Physical-chemical parameters and validation of a colorimetric method for deoxycholic and ursodeoxycholic acids: Kit reagent and optical sensor

Article abstract of DOI:10.1016/j.chemphyslip.2010.11.004

The simple and low cost β-cyclodextrin (β-CD)-phenolphthalein (PHP) inclusion complex was used for both the study of physical-chemical parameters and validation of analytical procedures for deoxycholic acid (DCA) and ursodeoxycholic acid (UDCA) determinations in different formulations. The usefulness of this inclusion complex is proposed either in the form of kit reagent and as an original optical sensor for DCA and UDCA. The results showed that temperature had a negative effect on the equilibrium constant resulting in high negative values of enthalpy and positive values of entropy. The half-life values for DCA and UDCA measurements were 68.71 and 294.71 days, respectively. The method was validated showing limits of detection and quantification of 4.92 × 10^-5 mol L^-1 and 1.64 × 10^-4 mol L^-1 for DCA, 1.14 × 10^-5 mol L^-1 and 3.79 × 10^-5 mol L^-1 for UDCA, respectively. The developed optical sensor also showed response linearity, ease of implementation and potential application in fast screening tasks even out of the laboratory.

Full text of DOI:10.1016/j.chemphyslip.2010.11.004

Chemistry and Physics of Lipids 164 (2011) 99–105
Contents lists available at ScienceDirect
Chemistry and Physics of Lipids
journal homepage: www.elsevier.com/locate/chemphyslip
Physical–chemical parameters and validation of a colorimetric method for
deoxycholic and ursodeoxycholic acids: kit reagent and optical sensor
Pabyton G. Cadenaa,b,∗, Alberto N. Araújoc, Maria C.B.S.M. Montenegroc,
Maria C.B. Pimentela, José L. Lima Filhoa, Valdinete L. Silvab
a Lab. de Imunopatologia Keizo Asami (LIKA), Universidade Federal de Pernambuco (UFPE), Av. Prof. Moraes Rego, s/n 50780-901 Recife, Pernambuco, Brazil
b Lab. de Engenharia Ambiental e da Qualidade (LEAQ), Universidade Federal de Pernambuco (UFPE), Av. Prof. Arthur de Sá,
s/n Cidade Universitária, 50740-521 Recife, Pernambuco, Brazil
c REQUIMTE, Dep. de Química-Física, Faculdade de Farmácia, Rua Aníbal Cunha, 164, 4099-030 Porto, Portugal
a r t i c l e i n f o
a b s t r a c t
Article history:
The simple and low cost ␤-cyclodextrin (␤-CD)–phenolphthalein (PHP) inclusion complex was used for
both the study of physical–chemical parameters and validation of analytical procedures for deoxycholic
acid (DCA) and ursodeoxycholic acid (UDCA) determinations in different formulations. The usefulness of
this inclusion complex is proposed either in the form of kit reagent and as an original optical sensor for
DCA and UDCA. The results showed that temperature had a negative effect on the equilibrium constant
resulting in high negative values of enthalpy and positive values of entropy. The half-life values for
DCA and UDCA measurements were 68.71 and 294.71 days, respectively. The method was validated
showing limits of detection and quantification of 4.92 × 10−5 mol L−1 and 1.64 × 10−4 mol L−1 for DCA,
1.14 × 10−5 mol L−1 and 3.79 × 10−5 mol L−1 for UDCA, respectively. The developed optical sensor also
showed response linearity, ease of implementation and potential application in fast screening tasks even
out of the laboratory.
Received 2 December 2009
Received in revised form
22 November 2010
Accepted 25 November 2010
Available online 1 December 2010
Keywords:
Deoxycholic acid
Ursodeoxycholic acid
␤-Cyclodextrin
Inclusion complex
Phenolphthalein
Optical sensor
© 2010 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
and FDA in US. Another important bile acid, the ursodeoxycholic
acid (UDCA) is also used in the treatment of primary biliary cir-
Besides playing an important physiological role in biological sys-
tems some bile acids such as the deoxycholic acid (DCA), are also
employed to enhance oral availability of biodegradable nanoparti-
cles (Samstein et al., 2008), as choleretic agent in liver dysfunctions
(Rodriguez et al., 2000), as N-(2-dimethylamino)ethyl derivatives
in malaria treatment (Terzic et al., 2007) and in several cosmetic
preparations (Valenta et al., 1999). Injection lipolysis for sub-
cutaneous application primarily aiming cosmetic proposes were
initially used in Brazil. They were based on a mixture of phos-
phatidylcholine and DCA as major active components (Duncan
et al., 2009; Rotunda et al., 2004), being lately replaced by formu-
lations containing only DCA (Rotunda and Kolodney, 2006; Yagima
Odo et al., 2007). However, due to the associated health risks
both of them have been forbidden by national sanitary authorities
like ANVISA (the Brazilian National Health Surveillance Agency)
rhosis, primary cholangitic sclerosis (Lindor, 1997), cholelithiasis
(Petroni et al., 2001), to prevent the relapse of acute pancreatitis
caused by microlithiasis (Okoro et al., 2008) and to reduce ala-
nine aminotransferase levels in hepatitis C (Ikegami and Matsuzaki,
2008).
