of the glycine conjugate of flurbiprofen.

Targeted prodrug delivery of flurbiprofen to the colon
609
with hot methanol by adding 20–25 mL diethyl ether followed were validated according to the International Conference on
by cooling at 0°C (Dhaneshwar etal 2006).
Harmonization ICH Q2B guidelines.
The second step was synthesis of the glycine conjugate of
flurbiprofen. Twenty mmol of flurbiprofen, was dissolved in Deterination of R value
M
60mL N,N-dimethylformamide in conical flask A; 20mmol RM values were determined in triplicate by reverse-phase
N-N-dicyclohexylcarbodiimide was added to this with continu- TLC. Silica gel GF254 TLC plates were soaked for 5 h in ace-
ous stirring for 15min. Separately, 20mmol methyl ester tone containing 3% v/v n-octanol and left to dry overnight
hydrochloride of L-glycine was dissolved in 60mL N,N- (Saha et al 2002). Sample spots of 5mL of compound solution
dimethylformamide in conical flask B, and 42mmol triethyl- were loaded at 1.5 cm intervals. The compounds were
amine was added to it at 0°C. The contents of flask A were then allowed to develop by an ascending technique in a chromato-
added into flask B. The mixture was filtered, and an equal vol- graphic chamber under conditions of equilibrium using a
ume of distilled water was added to the filtrate, followed by mobile phase of methanol, water and chloroform (ratio 14:5:1
extraction of the drug with ether. Anhydrous sodium sulfate v/v/v). The plates were dried and the developed spots were
(15g) was added to the ether layer, and the crude product was localized under ultraviolet fluorescence at 254 nm. The Rf
recrystallized with methanol.
values were determined for the compound as the average of
three readings and the corresponding RM values were calcu-
lated using the formula R = log (1/R − 1).
M
f
Confirmation of the prodrug structure
Elemental analysis and mass spectra of flurbiprofen (C, H, O)
and FLU-GLY (C, H, N, O) were carried out on an elemental
Preformulation studies
analyser (Elementar vario ELIII, Carlo Erba 1108, (Elemen- Partition coefficient
tar Analysensysteme GmbH, Hanau, Germany) and mass
spectrometer (Jeol SX 102/DA-6000, Jeol, Tokyo, Japan) at
the Central Drug Research Institute (CDRI), Lucknow, Uttar
Pradesh, India.
Partition coefficients were determined according to the
Hansch method (Avdeef et al 2000) between 10 mL n-octanol
and 10 mL buffer of varying pH (1.2, 4.0, 6.8 and 7.4).
Respective buffers and n-octanol were added to a separator
To identify the presence of organic functional groups, FT-IR
spectra of flurbiprofen and FLU-GLY were recorded in potas-
sium bromide (anhydrous grade) pellets, using a Simadzu-8400
S FTIR spectrophotometer (Simadzu, Tokyo, Japan).
(
Hicon Ltd, New Delhi, India) mounted on an automatic
shaker (Hicon). Both phases were saturated for 60 min with
intermittent shaking. Weighed amounts of flurbiprofen and
FLU-GLY were added to different separators, which were
then shaken for 30 min to achieve drug distribution into both
phases. The separators were allowed to stand for 5 min, and
the aqueous and organic layers were separated, suitably
diluted and analysed using the Cecil 4200 HPLC system
described above, with UV detector set at 249 nm and 247 nm
for FLU-GLY and flurbiprofen, respectively. Each experi-
ment was performed six times.
To determine the nature of protons and protonated groups
1
in flurbiprofen and FLU-GLY, the H-NMR spectra in CDCl
3
were recorded on a Jeol AL 300, FT-NMR spectrometer
Jeol) at 300MHz, using trimethylsilane as the internal standard;
(
the chemical shifts (d) were recorded in ppm.
Characterization of the prodrug
Thin-layer chromatography (TLC)
Aqueous solubility
TLC of flurbiprofen and FLU-GLY was performed in tripli-
cate using silica gel GF254 as the stationary phase. Sample
spots of the compound solution (5 mL) were loaded at 1.5 cm
intervals. The compounds were allowed to develop by an
ascending technique in a TLC jar, under conditions of equi-
librium using a mobile phase of methanol:water (3:1 v/v).
