Complexation of the sunscreen agent, phenylbenzimidazole sulphonic acid with cyclodextrins: Effect on stability and photo-induced free radical formation

Article abstract of DOI:10.1016/j.ejps.2004.03.014

The interaction between the sunscreen agent, phenylbenzimidazole sulphonic acid (PBSA) and hydrophilic α-, β-, and γ-cyclodextrin derivatives was investigated under acidic conditions (pH 4.0) by phase-solubility analysis. Among the available cyclodextrins, hydroxypropyl-β-cyclodextrin (HP-β-CD) and random methyl-β- cyclodextrin (RM-β-CD) had the greatest solubilizing activity. The complexation of the sunscreen agent with HP-β-CD and RM-β-CD was confirmed by nuclear magnetic resonance spectroscopy. Solid-phase characterization of the PBSA/cyclodextrin systems by X-ray diffractometry defined the most appropriate method (co-evaporation) and cyclodextrin concentration (10-fold molar excess) for the preparation of a stable complexed form of PBSA. Long-term stability studies demonstrated that the decrease of the sunscreen level in emulsion preparations (pH 4.0) was almost completely suppressed by HP-β-CD, RM-β-CD being less effective. Moreover, the irradiation-induced decomposition of PBSA in the emulsion vehicle was markedly reduced by complexation with HP-β-CD (the extent of degradation was 3.9% for the complex compared to 9.1% for uncomplexed PBSA), whereas RM-β-CD had no significant influence. In addition, electron paramagnetic resonance (EPR) spin-trapping studies showed that the inclusion of the sunscreen agent into the HP-β-CD cavity completely inhibited the formation of free-radicals generated by PBSA on exposure to simulated sunlight, thereby suppressing its photosensitising potential.

Full text of DOI:10.1016/j.ejps.2004.03.014

European Journal of Pharmaceutical Sciences 22 (2004) 241–249
Complexation of the sunscreen agent, phenylbenzimidazole sulphonic
acid with cyclodextrins: effect on stability and photo-induced
free radical formation
Santo Scaliaa,∗, Alessandra Molinarib, Alberto Casolaria, Andrea Maldottib
a
Dipartimento di Scienze Farmaceutiche, Università di Ferrara, via Fossato di Mortara 17, 44100 Ferrara, Italy
b
Dipartimento di Chimica, Università di Ferrara, via Borsari 46, Ferrara, Italy
Received 10 October 2003; received in revised form 18 February 2004; accepted 17 March 2004
Available online 14 May 2004
Abstract
The interaction between the sunscreen agent, phenylbenzimidazole sulphonic acid (PBSA) and hydrophilic ␣-, ␤-, and ␥-cyclodextrin deriva-
tives was investigated under acidic conditions (pH 4.0) by phase-solubility analysis. Among the available cyclodextrins, hydroxypropyl-␤-cy-
clodextrin (HP-␤-CD) and random methyl-␤-cyclodextrin (RM-␤-CD) had the greatest solubilizing activity. The complexation of the
sunscreen agent with HP-␤-CD and RM-␤-CD was confirmed by nuclear magnetic resonance spectroscopy. Solid-phase characterization
of the PBSA/cyclodextrin systems by X-ray diffractometry defined the most appropriate method (co-evaporation) and cyclodextrin concen-
tration (10-fold molar excess) for the preparation of a stable complexed form of PBSA. Long-term stability studies demonstrated that the
decrease of the sunscreen level in emulsion preparations (pH 4.0) was almost completely suppressed by HP-␤-CD, RM-␤-CD being less
effective. Moreover, the irradiation-induced decomposition of PBSA in the emulsion vehicle was markedly reduced by complexation with
HP-␤-CD (the extent of degradation was 3.9% for the complex compared to 9.1% for uncomplexed PBSA), whereas RM-␤-CD had no
significant influence. In addition, electron paramagnetic resonance (EPR) spin-trapping studies showed that the inclusion of the sunscreen
agent into the HP-␤-CD cavity completely inhibited the formation of free-radicals generated by PBSA on exposure to simulated sunlight,
thereby suppressing its photosensitising potential.
© 2004 Elsevier B.V. All rights reserved.