Due to stated applications of both bile acids, a variety of meth-
ods are routinely used to measure their content in pharmaceutical
formulations, such as: electrochemical (voltammetric), fluorimet-
ric or spectrophotometric (Arias De Fuentes et al., 2000), HPLC
(Scalia et al., 1989) and micellar electrokinetic chromatographic
(Rodriguez et al., 2000). Nevertheless, each of one has its own disad-
vantage (Arias De Fuentes et al., 2000; Lin et al., 2003) which added
to complexity of the samples treatment turn difficult the “in loco”
control by national sanitary authorities thus usually addressing the
formulation or cosmetic sample into the laboratory for further pro-
cessing and analysis. The research about validated simple and low
cost methods for measurement of bile acids in commercial formu-
lations can be also useful for screening tasks during technological
development replacing expensive technologies more difficult to
access in developing countries. In this way, inclusion complexes
formed between cyclodextrins and indicators provide the basis
∗
Corresponding author at: Lab. de Imunopatologia Keizo Asami (LIKA), Universi-
dade Federal de Pernambuco (UFPE), Av. Prof. Moraes Rego, s/n 50780-901 Recife,
Pernambuco, Brazil. Tel.: ++55 81 2126 8484; fax: +55 81 2126 8485.
E-mail address: pabyton@yahoo.com.br (P.G. Cadena).
0009-3084/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.chemphyslip.2010.11.004

100
P.G. Cadena et al. / Chemistry and Physics of Lipids 164 (2011) 99–105
for simplified methods of measurement of hydrophobic molecules
such as is the case of bile acids (Cadena et al., 2009).
was strongly mixed after each solution addition. The blank solution
was composed of 1 mL the same buffer plus 3 mL of water.
Absorption spectra were obtained at pH 10.5, containing
1.55 × 10−4 mol L−1 phenolphthalein (PHP) and different amounts
of ␤-cyclodextrin (␤-CD), 3.88 × 10−5 up to 6.20 × 10−4 mol L−1, at
25 ◦C. Spectra for DCA analysis were carried out using 4.38 × 10−5
up to 7.0 × 10−4 mol L−1 DCA solutions and ␤-CD–PHP complex
in the fixed proportion of 6.2 × 10−4:1.55 × 10−4 mol L−1. In turn,
for UDCA analysis 1.19 × 10−5 up to 1.9 × 10−4 mol L−1 solutions
of this bile acid were assayed with the ␤-CD–PHP complex
(3.1 × 10−4:7.75 × 10−5 mol −1).
The important property of cyclodextrins (CDs) – cyclic oligosac-
charides with six (␣-), seven (␤-) or eight (␥-) glucose residues
linked by ␣-(1–4) glycosidic bonds (Yanez et al., 2004) – and their
numerous derivatives is the ability to form inclusion complexes
with inorganic and organic guests (Brewster and Loftsson, 2007).
Concomitantly, inclusion in cyclodextrins exerts a profound effect
on the physicochemical properties of guest molecules such as sol-
ubility, chemical stability, absorption and bioavailability (Zhang
et al., 2009). Guests reaction with CDs by competitive complexation
with indicators has been used to assess the respective equilibrium
constants (Kc) and other related thermodynamic parameters due to
the fails in the direct determination of the complex. For example,
phenolphthalein (PHP) is a typical acid/base indicator that forms a
colourless 1:1 inclusion complex with ␤-CD and by this used in indi-
rect determinations of other colourless compounds by competitive
complexation reaction (Afkhami et al., 2006; Glazyrin et al., 2004).
Herein, the temperature dependence of the equilibrium con-
stants of the PHP–, DCA– and UDCA–␤-cyclodextrin were used
to obtain thermodynamic parameters, i.e. the free energy change
(ꢀG), the enthalpy change (ꢀH) and the entropy change
(ꢀS). The usefulness of the obtained data is evidenced in the
subsequent validation of analytical procedures based on the ␤-
cyclodextrin–phenolphthalein (␤-CD–PHP) inclusion complex as
kit reagent or optical sensor both providing rapid, inexpensive and
simple alternative method for “in loco” measurements and fast
screening.
Temperature effect was assessed in the range from 10 up to 55 ◦C
for the 8.75 × 10−5 up to 1.40 × 10−3 mol L−1 DCA and 2.38 × 10−5
up to 1.9 × 10−4 mol L−1 UDCA concentrations, while keeping pH
constant at the value of 10.5.
The effect of the ionic strength on inclusion complex forma-
tion with DCA or UDCA and ␤-CD was examined in the range from
0.11 up to 0.40 using ␤-CD–PHP – 6.2 × 10−4:1.55 × 10−4 mol L−1
for DCA and 3.1 × 10−4:7.75 × 10−5 mol L−1 for UDCA, according to
Wang et al. (2007).
Storage stability of the kit reagent developed was evaluated
by maintaining two solutions of this complex at 25 ◦C during 60
days in dark flasks. At defined time intervals, samples were with-
drawn from each of one for testing with solutions containing
7.0 × 10−4 mol L−1 of DCA and 1.9 × 10−4 mol L−1 of UDCA, respec-
tively.
2.2.3. Validation of the method (kit reagent)
The methodology used to validate the determination of DCA
and UDCA followed the procedures presented by the EMEA (Euro-
pean Medicines Agency – CPMP/ICH/381/95) and ANVISA (Brazilian
National Health Surveillance Agency – RE 899, May 2003).
Absorbances vs. concentration linearity was evaluated using
authentic solutions of each bile acid and from them 10 samples at
the range 8.3 × 10−6 up to 3.36 × 10−3 mol L−1 for DCA or 8.0 × 10−6
up to 4.0 × 10−4 mol L−1 for UDCA were measured. Assays with each
solution were performed in triplicate. At the experimental condi-
tions previously established, the linearity of the calibration graphs
were validated for both bile acids by the least squares method and
the one-way analysis of variance (ANOVA) for p < 0.05.