The plates were dried and the developed spots were localized
using ultraviolet fluorescence at 254 nm. The R values for
the compound were determined as the average of three read-
ings.
The aqueous solubilities of flurbiprofen and FLU-GLY were
determined (n = 6) by adding excess amounts of the solutes in
HCl buffer pH 1.2, acid phthalate buffer pH 4.0 and phos-
phate buffer pH 6.8 and 7.4, and equilibrating them at 37°C in
a water bath shaker. After 72 h of shaking, the samples were
withdrawn, filtered, suitably diluted and analysed using the
Cecil 4200 HPLC system described above, with UV detection
at 247 nm and 217.5 nm for FLU-GLY and flurbiprofen,
respectively.
f
Reversion studies
HPLC
The reversion of the FLU-GLY to flurbiprofen was studied in
HCl buffer pH 1.2, phthalate buffer pH 4.0, phosphate buffer
pH 6.8, phosphate buffer pH 7.4 and phosphate buffer pH 6.8
containing fresh rat faecal content (20% w/v – to mimic the
colonic environment) at 37 ± 0.5°C. The ionic strength
HPLC analysis was carried out in triplicate on a Cecil 4200
system (Cecil Instruments Ltd, Cambridge, UK) using a
2
50 × 4.6 cm C reverse-phase column, particle size 5 mm
18
(
Thermo Electron, Fife, Scotland, UK). The mobile phase
was acetonitrile, methanol and water (ratio 2:1:2 by volume),
−
1
(
m = 0.5) was kept constant for each buffer by adjusting with a
delivered at a flow rate of 1 mL min . Samples were injected
onto the system using a 25 mL Hamilton syringe. Detection
was at 247 nm (Shorbagi & El Aboul 1996). The analytical
performance parameters (specificity, linearity, range, preci-
sion, accuracy, limit of detection and limit of quantification)
calculated amount of potassium chloride (Martin 1999). FLU-
GLY was dissolved in sufficient volume of buffer to give a
−
1
concentration of 1000 mgmL ; 1 mL of the prodrug solution
was taken into glass vials and diluted to 10 mL with the buffer

6
10
Philip et al
−
1
to give a final concentration of 100 mgmL . The vials were the 14th day, and the change in weight calculated and
kept in a water bath at 37 ± 0.5°C. For analysis, 5 mL of solu- recorded. Rats were killed by cervical dislocation on day 14.
tion was withdrawn from the vials at different time points and
Gross pathological evaluation was performed, and the
shaken with equal amount of n-butanol, to extract free flurbi- weights of the heart, liver, kidney, stomach and spleen were
profen reverted from FLU-GLY. The concentration of flurbi- measured and recorded immediately afterwards. Relative
profen was estimated directly from the n-butanol layer using organ weight (organ to body weight ratio) was also calculated
the Cecil 4200 HPLC system described above with UV detec- for each organ. Histopathological examination was performed
tion at 315.5 nm. Each measurement was repeated six times.
on routinely prepared sections of stomach tissues; the tissues
were fixed in 10% v/v formalin immediately after removal to
avoid autolysis, and weighed.
In-vivo evaluation of FLU-GLY
Ulcerogenicity
Ulcerogenic activity was determined by the cold stress method
Statistical analysis
(
Rainsford & Whitehouse 1980), an acute model, used to The cumulative percentage of flurbiprofen released from all
determine ulcerogenic potency of a drug at 10 times the nor- the formulations (n= 6) in the dissolution medium, with and
mal dose. Albino rats weighing 150–200 g were fasted over- without rat caecal contents, was compared, and the statistical
night before administration of the compounds; water was significance tested using two-way analysis of variance fol-
available ad libitum. The animals were randomly distributed lowed by Dunnett’s multiple comparison t-test. P < 0.05 was
into control, drug and prodrug groups of six animals each. The considered significant.