Keywords: Sunscreen agent; Phenylbenzimidazole sulphonic acid; Cyclodextrin; Complexation; Stability; Photo-induced free radicals
1. Introduction
Green et al., 1999). The active sunscreen constituents are of
two types, inorganic substances that act mainly by reflect-
ing or scattering the UV radiation and organic agents which
decrease the dose of UV rays reaching the skin by absorb-
ing the radiation, the latter being used most commonly
(Gasparro et al., 1998). When exposed to sunlight, the sun-
screen molecule is photoactivated and the excitation energy
is eventually dissipated in the form of thermal energy, fluo-
rescence, phosphorescence or transferred to the surrounding
molecules. In addition, photo-induced decomposition of the
sunscreen agent can also occur leading to a decrease of its
UV-protective capacity and to the accumulation on the skin
of potentially harmful photolytic products (Berset et al.,
1996; Schauder and Ippen, 1997; Tarras-Wahlberg et al.,
1999; Maier et al., 2001). Consequently, a high photochem-
ical stability is an essential requirement for an effective
UV filter (Berset et al., 1996; Maier et al., 2001). Another
It is well-documented that exposure to the UV radia-
tion (290–400 nm) from the sun plays a causal role in skin
photodamage including acute inflammatory responses (i.e.,
erythema, oedema) and long-term adverse reactions such
as cutaneous photoaging, immune suppression and various
forms of skin cancers (National Institute of Health, 1989;
Ziegler et al., 1994; Serre et al., 1997; Tarras-Wahlberg
et al., 1999). The increasing knowledge on the hazards of
exposure to sunlight has fuelled the widespread use of topi-
cal sunscreening preparations (National Institute of Health,
1989; Schauder and Ippen, 1997; Gasparro et al., 1998;
∗
Corresponding author. Tel.: +39-0532-291277;
fax: +39-0532-291296.
E-mail address: sls@unife.it (S. Scalia).
0928-0987/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.ejps.2004.03.014

242
S. Scalia et al. / European Journal of Pharmaceutical Sciences 22 (2004) 241–249
l-lysine) on the complexation efficiency. The influence of
4
H
HO3S
N 3
6'
5'
3'
complexation on the light-induced free radical generation
by the sunscreen and on its photostability and solubility at
acidic pH are also presented.
5
4'
2
6
N
1
2'
2. Materials and methods
7
Fig. 1. Chemical structure of PBSA.
2.1. Materials
Phenylbenzimidazole sulphonic acid was supplied by
Merck (Darmstadt, Germany). The cyclodextrins used
in this study included hydroxypropyl-␤-cyclodextrin,
hydroxypropyl-␣-cyclodextrin (HP-␣-CD), hydroxypropyl-
␥-cyclodextrin (HP-␥-CD) and random methyl-␤-cyclo-
dextrin. They were purchased from Aldrich Chimica (Milan,
Italy). Polyvinylpyrrolidone (PVP; MW ca. 55,000), hy-
droxyethyl cellulose (HEC; MW ca. 90,000), l-lysine and
5,5-dimethyl-1-pyrroline-N-oxide (DMPO) were obtained
from Aldrich Chimica. Methanol, acetonitrile and water
were high-performance liquid chromatography (HPLC)-
grade from Merck. All other chemicals were of analytical-
reagent grade (Sigma Milan, Italy).
important characteristic that sunscreens should have is com-
plete solubilization in the finished preparation, since the pre-
cipitation of the UV filter will lead to its uneven distribution
over the skin surface and consequently to a reduction of the
protective effect of the sun-care product (Johncock, 1999).
Phenylbenzimidazole sulphonic acid (PBSA; Fig. 1) is a
widely used sunscreen agent (Inbaraj et al., 2002; Hayden et
al., 1998) which absorbs most efficiently in the 290–320 nm
region (UV-B) of the solar UV radiation. It is included in
the list of authorized UV filters in Europe (EEC Directive,
1976), USA (US Food and Drug Administration, 1999) and
Australia (Hayden et al., 1998). PBSA is used, after con-
version into a salt, as a water-soluble sunscreen in aqueous
preparations and in the hydrophilic phase of emulsions. In
the latter case, the association of PBSA with an oil-soluble
UV absorber achieves a synergic increase of the sun protec-
tion factor (SPF) which makes it possible to employ a lower
concentration of UV filters (Siemer, 1991; Johncock, 1999).
