Characterization of mefenamic acid-guaiacol ester: stability and transport across Caco-2 cell monolayers.

Article abstract of DOI:10.1023/A:1016470523923

PURPOSE: Prodrug of non-steroidal anti-inflammatory drugs (NSAIDs) or NSAIDs linked with guaiacol have been reported to suppress gastrointestinal (GI) toxicity or induce GI protective effect. In this study. mefenamic-guaiacol ester was synthesized and its physicochemical properties. stability, and transport across Caco-2 monolayers were investigated. METHODS: Synthesis of the ester was carried out using mefenamic acid, guaiacol. N. N'-dimethylaminopyridine, and N,N'dicyclohexylcarbodiimide. The hydrolysis of the ester was investigated in aqueous buffer solutions pH 1-12 as well as in Caco-2 homogenate, human plasma, and porcine liver esterase. Caco-2 cell monolayers were utilized for transport studies. Due to the high lipophilicity of the ester with a calculated logP of 6.15, bovine serum albumin (BSA, 4%) was included in the receiver compartment to obtain a good in vitro-in vivo correlation. Permeation of the ester was assessed with or without the exposure of cells to PMSF, an inhibitor of esterase. RESULTS: The ester was stable at a wide pH range from 1-10. However, it was hydrolyzed by enzymes from porcine liver esterase and Caco-2 homogenate. With the PMSF exposure on the apical (AP) side and in the presence of 4% BSA on the basolateral (BL) side, the transported amount of the ester from AP-to-BL direction was 14.63% after 3 hr with a lag time of 23 min. The Papp for the ester was 4.72 x 10(-6) cm s(-1). CONCLUSION: The results from hydrolysis studies indicate that this ester is a prodrug. The Papp value suggests the moderate absorption characteristic of the compound. The accumulation of this highly lipophilic ester in Caco-2 cells is reduced in the presence of BSA.

Full text of DOI:10.1023/A:1016470523923

Pharmaceutical Research, Vol. 19, No. 7, July 2002 (© 2002)
Research Paper
for the conversion of arachidonic acid into prostaglandins.
NSAIDs are frequently associated with high incidences of
gastroenteropathy, ranging from mild gastric upset to life-
threatening ulceration and hemorrhage (1). GI injury pro-
duced by NSAIDs is generally believed to be a result of two
different mechanisms. The first mechanism involves a local
irritant produced by acidic group of the NSAIDs. The second
effect is attributed to blockage of prostagandin biosynthesis in
the GI tract, which inhibits its cytoprotective effect.
Prodrugs that temporarily mask the acidic group of
NSAIDs have been reported to suppress GI toxicity due to
the local action mechanism (2). However, esterified NSAIDs
and some related molecules have proven to retain anti-
inflammatory activities. It has been reported that guaiacol is
able to inhibit prostaglandin biosynthesis like the classic
NSAID but it does not induce gastric damage (3). The guaia-
col-esterified ibuprofen (methoxybutropate) has been avail-
able for sometimes as analgesic (4), and its water-soluble for-
mulations have recently been developed (5–6). Gastric dam-
age caused by this drug has been reported to be very low (3).
Amtolmetin guacyl is another analgesic derived from an ami-
dation of tolmetin by glycine, which is in turn linked to guaia-
col by an ester bond (7). Amtolmetin guacyl has been re-
Characterization of Mefenamic
Acid-Guaiacol Ester: Stability and
Transport across Caco-2
Cell Monolayers
1
,4
Vimon Tantishaiyakul,
Kamonthip Wiwattanawongsa, Sirirat Pinsuwan,
1
2
1
1
Srirat Kasiwong, Narubodee Phadoongsombut,
2
2
Sanae Kaewnopparat, Nattha Kaewnopparat, and
Yon Rojanasakul
3
Received March 15, 2002; accepted March 26, 2002
Purpose. Prodrug of non-steroidal anti-inflammatory drugs
(NSAIDs) or NSAIDs linked with guaiacol have been reported to
suppress gastrointestinal (GI) toxicity or induce GI protective effect.
In this study, mefenamic-guaiacol ester was synthesized and its phys-
icochemical properties, stability, and transport across Caco-2 mono-
layers were investigated.