control group received the drug vehicle – carboxymethyl cel-
lulose (CMC, 1% w/v) by gavage. The drug (flurbiprofen) and
prodrug (FLU-GLY) were administered orally, as fine parti-
cles suspended in 1% w/v CMC by continuous stirring. Fol-
Results and Discussion
lowing oral administration of 5 mL of the aqueous drug Characterization of flurbiprofen and the FLU-
suspensions, the animals were stressed by exposure to cold GLY conjugate
(
−15°C for 1 h), with each in a separate polypropylene cage to
The FLU-GLY prodrug was synthesized in two steps: the
intermediate GME HCl was synthesized from glycine, and
then coupled with flurbiprofen to form the FLU-GLY prod-
rug. Flurbiprofen and FLU-GLY were characterized from Rf,
R , and R values. R was 0.26 ± 0.13 for GLY, 0.39 ± 0.22
ensure equal cold exposure. Animals were killed by cervical
dislocation 2 h after drug administration.
The stomach and duodenal parts were opened along the
greater curvature and the number of lesions was examined by
means of a magnifying lens. The size of ulcers was measured by
Microimage process software (DA1-180M v 2.01; Sunny Inter-
national United Co., Ltd, Zhejiang, China), using an Olympus
SP 350 camera (Olympus, Tokyo, Japan). Ulcers were scored as:
M
t
f
for GME, 0.70 ± 0.09 for flurbiprofen and 0.75 ± 0.13 for
FLU-GLY. R values were −0.486 ± 0.11 for flurbiprofen
M
and −0.276 ± 0.18 for FLU-GLY.
R
values were
t
4
.16 ± 0.16 min for flurbiprofen and 2.07± 0.21 min for FLU-
0
for normal-coloured stomach; 0.5 for red colouration; 1 for
GLY, indicating the formation of a new product.
spot ulcers; 1.5 for haemorrhagic streaks; 2 for ulcers of 3mm up
to 5mm; and 3 for ulcers of 5mm and greater (Kulkarni 1999).
The synthesized prodrug was further characterized by ana-
lytical techniques. The theoretically elemental analysis calcu-
lated for FLU-GLY (C H FNO ) was C: 68.59%, H:
1
8
18
3
Acute toxicity study
5.71%, N: 4.44%, O: 15.22%; practically, elemental analysis
was C: 68.51%, H: 5.69%, N: 4.40%, O: 15.19%. FT-IR (cm )
−1
Single-dose acute toxicity studies were carried out following
OECD guidelines No. 401 (OECD 1987). Animals were
fasted overnight but with free access to water. Rats were
divided into three groups (control, drug and prodrug) each
comprising three males and three females and were weighed
just before the experiment. A dose of 2000 mg kg of drug or
prodrug, suspended in demineralized water using 1% w/v
CMC as suspending agent, was administered orally by gav-
age. The animals in the control group received 1% w/v CMC
in demineralized water. In each case, the volume adminis-
tered was 5 mL. Any toxicity or mortality was recorded at 0.5,
(
KBr): 2990 (N-H stretching amide), 1735 (C = O stretching
ester), 1650 (C = O stretching amide), 1535 (N-H bending
amide), 1250 (C-N stretching), 1090 (C-O stretching ester).
1
H-NMR (CDCl ): 7.6 (benzene, 3H, m),7.48 (benzene, 5H, m),
3
−
1
3.9 (CH, 1H, q), 1.6 (CH3, 3H, d), 7.27–7.42 (CONH 1H, m),
3
.7 (CH of ester 3H, s),1.48 (GH , 2H, d).
3
2
High-resolution FAB-MS theoretical calculation of M+
for C H FNO resulted in a value of 315.15; the observed
1
8
18
3
value was 315.
Thus, these analytical techniques confirmed the structure,
molecular formula and molecular weight of the synthesized
FLU-GLY conjugate.
