Chemical and in vitro enzymatic stability of newly synthesized celecoxib lipophilic and hydrophilic amides

Article abstract of DOI:10.1016/j.ijpharm.2011.06.013

Five celecoxib (CXB) acylamide sodium salts, MP-CXB, Cy-CXB, Bz-CXB, CBz-CXB and FBz-CXB were synthesized and characterized. Two simple, fast and validated RP-HPLC methods were developed for simultaneous quantitative determination of the amides and celecoxib in aqueous and biological samples and LOD and LOQ were ≤13.6 and ≤40 ng/mL, respectively. The solubility and log P_app of the amides, in relevant media, were determined. The chemical hydrolysis, at 60, 70 and 80 °C, of MP-CXB was studied at GIT-relevant pH (1.2, 6.8 and 7.4) and of CY-CXB was studied at skin relative pH (5.4 and 7.4). Significant hydrolysis was observed for MP-CXB at pH 1.2 only with half-lives 28.28, 11.64 and 3.53 h at 60, 70 and 80 °C, respectively, with extrapolated half-lives of 2060 and 443 h at 25 and 37 °C, respectively. The hydrolysis of all amides was studied in rat live homogenate and only Cy-CXB was hydrolyzed with half-life of 3.79 h. The hydrolysis of MP-CXB and Cy-CXB was studied in human plasma and neither was hydrolyzed. It is finally suggested that hydrophobic interactions plays a role in the binding of susceptible acylamides to the hepatic hydrolyzing enzyme since only amides with saturated hydrocarbon chains underwent hydrolysis.

Full text of DOI:10.1016/j.ijpharm.2011.06.013

International Journal of Pharmaceutics 416 (2011) 85–96
Contents lists available at ScienceDirect
International Journal of Pharmaceutics
journal homepage: www.elsevier.com/locate/ijpharm
Chemical and in vitro enzymatic stability of newly synthesized celecoxib
lipophilic and hydrophilic amides
Amjad M. Qandila,∗, Farah H. El Mohtadib, Bassam M. Tashtoushb
a Department of Medicinal Chemistry and Pharmacognosy, Faculty of Pharmacy, Jordan University of Science and Technology, Irbid 22110, Jordan
b Department of Pharmaceutical Technology, Faculty of Pharmacy, Jordan University of Science and Technology, Irbid 22110, Jordan
a r t i c l e i n f o
a b s t r a c t
Article history:
Five celecoxib (CXB) acylamide sodium salts, MP-CXB, Cy-CXB, Bz-CXB, CBz-CXB and FBz-CXB were
synthesized and characterized. Two simple, fast and validated RP-HPLC methods were developed for
simultaneous quantitative determination of the amides and celecoxib in aqueous and biological samples
and LOD and LOQ were ≤13.6 and ≤40 ng/mL, respectively. The solubility and log Papp of the amides, in
relevant media, were determined. The chemical hydrolysis, at 60, 70 and 80 ◦C, of MP-CXB was studied at
GIT-relevant pH (1.2, 6.8 and 7.4) and of CY-CXB was studied at skin relative pH (5.4 and 7.4). Significant
hydrolysis was observed for MP-CXB at pH 1.2 only with half-lives 28.28, 11.64 and 3.53 h at 60, 70
and 80 ◦C, respectively, with extrapolated half-lives of 2060 and 443 h at 25 and 37 ◦C, respectively. The
hydrolysis of all amides was studied in rat live homogenate and only Cy-CXB was hydrolyzed with half-
life of 3.79 h. The hydrolysis of MP-CXB and Cy-CXB was studied in human plasma and neither was
hydrolyzed. It is finally suggested that hydrophobic interactions plays a role in the binding of susceptible
acylamides to the hepatic hydrolyzing enzyme since only amides with saturated hydrocarbon chains
underwent hydrolysis.
Received 14 March 2011
Received in revised form 6 June 2011
Accepted 9 June 2011
Available online 15 June 2011
Keywords:
Celecoxib
HPLC
Hydrolysis
Liver homogenate
Hydrophobic interactions
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
COX-2 inhibiting NSAIDs is a weak acid with a pKa value of about
11, very lipophilic and is practically insoluble in acidic, neural and
Non steroidal anti-inflammatory drugs (NSAIDs) are inhibitors
of the cyclooxygenase (COX) enzyme in the biosynthesis of
prostaglandins and they are generally used for the management
of acute and chronic pain, inflammation and fever (Duttaa et al.,
2009; Kısmet et al., 2004). To date, cyclooxygenases exit in three
isoforms, COX-1, COX-2 and COX-3, among which, only COX-1
and COX-2 are utilized clinically (Vane and Botting, 1998). COX-
1 is a constitutive enzyme that, mainly, plays a cytoprotective role
(El-Medanya et al., 2005). On the other hand, COX-2 is primarily
inducible by a variety of inflammatory stimuli including expo-
sure to growth factors, cytokines, tumor promoters, carcinogens,
and endotoxins and so it induces inflammation, pain, and fever
(Radi and Khan, 2006). Traditional NSAIDs, which are nonselective
COX inhibitors, exhibit an anti-inflammatory action and can, some-
times, cause severe gastrointestinal side effects. On the other hand,
selective COX-2 inhibitors have been demonstrated to produces
anti-inflammatory action while at the same time exert less unto-
ward effects on the gastrointestinal tract tissues (Bombardier et al.,
2000; Silverstein et al., 2000; Vonkeman and van de Laar, 2010).
Celecoxib (CXB), Fig. 1, the prototype of the Coxib family of selective
moderately alkaline environment including the physiological pH
(Subramanian et al., 2004; Ventura et al., 2005). CXB capsules have
a low absolute bioavailability of only 22–40% (Paulson et al., 2001)
due to celecoxib’s very low aqueous solubility, low wettability and
low dissolution rate. CXB, also, has low intrinsic permeation capac-
ity through skin which is mainly due to its poor aqueous solubility
(Baboota et al., 2007).
Prodrug design can be employed to optimize a drug’s perme-
ation properties and consequently affecting its oral bioavailability
and transdermal penetration. A product of such design is parecoxib,
a sodium salt of the propionylamide of valdecoxib which has been
shown to be suitable for of IV administration (Talley et al., 2000).
Like valdecoxib, CXB is an arylsulfonamide that can be transformed
into a hydrolysable amide by acylation (Larsen and Bundgaard,
1987; Larsen et al., 1988). For example, CXB butyrylamide was
shown to exhibit 5-fold enhancement in oral bioavailability com-
pared to its parent drug (Mamidi et al., 2002).
Herein, we report the design, synthesis and in vitro hydroly-
sis of a relatively diverse group of CXB acylamides with chemical
structures shown in Fig. 1. Due to the differences in the chemi-
cal structure of MP-CXB compared to the rest of the acylamides,
two different validated RF-HPLC methods have been developed and
employed for the quantitative determination of CXB and its acy-
lamides. There are limited reports of HPLC methods in which CXB
∗
Corresponding author. Tel.: +962 2 720 1000x26776; fax: +962 2 720 1075.
E-mail address: drqandil@just.edu.jo (A.M. Qandil).
0378-5173/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.ijpharm.2011.06.013