However, as PBSA is poorly soluble in acidic media, the pH
of the formulations should remain at values higher than 7
(Siemer, 1991; Johncock, 1999) and hence above the desir-
able acid range of 4–6 (Block, 2000). Moreover, although
PBSA is considered photostable (Berset et al., 1996), recent
studies have demonstrated that this sunscreen agent under
solar-simulated radiation generates a variety of free radicals
and active oxygen species that cause photo-induced damage
to DNA in vitro (Stevenson and Davies, 1999; Inbaraj et al.,
2002).
2.2. High-performance liquid chromatography
The HPLC apparatus comprised a Model LabFlow 3000
pump (LabService Analytica, Bologna, Italy), a Model 7125
injection valve with a 20 ␮l sample loop (Rheodyne, Cotati,
CA, USA) and a Model 975-UV variable wavelength UV-
Vis detector (Jasco, Tokyo, Japan) set at 305 nm. Data ac-
quisition and processing were accomplished with a personal
computer using Borwin software (JBMS Developpements,
Le Fontanil, France). Sample injections were effected with
a Model 701 syringe (10 ␮l; Hamilton, Bonaduz, Switzer-
land). Separations were performed on a 5 ␮m Zorbax SB-CN
column (150 mm × 4.6 mm i.d.) fitted with a guard column
(5 ␮m particles, 4 mm × 2 mm i.d.) and eluted isocratically,
at a flow-rate of 1.0 ml/min, with aqueous triethylammonium
phosphate (pH 2.8)–methanol–acetonitrile (70:25:5, v/v/v).
The column temperature was maintained at 40 ◦C using a
Model 7990 Space Column Heater (Jones Chromatography,
Hangoed, UK). The identity of PBSA peak was assigned by
co-chromatography with the authentic standard. Quantifica-
tion was carried out by integration of the peak areas using
the external standardization method.
Because of the growing interest in cyclodextrins and their
ability to improve both the solubility and stability of active
ˆ
ingredients (Duchene et al., 1999), the present study was
carried out to evaluate the influence of cyclodextrin-based
systems on the performance of PBSA. Cyclodextrins are
cyclic oligosaccharides with a hydrophilic outer surface and
an apolar cavity. They can encapsulate appropriately sized
lipophilic molecules into their hydrophobic interior by form-
ing non-covalent complexes (Loftsson and Brewster, 1996).
This phenomenon affects the physico-chemical properties of
the included substance and can increase its stability to air
and light and the apparent aqueous solubility (Loftsson and
Brewster, 1996; Rajewski and Stella, 1996).
The present work reports on the preparation and
characterization of the complexes between PBSA and
hydroxypropyl-␤-cyclodextrin (HP-␤-CD) or random
methyl-␤-cyclodextrin (RM-␤-CD) and investigates the ef-
fect of a third component (i.e., water-soluble polymers or
2.3. Phase-solubility studies
Solubility analyses were carried out according to Higuchi
and Connors (1965). An excess amount of PBSA was added
to 5 ml acetate buffer solutions (50 mM; pH 4.0) contain-
ing the different cyclodextrins examined (0–90 mM) with or
without a water-soluble polymer (0.25%, w/v) or l-lysine
(90 mM). The samples with the polymers were heated in a
sealed container at 120 ◦C for 20 min. The obtained suspen-
sions were stirred in 10 ml screw-capped vials at 25 1 ◦C

S. Scalia et al. / European Journal of Pharmaceutical Sciences 22 (2004) 241–249
243
and shielded from light. After equilibrium had been reached
(3 days, as demonstrated by a constant PBSA concentration
of three successive samples at 3, 5 and 7 days), the content
of each vial was filtered through 0.45 ␮m membrane filters
(Whatman, Clifton, NJ, USA) and analysed for PBSA by
HPLC as outlined above. Data were determined from the
average of at least three determinations. Solubility diagrams
were constructed for each cyclodextrin by plotting the molar
concentration of PBSA in solution against the molar concen-
tration of cyclodextrin. The stability constant values were
calculated with the following equation.
ca. 10 mM. Chemical shifts are reported in ppm (δ) relative
1
to TMS. Typical parameters for the H NMR spectra were
0.35 Hz/pt resolution, 16 scans, 18 s relaxation delay, 90◦
pulse.
2.7. UV spectrophotometry
UV spectra were recorded in MeOH:DMSO:H2O
(50:35:15, v/v) on a UV-Vis/NIR Spectrometer (Lambda
19; Perkin-Elmer, Norwalk, USA).