Methods. Synthesis of the ester was carried out using mefenamic acid, ported to exhibit a good gastric tolerability and produce a
guaiacol, N, NЈ-dimethylaminopyridine, and N, NЈdicyclohexylcarbo- gastroprotective effect (7–8). These studies provide the basis
diimide. The hydrolysis of the ester was investigated in aqueous
for combining guaiacol with other molecules of NSAIDs to
buffer solutions pH 1-12 as well as in Caco-2 homogenate, human
obtain an analgesic with low GI toxicity. In this study, mefe-
plasma, and porcine liver esterase. Caco-2 cell monolayers were uti-
namic-guaiacol ester was synthesized and its physicochemical
lized for transport studies. Due to the high lipophilicity of the ester
properties, stability, and transport across Caco-2 monolayers
with a calculated logP of 6.15, bovine serum albumin (BSA, 4%) was
were investigated.
included in the receiver compartment to obtain a good in vitro-in vivo
correlation. Permeation of the ester was assessed with or without the
exposure of cells to PMSF, an inhibitor of esterase.
Results. The ester was stable at a wide pH range from 1-10. However,
it was hydrolyzed by enzymes from porcine liver esterase and Caco-2
homogenate. With the PMSF exposure on the apical (AP) side and in
the presence of 4% BSA on the basolateral (BL) side, the transported
amount of the ester from AP-to-BL direction was 14.63% after 3 hr
−6
−1
with a lag time of 23 min. The Papp for the ester was 4.72 × 10 cm s
.
Conclusion. The results from hydrolysis studies indicate that this
ester is a prodrug. The Papp value suggests the moderate absorption
characteristic of the compound. The accumulation of this highly li-
pophilic ester in Caco-2 cells is reduced in the presence of BSA.
KEY WORDS: mefenamic acid; guaiacol; prodrug; caco-2 cells; hy-
drolysis; transport.
MATERIALS AND METHODS
INTRODUCTION
Materials
Non-steroidal anti-inflammatory drugs (NSAIDs) are
Caco-2 cells were obtained from the American Tissue
widely used for the treatment of inflammatory diseases. The Culture Collection (Rockville, Maryland) at passage 18. All
mechanism of action of NSAIDs involves the inhibition of chemicals used for culturing the Caco-2 cells were obtained
cyclooxygenase (COX), the rate-limiting enzyme responsible from Gibco BRL (Life Technologies, New York). Porcine
liver esterase (#E-3128) and phenylmethyl sulfonylfluoride
(
PMSF) were obtained from Sigma (St. Louis, Missouri).
1
Department of Pharmaceutical Chemistry, Faculty of Pharmaceuti-
cal Sciences, Prince of Songkla University, Hat-Yai, Songkhla 90112
Thailand.
Department of Pharmaceutical Technology, Faculty of Pharmaceu-
tical Sciences, Prince of Songkla University, Hat-Yai, Songkhla
Other reagents and solvents were purchased from common
suppliers and were used as received.
Melting point was determined on a MEL-TEMP II cap-
illary melting point apparatus and is uncorrected. Proton
magnetic resonance spectra were recorded on a Bruker Spec-
trospin 300 NMR. Chemical shifts are reported in parts per
million (ppm, ␦ units) and peak multiplicities are expressed as
2
90112 Thailand.
3
4
Department of Basic Pharmaceutical Sciences, School of Pharmacy,
West Virginia University, Morgantown, West Virginia 26506.
To whom Correspondence should be addressed. (e-mail: follows: s, singlet; d, doublet; m, multiplet. The IR spectra
tvimon@ratree.psu.ac.th)
were recorded on a Perkin Elmer Model 1600 FT-IR spec-
1013
0724-8741/02/0700-1013/0 © 2002 Plenum Publishing Corporation

1
014
Tantishaiyakul et al.
trometer. Mass spectrum was recorded on a Micromass Plat- tonitrile to 10 ml of buffer solutions, pre-equilibrated at 37°C,
form II mass spectrometer. Yield is of purified product and in screw-capped test tubes. The initial concentration was 2 ×
−
4
was not optimized. The purity of the ester was assessed by 10 M. At appropriate times, the reaction samples were with-
analytical HPLC as described below.
drawn and analyzed by HPLC as described above. Pseudo-
first order rate constants (k) were calculated from the slopes
of linear plots of the logarithm of residual ester concentration
against time, and the corresponding half-life obtained from
the equation: t1/2ס 0.693/k. Triplicate samples were analyzed,
and the mean value of the rate constant was calculated.