86
A.M. Qandil et al. / International Journal of Pharmaceutics 416 (2011) 85–96
Fig. 1. The chemical structures of the target CXB acylamides.
is being analyzed with its derivatives and none if that derivative is
not lipophilic. One method for the analysis of CXB and three acy-
lamides was long with retention times up to 22.3 min and uses a
25 cm column (Mamidi et al., 2002) and another for the analysis
for CXB alone uses a similar length column with no mention of
retention time (Baboota et al., 2007).
(Spain) and BDH organics (England). The deuterated solvents were
obtained from Acros Chemicals (Belgium). Deionized water was
used in the HPLC separation. Melting points were determined using
Stuart Scientific-melting point apparatus (UK) and were uncor-
rected. 1H and 13C NMR spectra were obtained using 400 MHz
Bruker Avance UltraShield Spectrometer (Switzerland). NMR data
are reported in ppm (ı) using automatic calibration to the residual
proton peak of the solvent, CD3OD and DMSO-d6; the splitting pat-
tern abbreviations are as follows: s, singlet; d, doublet; t, triplet;
q, quartet; m, multiplet; dd, doublet of doublet; bs, broad sin-
glet. Atomic pressure chemical ionization (APCI) mass spectra were
obtained using Agilent 1100 series LC–MS (USA) at the Pharma-
ceutical Research Center (PRC), Jordan University of Science and
Technology. The FT-IR spectra were recorded on Nicolet avatar 360
FT-IR (USA) using KBr disks. TLC analysis was performed on Albet
aluminium TLC plate (Silica 60 and UV 254 (Spain). Log pKa values
for the compounds were calculated using Pipline Pilot by Accelrys,
USA (http://accelrys.com).
MP-CXB is designed with the main goal of enhancement the sol-
ubility of CXB beyond what a sodium salt of its alkanoylamides can
achieve which is reported to be approximately 15 mg/mL (Mamidi
et al., 2002). In the GIT MP-CXB which contains a piprazine ring is
expected to be ionizable both, in basic and acidic media and hence,
maintaining high solubility over a large pH range. Cy-CXB is an ali-
cyclic amide that is designed as a prodrug for transdermal delivery.
The promoiety was chosen because it has a similar size to that in
MP-CXB, hence allows for direct comparison specially in enzymatic
hydrolysis. Accordingly, in addition to enzymatic hydrolysis, the
hydrolysis of MP-CXB will be studied in GIT-relevant pH values and
the hydrolysis of Cy-CXB will be studied in skin-relevant pH values.
In addition, Cy-CXB is an alicylic amide rather than a straight chain
amide and it will be interesting to compare its enzymatic hydrol-
ysis to that of the other CXB aliphatic acylamides reported in the
literature. The benzamides are synthesized to give a more compre-
hensive idea about the in vitro hydrolysis of this class of prodrugs in
liver homogenate. Finally it is worth mentioning that the hydrolysis
in liver homogenate of the labile bond in these prodrugs, sulfonyl
acylamide, has not been fully studied before and this work provide
an insight into this accept of the functional group.
2.1. Synthesis and characterization
2.1.1. Sodium 2-chloro-N-(4-(5-p-tolyl-3-(trifluoromethyl)-1H-
pyrazol-1-yl)-phenylsulfonyl) acetamide
(CAc-CXB)
To
a stirring solution of triethylamine (1.09 mL, 0.79 g,
7.86 mmol) and CXB (3.00 g, 7.86 mmol) in dichloromethane
(200 mL), chloroacetyl chloride (0.62 mL, 0.88 g, 7.86 mmol) was
added drop wise and the reaction mixture was allowed to stir at
room temperature for 24 h. The reaction progress was followed by
TLC (50% ethyl acetate in hexane). Upon completion of the reac-
tion, the solution was washed with distilled water (200 mL× 3),
then with 0.1 N HCl solution (300 mL× 3). The organic layer was
then dried over sodium sulfate and the solvent was evaporated. The
resultant compound was dissolved in absolute ethanol (200 mL)
and was then titrated with 2 M sodium hydroxide. The mixture
2. Experimental
CXB was kindly donated by the Jordanian Pharmaceutical Man-
ufacturing Company (JPM). Bulk solvents were purchased through
local vendors. The acyl and benzoyl chlorides were obtained from
Aldrich Chemical Company (USA). Other reagent grade, fine chem-
icals and HPLC solvents were obtained from Scharlau chemicals

A.M. Qandil et al. / International Journal of Pharmaceutics 416 (2011) 85–96
87
was then concentrated and the residue was crystallized from ethyl
acetate to afford 1.23 g of CAc-CXB as off-white crystals (30.20%).
1H NMR (DMSO-d6, 400 MHz), ı (ppm): 8 (2H, d, J = 8.6 Hz, Ar-H),
7.42 (2H, d, J = 8.6 Hz, Ar-H), 7.18 (4H, q, J = 8 Hz, Ar-H), 6.9 (1H,
s, pyr-H), 3.98 (2H, s, CH2), 2.36 (3H, s, Ar-CH3). IR (KBr, cm−1):
1612.49, 1408.04, 1342.49, 1236.3, 1114.8. LC–MS (APCI) m/z: M
(479.9, 100%), MH (480.9, 52.3%), MH + 1 (481.9, 48%). Elemental
analysis (calculated) C: 47.56; H: 2.94; N: 8.76; (found) C: 46.72;
H: 3.15; N: 9.34.
analysis (calculated) C: 56.13; H: 4.51; N: 8.18; (found) C: 54.76;
H: 4.59; N: 7.99.
2.1.3.2. Sodium N-(4-(5-p-tolyl-3-(trifluoromethyl)-1-H-pyrazol-1-
yl)phenylsulfonyl) benzamide (Bz-CXB). Benzoyl chloride (0.91 mL,
1.10 g, 7.86 mmol). Yield = 83.00%. 1H NMR (DMSO-d6, 400 MHz),
ı (ppm): 7.82–7.92 (4H, m, J = Hz, Ar-H), 7.24–7.39 (5H, m, Ar-H),
7.17 (4H, m, Ar-H), 7.13 (1H, s, pyr-H), 2.28 (3H, s, CH3). 13C NMR
(DMSO-d6, 100 MHz), ı (ppm): 172.67, 147.72, 144.44 (q, J = 37.5 Hz,
CF3), 142.31, 141.68, 140.11, 132.63, 132.03, 131.64, 131.31, 130.99,
130.59, 130.06, 129.70, 128.13, 127.42, 108.32, 23.43. FT-IR (KBr,
cm−1): 1595.1, 1409.96, 1330.88, 1238.30, 1170.79. LC–MS (APCI)
m/z: MH (508.2, 100%), MH + 1 (509.2, 35.8%), MH + 2 (510.2,
34.48%). Elemental analysis (calculated) C: 56.80; H: 3.38; N: 8.28;
(found) C: 52.49; H: 3.28; N: 7.81.
2.1.2. Sodium 2-(4-methylpiperazin-1-yl)-N-(4-(5-p-tolyl-3-
(trifluoromethyl)-1H-pyrazol-1-yl)-phenyl sulfonyl) acetamide
(MP-CXB)
CAc-CXB (1.00 g, 2.08 mmol) was dissolved in acetonitrile and
to the solution, 1-methylpiperazine (0.46 mL, 0.41 g, 4.16 mmol)
and sodium iodide (0.44 g, 2.08 mmol) were added. The reaction
mixture was allowed to stir at room temperature for 3 days.
The reaction progress was followed by TLC (10% methanol in
dichloromethane). Upon completion of the reaction it was filtered
and the solvent was evaporated under vacuum and the obtained
solid was crystallized from acetonitrile to yield 0.90 g of MP-CXB
as white crystals (90.00%). 1H NMR (CD3OD, 400 MHz), ı (ppm):
7.99 (2H, d, J = 8.6 Hz, Ar-H), 7.42 (2H, d, J = 8.6 Hz, Ar-H), 7.2 (4H, q,
J = 8.3 Hz, Ar-H), 6.91 (1H, s, pyr-H), 3.04 (2H, s, CH2), 2.50 (8H, bs,
pip-H), 2.38 (3H, s, N-CH3), 2.28 (3H, s, Ar-CH3). 13C NMR (CD3OD,
100 MHz), ı (ppm): 179.23, 147.67, 146.33, 145.38 (q, J = 38 Hz,
CF3), 143.39, 141.61, 131.40, 130.81, 129.97, 128.06, 127.04, 124.89,
122.22, 107.54, 65.33, 56.20, 54.52, 46.80, 22.17. FT-IR (KBr, cm−1):
1600.92, 1408.04, 1303.88, 1263.37, 1151.50. LC–MS (APCI) m/z:
MH (544.1, 100%), MH + 1 (545.1, 42.5%), MH + 2 (546.1, 46.4%). Ele-
mental analysis (calculated) C: 53.03; H: 4.64; N: 12.88; (found) C:
53.8; H: 4.66; N: 13.84.
2.1.3.3. Sodium
1-H-pyrazol-1-yl)phenylsulfonyl)
p-Chlorobenzoyl chloride (0.99 mL,
4-chloro-N-(4-(5-p-tolyl-3-(trifluoromethyl)-
benzamide
1.37 g,
(CBz-CXB).
7.86 mmol).
Yield = 76.6%. 1H NMR (DMSO-d6, 400 MHz), ı (ppm): 7.78–7.98
(4H, m, Ar-H), 7.26–7.42 (4H, m, Ar-H), 7.06–7.24 (5H, m, Ar-H
and pyr-CH), 2.28 (3H, s, Ar-CH3). 13C NMR (DMSO-d6, 100 MHz),
ı (ppm): 171.64, 148.8, 147.7, 144.47 (q, J = 37.7 Hz, CF3), 142.44,
141.6, 140.49, 137.5, 133.54, 132.87, 132.03, 131.3, 130.58, 130.16,
129.76, 128.12, 127.55, 108.37, 23.43. FT-IR (KBr, cm−1): 1597.06,
1409.96, 1332.8, 1238.3, 1136. LC–MS (APCI) m/z: MH (543.2,
100%), MH + 1 (544.2, 93.3%). Elemental analysis (calculated) C:
53.19; H: 2.98; N: 7.75; (found) C: 49.93; H: 2.92; N:7.09.
2.1.3.4. Sodium
1-H-pyrazol-1-yl)phenylsulfonyl)
4-fluoro-N-(4-(5-p-tolyl-3-(trifluoromethyl)-
benzamide (FBz-CXB).
p-Fluorobenzoyl chloride (1.67, 1.24 g, 7.86 mmol). Yield = 92.00%.
1H NMR (DMSO-d6, 400 MHz), ı (ppm): 7.88–7.97 (2H, m, Ar-H),
7.85 (2H, d, J = 8.4 Hz, Ar-H), 7.34 (2H, d, J = 8.4 Hz, Ar-H), 6.99–7.23
(7H, m, Ar-H and pyr-H), 2.28 (3H, s, Ar-CH3). 13C NMR (DMSO-
d6, 100 MHz), ı (ppm): 166.17 (d, J = 246.3 Hz), 148.88, 147.72,
144.47 (q, J = 37.5 Hz), 142.43, 141.6, 138.08, 133.46 (d, J = 8.7 Hz),
132.35, 131.31, 130.54, 128.1, 127.55, 125.35, 122.69, 116.82 (d,
J = 29.3 Hz), 108.35, 23.42. FT-IR (KBr, cm−1) 1608.63, 1409.96,
1325.10, 1238.86, 1168.86. LC–MS (APCI) m/z: MH + (526, 100%).
Elemental analysis (calculated) C: 54.86; H: 3.07; N: 8.00; and
(found) C: 53.05; H: 3.05; N: 8.40.
2.1.3. General procedure for the synthesis of Cy-CXB, Bz-CXB,
CBz-CXB and FBz-CXB as the sodium salts
To
a stirring solution of triethylamine (1.09 mL, 0.79 g,
7.86 mmol) and CXB (3.00 g, 7.86 mmol) in dichloromethane
(200 mL), the appropriate acylchloride or benzoylchloride
(7.86 mmol) was added drop wise and the reaction mixture
was allowed to stir at room temperature for 24 h. The reaction
progress was followed by TLC (5% methanol in dichloromethane).
Upon completion of the reaction the solvent was evaporated and
300 mL of ether was added and the solution washed with 1N HCl
aqueous solution (200 mL× 3), then with 1 M sodium bicarbonate
(300 mL× 3). The organic layer was then dried over sodium sulfate
and the solvent was evaporated. The resultant residue was dis-
solved in absolute ethanol (200 mL) and then was titrated with 2 M
sodium hydroxide. The resultant mixture was then concentrated
and the residue was washed extensively with boiling diethylether
to afford the desired compound as white solids with 99–96% purity
according to elemental analysis.
2.2. High performance liquid chromatography (HPLC)
The HPLC system consisted of a SCL-10 AVP system controller,
LC-10 ADVP isocratic pump, DGV-14AVP degasser, SPD-M 10AVP
diode array detector, SIL-10ADVP auto injector, and CTO-10ASVP
column oven connected to a computer prepared with appropriate
software. Chromatographic separation and quantitative analysis
were performed using a LiChroCART C18 (125 mm × 4.6 column
mm ID), 5 ␮m (RP-select B) with a flow rate of 2 mL/min. The tem-
perature of the column oven was set to 25 ◦C 1. The wavelength
was set at 254 nm and the injection volume was 30 ␮L. In method
A, for MP-CXB, the mobile phase for isocratic elution was a mixture
of 0.02 M phosphate buffer and acetonitrile (60:40, v/v) and 1% tri-
ethylamine, pH 3.5. In method B, for Cy-CXB, Bz-CXB, CBz-CXB and
FBz-CXB, the mobile phase for isocratic elution consisted of 0.01 M
phosphate buffer and acetonitrile (45:55, v/v), pH 3.2. The mobile
phases were filtered through 0.45 ␮m regenerated cellulose mem-
brane filter (VIVID, USA). CXB was analyzed simultaneously with
its acylamides and benzamides in each of the two methods.
2.1.3.1. Sodium N-(4-(5-p-tolyl-3-(trifluoromethyl)-1-H-pyrazol-1-
yl)-phenylsulfonyl) cyclohexane carboxamide (Cy-CXB). Cyclohex-
ane carbonyl chloride (1.05 mL, 1.15 g, 7.86 mmol). Yield = 66.00%.
1H NMR (DMSO-d6, 400 MHz), ı (ppm): 7.76 (2H, d, J = 8.6 Hz, Ar-
H), 7.32 (2H, d, J = 8.6 Hz, Ar-H), 7.23–7.13 (5H, m, Ar-H), 7.15 (1H,
s, pyr-H), 2.31 (3H, s, Ar-CH3), 1.89 (1H, m, Cyc-H), (5H, m, Cyc-
H), 1.16 (5H, m, Cyc-H). 13C NMR (DMSO-d6, 100 MHz), ı (ppm):
183.51, 149.54, 147.70, 144.38 (q, J = 37.8 Hz, CF3), 142.03, 141.52,
131.97, 131.28, 130.25, 128.12, 127.46, 125.37, 122.70, 108.3, 49.70,
32.90, 28.56, 28.22, 23.42. FT-IR (KBr, cm−1): 2926.01, 2852.72,
1593.20, 1409.96, 1321.24, 1238.3, 1134.1. LC–MS (APCI) m/z: MH
(514.1, 100%). MH + 1 (515.1, 15.4%), MH + 2 (516.1, 16%). Elemental
2.2.1. Calibration curves
A stock solution (400 ␮g/mL) was prepared by dissolving 40 mg
of each of the analytes in 100 mL HPLC grade methanol. Standard