2.8. EPR measurements
slope
K =
S0(1 − slope)
Electron paramagnetic resonance (EPR) measurements
were performed with a X-band Bruker 220 SE spectrome-
ter (Bruker, Karlsruhe, Germany) equipped with a TE 201
resonator (Bruker OR 4104, 100% optical transmittance).
Spectra were recorded with the following instrumental set-
tings: 2 mW microwave power, 1 G modulation amplitude
and 100 kHz field modulation. A quartz flat cell was used
as a reaction vessel and samples were irradiated directly in-
side the microwave cavity employing a 350 W medium pres-
sure Hg lamp. Irradiation wavelengths were selected using
an Oriel 59814 (Oriel Corporation, USA) band pass filter
(290 nm < λ < 410 nm) coupled with an Oriel IR-block fil-
ter to avoid thermal effects. EPR spin-trapping experiments
were performed in oxygen saturated phosphate buffer so-
lutions (pH 7.4), containing DMPO (100 mM) as spin-trap
and free or cyclodextrin-complexed PBSA (10 mM). The
samples were irradiated in the EPR cavity for 25 min and
spectra of the paramagnetic adducts were accumulated and
recorded at appropriate time intervals. EPR signal intensities
of the obtained paramagnetic adducts were measured at a
fixed field position. Each series of experiments was repeated
at least three times.
where S0 represents the solubility of the sunscreen in the ab-
sence of cyclodextrins and slope is the slope of the obtained
phase-solubility diagram.
2.4. Inclusion complex preparation
The inclusion complexes were prepared at a 1:1 and 1:10
molar ratios of PBSA to cyclodextrin using two different
methods, namely kneading and co-evaporation. For knead-
ing, the calculated amounts of PBSA and corresponding cy-
clodextrin were weighed, wetted with a small volume (5 ml)
of methanol/water solution (30:70, v/v) and the slurry was
kneaded throughly for ca. 30 min. After evaporation of the
solvent, the samples obtained were dried under low pressure
at room temperature for 3 days and stored in a desiccator. In
the co-evaporation method, the calculated amount of PBSA
was dissolved in methanol (4 ml) and added to 6 ml of an
aqueous solution containing an equimolar or a 10-fold mo-
lar quantity of the corresponding cyclodextrin. The mixture
was maintained under stirring for 24 h at room temperature
and shielded from light. The solvent was then evaporated
under vacuum at 40 ◦C by rotary evaporation and the residue
was kept in a desiccator until used. The amount of PBSA
incorporated into each complex was determined by HPLC
after proper dilution.
2.9. Long-term stability studies
The stability of PBSA was tested in cream preparations
(oil-in-water emulsion) containing the sunscreen (0.5%,
w/w) alone (after neutralization with NaOH) or complexed
with cyclodextrins. The cream excipients were: sorbitan
monostearate, polyoxyethylene sorbitan monostearate, buty-
lated hydroxyanisole, isopropyl isostearate (Henkel, Fino
Mornasco, Italy), cetearyl isononanoate (Henkel), cetearyl
alcohol (Henkel), sodium benzoate, glycerin, dehydroacetic
acid, EDTA, water. To prepare the cream, butylated hy-
droxyanisole, free or complexed PBSA were dissolved at
room temperature in isopropyl isostearate, aqueous sodium
hydroxide or deionized water, respectively. The other oil-
and aqueous-soluble components were separately heated
to approximately 60 ◦C and the aqueous phase was slowly
added to the oil phase while stirring with a Silverson mixer
(Chesham, England). Mild agitation was continued until
the emulsion cooled at room temperature. Butylated hy-
droxyanisole, free or complexed PBSA were added in the
2.5. X-ray diffractometry
The powder X-ray diffraction patterns were recorded on
a D 5000 powder diffractometer (Siemens, Munich, Ger-
many) using a voltage of 45 kV and a current of 25 mA
for the generator, with Cu anode material. The wavelength
of the graphite-monocromated radiation was 1.5406 Å. The
diffractograms were recorded from 3◦ (2θ) to 50◦ (2θ) at an
angular speed of 1◦ (2θ) per minute using 1–1–1–0.15◦ slits.
2.6. NMR spectroscopy
1H NMR spectra were recorded on a Varian Mercury
Plus spectrometer (400 MHz). Since the aqueous solubility
of PBSA is too low for the recording of NMR spectra, sam-
ples were solubilized in DMSO-d6, at a concentration of

244
S. Scalia et al. / European Journal of Pharmaceutical Sciences 22 (2004) 241–249
cooling phase of the production process at about 40 ◦C.