Standard curves were then established to 20% over the spec-
ified range, 5.6 × 10−4 up to 8.4 × 10−4 mol L−1 for the DCA and
from 1.52 × 10−4 up to 2.28 × 10−4 mol L−1 for the UDCA, respec-
tively. Therefore, results of the triplicate assays at five different
concentrations were considered.
For evaluation of the repeatability, three concentrations of
the DCA (5.6 × 10−4, 7.0 × 10−4 and 8.4 × 10−4 mol L−1) and UDCA
(1.52 × 10−4, 1.9 × 10−4 and 2.28 × 10−4 mol L−1) were assayed in
nine determinations (3 concentrations/3 replicates).Intermediate
precision was assessed with solutions of the same concentration in
a 22 full factorial design considering different analysts and equip-
ments (Pharmacia Ultrospec 3000pro and Micronal B582). The limit
of detection was defined as CL = 3.3SB/m, where CL, SB, and m are
respectively the limit of detection, standard deviation and slope,
also the same approach was applied for the limit of quantification
defined by CLQ = 10SB/m.
Accuracy was established by comparison of the concentrations
found in pharmaceutical formulations: Injectable Phosphatidyl-
choline formula (Rotunda et al., 2004) – phosphatidylcholine 5%
(w/v), deoxycholic acid (sodium salt) 4.75% (w/v), benzyl alcohol
0.9% (v/v), water 100 mL; injectable deoxycholate formula (Yagima
Odo et al., 2007) – deoxycholic acid (sodium salt) 2.5% (w/v), ben-
zyl alcohol 1% (v/v), propylene glycol 10% (v/v), water 100 mL;
Ursacol® – Ursodeoxycholic acid 300 mg and excipients (lac-
tose, povidone, crospovidone, magnesium stearate) with the ones
obtained for standardized solutions of DCA (5.6 × 10−4, 7.0 × 10−4
2. Experimental
2.1. Materials
Analytical grade chemicals without any further purification
treatment and double deionised water were used thoroughly.
␤-cyclodextrin was obtained from Fluka (Steinheim, Germany).
Phenolphthalein, deoxycholic acid (sodium salt) and ursodeoxy-
cholic acid were obtained from Sigma (St. Louis, MO, USA).
Absorption spectra were collected from a Pharmacia Ultrospec
3000pro UV/Vis spectrophotometer using 1-cm path length quartz
cells. Statistical evaluations were carried out by Statistica software
(StatSoft Inc., Tulsa, OK, USA) and the images were analysed by trial
version of Adobe Photoshop CS2 software (Adobe Systems, USA).
2.2. Methods
2.2.1. Preparation of analytical reagent
The co-precipitation technique (Del Valle, 2004) enabled to
prepare the inclusion complexes in batch conditions using the
following solutions: 1 mL of phenolphthalein solution, 1 mL of car-
bonate buffer solution (pH 10.5; 150 mmol L−1) (Afkhami et al.,
2006) and 1 mL of ␤-CD solution. For having a homogeneous system
the mixture was vigorously mixed after each solution addition (the
order did not interferes in the result), then readily transferred into
a dark flask and kept at room temperature (25 ◦C). Humidity and
temperature effects were considered neglected because all exper-
iments and kit storage were performed in the same acclimatized
room.
2.2.2. Development of the method (kit reagent)
The competitive inclusion complexes above obtained were just
homogenized with either 1 mL of water (control) or 1 mL of bile
acid solution (sample). The positive control to study the absorbance
of the phenolphthalein was made using 1 mL of PHP solution, 1 mL
carbonate buffer (pH 10.5; 150 mmol L−1) and 2 mL of water, which

P.G. Cadena et al. / Chemistry and Physics of Lipids 164 (2011) 99–105
101
Fig. 1. Absorption spectrum of phenolphthalein (PHP
–
1A) (1.55 × 10−4 mol L−1
)
at pH 10.5 in different ␤-cyclodextrin (␤-CD) concentrations (3.88 × 10−5
up to 6.20 × 10−4 mol L−1). Determination of (4.38 × 10−5 up to 7.0 × 10−4 mol L−1
) deoxycholic acid (DCA – 1B) by the inclusion complex of ␤-CD–PHP
(6.2 × 10−4:1.55 × 10−4 mol L−1) and (1.19 × 10−5 up to 1.9 × 10−4 mol L−1) of ursodeoxycholic acid concentrations (UDCA – 1C) by the inclusion complex of ␤-CD–PHP
(3.1 × 10−4:7.75 × 10−5 mol L−1).
and 8.4 × 10−4 mol L−1) and UDCA (1.52 × 10−4, 1.9 × 10−4 and
solution colour change can then be easily determined by standard
curve of the free PHP and the corresponding concentration related
with the amount of bile acid.
2.28 × 10−4 mol L−1) in nine determinations (3 concentrations/3
replicates).
The effect of temperature on the absorbance of solutions con-
taining the complexes with PHP, DCA and UDCA is presented in
Fig. 2, where can be observed that with the increase of tempera-
ture an increase in absorbance was induced. Similar findings were
previously described by Zarzycki and Lamparczyk (1998) being this
effect caused by destabilization of the complex (Del Valle, 2004),
with consequent phenolphthalein release. Hence, it becomes dif-
ficult to distinguish between the absorbance increase caused by
an increase in the bile acid affinity by the ␤-CD cavity or the sim-
ple destabilization of the inclusion complex. For this reason, the
temperature for the following studies was fixed at 25 ◦C, once it
was enabled in both extended absorbance range for free PHP and
␤-CD–PHP inclusion complex, and there was no significant differ-
ences for robustness (p < 0.05 by Turkey’s test) working between
20 and 30 ◦C.