88
A.M. Qandil et al. / International Journal of Pharmaceutics 416 (2011) 85–96
solutions for the calibration curves of each of the analyzed com-
pounds were obtained by serial dilution of the stock solution of
the corresponding compound with HPLC grade methanol. Calibra-
tion curves for CXB and five amides were constructed by injecting
samples from the standard solutions of each of the compounds in
the concentration range from 1.5 to 200 ␮g/mL. The area under the
curve (AUC) for each compound was determined and plotted versus
the concentration (␮g/mL).
phosphate buffer at pH 1.2 at pH 6.8 and pH 7.4 (except for pH
1.2 at 80 ◦C where a 50 mg/mL initial concentration was used) and
the solutions were placed in a thermostatically controlled water
bath at the respective temperatures. At appropriate time intervals,
samples were withdrawn, cooled with iced water and immediately
analyzed for CXB and remaining MP-CXB by HPLC. All experiments
were carried out in triplicate.
2.4.2. Hydrolysis of Cy-CXB in aqueous phosphate buffer
2.2.2. Method validation
Hydrolysis of Cy-CXB was investigated in aqueous 0.16 M phos-
phate buffer solution in the presence of 30% acetonitrile as a
cosolvent at pH 5.4 and 7.4 at 80 0.2 ◦C. The ionic strength was
maintained at 0.44 by adding a calculated amount of sodium chlo-
ride. The reactions were initiated by preparing 50 ␮g/mL solutions
of Cy-CXB in 0.16 M phosphate buffer at pH 5.4 and pH 7.4 in the
presence of 30% acetonitrile and the solutions were placed in a
thermostatically controlled water bath 80 ◦C 0.2. At appropriate
time interval, samples were withdrawn, cooled with iced water and
immediately analyzed for CXB and remaining Cy-CXB by HPLC. All
experiments were carried out in triplicate.
HPLC method A was validated for the quantitation of CXB and
MP-CXB while method B was validated for the quantitation of CXB,
Cy-CXB, Bz-CXB, CBz-CXB and FBz-CXB in terms of linearity, accu-
racy, precision, selectivity, and recovery, limit of detection (LOD)
and limit of quantitation (LOQ).
Linearity was performed by constructing six calibration of curve
for all the analytes in each method. Accuracy and precision were
carried out by injecting freshly prepared control solutions at 10, 80,
and 160 ␮g/mL six times a day for three consecutive days. Signal
to noise ratio was used to determine the LOD and LOQ. Relative
recovery of the analytes from biological samples was calculated
by comparing the theoretical concentration to that recovered from
spiking liver homogenate and plasma with known concentrations
of the analytes.
2.5. In vitro enzymatic hydrolysis
2.5.1. Hydrolysis in liver homogenate
The liver homogenate was prepared according to published pro-
cedure (Mamidi et al., 2002). The hydrolysis of Cy-CXB, Bz-CXB,
CBz-CXB and FBz-CXB was assessed in liver homogenate by placing
980 ␮L of the supernatant from a liver homogenate in an eppondorf
test tube and incubating it for 15 min in a shaking water bath equi-
librated at 37 1 ◦C. Then, 20 ␮L from a 2 mg/mL solution of each
of the amides in methanol was mixed well with supernatant using
vortex mixer leading to an initial concentration of 40 ␮g/mL. For
the more water soluble MP-CXB, 900 ␮L of the supernatant from
liver homogenate was placed in an eppondorf test tube and incu-
bating it for 15 min in a shaking water bath equilibrated at 37 1 ◦C.
Then 100 ␮L (MP-CXB) from a 2 mg/mL solution in methanol was
mixed well with supernatant using vortex mixer leading to an
initial concentration of 200 ␮g/mL. At appropriate time intervals,
100 ␮L samples were withdrawn by micropipette and added to
300 ␮L methanol to precipitate the proteins. The mixture was then
centrifuged for 5 min at 10,000 rpm and the clear supernatants
obtained were analyzed for remaining amide and released CXB by
HPLC. The rate of hydrolysis of each amide was determined from
the slope of the linear plot of Ln (residual amide concentration)
versus time. The half-life of the hydrolysis was also calculated.
2.3. Physico-chemical properties
2.3.1. Aqueous solubility
All solubility determinations was done at room temperature.
The aqueous solubility of the five amides was determined both in
water and in aqueous phosphate buffer solution (0.16 M) at pH 7.4.
In addition, the solubilities of Cy-CXB and MP-CXB werewas also
determined in aqueous phosphate buffer solution (0.16 M) at pH
5.4 or 6.8, respectively. In each case, excess amounts of compound
were added to 1 mL of the dissolution media and were incubated
at 25 1 ◦C for 48 h. The saturated solutions were centrifuged at
4000 rpm for 5 min, and filtered using Millipore filters 0.45 ␮m and
the concentration of each compound was determined by HPLC.
2.3.2. Partition coefficient and log Papp
The apparent partition coefficients expressed as log Papp of the
five amides was determined, at room temperature, in n-octanol and
water or aqueous phosphate buffer solution (0.16 M) pH 7.4 sys-
tems. The log Papp of Cy-CXB and MP-CXB were also determined
using n-octanol and aqueous phosphate buffer solution (0.16 M)
pH 5.4 or 6.8 systems, respectively. n-Octanol was firstly satu-
rated with aqueous phosphate buffer solution of the appropriate
pH by vigorous stirring for 24 h. Log Papp was measured by initially
suspending 2 mg of each compound in 10 mL of the mixture of
n-octanol water or aqueous phosphate buffer solution in a glass
screw-capped test tube. The test tube was shaken at 25 ◦C in water
bath for 2 h and then centrifuged at 5000 rpm for 8 min to assure
complete separation of the two phases. The partition coefficient
was calculated by dividing the concentration of each CXB acylamide
or benzamide in the n-octanol layer by its concentration in aqueous
phase. The experiments were performed in triplicate.
2.5.2. Hydrolysis in 10% human plasma
The rate of hydrolysis of MP-CXB and Cy-CXB was studied in
human plasma at 37 ◦C. For Cy-CXB, from a solution of 2 mg/mL
solution in methanol, 20 ␮L were added to 980 ␮L of 10% human
plasma. The mixture was kept in a water bath for 48 hat 37 ◦C. While
for MP-CXB, from a 2 mg/mL solution in methanol, 100 ␮L were
added to 900 ␮L of 10% human plasma. The mixture was kept in a
water bath for 48 hat 37 ◦C. At appropriate time intervals, 100 ␮L
of the mixture was withdrawn and added to 300 ␮L of absolute
methanol to precipitate proteins and then the mixture was cen-
trifuged for 5 min at 10,000 rpm. The supernatant was analyzed for
the remaining MP-CXB and Cy-CXB and released CXB by the HPLC.
2.4. Chemical hydrolysis
2.4.1. Hydrolysis of MP-CXB in aqueous phosphate buffer
3. Results and discussion
Hydrolysis of MP-CXB was investigated in aqueous 0.16 M phos-
phate buffer solution at GIT relevant pH values of 1.2, 6.8 and 7.4
at 80 ◦C 0.2. The hydrolysis was furthered studied at pH 1.2 at 60
and 70 ◦C 0.2. The ionic strength was always adjusted to 0.44 by
adding a calculated amount of sodium chloride. The reactions were
initiated by preparing 100 mg/mL solution of MP-CXB in 0.16 M
3.1. Chemistry
As seen in Scheme 1, the first step in the synthesis of MP-
CXB was to acetylate CXB with ␣-chloroacetyl chloride followed
by conversion of the resultant amide to the sodium salt to afford