The pH of the formulations was adjusted to 4.0 with citric
acid. The emulsions were placed into stoppered containers
and stored at room temperature for 6 months, in the dark.
At appropriate time intervals, aliquots (300–320 mg) were
withdrawn from the emulsions and transferred into 10 ml
calibrated flasks, diluted to volume with methanol and fil-
tered (0.45 ␮m membrane filter). A portion (5 ␮l) of the
resulting solution was analysed by HPLC for the assay of
the remaining amount of PBSA.
2.10. Photodegradation studies
Photodecomposition experiments were performed in the
same formulations utilized for the long-term stability stud-
ies. A portion (200–220 mg) of the test sample contain-
ing uncomplexed or complexed PBSA (0.5%, w/w) was
transferred by means of a syringe onto the bottom of a
beaker and then irradiated for 2 h with a solar simulator
(Suntest CPS+; Atlas, Linsengericht, Germany) equipped
with a Xenon lamp, an optical filter to cut off wavelengths
shorter than 290 nm and an IR-block filter to avoid ther-
mal effects. The solar simulator emission was maintained
at 500 W/m2. After the exposure interval (2 h), the beaker
was removed and its content quantitatively transferred into
a 10 ml calibrated flask with methanol and the remaining
PBSA concentration was quantified by HPLC as outlined
above. All samples were protected from light both before
and after irradiation. The degree of photodegradation was
evaluated by comparing the peak areas of PBSA from the
irradiated samples, with those obtained by analysis of an
equivalent amount of the unirradiated preparations.
Fig. 2. (A) Phase-solubility diagrams for PBSA with different cyclodex-
trins in pH 4.0 acetate buffer at 25 ◦C. (ꢀ) HP-␣-CD; (᭺) HP-␥-CD;
(ᮀ) RM-␤-CD; (᭛) HP-␤-CD. (B) Phase-solubility diagrams in pH 4.0
acetate buffer at 25 ◦C of PBSA in the presence of HP-␤-CD without
(᭛) and with 0.25% (w/v) PVP (ꢀ), 0.25% (w/v) HEC (᭺) or 90 mM
l-lysine (ᮀ). Each point represents the mean S.D. of at least three ex-
periments. The lines are the regression lines obtained using least squares
linear regression analysis.
2.11. In vitro sun protection factor measurement
free acid form of PBSA (pKa 4.5) occurs with loss of screen-
ing efficacy (Johncock, 1999). Moreover, 4.0 represents the
lowest value in the normal pH range for cosmetic prod-
ucts and for the skin surface (Block, 2000). The diagrams
obtained (Fig. 2A) demonstrated that the major solubility
enhancements were produced by RM-␤-CD (8.9-fold in-
crease) and HP-␤-CD (7.7-fold increase) which indicated
that these cyclodextrins exhibited stronger interactions with
PBSA than HP-␣-CD and HP-␥-CD. Over the concentration
range used in this study, the apparent solubility of PBSA
increased linearly (r > 0.99) with increasing RM-␤-CD or
HP-␤-CD concentrations showing Al-type profiles (Higuchi
and Connors, 1965). These plots (Fig. 2A) with a slope <1
suggested that the solubility enhancement can be attributed
to the formation of a complex with a 1:1 stoichiometry
The in vitro determination of the cream sun protection
factor was carried out according to the Diffey and Robson
(1989) technique, with minor modifications as reported in
an earlier study (Scalia et al., 1999).
3. Results and discussion
3.1. Solubility studies
The solubility method was used initially for studying the
interaction of PBSA with cyclodextrins. Because of the lim-
ited aqueous solubility of natural cyclodextrins (Loftsson
and Brewster, 1996), the highly hydrophilic hydroxypropy-
lated (i.e., HP-␣-CD, HP-␤-CD, HP-␥-CD) and randomly
methylated (i.e., RM-␤-CD) derivatives were selected for
this investigation. Fig. 2A illustrates the phase-solubility
curves of the UV-filter with the examined cyclodextrins in
pH 4.0 acetate buffer (0.05 M). The solubilizing activity of
cyclodextrins was evaluated at this pH, because it is under
acidic conditions that precipitation of the water-insoluble
(Higuchi and Connors, 1965). The stability constants (K1:1
)
calculated according to the method of Higuchi and Connors
(1965) were 91.9 2.1 and 78.6 2.5 M−1 for the RM-␤-
CD/PBSA and HP-␤-CD/PBSA complex, respectively. Pub-
lished studies (Loftsson and Brewster, 1996; Loftsson, 1998)
have demonstrated that the efficiency of complexation can
be improved by formation of ternary complexes between

S. Scalia et al. / European Journal of Pharmaceutical Sciences 22 (2004) 241–249
245
the cyclodextrin, the guest molecule and a third component.