The robustness was evaluated based on the variations in the
analytical conditions, such as pH (10.3–10.7) and water (distilled,
deionised, double deionised), using different settling conditions for
each variation.
2.3. Optical chemical sensor for bile acids
Optical chemical sensor strips were implemented using
2 cm × 3.5 cm Bristol-paper strips afterwards soaked for 2 min in
5 mL of 1% (w/v) sodium alginate gel containing the ␤-CD–PHP
inclusion complex (6.2 × 10−4:1.55 × 10−4 mol L−1 for DCA and
3.1 × 10−4:7.75 × 10−5 mol L−1 for UDCA) and finally dried for 24 h
(25 ◦C). Later, 20 ␮L of DCA (1.68 × 10−3 up to 8.40 × 10−3 mol L−1
)
or UDCA (4.56 × 10−4 up to 2.28 × 10−3 mol L−1) solution was
added. Colour changes were evaluated using a scanner (Model HP
5590) for obtaining images with 600dpi definition. All the images
were downloaded to a computer, each image was converted to 8-bit
grayscale by Adobe Photoshop software and the arithmetic mean
of pixel intensity within each test zone was used to quantify the
colorimetric response by the observed RGB (red, green, and blue
colour system) values of digital images (Yang et al., 2007). Finally,
the background-corrected response was calculated by subtraction
of the each sample and control (without bile acids) values (Martinez
et al., 2008a).
The equilibrium constants (Kc1:1 bile acid/cyclodextrin) for the
␤-CD–PHP, ␤-CD–DCA and ␤-CD–UDCA inclusion complexes were
obtained by means of the Benesi–Hildbrand plot (Abdel-Shafi,
2007; Benesi and Hildebrand, 1949) according to Eqs. (1) and (2):
[␤-CD–PHP]
[PHP][␤-CD]
PHP + ␤-CD ꢀ ␤-CD–PHP, Kc =
(1)
3. Results and discussion
3.1. Development of the method (kit reagent)
The spectral changes obtained for phenolphthalein solutions in
the presence of different ␤-CD concentrations are shown in Fig. 1A.
Phenolphthalein interacted with the ␤-CD in alkaline pH causing a
decrease of more than 95% in the absorption band at the wavelength
of 553 nm. The ionised red form of PHP was forced into ␤-CD cavity,
forming its colourless lactone structure without protonation of the
phenolic groups (Afkhami et al., 2006). Spectra of phenolphthalein
solutions with increasing amount of ␤-CD revealed a proportional
decrease of the free PHP up to concentration of 6.2 × 10−4 mol L−1
at pH 10.5 resulting in the ratio of 1:4 for PHP:␤-CD. Higher con-
centrations of ␤-CD did not cause additional observable decrease
in absorbance. However, the addition of bile acids forced the PHP
leaves the cavity of ␤-CD molecules returning to its ionised red
form in the alkaline solution (Fig. 1B and C). Thus, the extent of the
Fig. 2. Temperature effect (10–55 ◦C) on the phenolphthalein (PHP); ␤-CD–PHP
inclusion complex (6.2 × 10−4:1.55 × 10−4 mol L−1) formation and complex interac-
tion with deoxycholic (DCA – 7.0 × 10−4 mol L−1) and ursodeoxycholic acids (UDCA
– 1.9 × 10−4 mol L−1).

102
P.G. Cadena et al. / Chemistry and Physics of Lipids 164 (2011) 99–105
Table 1
Equilibrium constants (Kc) of the inclusion complexes: ␤-cyclodextrin–phenolphthalein (␤-CD–PHP) without or with the deoxycholic acid (DCA) or ursodeoxycholic acid
(UDCA) at different temperatures and the thermodynamic parameters.
Inclusion complexes
T
(K)
Kc
ꢀG
ꢀH
ꢀS
(×104 L mol−1
)
(kJ mol−1
)
(kJ mol−1
)
(J mol−1 K−1
)
298
308
318
328
298
308
318
328
298
308
318
328
1.7( 0.1)
1.4( 0.1)
−2.4( 0.2)
−16( 1)
26( 3)
–
–
–
50( 5)
–
–
–
␤-CD–PHP
−2.5( 0.2)
–
1.17( 0.07)
0.93( 0.04)
2.60( 0.01)
2.39( 0.01)
2.12( 0.01)
1.77( 0.01)
2.81( 0.01)
2.43( 0.01)
2.13( 0.01)
1.76( 0.01)
−2.5( 0.2)
–
–
−2.5( 0.1)
−2.53( 0.01)
−2.59( 0.01)
−2.64( 0.01)
−2.68( 0.01)
−2.55( 0.01)
−2.59( 0.01)
−2.64( 0.01)
−2.67( 0.01)
−10( 1)
␤-CD–DCA
–
–
–
−13( 1)
43( 3)
␤-CD–UDCA
–
–
–
–
–
–
1
1
a
1
Apparent free energy change (ꢀG) was obtained according to
the Eq. (8):
=
+
(2)
A − A0
aKc[␤-CD0]
where A and A0 are the absorbances of PHP in the presence and
absence of ␤-CD, respectively, a is a constant related to the molar
absorption coefficient changes, and [␤-CD0] is the initial concen-
tration of ␤-CD. Yuexian et al. (2005) described the competitive
complexation equilibrium constant (Kc1:1) between amino acid and
methyl orange that can be applied for Kc determination between
DCA or UDCA with ␤-CD by Eqs. (3)–(6):
ꢀG = −RT ln K
(8)
The Van’t Hoff plots were linear and exhibited large negative ꢀH
and positive ꢀS for the developed complexes; these results were
similar to those reported by Holm et al. (2009) for conjugated bile
acids, suggesting an exothermic inclusion processes.