A.M. Qandil et al. / International Journal of Pharmaceutics 416 (2011) 85–96
89
Scheme 1. Synthesis of the target CXB acyl and benzamide.
CAc-CXB in modest yield. CAc-CXB was then reacted with 1-
methylpiperazine in the presence of sodium iodide as the catalyst
to afford MP-CXB, after crystallization, in excellent yield. For the
synthesis of the rest of the amides, CXB was acylated with either
cyclohexane carbonyl chloride, benzoyl chloride, 4-chlorobenzoyl
chloride or 4-fluorobenzoyl chloride in the presence of triethy-
lamine to afford Cy-CXB, Bz-CXB, CBz-CXB and FBz-CXB in good
to excellent yields, Scheme 1. The crude amides were converted to
the sodium salts but failed to crystallize after many attempts. The
best way to obtain them with acceptable purity was to extensively
wash the sodium salts with boiling diethylether. HPLC and elemen-
tal analyses clearly showed that the purity of all the synthesized
amides was 96–99%.
or in plasma. The chromatograms form methanol is shown in
Figs. 2 and 3. The chromatograms for the same compounds in liver
homogenate and plasma are essentially identical expect for the
appearance of one sharp peak at around 0.6 min, which corresponds
for the solvent front.
3.2.3. Accuracy
The accuracy of an analytical method expresses the closeness of
agreement between true value of analyte in samples and test result
obtained by the method (Ermer and Miller, 2005). The accuracy as
confirmed by triplicate analysis of three control samples at concen-
tration level of 10, 80 and 160 ␮g/mL for CXB and all the amides.
The samples were injected daily for three days. The measured value
for each compound was found to be within 15% of the true value
during the three days of study which indicates accurate method of
analysis (Center for Drug Evaluation and Research (CDER), 2001).
Tables 2 and 3 summarize the results of accuracy of methods A and
B for CXB and its amides.
3.2. HPLC method validation
3.2.1. Linearity
Acceptability of linearity data is often judged by examining
the correlation coefficient of the linear regression line for the
area under the curve versus concentrations. For establishment
of linearity, a minimum of five concentrations is recommended
(International Conference on Harmonisation (ICH), 1997) and a cor-
relation coefficient of 0.999 is generally considered as evidence of
acceptable fit of the data to the regression line (Shabir, 2003). For
method A, six calibration curves were constructed for CXB and MP-
CXB over the concentration, ranges of 1.56–400 ␮g/mL by triplicate
injection of nine freshly prepared standard solutions. In the same
manner, for method B, six calibration curves were constructed for
CXB and Cy-CXB, Bz-CXB, CBz-CXB, FBz-CXB of the same concen-
tration range. The linear regression equation and its coefficient of
correlation, R2, for each of the calibration curves is summarized
in Table 1. It can be seen that R2 was always greater than 0.999
indicating the methods linearity.
3.2.4. Precision
The precision of an analytical method describes the closeness
of individual measures of the same analyte when the procedure
is applied repeatedly to multiple aliquots of a single homoge-
neous sample (Center for Drug Evaluation and Research (CDER),
Table 1
The average linear regression equations from the six calibration curves for the
analytes.
Compound
Calibration curve
equation (average of
six curves)
R2
Celecoxib (method A)
Celecoxib (method B)
MP-CXB
Cy-CXB
Bz-CXB
y = 31219x − 3201.4
y = 45739x + 4080.6
y = 30300x − 416.72
y = 32445x + 13159
y = 31960x − 11883
y = 45277x − 5651.2
y = 31671x − 14274
1
0.99998
0.99999
0.99999
0.99999
0.99999
0.99994
3.2.2. Selectivity
The selectivity of the method was confirmed by analysis of
samples containing CXB and each of its amides spiked in either
methanol, in supernatant obtained from the liver homogenate
CBz-CXB
FBz-CXB