Synthesis of Mefenamic Acid - Guaiacol Ester
To a solution of mefenamic acid (3.0 g, 12.43 Mmol) in
dry CH Cl (150 ml) was added guaiacol (1.39 g, 11.19 Mmol),
2
2
the mixture was cooled at 0°C, followed by the addition of N,
NЈ-dimethylaminopyridine (0.06 g, 0.5 Mmol), and N, NЈdi-
cyclohexylcarbodiimide (3.10 g, 15.08 Mmol). The reaction Enzymatic Stability
mixture was stirred at room temperature overnight. The pre-
cipitated N, NЈ-dicyclohexylurea was removed by filtration.
The degradation of the ester was studied at 37°C in
The filtrate was concentrated in vacuo, and the residue was Caco-2 homogenate, human plasma, and porcine liver ester-
purified by column chromatography using CH Cl :hexane (1: ase preparations. Caco-2 homogenate was prepared accord-
2
2
1
) as the eluent. The product was recrystallized from CH Cl / ing to the method of Augustijns et al. (10) with slight modi-
2 2
2
MeOH to afford 1.9 g (51%) of mefenamic-guaiacol ester as fications. Briefly, Caco-2 cells grown in 75-cm culture flask
yellow crystals: >99% purity; mp 111.5–111.9°C; IR (KBr) for 21–25 days were washed with ice-cold phosphate-buffered
−
1 1
3
337, 1693 cm ; H NMR (300 MHz, CDCl ) ␦ 2.08 (s, 3H, saline (PBS), pH 7.4. The cells were scraped off, homogenized
3
ArCH ), 2.23 (s, 3H, ArCH ), 3.76 (s, 3H, ArOCH ), 6.69 (m, in 4 ml ice-cold PBS using a glass-teflon polytron homog-
3
3
3
2
1
3
H, aromatic), 7.25-6.90 (m, 8H, aromatic), 8.15 (d, J ס 8 Hz, enizer (Thomas, Pennsylvania), and centrifuged at 12,000 rpm
H, aromatic), 9.06 (s, 1H, NH, broad); MS: (IE, 70eV) m/z for 10 min at 4°C. The resultant supernatant was used as
−
1
47 [M].
Caco-2 homogenate (protein content ס 0.63 mg ml ). Hu-
man plasma was obtained from the Songklanakarin Hospital
Blood Bank (Hat-Yai, Songkhla). For stability studies, hu-
man plasma was diluted to 80% (v/v) with phosphate buffer,
Analysis of Samples
For the detection and quantitation of mefenamic acid
and its ester, a high-performance liquid chromatography
−1
pH 7.4 (protein content ס 49.57 mg ml ). Porcine liver es-
−1
terase was used at 0.06 unit ml
.
(HPLC) system capable of separating mefenamic acid and
The ester was incubated at the concentration of 12 ␮M
containing 0.6% acetonitrile), 8 ␮M (containing 0.6% aceto-
guaiacol from the ester was used. The HPLC system (Waters,
Milford, Massachusetts) consisted of Waters 600 Pump, Wa-
ters 717 plus Autosampler, Waters 486 Tunable Absorbance
Detector, and Waters 746 Data Module. The analytical col-
umn was Rexchrom ODS, 5␮m, 15cm × 4.6 mm, (Regis Tech-
nologies, IL, USA) equipped with a precolumn packed with
Novapak RP-8, 5 ␮m, 3.9 × 20 mm (Waters, Milford, Massa-
chusetts). The mobile phase was a mixture of 60% acetonitrile
and 40% phosphate buffer pH 3. The flow rate was 1.0-ml
(
nitrile), and 50 ␮M (containing 4% DMSO) in prewarmed
0% human plasma in phosphate buffer pH 7.4, porcine liver
8
esterase and Caco-2 homogenate, respectively. The mixtures
were kept in a water bath at 37°C, and at appropriate inter-
vals, samples (250 ␮l) of the reaction solution were withdrawn
and analyzed by HPLC as described above. Reactions were
followed for at least two half-lives. Apparent half-lives for the
disappearance of ester and pseudo-first order rate constants
were calculated as previously described.
−
1
min . The eluents were detected at 280 nm. The retention
times were 6.7 and 19.8 min for mefenamic acid and the ester,
respectively.
For determination in biologic media, samples were de- Solubility Determination
proteinized by adding 500 ␮l of acetonitrile. After mixing and,
in the case of determination of accumulated amount of the
The solubility of the ester and mefenamic acid were de-
ester in Caco-2 cell monolayers, sonication for 5 min, the termined at room temperature by adding an excess amount of
samples were centrifuged for 10 min at 12,000 rpm at 4°C. ester to the solvent in screw-capped test tubes. The mixture
Aliquots of clear supernatant were then injected into the was sonicated for 15 min and then rotated for 48–72 h to
HPLC system and analyzed as described above.
ensure equilibrium. Saturated solution of the ester was fil-
®
Protein content was determined using a Coomassie pro- tered through a 0.5-␮m membrane filter (Millipore Co., Bed-
tein assay reagent (Pierce, Rockford, Illinois) according to the ford, Massachusetts), and an aliquot of the filtrate was diluted
method of Bradford (9). BSA was used as a standard.
with a mobile phase and analyzed by HPLC. Experiments
were performed in triplicate.