[␤-CD–DCA]
[DCA][␤-CD]
It was also observed an enthalpy decrease and entropy increase
to the inclusion complex with the bile acids compared to ␤-CD–PHP
inclusion complex. These results suggested that the inclusion pro-
cesses of DCA and UDCA into the ␤-CD cavity were more favourable
and spontaneous than that of PHP. In all cases ꢀG was negative
showing that the inclusion complexes formation can be sponta-
neous.
According to Jullian et al. (2008), the formation of an inclusion
complex with cyclodextrin occurs due to the hydrogen bonding
with the OH groups at the periphery of the cavity, Van der Waals
and hydrophobic interactions. Generally, solute inclusion in the
cyclodextrin cavity is associated with large negative values of ꢀH.
Either negative or slightly positive ꢀS values indicating inclu-
sion complexation of the guest without extensive desolvation in
a primarily enthalpy-driven process. The same behaviour has been
reported by several authors which confirmed the great contribution
from Van der Walls forces for the complexes formation (Brewster
and Loftsson, 2007; Castronuovo and Niccoli, 2006).
DCA + ␤-CD ꢀ ␤-CD–DCA, Kc-DCA
=
(3)
(4)
(5)
[␤-CD–UDCA]
=
[UDCA][␤-CD]
UDCA + ␤-CD ꢀ ␤-CD–UDCA, Kc-UDCA
[␤-CD0] − [CD]
KcBA
=
[CD]([BA]0 − [␤-CD0] + [CD])
where [BA]0 represented the initial concentration bile acid. KcBA
represented the equilibrium constant of bile acid–␤-CD, [CD] was
the uncomplexing ␤-CD concentration when bile acid, PHP and ␤-
CD coexisted in solution and it was related to the absorbance variety
of phenolphthalein in solution. [CD] could be obtained from:
EPHP − E
[CD] =
(6)
Kc(E − E␤-CD–PHP
)
where EPHP and E␤-CD–PHP represented the absorbance of uncom-
plexing phenolphthalein and ␤-CD–PHP inclusion complex,
respectively; E was the absorbance of system in which bile acid, PHP
and ␤-CD coexisted (Yuexian et al., 2005). For the selected wave-
length of 553 nm the corresponding molar absorptivity ␤-CD–PHP
complex is negligible, the absorbance of the solution is mainly
due to free PHP (uncomplexed). In turn, PHP releasing causes an
absorbance increase which is proportional to bile acids concentra-
tion in solution, allowing the Kc determination (Cadena et al., 2009).
The values of Kc were determined at various temperatures (25 up
to 55 ◦C) and at pH 10.5. Table 1 shows the temperature effect in
the equilibrium constant (Kc1:1 bile acid/cyclodextrin).
The thermodynamic parameters (Table 1) were calculated
according to the Van’t Hoff Eq. (7) which describes the temperature
dependence as function of K (Kc1:1). The ln K values were plotted as
a function of the inverse temperature to give a linear relationship.
Then, the apparent enthalpy (ꢀH) and entropy (ꢀS) changes were
obtained from the slope and the intercept of the curve (Del Valle,
2004; Wang et al., 2007; Yuexian et al., 2005).
The effect of ionic strength on inclusion of the ␤-CD–PHP com-
plex at constant concentration was examined in the range from
0.11 up to 0.40. For this study, different concentrations of NaCl solu-
tion were prepared leading to systems containing H+, Na+, CO3
,
−2
Cl−, PHP, and ␤-CD. Due to the lower concentration of PHP and the
neutral molecule ␤-CD, their contributions to ionic strength were
negligible, so the H+, Na+, CO3
and Cl− are only responsible to
−2
ionic strength in the systems. The changes in the absorbance of
inclusion complex as a function of ionic strength were carried out.
The absorbance of inclusion complex gradually decreased with a
rise in the ionic strength with significant difference by the Tukey
test (p < 0.05). May be due to the system polarities were enhanced
by ionic strength. Second Wang et al. (2007), the enhancement of
polarity of solution is favourable to the hydrophobic interaction
of ␤-CD molecules, which shields the interaction between ␤-CD
molecule and PHP molecule and leads to decrease of the absorbance
inclusion complex.
The study of the ␤-CD–PHP inclusion complex as kit reagent
showed that it was stable for bile acids determination up to 12 days.
A loss of about 30% for DCA and 12% regarding UDCA determinations
were observed after 30 days of storage at 25 ◦C. The inactivation
ꢀH
RT
ꢀS
R
ln K = −
+
(7)

P.G. Cadena et al. / Chemistry and Physics of Lipids 164 (2011) 99–105
103
Table 2
Validation data (p < 0.05).
Linearity
DCA
UDCA
Linearity range Standard curve
Correlation coefficient
8.30 × 10−6–1.68 × 10−3 mol L−1
ABS = −0.0069( 0.0091) + 870.0373( 12.4216)DCAmol/L
0.99919
8.00 × 10−6–2.28 × 10−3 mol L−1
ABS0.5 = 0.0699( 0.011) + 4506.688( 77.1179)UDCAmol/L
0.99898
Robustnessa
pH
DCA
–
UDCA
Water
DCA
102.38( 1.52)
100.11( 4.48)
100( 2.18)
UDCA
100.42( 0.95)
98.04( 1.86)
100( 0.66)
10.3
10.4
10.5
10.6
10.7
97.75( 1.49)
99.19( 1.76)
100.0( 2.15)
99.01( 1.85)
98.46( 0.97)
Distilled
Deionised
Double deionised
97.84( 3.53)
100.00( 3.48)
102.62( 2.48)
104.50( 0.46)
a
%Mean %R.S.D.