90
A.M. Qandil et al. / International Journal of Pharmaceutics 416 (2011) 85–96
Table 3
Accuracy of method B for the determination of CXB, Cy-CXB, Bz-CXB, CBz-CXB and
FBz-CXB.
Compound
Measured concentration (␮g/mL) SDa
10
80
160
Celecoxib
Day 1
Day 2
9.957 0.03
10.08 0.01
10.16 0.02
78.09 0.07
77.99 0.06
78.17 0.03
162.78 0.1
163.11 0.4
162.84 0.09
Day 3
Cy-CXB
Day 1
Day 2
10.26 0.02
10.09 0.02
10.12 0.02
78.47 0.15
77.76 0.45
78.02 0.46
160.16 0.22
159.84 0.45
161.14 0.65
Day 3
Bz-CXB
Day 1
Day 2
10.18 0.05
9.78 0.06
10.16 0.03
80.92 0.29
80.67 0.37
78.76 0.47
161.88 0.22
160.34 0.35
160.16 0.66
Day 3
CBz-CXB
Day 1
Day 2
10.04 0.14
10.22 0.02
9.85 0.27
82.26 0.17
80.44 0.42
79.61 0.07
160.90 0.67
161.59 0.50
159.43 0.21
Day 3
FBz-CXB
Day 1
Day 2
10.04 0.14
10.22 0.02
9.85 0.27
82.26 0.17
80.44 0.42
79.61 0.07
160.90 0.67
161.59 0.50
159.43 0.21
Day 3
Fig. 2. HPLC chromatogram from methods A for CXB and MP-CXB from methanol.
a
Standard deviation (n = 3).
Table 2
Accuracy of method A for the determination of CXB and MP-CXB.
short interval of time (International Conference on Harmonisation
(ICH), 1997). Repeatability can be obtained by analysis of a mini-
mum of three concentrations (low, medium and high) representing
the entire range of the standard curve for at least five injections
(n ≥ 5) (Center for Drug Evaluation and Research (CDER), 2001).
Intermediate precision shows variation affected in day to day analy-
sis, at least two days is recommended for inter-day precision testing
(International Conference on Harmonisation (ICH), 1995). Precision
acceptance criteria for a method are that the intra-day precision
(%RSD) will be ≤2.0% and the inter-day precision will be ≤3% (Ermer
and Miller, 2005). The intra- and inter-day precision was deter-
mined by replicate injection (n = 6) for three different concentration
(10, 80 and 160 ␮g/mL) of CXB and its amides for three consecu-
tive days. The results obtained for intra- and inter-day precision
studies are presented in Tables 4–10, respectively. As it can be seen
method precision has a relative standard deviation (%RSD) below
1% for repeatability and 2% for intermediate precision.
Compound
Measured concentration (␮g/mL) SDa
10 80
160
Celecoxib
Day 1
9.85 0.03
78.69 0.38
162.40 0.46
162.48 0.3
160.60 0.11
Day 2
Day 3
10.06 0.02
10.11 0.06
82.643 0.28
81.88 0.1
MP-CXB
Day 1
Day 2
10.01 0.01
10.05 0.02
10.11 0.02
81.62 0.06
81.81 0.13
82.18 0.09
160.32 0.29
161.16 0.31
161.72 0.11
Day 3
a
Standard deviation (n = 3).
2001). The precision is expressed as relative standard deviation
(RSD) of the replicate measurement (Lindholm, 2004), where
%RSD = (SD/mean) × 100 (Karnes and March, 1993). The precision
is divided into three categories: repeatability (intra-day precision),
intermediate precision (inter-day precision) and reproducibility
(between laboratories precision) which is not normally performed
if intermediateprecisionis accomplished(USP, 1994). Repeatability
expresses the precision under the same operating conditions over a
3.2.5. Recovery
Recovery is expressed as the amount of the compound of inter-
est analyzed as a percentage to the theoretical amount present
Fig. 3. HPLC chromatograms from methods B for CXB, Cy-CXB, Bz-CXB, CBz-CXB and FBz-CXB from methanol.

A.M. Qandil et al. / International Journal of Pharmaceutics 416 (2011) 85–96
91
Table 4
Inter- and intra-day precision of method A for the determination of CXB.
Concentration
Inter-day precision
Day 1
Day 2
Day 3
10 ␮g/mL intra-day precision
Average (n = 6)
%RSD
9.96 0.03
0.28%
10.08 0.01
0.13%
10.16 0.02
0.25%
10.06 0.1
1.00%
80 ␮g/mL intra-day precision
Average (n = 6)
%RSD
78.09 0.07
0.08%
77.99 0.06
0.08%
78.17 0.03
0.04%
78.08 0.09
0.12%
160 ␮g/mL intra-day precision
Average (n = 6)
%RSD
162.78 0.10
0.06%
163.11 0.4
0.24%
162.84 0.09
0.05%
162.91 0.17
0.10%
Table 5
Inter- and intra-day precision of method A for the determination of MP-CXB.
Concentration
Inter-day precision
Day 1
Day 2
Day 3
10 ␮g/mL intra-day precision
Average (n = 6)
%RSD
10.02 0.01
0.13%
10.05 0.02
0.23%
10.12 0.02
0.19%
10.06 0.05
0.51%
80 ␮g/mL intra-day precision
Average (n = 6)
%RSD
81.62 0.06
0.07%
81.82 0.13
0.15%
82.19 0.09
0.11%
81.87 0.29
0.35%
160 ␮g/mL intra-day precision
Average (n = 6)
%RSD
160.32 0.29
0.18%
161.16 0.3
0.07%
161.72 0.11
0.19%
161.07 0.70
0.43%
Table 6
Inter- and intra-day precision of method B for the determination of CXB.
Concentration
Inter-day precision
Day 1
Day 2
Day 3
10 ␮g/mL intra-day precision
Average (n = 6)
%RSD
9.85 0.03
0.35%
10.53 0.03
0.49%
10.11 0.06
0.58%
10.01 0.14
1.38%
80 ␮g/mL intra-day precision
Average (n = 6)
%RSD
78.70 0.38
0.49%
82.64 0.28
0.34%
81.88 0.1
0.12%
81.07 2.09
2.58%
160 ␮g/mL intra-day precision
Average (n = 6)
%RSD
162.40 0.46
0.28%
162.48 0.29
0.18%
160.62 0.11
0.07%
161.82 1.06
0.657%
in the medium (Center for Drug Evaluation and Research (CDER),
1994). The percentage of recovery of CXB and its amides from liver
homogenate was performed using the concentrations in Table 11.
For amides that showed no hydrolysis in liver homogenate only
one concentration was used. While in plasma, Table 12, only recov-
ery for Cy-CXB and MP-CXB but not CXB itself because neither
amide showed hydrolysis in plasma. In all cases recoveries were
calculated by comparing recovered concentration to the theoret-
ical one from spiked samples in a solution of equal volume of
homogenates and plasma. The lower recovery of Cy-CXB in plasma
compared to MP-CXB is most likely due to the predominant acidic
nature of former, which results in higher percentage of protein
binding.
3.2.6. Limit of detection (LOD) and limit of quantification (LOQ)
LOD and LOQ of an individual analytical procedure is the lowest
amount of analyte in a sample, which can be detected or quan-
titatively determined, respectively, with suitable precision and
accuracy (International Conference on Harmonisation (ICH), 1995).
LOD and LOQ are calculated according to signal-to-noise ratio.
Signal-to-noise ratios of 3:1 and 10:1 are generally considered
acceptable for estimating the detection limit and quantification
Table 7
Inter- and intra-day precision of method B for the determination of Cy-CXB.
Concentration
Inter-day precision
Day 1
Day 2
Day 3
10 ␮g/mL intra-day precision
Average (n = 6)
%RSD
10.27 0.02
0.20%
10.09 0.02
0.22%
10.12 0.02
0.17%
10.16 0.09
0.90%
80 ␮g/mL intra-day precision
Average (n = 6)
%RSD
78.48 0.15
0.18%
77.77 0.45
0.57%
78.03 0.46
0.59%
78.09 0.36
0.46%
160 ␮g/mL intra-day precision
Average (n = 6)
%RSD
160.17 0.22
0.14%
159.84 0.45
0.27%
161.15 0.66
0.402%
160.39 0.68
0.42%