1
, 2, 4 and 6 h after dosing and twice a day thereafter for 14
days. All observations were recorded systematically, with
individual records being maintained for each animal. Cage-
side observations included evaluation of skin and fur; eyes;
respiratory effects; autonomic effects, such as salivation,
Preformulation studies
diarrhoea and urination; and central nervous system effects, Aqueous solubility and partition coefficient
including tremors and convulsions, changes in the level of The aqueous solubility and partition coefficients of flurbiprofen
activity, gait and posture, reactivity to handling or sensory and FLU-GLY in different buffers (pH 1.2, 4.0, 6.8 and 7.4) are
stimuli and altered strength. The rats were weighed again on shown in Table 1. Aqueous solubilities of flurbiprofen and

Targeted prodrug delivery of flurbiprofen to the colon
611
Table 1 pH solubility profile and partition studies of flurbiprofen and flurbiprofen–glycine (FLU-GLY)
Solubility (mg mL-1)
log P
FLU
FLU-GLY
FLU
FLU-GLY
HCl buffer, pH 1.2
0.22 ± 0.12
0.28 ± 0.23
4.87 ± 0.15
7.35 ± 0.18
1.38 ± 0.09
1.66 ± 0.13
5.77 ± 0.22
8.03 ± 0.10
3.55 ± 0.11
2.66 ± 0.18
2.32 ± 0.24
2.22 ± 0.14
3.69 ± 0.12
2.85 ± 0.15
2.55 ± 0.19
2.25 ± 0.11
Phthalate buffer, pH 4.0
Phosphate buffer, pH 6.8
Phosphate buffer, pH 7.4
Data are mean ± s.d. (n = 6).
FLU-GLY increased with increasing pH, which is probably due
to an increase in the ionization of the compounds as the pH
increased. The solubility of FLU-GLY was higher than that of
flurbiprofen across this pH range. This may be due to the pres-
8
7
6
0
0
0
HCl buffer pH 1.2
ence of highly polar groups, namely −NH and −COOH in gly-
50
2
Acid phthalate buffer pH 4.0
Phosphate buffer pH 6.8
Phosphate buffer pH 7.4
Phosphate buffer
cine, which increase the polarity of the conjugate much more so
than the less polar carboxylic group of flurbiprofen. This theory
is supported by reverse-phase HPLC: the C18 column used as
the stationary phase had less affinity for polar drugs (i.e. polar
4
3
0
0
pH 6.8 + rat faecal contents
2
0
drug would elute first). R values for flurbiprofen and FLU-
10
0
t
GLY were 4.16 and 2.07min, respectively, confirming the
higher polarity of the conjugate compared with flurbiprofen.
The enhancement of log P values of the prodrug compared
with flurbiprofen was negligible at 95% confidence interval
0
10 20 30 40 50 60
Time (h)
Figure 2 In-vitro reversion of flurbiprofen–glycine to flurbiprofen
FLU) at different pHs and in a colonic environment.
(
P > 0.3241). The slight increase might be due to the more
(
lipophilic character of the prodrug, which is supported by the
RM values of the conjugate and the parent compound. The
higher the RM value, the higher the lipophilicity of the com-
pound. RM values of FLU-GLY and flurbiprofen were −0.276
and −0.486, respectively; thus, the prodrug is slightly more
lipophilic than flurbiprofen.
index (median ± range) were 2 ± 1 for control animals,
6.2 ± 4.1 for the flurbiprofen group and 17.0 ± 1.2 for the
6
FLU-GLY group, indicating marked differences in the ulcer-
ogenic activity of the prodrug vs flurbiprofen (P = 3.234). No
haemorrhagic or red spots were found on the stomach walls
of control animals (Figure 3A).The stomach walls of animals
treated with flurbiprofen showed severe congestion, numer-
ous haemorrhagic spots, streaks, erosion of the gastric
mucosa, deep ulceration and necrotic cells (Figure 3B). Stom-
achs from animals treated with FLU-GLY showed haemor-
rhagic and red spots but no necrosis of the cells (Figure 3C).
On comparing histopathology of the stomachs of control rats
of (Figure 4A) and those treated with drug (Figure 4B) and
prodrug (Figure 4C), more severe haemorrhage, ulcers and
necrosis were evident in the drug group than the prodrug
group.