Table 3
Accuracy of the kit reagent (␤-CD–PHP) developed for different pharmaceutical formulations.
Drugs
Accuracy (%mean %R.S.D.)
Deoxycholic acid standard
98.97 ( 0.84)
97.11( 1.43)
98.08( 0.43)
99.43( 1.17)
98.1( 0.49)
101.36( 0.36)
101.33( 1.28)
102.77( 0.63)
100.89( 0.54)
98.7( 1.86)
99.31( 1.48)
100.51( 0.99)
102.52( 0.38)
99.63( 0.21)
99.22( 0.81)
Injectable Phosphatidylcholine formula (Rotunda et al., 2004)
Injectable deoxycholate formula (Yagima Odo et al., 2007)
Ursodeoxycholic acid standard
Ursacol®
constant (ki) of the kit reagent was obtained by:
ln A = ln A0 − kit
3.2. Validation of the method (kit reagent)
The spectrophotometric method using the developed kit
reagent for DCA and UDCA was validated according to the EMEA and
ANVISA guidelines. Thus the validation characteristics addressed
were linearity, accuracy, precision, specificity, limits of detection
and quantification and robustness (Table 2). The standard curves for
the linearity assays were constructed with 10 concentrations and
validated by the least squares method showing correlation coef-
ficients higher than 0.998 and the one-way analysis of variance
(ANOVA) showing F and p values of 4905.94 and less than 10−10
for DCA and 3415.11 and less than 10−9 for UDCA, respectively. For
the range (80–120%), the standard curves were constructed with 5
concentrations showing correlation coefficients of the same order
of magnitude.
The precision (repeatability) was determined by the per-
centage of relative standard deviation (%R.S.D.) at three levels
(DCA – 5.6 × 10−4, 7.0 × 10−4 and 8.4 × 10−4 mol L−1 and UDCA –
1.52 × 10−4, 1.9 × 10−4 and 2.28 × 10−4 mol L−1). The %R.S.D. values
were less than 3% obtained in the concentrations studied indicating
a good repeatability. The use of an experimental design is encour-
aged by EMEA guidelines, in this way the intermediate precision
(9)
where A0 is the initial absorbance of the bile acids and A is the final
absorbance of the bile acids after 60 days. Furthermore, the half-life
(t1/2) can be obtained by:
ln 2
ki
t1/2
=
(10)
The inactivation constants (ki) of the kit reagent were of
1.01 × 10−2 day−1 for DCA and 2.35 × 10−3 day−1 for UDCA deter-
minations. The half-life (t1/2) for the kit reagent was 68.71 days
for DCA and 294.71 days for the UDCA measurements. Despite
the greater stability of the kit reagent for UDCA compared to that
one for DCA, the last one allowed determinations in more concen-
trated solutions (data not shown) and pharmaceutical formulations
containing high DCA concentrations (Rotunda et al., 2004; Schuller-
Petrovic et al., 2008; Yagima Odo et al., 2007).
Fig. 3. Standard curve of the optical chemical sensor for deoxycholic and ursodeoxycholic acids (␤-CD–PHP
3.1 × 10−4:7.75 × 10−5 mol L−1 for UDCA) determinations.
–
6.2 × 10−4:1.55 × 10−4 mol L−1 for DCA and

104
P.G. Cadena et al. / Chemistry and Physics of Lipids 164 (2011) 99–105
was studied by a 22 full factorial design which did not present
significant effects and interactions (p < 0.05). According to these
results, the different analysts and equipments and your interactions
did not interfere in the method.
sion complexes as kit reagent or optical sensor was carried out.
The proposed method had good storage stability, linearity and pre-
cision and can be applied to the analysis of a wide concentration
range of DCA and UDCA in real samples with satisfactory results as
kit reagent. The original optical chemical sensor developed showed
good linearity suggesting that can be used for fast screening and the
analyses “in loco” of bile acids in pharmaceutical formulations or
raw materials.
The accuracy results demonstrated the method effectiveness for
quantitative determination of DCA and UDCA in the pharmaceutical
formulation. Accuracy was showed as percentage recovery of the
target value, exhibited excellent results with bias lower than 2%
throughout the tested range (Table 3). In the concentrations stud-
ied, the excipients used in the formulations and water type did not
interfere in the results. The optimum pH was fixed at 10.5, condi-
tion that was also reported by Afkhami et al. (2006) and Glazyrin
et al. (2004). Determinations performed at high pH conditions did
not cause significant changes of the results and were limited by the
buffering power of carbonate buffer (Table 2).
Acknowledgements
The authors thank FACEPE, CAPES-GRICES, LIKA/UFPE, CNPq.
and Dr. Marta M.M.B. Duarte for her appreciated suggestions in
methods validation.
The limits of detection were of 4.92 × 10−5 mol L−1 for DCA
and 1.14 × 10−5 mol L−1 for UDCA, respectively. The correspond-
ing limits of quantification were of 1.64 × 10−4 mol L−1 and
References
Abdel-Shafi, A.A., 2007. Spectroscopic studies on the inclusion complex of 2-
naphthol-6-sulfonate with ␤-cyclodextrin. Spectrochim. Acta A 66 (3), 732–738.
Afkhami, A., Madrakian, T., Khalafi, L., 2006. Spectrophotometric determination of
fluoxetine by batch and flow injection methods. Chem. Pharm. Bull. 54 (12),
1642–1646.
3.79 × 10−5 mol L−1
.
3.3. Optical chemical sensor for bile acids
Arias De Fuentes, O., Campanella, L., Crescentini, G., Falcioni, A., Sammartino, M.P.,
Tomassetti, M., 2000. Flow injection analysis of cholic acids in pharmaceutical
preparations using a polymeric membrane ISE as detector. J. Pharm. Biomed.