92
A.M. Qandil et al. / International Journal of Pharmaceutics 416 (2011) 85–96
Table 8
Inter- and intra-day precision of method B for the determination of Bz-CXB.
Concentration
Inter-day precision
Day 1
Day 2
Day 3
10 ␮g/mL intra-day precision
Average (n = 6)
%RSD
10.19 0.05
0.54%
9.78 0.06
0.58%
10.16 0.03
0.34%
10.17 0.01
0.12%
80 ␮g/mL intra-day precision
Average (n = 6)
%RSD
80.92 0.29
0.36%
80.67 0.37
0.46%
78.76 0.47
0.60%
80.12 1.18
1.47%
160 ␮g/mL intra-day precision
Average (n = 6)
%RSD
161.89 0.22
0.14%
160.35 0.35
0.22%
160.17 0.65
0.41%
160.78 0.91
0.56%
Table 9
Inter- and intra-day precision of method B for the determination of CBz-CXB.
Concentration
Inter-day precision
Day 1
Day 2
Day 3
10 ␮g/mL intra-day precision
Average (n = 6)
%RSD
10.04 0.14
0.14%
10.22 0.23
0.23%
9.86 0.27
0.27%
10.04 0.18
1.79%
80 ␮g/mL intra-day precision
Average (n = 6)
%RSD
82.27 0.17
0.2%
80.44 0.42
0.5%
79.61 0.07
0.08%
80.77 1.36
1.68%
160 ␮g/mL intra-day precision
Average (n = 6)
%RSD
160.91 0.67
0.42%
161.60 0.50
0.31%
159.43 0.21
0.13%
160.64 1.10
0.68%
Table 10
Inter- and intra-day precision of method B for the determination of FBz-CXB.
Concentration
Inter-day precision
Day 1
Day 2
Day3
10 ␮g/mL intra-day precision
Average (n = 6)
%RSD
9.92 0.02
0.26%
10.03 0.02
0.23%
10.10 0.67
0.66%
10.02 0.09
0.93%
80 ␮g/mL intra-day precision
Average (n = 6)
%RSD
77.65 0.41
0.53%
80.51 0.12
0.15%
80.43 0.20
0.25%
79.53 1.63
2.04%
160 ␮g/mL intra-day precision
Average (n = 6)
%RSD
159.71 0.19
0.12%
160.50 1.24
0.77%
161.61 0.20
0.12%
160.60 0.95
0.59%
Table 11
Recovery data for CXB and its amides in liver homogenate.
Compound
CXB
MP-CXB
Cy-CXB
Bz-CXB
CBz-CXB
FBz-CXB
Actual (␮g/mL)
10
200
10
40
10
10
10
Mean recovered (␮g/mL)
9.79
97.9
0.21
197.91
98.95
0.44
9.492
94.92
0.5
39.70
99.26
0.05
9.492
94.92
0.5
9.776
97.76
0.204
8.40
84
0.322
%Recovery
SDa
a
Standard deviation (n = 3).
Table 12
Recovery data for MP-CXB and Cy-CXB from human plasma.
Compound
MP-CXB
Cy-CXB
10
6.621
66.21
0.08
Actual (␮g/mL)
200
40
Mean recovered (␮g/mL)
196.93
98.46
0.36
26.77
66.92
0.51
%Recovery
SDa
a
Standard deviation (n = 3).

A.M. Qandil et al. / International Journal of Pharmaceutics 416 (2011) 85–96
93
Table 13
LOD and LOQ for CXB and its amides.
CXBa
CXBb
MP-CXBa
Cy-CXBb
Bz-CXBb
CBz-CXBb
FBz-CXBb
LOD (ng/mL)
LOQ (ng/mL)
9.74
29.50
1.06
3.2
13.6
40.00
0.40
1.20
0.59
1.80
0.36
1.10
1.39
4.20
a
Method A.
Method B.
b
Table 14
Physico-chemical properties of CXB, MP-CXB, Cy-CXB and the benzamides.
Cmpd
MP (◦C)
log Papp
Solubility (␮g/mL) at rt
Water
pH 5.4 (6.8)
pH 7.4
–
Water
pH 5.4 (6.8)
pH 7.4
CXBa
157–159
296–298
289–290
332–334
342–344
337–339
3.50
–
3.00–7.00
–
–
MP-CXB
Cy-CXB
Bz-CXB
ClBz-CXB
FBz-CXB
0.99
1.5 (pH 6.8)
4.50 0.68
ND
ND
ND
1.40 0.04
3.59 0.57
4.74 0.34
4.62
>45,000
>45,000 (pH 6.8)
0.53 0.05
ND
ND
ND
>45,000
2.52 0.21
1.78 0.03
2.80
1587.60 38.9
1125.20 69.36
234.37 9.55
624.71 18.92
53.3 9.84
0.17 0.01
5.83 0.61
0.27 0.05
2.13 0.03
5.35 0.371
a
Literature (Subramanian et al., 2004). ND: not determined.
limit, respectively (International Conference on Harmonisation
(ICH), 1997). Table 13 shows the calculated LOD and LOQ for CXB
and its amides.
3.3. Physico-chemical properties of CXB acylamides
3.3.1. Aqueous solubility
The MP-CXB was synthesized with the intention of enhancing
solubility in the GIT over all the pH range. The solubility of this
amide, CXB, Cy-CXB and the other benzamide in some dissolution
media is summarized in Table 14.
It is important to mention that the solubility of MP-CXB is higher
than 45 mg/mL regardless of the pH of the solution. This is due
to the fact that it contains functional groups that are ionizable at
Fig. 4. The hydrolysis of MP-CXB in pH 1.2, 6.8, and 7.4, 0.16 M phosphate buffer at
80 ◦C, ꢀ = 0.44.
both acidic and basic pH (calculated pKa: Npip-CH3 = 7.44, Npip
-
CH2-CO = 4.53 and NH-SO2 = 5.40). The halogenated benzamides
(CBz-CXB and FBz-CXB) have the lowest water solubility, which
is expected due the presence of the chlorine and fluorine atoms.
The solubility of Cy-CXB in basic media, were the drug is expected
to be ionized, is 100 more than that in a more acidic media, pH 5.4. It
must be mentioned here that in general, the solubility of all amides,
except MP-CXB, is lower at pH 7.4 compared to that in water (pH
6.9), which is may be due to the fact the buffer species have much
higher solubility in water than the tested amides and hence can salt
them out.
Table 15
The first order degradation rate constants (kobs) and half-lives of MP-CXB in pH 1.2,
6.8 and 7.4 at 80 ◦C.
pH
kobs (h−1
)
t1/2 (h)
3.53 0.05
1732.40 0.00
2310.00 0.00
1.2
6.8
7.4
1.96 × 10−1
4.00 × 10−4
3.00 × 10−4
3.00 × 10−3
0.00
0.00
with half-life of 3.53 h, compared to 72 and 96 days in pH 6.8 and
7.4, respectively.
3.3.2. Partition coefficient
The stability of MP-CXB in relatively basic media was sur-
prising, at first, since an extensive report by Larsen et al. (1988)
have previously demonstrated that the introduction of an ␣-amino
group (even a tertiary one) will facilitate the hydrolytic reaction
both at acidic and basic media by intramolecular catalysis (Larsen
et al., 1988). It is important to consider that while the amides
reported by Larsen et al. are tertiary, MP-CXB is secondary and
hence its amide group is prone to ionization, which renders it resis-
tant to hydrolysis, Fig. 5. Another additional explanation can be
attributed to the presence of another nitrogen atom in piperazine
ring. The most basic nitrogen atom in MP-CXB, due to steric and
electronic factors, will most likely be the N-methylated one and
Again, as seen in Table 14, MP-CXB has the lowest log Papp due
to the fact that it has a significant ionized fraction in the studied
pH range. The rest of the amides showed high log Papp values at pH
7.4 (3.59–5.35) and the most lipophilic were the benzamides, which
correlates with the solubility data in Table 14. This data is consistent
with reported hydrophobicity constants of cyclohexane, benzene,
chlorobenzene and fluorobenzene (Hung et al., 2010; Ruelle, 2000).
3.4. Chemical hydrolysis of MP-CXB in aqueous phosphate buffer
The chemical hydrolysis of MP-CXB was studied at GIT-relevant
pH values, i.e. 1.2, and 6.8 and 7.4. In all experiments, 0.16 M aque-
ous phosphate buffer was used and the ionic strength was adjusted
to 0.44. First, the hydrolysis was studied at 80 ◦C and as seen in
Fig. 4, the chemical hydrolysis follows pseudo first-order kinetics.
The first order degradation rate constants (kobs) and half-lives of
MP-CXB in pH 1.2, 6.8 and 7.4 at 80 ◦C are summarized in Table 15.
It can be seen that MP-CXB was hydrolyzed the fastest in pH 1.2
Fig. 5. The resonance stabilization of the deprotonated sulfonyl acylamide groups.