Chemical Stability
The hydrolysis of ester was carried out in aqueous buffer Transport Studies
solutions, pH 1–12, at 37°C. The buffer systems used were
HCl (pH 1–2), acetate (pH 3-5), phosphate (pH 6–8), and
NaOH (10–12). The ionic strength of 0.04 was maintained for 60,000 cells cm on Transwell (Costar, USA) polycarbon-
each buffer by adding an appropriate amount of KCl. To ate membrane (24 mm diameter, 4.7 cm , 3.0 ␮m pore size).
For transport studies, cells were seeded at a density of
−
2
®
2
check for buffer catalysis, three different buffer concentra- Cells were grown at 37°C in an atmosphere of 5% CO and
2
tions were used (0.01, 0.02, and 0.05 M). The reaction was 90% relative humidity. Cells were maintained in Dulbecco’s
initiated by adding 75 ␮l of a stock solution of ester in ace- Modified Eagles Medium (DMEM) containing 10% fetal calf

Transport and Stability of Mefenamic Acid-Guaiacol Ester
1015
serum, 1% nonessential amino acids, 1% L-glutamine, 1% of remaining ester and mefenamic acid, mass balance (>91%)
−
1
antibiotic-antimycotic mixture (10,000 U ml benzylpenicil- was achieved in all experiments. The observed first order rate
−
1
−1
lin, 10,000 ␮g ml streptomycin, and 25 ␮g ml amphotericin constants and apparent half-lives in aqueous buffer are shown
B). The medium was changed every other day. The cells from in Table I. There is no buffer catalysis in buffer systems used.
a passage number between 42 and 48 were used in the ex- The pH-rate profile of ester hydrolysis is shown in Fig. 1. The
periments. After 21–23 days in culture, the Caco-2 cell mono- total rate equation for hydrolysis can be written as:
layers were used for transport studies. Transport medium
dc
(
TM) consisted of Hanks Balanced Salt Solution (HBSS)
−
= kobsc
dt
supplemented with 25 mM glucose, adjusting to pH 6.5 for the
apical (AP) side and 7.4 for the basolateral (BL) side.
The integrity of each batch of cells was tested by mea-
suring the transepithelial electrical resistance (TEER) using
kobs = k + k
͓OH−͔
−
OH
0
Where kobs is the apparent first-order rate constants, [OH−]
is the hydroxide ion concentration, kOH is the rate constant
for specific base-catalyzed hydrolysis, and k is the apparent
®
the Millicell -ERS system (Millipore Co., Bedford, Massa-
−
chusetts) before and after the experiments. Before the experi-
ments, cells were equilibrated with HBSS for 20 min at 37°C,
and the TEER was determined. The TEER value corrected
0
first-order rate constant for spontaneous or water-catalyzed
hydrolysis. As seen in Fig. 1, the minimum rate of hydrolysis
occurs at pH below 10, where the rate of reaction is appar-
ently pH-independent. The average of k (or k ) at this pH
for the blank filter resistance was in the range of 250–300
2
⍀
.cm .
obs
0
Transport studies were performed in triplicate for a com-
pound in the AP-to-BL or BL-to-AP direction at 37°C in a
−4
−1
range is 2.65 × 10 min (Table I). With increasing pH
above 10, the ester degrades progressively faster to yield me-
fenamic acid and guaiacol. The plot of log kobs vs. pH (10–12)
gave a straight line with a slope of +1.11, indicating that the
hydroxide ion concentration is the major factor that influ-
−
1
shaking water bath (85 strokes min ). Permeation of the
ester through Caco-2 cell monolayers was assessed with or
without the exposure of cells to PMSF (0.5 mM), an inhibitor
of serine protease/mammalian esterase, for 15 min on the AP
side prior to transport experiments.
ences the hydrolysis rate of ester. The rate equation at these
pHs can be written as: kobs = kOH
determined from this kinetic model is 4.78 × 10 min
−
−
[OH ]. The k
−
value
.