In-vitro reversion study
In-vitro kinetic studies confirmed negligible reversion of FLU-
GLY in the gastric environment (HCl buffer pH 1.2) and caecal
environment (acid phthalate buffer pH 4.0), and only 6.05% and
8.14% of the prodrug reverted to flurbiprofen in the intestinal
environment (phosphate buffer pH 6.8 and 7.4, respectively)
over a period of 8h (P=0.3452, t=0.876; Figure 2). Reversion
was also studied in phosphate buffer pH 6.8 in the presence of
fresh rat faecal matter (20% w/v), to confirm the colonic break-
down of the prodrug. In this simulated colonic environment,
67.83% of the prodrug reverted to flurbiprofen over a period of
48h, with first-order kinetics. The marked reversion (P=3.657,
t=2.891) in the colonic environment was due to hydrolysis of
the prodrug catalysed by the amidase enzyme released by
colonic microflora. In-vitro reversion studies suggested that the
prodrug might bypass the GI tract and would be reverted to flur-
biprofen by amidase enzyme in the colon.
Acute toxicity
Following administration of single doses of flurbiprofen and
FLU-GLY, there were no overt signs and symptoms of toxic-
ity in any of the dosed groups on the first day, except for
weakness. One rat in the drug group was found dead on the
first day of study. The anomalies waned over time and all sur-
viving animals became overtly normal until termination of
the study. There was no significant difference in the activities
observed and no significant difference (one-way ANOVA,
In-vivo evaluation of FLU-GLY
Ulcerogenic study
The ulcerogenic activity of flurbiprofen and FLU-GLY was P < 0.05) in body weight gain (Table 2) or organ weight at the
determined by the cold stress method. Values for the ulcer end of the study (P = 0.3345, t = 1.23).

6
12
Philip et al
A
B
C
Figure 3 Evaluation of ulcerogenic activity in the stomachs of albino rats treated with drug vehicle (1% w/v CMC (A); flurbiprofen (1000 mgkg−1)
−
1
(
B) and flurbiprofen–glycine (1000 mgkg ) (C).
A
B
C
D
E
F
Figure 4 Histopathology of stomachs from albino rats treated with drug vehicle (1% w/v CMC (A); flurbiprofen (1000 mgkg−1) (B) and flurbipro-
−
1
fen-–glycine (1000 mgkg ) (C), and in stomachs from rats on the 14th day of the acute toxicity study, treated with drug vehicle (2% w/v CMC (D);
−
1
−1
flurbiprofen (2000 mgkg ) (E) and flurbiprofen–glycine (2000 mgkg ) (F).
Table 2 Weight (in grams) of the albino rats during the toxicity study
The microscopic appearance of liver, kidney, heart,
spleen, stomach and intestine was essentially similar in all the
animals.
Treatment day 0
day 14
Weight gain % increase
in weight
Result obtained in albino rats showed that FLU-GLY
exhibited moderately less acute toxicity than flurbiprofen.
The 50% lethal dose (LD50) was estimated to exceed
Control
Drug
Prodrug
161 ± 11.31 173 ± 12.72 12 ± 1.41
173 ± 16.26 184 ± 18.38 10 ± 2.12
171 ± 14.84 183 ± 16.97 11 ± 2.13
7.45
6.05
6.70
−
1
2000 mgkg for both FLU-GLY and flurbiprofen. The
marked reduction in the ulcerogenic activity and less toxic
effects of the prodrug (FLU-GLY) compared with flurbipro-
fen (one-way ANOVA, P < 0.05) may be due to temporary
masking of the carboxylic group of flurbiprofen, making the
prodrug safe for oral use. These results, coupled with the esti-
mation of the higher LD50, indicate that the prodrug us safe,
even at higher doses.
Values are mean ± s.d. (n = 6).