94
A.M. Qandil et al. / International Journal of Pharmaceutics 416 (2011) 85–96
Fig. 7. Arrhenius plot of MP-CXB in 0.16 M phosphate buffer pH 1.2.
Fig. 6. The hydrolysis of MP-CXB in pH 1.2 in 0.16 M phosphate buffer at 60, 70,
80 ◦C, ꢀ = 0.44.
Table 16
The observed rate constants and half-lives for degradation of MP-CXB in aqueous
buffer solutions pH 1.2 at different temperatures.
Temperature, T (K)
kobs (h−1
)
R2
t1/2 (h)
333
343
353
2.45 × 10−2
5.95 × 10−2
1.96 × 10−1
3.00 × 10−4
6.00 × 10−5
2.80 × 10−3
0.999
0.999
0.9995
28.28 0.34
11.64 0.13
3.53 0.05
hence, undermines the ␣-nitogen’s participation in hydrolysis. On
the other hand, in acidic media, MP-CXB is hydrolyzed in a faster
rate and intramolecular catalysis described by Larsen et al. (1988)
will play a role since both nitrogens of the piperazine can be ion-
ized at this pH. In addition, at this pH, the amide of MP-CXB will
not be ionized and there will be much less resistance to hydroly-
sis.
Since the hydrolysis of MP-CXB at pH 6.8 and 7.4 was slow,
only the rate of its hydrolysis at pH 1.2 was investigated further by
determining its hydrolysis rate at 60 and 70 ◦C in 0.16 M phosphate
buffer of ionic strength of 0.44, Fig. 6 and Table 16.
Fig. 8. Degradation of Cy-CXB in 0.16 M phosphate buffer pH 5.4 and 7.4 at 80 ◦C,
ꢀ = 0.44.
3.5. Chemical hydrolysis of Cy-CXB in aqueous phosphate buffer
The effect of pH on the chemical hydrolysis of the synthesized
Cy-CXB was studied at skin-relevant pH values, i.e. 5.4 and 7.4 in
0.16 M aqueous phosphate buffer solutions of ionic strength of 0.44
at 80 ◦C (Fig. 8).
The observed rate constants of Cy-CXB in pH 5.4 and 7.4 were
1.30 × 10−3 (R2 = 0.9675) and 2.00 × 10−4 (R2 = 0.9386), respec-
tively. It will be difficult to compare these data to those of MP-CXB
because of the difference in the dissolution media, i.e. the use of
acetonitrile as a cosolvent in the study of the hydrolysis of Cy-CXB.
The hydrolysis of Cy-CXB was slow at the selected pH values; hence,
no further studies were performed.
It is obvious that the hydrolysis is influenced by temperature
and the relation between ln kobs and 1/T followed the Arrhe-
nius relationship, R2 = 0.9911, over the temperature region studied
according to the following equation: ln kobs = ln A − (Ea/RT). Where
(Ea) is the energy of activation and (A) is the pre-exponential
coefficient for the degradation reaction. The activation energy for
the hydrolysis of MP-CXB in aqueous solution was found to be
23.83 kcal mol−1 while the pre-exponential coefficient was found
to be 32.27. The observed first-order constant and half-life at room
temperature 298 K and at 310 K are also calculated from the lin-
ear plot of Arrhenius equation and found to be 3.36 × 10−4 h−1
and 2059.82 h (86 days), 1.60 × 10−3 h−1 and 433.42 h (18 days),
respectively (Fig. 7).
3.6. In vitro enzymatic hydrolysis study
3.6.1. Hydrolysis in liver homogenate
The enzymatic hydrolysis of all acylamide was studied in 20%
liver homogenate at 37 ◦C. Of all the studied amides only the ali-
Fig. 9. Enzymatic hydrolysis of Cy-CXB in 20% liver homogenate at 37 ◦C.

A.M. Qandil et al. / International Journal of Pharmaceutics 416 (2011) 85–96
95
Fig. 11. Enzymatic hydrolysis of MP-CXB and Cy-CBX in 80% human plasma at 37 ◦C.
Fig. 10. Enzymatic hydrolysis of MP-CXB, Bz-CXB, CBz-CXB and FBz-CXB in 20%
liver homogenate at 37 ◦C.
The fact that only Cy-CXB underwent hydrolysis in liver
homogenate and taking in consideration the report by Mamidi et al.
(2002) leads to important observations. Firstly, it is obvious that
the enzymes in liver homogenate can hydrolyze aliphatic amides
of celecoxib whether they are straight chain or cycles. Secondly, the
hydrolysis rate increases as the length of the side chain increases,
but in the case of the cyclohexyl moiety it appears that its steric bulk
lead to slower hydrolysis. Thirdly, from the data presented in this
work, it is clear that lipophilicity is not the decisive factor in the rate
of hydrolysis since lipophilic aromatic amide (Bz-CXB, CBz-CBX
and FBz-CZB) did not show any hydrolysis. These observations can
lead to the suggestion that the binding of the amide to its hydrolyz-
ing enzyme might involve hydrophobic interactions rather than van
der Waal interactions since amides bearing saturated hydrocarbon
side chains were selectively hydrolyzed. Hydrophobic interactions
increases in strength in a proportional manner to the side chain size
(Davey et al., 1976; Ho et al., 1973; Lew et al., 2000) and it becomes
weaker when the side chain is branched or cyclic (Imai et al., 2006).
In addition, these interactions are entropy driven and the aromatic
rings are generally much better hydrated than aliphatic or alicyclic
side chains (Raschke and Levitt, 2004), hence aromatic rings do not
form hydrophobic interaction as readily as saturated hydrocarbons.
The lack of hydrolysis of the arylamides also suggests that van der
Waal forces may not play a significant role in the amide-enzyme
interaction. Finally, the above conclusion can easily predict that
MP-CXB will not be hydrolyzed in liver homogenate. If this infor-
mation were available to us before hand, our choice in a candidate
for a prodrug with enhanced water solubility in the GIT conditions
would have taken a different direction, which further emphasizes
the importance of this work. It is worth mentioning that a report
for the liver homogenate hydrolysis of amide prodrugs of ketorolac
showed that the hydrolysis reaction can be selective with regard
to the chemical structure of the promoiety distinguishing between
cyclic amide Cy-CXB was hydrolyzed under these conditions. The
disappearance of Cy-CXB and the appearance of the parent drug,
CXB, are shown in Fig. 9(a). As seen in Fig. 9(b), the hydrolysis
of Cy-CXB follows pseudo-first order kinetic with a kobs equal to
1.83 × 10−1 and half-life of 3.79 h (R2 = 0.9977).
As mentioned earlier, MP-CXB, Bz-CXB, CBz-CXB and FBz-CXB
did not show visible hydrolysis in 20% liver homogenate, Fig. 10.
Further discussion of the point can be seen in the next section.
3.6.2. Hydrolysis in human plasma
The hydrolysis of Cy-CXB, which was the only amide that
showed hydrolysis in liver homogenate, and MP-CXB were stud-
ied in 10% human plasma. As seen in Fig. 11, there is no detectable
hydrolysis even after 18–24 h.
Examining the literature concerning prodrugs reveals that
they are most commonly esters. The ester linkage can easily be
hydrolyzed by various esterases, amidases, and proteases and
sometimes can be rapidly hydrolyzed at pH 7.4 without the aid
of enzymes (Qandil et al., 2008). The secondary sulfonyl acylamide
moiety, on the other hand, is much more resistant to hydrolysis
chemically and in plasma as demonstrated by this work and ear-
lier reports, most likely due to ionization of the sulfonamide amide
group (Larsen and Bundgaard, 1987; Mamidi et al., 2002) and only
tertiary sulfonyl acylamide, which is unionizable due to the lack
of the acidic proton, were shown to undergo chemical hydroly-
sis and in human plasma with relative ease (Larsen et al., 1988).
On the other hand, Mamidi et al. (2002) shown that simple acy-
lamides of CXB, Fig. 12, were hydrolyzed to variable degrees in
live homogenates with the following approximate half-lives, Ac-
CXB < 2 h and Pr-CXB = Bu-CXB < 30 min. It has also been reported
that parecoxib, Fig. 12, is hydrolyzed to its parent drug valdecoxib
by a non cytochrome P-450 hepatic enzyme (Karim et al., 2001).
Fig. 12. The chemical structures of Mamidi et al. CXB acylamides and parecoxib.