Anal. 23 (1), 89–98.
Benesi, H.A., Hildebrand, J.H., 1949. A spectrophotometric investigation of the
interaction of iodine with aromatic hydrocarbons. J. Am. Chem. Soc. 71 (8),
2703–2707.
Brewster, M.E., Loftsson, T., 2007. Cyclodextrins as pharmaceutical solubilizers. Adv.
Drug Deliver. Rev. 59 (7), 645–666.
Cadena, P.G., Oliveira, E.C., Araújo, A.N., Montenegro, M.C.B.S.M., Pimentel, M.C.B.,
Lima Filho, J.L., Silva, V.L., 2009. Simple determination of deoxycholic and
ursodeoxycholic acids by phenolphthalein–␤-cyclodextrin inclusion complex.
Lipids 44 (11), 1063–1070.
Castronuovo, G., Niccoli, M., 2006. Thermodynamics of inclusion complexes of nat-
ural and modified cyclodextrins with propranolol in aqueous solution at 298 K.
Bioorg. Med. Chem. 14 (11), 3883–3887.
Del Valle, E.M.M., 2004. Cyclodextrins and their uses: a review. Process Biochem. 39
(9), 1033–1046.
Duncan, D., Rubin, J.P., Golitz, L., Badylak, S., Kesel, L., Freund, J., Duncan, D., 2009.
Refinement of technique in injection lipolysis based on scientific studies and
clinical evaluation. Clin. Plast. Surg. 36 (2), 195–209.
Gaiao, E.d.N., Martins, V.L., Lyra, W.d.S., Almeida, L.F.d., Silva, E.C.d., Araújo, M.C.U.,
2006. Digital image-based titrations. Anal. Chim. Acta 570 (2), 283–290.
Glazyrin, A.E., Grachev, M.K., Kurochkina, G.I., Nifant’ev, E.E., 2004. Inclusion
compounds of some water-soluble ␤-cyclodextrin derivatives with phenolph-
thalein. Russ. J. Gen. Chem. 74 (12), 1922–1925.
Ha, J., Engler, C.R., Wild, J.R., 2009. Biodegradation of coumaphos, chlorferon,
and diethylthiophosphate using bacteria immobilized in Ca-alginate gel beads.
Bioresour. Technol. 100 (3), 1138–1142.
Holm, R., Shi, W., Hartvig, R.A., Askaer, S., Christian Madsen, J., Westh, P., 2009.
Thermodynamics and structure of inclusion compounds of tauro- and glyco-
conjugated bile salts and ␤-cyclodextrin. PCCP Phys. Chem. Chem. Phys. 11 (25),
5070–5078.
Ikegami, T., Matsuzaki, Y., 2008. Ursodeoxycholic acid: mechanism of action and
novel clinical applications. Hepatol. Res. 38 (2), 123–131.
Jullian, C., Orosteguis, T., Pérez-Cruz, F., Sánchez, P., Mendizabal, F., Olea-Azar, C.,
2008. Complexation of morin with threekinds of cyclodextrin. A thermodynamic
and reactivity study. Spectrochim. Acta A 71 (1), 269–275.
Lin, M.-C., Wu, H.-L., Kou, H.-S., Wu, S.-M., Chen, S.-H., 2003. Simple and sensitive
fluorimetric liquid chromatography for simultaneous analysis of chenodiol and
ursodiol in pharmaceutical formulations. Anal. Chim. Acta 493 (2), 159–166.
Lindor, K.D., 1997. Ursodiol for primary sclerosing cholangitis. N. Engl. J. Med. 336
(10), 691–695.
Martinez, A.W., Phillips, S.T., Carrilho, E., Thomas Iii, S.W., Sindi, H., Whitesides, G.M.,
2008a. Simple telemedicine for developing regions: camera phones and paper-
based microfluidic devices for real-time, off-site diagnosis. Anal. Chem. 80 (10),
3699–3707.
Martinez, A.W., Phillips, S.T., Wiley, B.J., Gupta, M., Whitesides, G.M., 2008b. FLASH:
a rapid method for prototyping paper-based microfluidic devices. Lab Chip 8
(12), 2146–2150.
Okoro, N., Patel, A., Goldstein, M., Narahari, N., Cai, Q., 2008. Ursodeoxycholic acid
treatment for patients with postcholecystectomy pain and bile microlithiasis.
Gastrointest. Endosc. 68 (1), 69–74.
The results obtained concerning robustness of ␤-CD–PHP inclu-
sion complex suggested its biotechnological application in the
form of optical strip sensor for both qualitative and quantitative
measurements of bile acids. Therefore, a mixture of alginate and ␤-
CD–PHP inclusion complex was used to impregnate strips of paper.
Alginate is largely used in biomolecule immobilization because it
provides simple implementation using a biodegradable and non
toxic material (Ha et al., 2009). Furthermore, it did not interfere
either with the competitive complexation reaction or with the dig-
ital images obtained after colour change induced by the reaction.
The quantitative analysis of digital images obtained from a scan-
ner of the optical sensor is a recent tool to measure the amount of
analytes. This new technology has been showed low cost and sim-
plicity, very important for the developing world (Martinez et al.,
2008b). The scanners are inexpensive, have high resolution, the
scanned image is always in focus, the intensity of the image is
not affected by lighting conditions, are portables (business card
scanners or pen scanners), and they can be linked to personal dig-
ital assistants (PDAs) by wireless communication (Martinez et al.,
2008a). In this technique, the analytical signal corresponds to the
RGB-based value that was calculated from each digital image, using
the proposed procedure based on the red, green, and blue colour
system (Martinez et al., 2008a; Gaiao et al., 2006; Yang et al., 2007).