OH
TM containing the ester at the concentration of about 30
−2
−1
␮
M (containing 4% DMSO) was added to the donor com-
The susceptibility of the ester to undergo potential en-
zymatic hydrolysis, was studied in vitro at 37°C in phosphate
buffer, pH 7.4, containing either 80% human plasma, porcine
liver esterase, or Caco-2 homogenate. The observed half-lives
for hydrolysis in these biologic media are listed in Table I.
The data indicate that enzymes from Caco-2 homogenate and
porcine liver esterase markedly increase the rate of hydrolysis
as compared to that from human plasma. Because the protein
content in human plasma is higher, it is possible that the
observed slow rate of hydrolysis in plasma may be due to
binding of the ester to plasma proteins, which in effect pro-
tects some of the substrate molecules from hydrolysis. This
result is consistent with previous findings by Brunner-Guenat
et al. (11) and Bundgaard et al. (12). A typical concentration-
time profile for the degradation of ester in Caco-2 homog-
enate is shown in Fig. 2. The result shows the rapid degrada-
tion of ester and the release of free mefenamic acid with a
partment (AP, 1.5 ml or BL, 2.6 ml). TM in the presence or
absence of BSA (4%) was added to the receiver compartment
(BL or AP), and samples were removed at various times (i.e.,
3
2
0, 60, 90, 120, 150, and 180 min) from both sides (receiver,
00 ␮l; donor 50 ␮l). The sample volume removed from the
receiver side was replaced with fresh, prewarmed TM in the
presence or absence of BSA (4%). Samples were kept at
−
80°C until HPLC analysis. The total amount of parent com-
pound and ester was used for the calculations. The perme-
−
1
ability [apparent permeability coefficient, Papp (cm s )] of
the compound was calculated according to the equation:
1
dQ
Papp
=
AC0 dt
Where dQ/dt is the rate of appearance of drug in the receiver
−
1
compartment (␮mole s ), C is the initial drug concentration
−1
0
half-live of 22.4 min. In the presence of 0.06 unit ml of
porcine liver esterase and at the initial ester concentration of
in the donor compartment (mM), and A is the surface area of
2
the monolayer (cm ).
8
␮M, the apparent half-live is reduced to 3.65 min.
The accumulation of the ester in Caco-2 cells was also
measured. After performing transport studies, the transport
medium was removed and the cells were washed with HBSS,
pH 7.4. The cells were then removed from the polycarbonate
filters, diluted using 100 ␮l PBS, and kept in Eppendorf tubes
at −80°C until HPLC analysis.
Table I. Hydrolysis Rates and Half-Lives of the Ester in Buffer
Solution, 80% Human Plasma, Porcine Liver Esterase and Caco-2
Homogenate at 37°C
−1
Incubation mixture
Kobs (min
)
Half-life
−
4
−5a
−5b
−4b
−4b
−4b
−2b
−3b
Buffer pH 1.21–10.59 2.65 × 10 ± 6.82 × 10
46.91 ± 1.61 h
29.41 ± 0.94 h
4.46 ± 0.53 h
RESULTS AND DISCUSSION
Chemical and Enzymatic Stability
−3
Buffer pH 11.23
Buffer pH 11.75
Buffer pH 12.00
1.20 × 10 ± 3.16 × 10
−
3
6.53 × 10 ± 7.44 × 10
−
2
1.02 × 10 ± 1.70 × 10
2.82 ± 0.05 h
−
3
8
0% human plasma
1.90 × 10 ± 1.00 × 10
15.23 ± 0.80 h
3.65 ± 0.54 min
22.40 ± 2.87 min
−
1
Porcine liver esterase 1.92 × 10 ± 2.83 × 10
The chemical and enzymatic degradation of ester was
followed for at least two half-lives. The hydrolysis follows
first-order kinetics. The degradation products of ester were
mefenamic acid and guaiacol. By analyzing for the presence
−2
Caco-2 homogenate
3.13 × 10 ± 4.32 × 10
a
Value represents mean ± S.D. for Kobs of 10 buffer pHs.
Value represents mean ± S.D. of three determinations.
b

1
016
Tantishaiyakul et al.
ceptors such as serum proteins are added to the receiving
medium (16). To obtain a good in vitro-in vivo correlation,
BSA was included in the BL side to provide an absorptive
driving force, resembling the in vivo sink condition (17–18).