96
A.M. Qandil et al. / International Journal of Pharmaceutics 416 (2011) 85–96
the bulkiness of the group as well as its electronic properties (Kim
et al., 2005).
Karim, A., Laurent, A., Slater, M., Kuss, M., Qian, J., Crosby-Sessoms, S., Hubbard,
R., 2001. A pharmacokinetic study of intramuscular (i.m.) parecoxib sodium in
normal subjects. Clin. Pharmacol. 41, 1111–1119.
Karnes, H.T., March, C., 1993. Precision, Accuracy, and Data Acceptance Criteria in
Biopharmaceutical Analysis. Springer Netherlands.
Kim, B., Doh, H., Le, T., Cho, W., Yong, C., Choi, H., Kim, J., Lee, C., Kim, D., 2005.
Ketorolac amide prodrugs for transdermal delivery: stability and invitro rat skin
permeation studies. Int. J. Pharm. 293, 193–202.
4. Conclusion
Various sodium celecoxib acylamides were successfully synthe-
sized. Two simple and relatively fast HPLC methods were developed
and validated for the determination of theses amides and CXB in
aqueous solution and from biological samples. The two methods
were linear, precise, accurate with very good limits of detection of
all analytes. These methods allowed for the simultaneous deter-
mination of celecoxib and its acylamides. The physico-chemical
properties of the synthesized amides were determined. This work
has shown that the sulfonyl acylamides are very stable at room
temperature and resistant to hydrolysis in basic conditions even
at high temperatures while they can be hydrolyzed in acidic pH
1.2, at high temperatures. Only aliphatic and alicylic acylamides
of celecoxib can be hydrolyzed in liver homogenate which lead
to the conclusion that hydrophobic interactions in the enzyme’s
catalytic site are a determining factor in this hydrolysis. The trans-
dermal permeation Cy-CXB as potential prodrug of CXB and the
oral bioavailability CXB from this amide will be fully investigated
in the near future. MP-CXB will not be investigated further because
of its resistance to enzymatic and chemical hydrolysis at physiolog-
ical conditions although it possesses the desired physico-chemical
properties.
Kısmet, K., Akay, M.T., Abbaso, O., Ercan, A., 2004. Celecoxib:
a potent
cyclooxygenase-2 inhibitor in cancer prevention. Cancer Detect. Prev. 28,
127–142.
Larsen, J.r.D., Bundgaard, H., 1987. Prodrug forms for the sulfonamide group. I. Eval-
uation of N-acyl derivatives, N-sulfonylamidines, N-sulfonylsulfilimines and
sulfonylureas as possible prodrug derivatives. Int. J. Pharm. 37, 87–95.
Larsen, J.r.D., Bundgaard, H., Lee, V.H.L., 1988. Prodrug forms for the sulfonamide
group. II. Water-soluble amino acid derivatives of N-methylsulfonamides as
possible prodrugs. Int. J. Pharm. 47, 103–110.
Lew, W., Chen, X., Kim, C.U., 2000. Discovery and development of GS 4104
(oseltamivir): an orally active influenza neuraminidase inhibitor. Curr. Med.
Chem. 7, 663–672.
Lindholm, J., 2004. Developement and Validation of HPLC methods for analytical and
preparative purposes. PhD Thesis, University of Uppsala, Uppsala.
Mamidi, R., Mullangi, R., Kota, J., Bhamidipati, R., Khan, A., Katneni, K., Datla, S.,
Singh, S., Rao, K., Rao, C., Srinivas, N., Rajagopalan, R., 2002. Pharmacological
and pharmacokinetic evaluation of celecoxib prodrugs in rats. Biopharm. Drug
Dispos. 23, 273–282.
Paulson, S., Vaughn, M., Jessen, S., Awal, Y., Gresk, C.J., Timothyg, B., Cook, M.C., Karim,
A., 2001. Pharmacokinetics of celecoxib after oral administration in dogs and
humans: effect of food and site of absorption. J. Pharm. Exp. Ther. 297, 638–645.
Qandil, A., Al-Nabulsi, S., Al-Taani, B., Tashtoush, B., 2008. Synthesis of piperaziny-
lalkyl ester prodrugs of ketorolac and their in vitro evaluation for transdermal
delivery. Drug Dev. Ind. Pharm. 34, 1054–1063.
Radi, Z.A., Khan, N.K., 2006. Effects of cyclooxygenase inhibition on the gastrointesti-
nal tract. Exp. Toxicol. Pathol. 58, 163–173.
Raschke, T.M., Levitt, M., 2004. Detailed hydration maps of benzene and cyclohexane
reveal distinct water structures. J. Phys. Chem. B 108, 13492–13500.
Ruelle, P., 2000. The n-octanol and n-hexane/water partition coefficient of envi-
ronmentally relevant chemicals predicted from the mobile order and disorder
(MOD) thermodynamics. Chemosphere 40, 457–512.
Shabir, G.A., 2003. Validation of high-performance liquid chromatography meth-
ods for pharmaceutical analysis: understanding the differences and similarities
between validation requirements of the US Food and Drug Administration, the
US Pharmacopeia and the International Conference on Harmonization. J. Chro-
matogr. A 987, 57–66.
Silverstein, F.E, Faich, G., Goldstein, J.L., Simon, L.S., Pincus, T., Whelton, A., Makuch,
R., Eisen, G., Agrawal, N.M., Stenson, W.F., Burr, A.M., Zhao, W.W., Kent,
J.D., Lefkowith, J.B., Verburg, K.M., Geis, G.S., 2000. Gastrointestinal toxicity
with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and
rheumatoid arthritis: The CLASS Study: A Randomized Controlled Trial. JAMA
284, 1247–1255.
Subramanian, N., Ray, S., Ghosal, S.K., Bhadra, R., Moulik, S.P., 2004. Formula-
tion design of self-microemulsifying drug delivery systems for improved oral
bioavailability of celecoxib. Biol. Pharm. Bull. 27, 1993–1999.
References
Baboota, S., Shakkel, F., Ahuja, A., Ali, J., Shfiq, S., 2007. Design, development and
evaluation of novel nanoemulsion formulations for transdermal potential of
celecoxib. Acta Pharm. 57, 315–332.
Bombardier, C., Laine, L., Reicin, A., Shapiro, D., Burgos-Vargas, R., Davis, B., Day, R.,
Ferraz, M.B., Hawkey, C.J., Hochberg, M.C., Kvien, T.K., Schnitzer, T.J., 2000. Com-
parison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients
with rheumatoid arthritis. N. Engl. J. Med. 343, 1520–1528.
Center for Drug Evaluation and Research (CDER), 1994. Reviewer Guidance. Valida-
tion of Chromatographic Methods, p. 30.
Center for Drug Evaluation and Research (CDER), 2001. Guidance for Industry. Bio-
analytical Method Validation, p. 25.
Davey, M.W., Sulkowski, E., Carter, W.A., 1976. Hydrophobic interaction of human,
mouse, and rabbit interferons with immobilized hydrocarbons. J. Biol. Chem.
251, 7620–7625.
Duttaa, N., Sarotraa, P., Guptaa, S., Aggarwal, R., Agnihotri, N., 2009. Mechanism of
action of celecoxib on normal and acid-challenged gastric mucosa. Exp. Toxicol.
Pathol. 61, 353–361.
El-Medanya, A., Mahgouba, A., Mustafa, A., Arafa, M., Morsi, M., 2005. The effects of
selective cyclooxygenase-2 inhibitors, celecoxib and rofecoxib, on experimental
colitis induced by acetic acid in rats. Eur. J. Pharmacol. 507, 291–299.
Ermer, J., Miller, J.H.M., 2005. Method Validation in Pharmaceutical Analysis: A Guide
to Best Practice. Wiley-VCH, Weinheim.
Talley, J.J., Bertenshaw, S.R., Brown, D.L., Carter, J.S., Graneto, M.J., Kellogg,
M.S., Koboldt, C.M., Yuan, J., Zhang, Y.Y., Seibert, K., 2000. N-[[(5-Methyl-
3-phenylisoxazol-4-yl)-phenyl]sulfonyl]propanamide, sodium salt, parecoxib
sodium: Äâ a potent and selective inhibitor of COX-2 for parenteral adminis-
tration. J. Med. Chem. 43, 1661–1663.
USP, 1994. Validation of Compendial Methods. United States Phamracopeial Con-
vention, Inc., Rockvill, MD, p. 1982.
Ho, Y.K., Zakrzewski, S.F., Mead, L.H., 1973. Hydrophobic interactions between the 5-
alkyl group of 2,4-diamino-6-methylpyrimidines and dihydrofolate reductase.
Biochemistry 12, 1003–1005.
Hung, H.-W., Lin, T.-F., Chiou, C.T., 2010. Partition coefficients of organic contami-
nants with carbohydrates. Environ. Sci. Technol. 44, 5430–5436.
Imai, T., Taketani, M., Shii, M., Hosokawa, M., Chiba, K., 2006. Substrate specificity of
carboxylesterase isozymes and their contribution to hydrolase activity in human
liver and small intestine. Drug Metab. Dispos. 34, 1741–1743.
International Conference on Harmonisation (ICH), 1995. Validation of Analytical
Procedures, ICH Q2A, p. 9.
Vane, J., Botting, R., 1998. Anti-inflammatory drugs and their mechanism of action.
Inflamm. Res. 47, S78–S87.
Ventura, C.A., Giannone, I., Paolino, D., Pistarà, V., Corsaro, A., Puglisi, G.,
2005. Preparation of celecoxib-dimethyl-b-cyclodextrin inclusion complex:
characterization and invitro permeation study. Eur. J. Med. Chem. 40,
624–631.
Vonkeman, H.E., van de Laar, M.A.F.J., 2010. Nonsteroidal anti-inflammatory drugs:
adverse effects and their prevention. Semin. Arthritis Rheum. 39, 294–312.
International Conference on Harmonisation (ICH), 1997. Validation of Analytical
Procedures Methodology, ICH Q2B, p. 13.