Gross pathological examination at autopsy showed cysts
in the stomach and haemorrhagic abdomen in rats that had
received a single dose of flurbiprofen. No changes were
observed in the other organs. Histopathological evaluation
of tissues from the stomachs of the drug group (Figure 4E)
Conclusions
revealed a number of lesions, such as eosinophilic reaction The synthesized prodrug of flurbiprofen had increased solu-
at the junction of the stomach and oesophagus, severe bility, lower toxicity and less ulcerogenic activity than the
haemorrhage, glandular destruction and severe necrosis. parent drug flurbiprofen. Thus, this prodrug approach solves
The prodrug group (Figure 3F) showed oedema, haemor- not only the formulation problem of flurbiprofen (lower
rhage and a small amount of glandular destruction. Normal aqueous solubility, BCS class II drug) and reduces adverse
glandular arrangement was observed in the control group effects, but also enables targeted delivery of the drug to
(
Figure 3A).
the colon. Selective delivery to the colon would be more

Targeted prodrug delivery of flurbiprofen to the colon
613
effective in therapy in terms of reduced dose and fewer Helper, D. J., Rex, D. K. (2001) Inflammatory bowel disease. Endos-
copy 33: 140–146
Hugot, J. P., Zouali, H., Lesage, S., Thomas, G. (1999) Etiology
side-effects.
of the inflammatory bowel diseases. Int. J. Colorectal Dis. 14:
2
–9
Kulkarni, S. K. (1999) Pharmacology of gastro-intestinal tract
GIT). In: Kulkarni, S. K. (ed) Handbook of experimental
pharmacology, 3rd edn. New Delhi: Vallabh Prakashan,
pp. 148–150
Martin, A. (1999) Ionic equilibria. In: Martin, A., Bustamante, P.,
Chun, A. H. C. (eds) Physical pharmacy, 4th edn. New Delhi:
B. I. Waverly Pvt Ltd, pp. 143–168
References
(
Avdeef, A., Berger, C. M., Brownell C. (2000) pH metric solubility:
correlation between the acid-base titration and the saturation shake
flask solubility-pH methods. Pharm. Res. 17: 85–89
Chourasia, M. K., Jain, S. K. (2003) Pharmaceutical approaches to
colon targeted drug delivery systems. J. Pharm. Sci. 6: 33–66
Cioli, V., Putzolu, S., Rossi, V., Carrandino, C. (1979) The role of direct OECD (1987) Acute oral toxicity test method. In: OECD Guidelines
tissue contact in the production of gastrointestinal ulcers by anti-
for testing of chemicals, No. 404. Organization for Economic
inflammatory drugs in rats. Toxicol. Appl. Pharmacol. 50: 283–289
Cooperation and Development, Paris, France
Dhaneshwar, S. S., Nagpal, D., Singh, R., Gairola, N., Bodhankar, S. Rainsford, K. D., Whitehouse, M. W. (1980) Anti-inflammatory/
(
2006) Mutual azo prodrug of 5-aminosalicylic acid for colon tar-
anti-pyretic salicylic esters with low gastric ulcerogenic activity.
Agents Actions 10: 451–456
Saha, R. N., Ravishankar, D., Jadhav, P. R., Sanjeev, C. (1992)
Determination of flurbiprofen in pharmaceutical formulations by
UV spectrophotometry and liquid chromatography. Analytica
Chimica Acta 463: 207–217
geted drug delivery: synthesis, kinetic studies and pharmacologi-
cal evaluation. Ind. J. Pharm. Sci. 68: 171–178
Dhaneshwar, S. S., Gairola, N., Kandpal, M., Vadnerkar, G., Bhatt, L.
(
acid with L-tryptophan: synthesis, kinetic studies and evaluation of
its mitigating effect in trinitrobenzenesulfonic acid-induced colitis Shorbagi, A., El Aboul, F. T. (1996) New prodrug approach for
2007) Colon-specific, mutual azo prodrug of 5-aminosalicylic
in rats. Bioorg. Med. Chem. 15: 4903–4909
Habib, M. M., Hodgson, H. J., Davidson, B. R. (2006) The role of
amino acids and amino-acid like drugs. Eur. J. Med. Chem. 31:
165–169
glycine in hepatic ischemia-reperfusion injury. Curr. Pharm. Sinha, V. R., Kumria, R. (2003) Microbially triggered drug delivery
Design 12: 2953–2967
to colon: review. Eur. J. Med. Chem. 18: 3–18