BSA can reduce the amount of lipophilic compounds ab-
sorbed into the cells and non-specific binding to the AP side,
thus increasing their diffusion out of the cells (17–19). Be-
sides, protein binding might also prevent non-specific adsorp-
tion of drugs to experimental apparatus such as supporting
filter or diffusion chamber in the BL side (20). These inter-
actions result in an enhanced transport of lipophilic com-
pounds and provide a reasonable estimate of drug permeabil-
ity under physiological conditions (17–18). Since the serosal
side is perfused with the blood that has about 4% albumin,
the inclusion of 4% BSA in the BL side was used to optimize
the in vitro experimental condition in this study.
Fig. 1. pH-rate profile for the chemical degradation of ester at 37°C.
Solubility and Partition Coefficient
Due to the short biologic half-life of the ester (22.4 min)
in Caco-2 homogenate, studies of ester transport across
Caco-2 cell monolayers were carried out in the presence or
absence of PMSF esterase inhibitor. And to increase the solu-
bility of poorly soluble compounds, DMSO as high as 5% was
used in the presence of 4% BSA in the BL side (17). The
TEER value obtained under this condition was lowered but
remained within an acceptable range. In this study, 4% of
DMSO was used to facilitate the solubilization of the ester.
After the transport experiments, the TEER values were
The ester of mefenamic acid-guaiacol is poorly soluble in
−
1
water (≈ 0.56 ␮g ml ). However, it is readily soluble in or-
ganic solvents such as ethanol, benzyl alcohol and methyl
benzoate. For mefenamic acid, a weak organic acid with pKa
of 4.2, its solubility is dependent on the solution pH; the
solubility increases with increasing pH. Its solubilities at pH
1
3
.16, 4.96, 5.92, 6.93, and 7.62 are 0.05, 0.24, 3.02, 33.62, and
−
1
29.75 ␮g ml , respectively.
The octanol-water partition (logP) of the ester could not
be determined since its value is too high to obtain a reliable
result by experimentation (13). The logP value was therefore
estimated using the ClogP program (Biobyte, California),
which calculates the value directly from its molecular struc-
ture. The calculated logP value for the ester is 6.15.
Transport across Caco-2 Cell Monolayers
Previous in vivo data suggest that intestinal absorption of
drugs may decline when their lipophilicity becomes too high
(14). Recent in vitro studies in HT29-18-C1 and Caco-2 cell
lines showed that when the log Do/b (octanol/buffer distribu-
tion coefficient) value was lower than 3.5, the permeability
increased with the lipophilicity, but when the log Do/b value
was between 3.5 and 5.2, the permeability decreased with
increasing lipophilicity (15). However, there has been an ar-
gument that the in vitro cell monolayer systems do not ad-
equately mimic the in vivo situation by underestimating the in
vivo bioavailability of highly lipophilic compounds, unless ac-
Fig. 3. A, Time course for the disappearance of ester and appearance
of mefenamic acid on the AP side when ester was applied to the AP
side of the Caco-2 cell monolayers, in the presence of 4% BSA in the
BL side and incubated up to 3 hours at 37°C after the exposure of
cells with PMSF; B, shows the ester transported to the BL side. The
Fig. 2. Stability of ester at 37°C in Caco-2 homogenate, showing the results were plotted as mass fractions, M(t), relative to the mass on
time course of the disappearance of ester and the appearance of the AP side (donor compartment) at time zero, Md(0). Experiments
mefenamic acid.
were performed in triplicate (average ±SD).

Transport and Stability of Mefenamic Acid-Guaiacol Ester
1017
No permeation to the BL side was observed with PMSF
exposure in the absence of BSA, although a significant
amount of the ester was accumulated in Caco-2 cells. The
accumulated amount was two-fold higher in the absence of
BSA compared to its presence. This result suggests that the
accumulated amount of this highly lipophilic compound de-
pends on protein binding, since protein binding facilitates the
permeation of the compound to the receiver side and cellular
accumulation of the compound is decreased. A similar obser-
vation was reported for cosalane uptake which was reduced in
the presence of BSA (22).
Without PMSF, the ester when applied to the AP side of
Caco-2 cell monolayers was converted to mefenamic acid to
some extent on the AP side and no ester was detected on the
BL side, although 4% BSA was added to this side (Fig. 4).
Due to the modest amount of initial ester concentration, the
length of transport study, and the detection limit of analytical
method, no permeation of the ester was observed.