The test was only based on dropwise 20 ␮L of each bile acid
solution and controls (buffer) in different places on the same strip,
after 5 min at room temperature (25 ◦C) when the tip was dried, fol-
lowed by immediate scanning. However, after the scanning, the tip
colour was observed for 3 h without any changes. Based on col-
lected images, it was possible to plot a standard curves (Fig. 3)
for DCA obtaining the equation: RGB = 1553 ( 32) DCAmol/L −0.8
(
0.2) with a correlation coefficient of 0.9994 and UDCA with the
corresponding equation RGB = 5671 ( 229) UDCAmol/L +0.6 ( 0.3)
showing the correlation coefficient of 0.998.
4. Conclusion
The development of low cost and simple method for determin-
ing DCA and UDCA are important in quality control of raw materials
and pharmaceutical formulations to prevent misuses and acci-
dents when used for aesthetic purposes, specifically for the DCA.
The ␤-CD–PHP inclusion complex showed to be a good alternative
method for bile acids measurement. The physical–chemical and
thermodynamic parameters studies for the application of inclu-
Petroni, M.L., Jazrawi, R.P., Pazzi, P., Lanzini, A., Zuin, M., Pigozzi, M.G., Fracchia, M.,
Galatola, G., Alvisi, V., Heaton, K.W., Podda, M., Northfield, T.C., 2001. Ursodeoxy-
cholic acid alone or with chenodeoxycholic acid for dissolution of cholesterol
gallstones: a randomized multicentre trial. Aliment. Pharmacol. Ther. 15 (1),
123–128.

P.G. Cadena et al. / Chemistry and Physics of Lipids 164 (2011) 99–105
105
Rodriguez, V.G., Lucangioli, S.E., Fernandez Otero, G.C., Carducci, C.N., 2000. Determi-
nation of bile acids in pharmaceutical formulations using micellar electrokinetic
chromatography. J. Pharm. Biomed. Anal. 23 (2–3), 375–381.
Wang, H.Y., Han, J., Feng, X.G., 2007. Spectroscopic study of orange G-␤-cyclodextrin
complex and its analytical application. Spectrochim. Acta
585.
A 66 (3), 578–
Rotunda, A.M., Kolodney, M.S., 2006. Mesotherapy and phosphatidylcholine injec-
tions: historical clarification and review. Dermatol. Surg. 32 (4), 465–480.
Rotunda, A.M., Suzuki, H., Moy, R.L., Kolodney, M.S., 2004. Detergent effects of
sodium deoxycholate are a major feature of an injectable phosphatidylcholine
formulation used for localized fat dissolution. Dermatol. Surg. 30 (7), 1001–1008.
Samstein, R.M., Perica, K., Balderrama, F., Look, M., Fahmy, T.M., 2008. The use of
deoxycholic acid to enhance the oral bioavailability of biodegradable nanopar-
ticles. Biomaterials 29 (6), 703–708.
Yagima Odo, M.E., Cuce, L.C., Odo, L.M., Natrielli, A., 2007. Action of sodium deoxy-
cholate on subcutaneous human tissue: local and systemic effects. Dermatol.
Surg. 33 (2), 178–188.
Yanez, C., Salazar, R., Nunez-Vergara, L.J., Squella, J.A., 2004. Spectrophotometric
and electrochemical study of the inclusion complex between ␤-cyclodextrin
and furnidipine. J. Pharm. Biomed. Anal. 35 (1), 51–56.
Yang, C.X., Sun, X.Y., Liu, B., Lian, H.T., 2007. Determination of total phosphorus
in water sample by digital imaging colorimetry. Chin. J. Anal. Chem. 35 (6),
850–853.
Yuexian, F., Yu, Y., Shaomin, S., Chuan, D., 2005. Molecular recognition of ␣-
cyclodextrin (CD) to choral amino acids based on methyl orange as a molecular
probe. Spectrochim. Acta A 61 (5), 953–959.
Zarzycki, P.K., Lamparczyk, H., 1998. The equilibrium constant of ␤-
cyclodextrin–phenolphthalein complex; influence of temperature and
tetrahydrofuran addition. J. Pharm. Biomed. Anal. 18 (1–2), 165–170.
Zhang, M., Li, J., Zhang, L., Chao, J., 2009. Preparation and spectral investigation
of inclusion complex of caffeic acid with hydroxypropyl-␤-cyclodextrin. Spec-
trochim. Acta A 71 (5), 1891–1895.
Scalia, S., Pazzi, P., Guarneri, M., 1989. Determination of bile acids in pharmaceutical
dosage forms by HPLC. Anal. Lett. 22 (4), 915–927.
Schuller-Petrovic, S., Wölkart, G., Höfler, G., Neuhold, N., Freisinger, F., Brunner, F.,
2008. Tissue-toxic effects of phosphatidylcholine/deoxycholate after subcuta-
neous injection for fat dissolution in rats and a human volunteer. Dermatol.
Surg. 34 (4), 529–542.
ˆ
Terzic, N., Opsenica, D., Milic, D., Tinant, B., Smith, K.S., Milhous, W.K., Solaja, B.A.,
2007. Deoxycholic acid-derived tetraoxane antimalarials and antiproliferatives.
J. Med. Chem. 50 (21), 5118–5127.
Valenta, C., Nowack, E., Bernkop-Schnurch, A., 1999. Deoxycholate-hydrogels: novel
drug carrier systems for topical use. Int. J. Pharm. 185 (1), 103–111.