Based on the chemical stability experiments, the ester
has sufficient chemical stability under the acidic conditions
encountered in the stomach with pH as low as 1 and in the
intestine with pH of about 8. The ester can, however, be
degraded to mefenamic acid and guaiacol by enzymes from
human plasma, porcine liver esterase and Caco-2 homog-
enate. This suggests that the ester might be hydrolyzed in vivo
before, during or after absorption, to release mefenamic acid
and guaiacol which would then exert its characteristic phar-
macological action. According to Yazdanian et al. (23) and
−
6
Yee (24), compounds with Papp of 0.4 or less than 1 × 10 cm
−
1
s
will be poorly absorbed. Compounds with a Papp varied
−
6
−1
between 7 (23) − 10 (24) × 10 cm s had well oral absorp-
Fig. 4. A, Time course for the disappearance of ester and appearance
of mefenamic acid on the AP side when ester was applied to the AP tion. If the addition of BSA provides a good estimation of
side of a Caco-2 cell monolayers, in the presence of 4% BSA in the Papp, the ester can probably be moderately absorbed. Fur-
BL side and incubated up to 3 hours at 37°C; B, shows mefenamic ther studies are needed to investigate the protective effect of
acid transported to the BL side. The results were plotted as mass
this ester on GI toxicity.
fractions, M(t), relative to the mass on the AP side (donor compart-
ment) at time zero, Md(0). Experiments were performed in triplicate
REFERENCES
(average ±SD).
found to be only slightly lower, i.e., 6% of the initial value,
indicating that the integrity of the monolayer was preserved
over a period of at least 3 h. In addition, the viability of the
Caco-2 cells after exposure to PMSF was accessed with a
MTT assay (21), which measured mitochondrial dehydroge-
nase activity of viable cells spectrophotometrically. This study
revealed that PMSF did not affect the cell viability at the
concentration used.
Time courses of the transport (from AP-to-BL) with and
without PMSF exposure on the AP side are shown in Fig. 3
and Fig. 4, respectively. With the exposure to PMSF (15 min)
and in the presence of 4% BSA on the BL side, the amount
of ester transported from the AP-to-BL direction increased
linearly with time after a lag time of 23 min (Fig. 3B). The
amount of ester transported in 3 h from AP-to-BL was
1. E. L. Semble and W. C. Wu. Anti-inflammatory drugs and gastric
mucosal damage. Semin. Arthritis Rheum. 16:271–286 (1987).
2
3
. V. K. Tammara, M. M. Narurkar, A. M. Crider, and M. A. Khan.
Morpholinoalkyl ester prodrugs of diclofenac: synthesis, in vitro
and in vivo evaluation. J. Pharm. Sci. 83:644–648 (1994).
. A. Fossati, A. M. G. Vimercati, M. Bozzi, C. Passarotti, G. L.
Bandi, and A. Formenti. Effects of metoxibutropate, ibuprofen
and guaiacol on the gastrointestinal system. Int J. Tissue React.
1
3:45–50 (1991).
4
. G. G. Melis, A. M. Paoletti, V. Mais, S. Ajossa, and S. Guerriero.
Controlled clinical study of the efficacy and tolerability of me-
thoxybutropate compared to nimesulide in gynecology. Minerva
Ginecol. 49:409–415 (1997).
5. G. F. Palmieri, D. Galli-Angeli, G. Giovannucci, and S. Martelli.
Inclusion of methoxybutropate in ␤-and hydroxypropyl ␤-cyclo-
dextrins: comparison of preparation methods. Drug Dev. Ind.
Pharm. 23:27–37 (1997).
6
. G. F. Palmieri, D. Lauri, S. Martelli, and P. Wehrle. Methoxy-
butropate microencapsulation by gelatin-acacia complex coacer-
vation. Drug Dev. Ind. Pharm. 25:399–407 (1999).
1
4.63%. Bi-directional transport studies in which the trans-
port of ester across the Caco-2 cell monolayer (AP-to-BL and
BL-to-AP) was carried out to determine whether the ester is
transported via an efflux mechanism. The apparent perme-
7. E. Tubaro, L. Belogi, C. M. Mezzadri, L. Ruco, and A. Stopac-
ciaro. Studies on the gastric tolerability of the new non-steroidal
anti-inflammatory drug amtolmetin guacyl. Arzneim-Forsch 45:
−
6
ability coefficients of ester from AP-to-BL (4.72 × 10 cm
1
298–1302 (1995).
−
1
−6
−1
s
) and BL-to-AP (5.37 × 10 cm s ) were found to be
8
. E. Tubaro, L. Belogi, and C. M. Mezzadri. The mechanism of
action of amtolmetin guacyl, a new gastroprotective nonsteroidal
anti-inflammatroy drug. Eur. J. Pharmacol. 387:233–244 (2000).
statistically indifferent (P < 0.05). Therefore, the ester is not
a substrate for the polarized efflux systems in Caco-2 cell
monolayer.
9. M. M. Bradford. A rapid and sensitive method for the determi-

1
018
Tantishaiyakul et al.
nation of microgram quantities of protein utilizing the principle
of protein-dye binding. Anal. Biochem. 72:248–254 (1976).
0. P. Augustijns, P. Annaert, P. Heylen, G. Van den Mooter, and R.
Kinget. Drug absorption studies of prodrug ester using the Caco2
model: evaluation of ester hydrolysis and transepithelial trans-
port. Int. J. Pharm. 166:45–53 (1998).
1. M. Brunner-Guenat, P. Carrupt, G. Lisa, B. Testa, S. Rose, K.
Thomas, P. Jenner, and P. Ventura. Esters of L-dopa: structure-
hydrolysis relationships and ability to induce circling behavior in
an experimental model of hemiparkinsonism. J. Pharm. Pharma-
col. 47:861–869 (1995).
diffusion of novel, highly lipophilicity antioxidants. J. Pharmacol.
Exp. Ther. 288:1317–1326 (1999).
1
17. G. Krishna and K. Chen. C. Lin, and A. A. Nomeir. Permeability
of lipophilic compounds in drug discovery using in-vitro human
absorption model, Caco-2. Int. J. Pharm. 222:77–89 (2001).
18. S. Yamashita, T. Furubayashi, M. Kataoka, T. Sakane, H. Sezaki,
and H. Tokuda. Optimized conditions for prediction of intestinal
drug permeability using Caco-2 cells. Eur. J. Pharm. Sci. 10:195–
1
2
04 (2000).
1
9. B. J. Aungst, N. H. Nguyen, J. P. Bulgarelli, and K. Oates-Lenz.
The influence of donor and reservoir additive on Caco-2 perme-
ability and secretory transport of HIV protease inhibitors and
other lipophilic compounds. Pharm. Res. 17:1175–1180 (2000).
0. Y. Lee, S. Chung, and C. Shim. The prevention of cyclosporin A
adsorption to Transwell surfaces by human plasma. Int. J. Pharm.
1
2. H. Bundgaard, A. Buur, S. Chang, and H. H. L. Lee. Timolol
prodrugs: synthesis, stability and lipophilicity of various alkyl,
cycloalkyl and aromatic esters of timolol. Int. J. Pharm. 46:77–88
2
2
(1988).
2
24:201–204 (2001).
1
3. W. A. Bruggeman, J. Van Der Steen, and O. Hutzinger. Re-
versed-phase thin-layer chromatography of polynuclear aromatic
hydrocarbons and chlorigated biphenyls relationship with hydro-
phobicity as measured by aqueous solubility and octanol-water
partition coefficient. J. Chromatogr. 238:335–346 (1982).
4. C. Hansch and T. Fujita. ␳-␴-␲ analysis. A method for the cor-
relation of biologic activity and chemical structures. J. Am. Chem.
Soc. 86:1616–1626 (1964).
1. Y. Quan, T. Fujita, D. Tohara, M. Jsuji, M. Kohyama, and A.
Yamamoto. Transport kinetics of leucine enkephalin across
Caco-2 monolayers: quantitative analysis for contribution of en-
zymatic and transport barrier. Life Sci. 64:1243–1252 (1999).
2. D. Pal, C. Udata, and A. K. Mitra. Transport of cosalane-a highly
lipophilic novel anti-HIV agent-across Caco-2 cell monolayers.
Pharm. Res. 89:826–833 (2000).
2
2
1
1
1
3. M. Yazdanian, S. L. Glynn, J. L. Wright, and A. Hawi. Correla-
tion partitioning and Caco-2 cell permeability of structurally di-
verse small molecular weight compounds. Pharm. Res. 15:1490–
1494 (1998).
5. P. Wils, A. Warnery, V. Phung-Ba, S. Legrain, and D. Scherman.
High lipophilicity decreases drug transport across intestinal epi-
thelial cells. J. Pharmacol. Exp. Ther. 269:654–658 (1994).
6. G. A. Sawada, C. L. Barsuhn, B. S. Lutzke, M. E. Houghton, G. 24. S. Yee. In vitro permeability across Caco-2 cells (colonic) can
E. Padbury, N. F. H. Ho, and T. J. Raub. Increased lipophilicity
and subsequent cell partitioning decrease passive transcellular
predict in vivo (small intestine) absorption in man-fact or myth.
Pharm. Res. 14:763–766 (1997).