Accordingly, the influence of the combined use of HP-␤-
CD and the water-soluble polymers, HEC and PVP, on the
hydrosolubility of PBSA was evaluated at pH 4.0 after acti-
vation by heating (Loftsson, 1998). In the presence of HEC
(0.25%, w/v) or PVP (0.25%, w/v) the solubilization of the
sunscreen induced by HP-␤-CD was only slightly improved
(Fig. 2B), the enhancement being not synergistic. The K1:1
calculated from the Al-type curves (Fig. 2B) for the HEC-
and PVP-systems were 70.5 1.4 and 58.2 1.2 M−1, re-
spectively. The lower values for the stability constants of
these ternary systems as compared to the PBSA/HP-␤-CD
binary complex (K1:1, 78.6 M−1) can be ascribed to the solu-
bilizing activity of the polymers themselves (Fig. 2B). Basic
aminoacids can be used for ternary cyclodextrin complexa-
tion of acidic drugs due to their potential ability to interact
simultaneously with the guest molecule (via salt formation)
and the cyclodextrin (Piel et al., 1997). The addition of l-
lysine (90 mM) to the pH 4.0 buffered medium produced a
greater increase of the HP-␤-CD solubilizing effect as com-
pared to the polymer systems, although the differences were
small (Fig. 2B). The stability constant value for the PBSA/l-
lysine/HP-␤-CD ternary complex was 66.3 1.0 M−1. The
decrease of the K1:1 value observed in the presence of l-
lysine can be traced to the higher initial drug solubility due
to salt formation. Based on the results of the solubility stud-
ies, the systems with the highest stability constants (i.e.,
PBSA/RM-␤-CD and PBSA/HP-␤-CD) were selected for
further experiments.
Fig. 3. Powder X-ray diffraction patterns of PBSA/HP-␤-CD (1:10) phys-
ical mixture (A), PBSA/HP-␤-CD (1:10) co-evaporated complex (B) and
PBSA/HP-␤-CD (1:10) kneaded complex (C).
3.2. Complex characterization
The interaction between PBSA and the cyclodextrins was
1
ascertained in solution by H NMR spectroscopic studies
phenyl substituent (see Fig. 1), suggesting that this moiety
is deeply inserted into the cyclodextrin hollow cone.
which provide the most conclusive evidence of complex
formation (Hedges, 1998). Table 1 lists the variations in
the chemical shift values of the aromatic protons of PBSA
(see Fig. 1 for PBSA structure and atom labels) induced by
The complexes of PBSA with HP-␤-CD and RM-␤-
CD were characterized in the solid-state by powder X-ray
diffraction. The solid complexes were prepared in molar
ratios of 1:1 and 1:10 (guest:host), using two different meth-
ods, kneading and co-evaporation. As illustrated in Fig. 3,
the diffractogram of the HP-␤-CD complex (1:10, sun-
screen:cyclodextrin) obtained by the co-evaporation process
(Fig. 3B) did not show the PBSA crystalline peaks which
were present in the X-ray diffraction pattern of the physical
mixture (Fig. 3A). This finding provided evidence of the
inclusion of PBSA into the cyclodextrin cavity. Conversely,
in the diffractogram of the kneaded HP-␤-CD complex
(1:10, guest:host), the main PBSA signals appeared, though
with a very low intensity (Fig. 3C). This indicated that the
complexation of the sunscreen agent is less efficient when
the kneading procedure is adopted and consequently the
co-evaporation method was utilized to prepare the com-
plexes for further studies. The same results were obtained
from the X-ray diffraction analysis of the PBSA/RM-␤-CD
complexes (1:10 molar ratio). In contrast with the foregoing
1
the presence of RM-␤-CD or HP-␤-CD. All the H NMR
signals were shifted upfield (negative ꢀδ values) indicating
that the aromatic rings were located inside the cyclodex-
trin cavity (Chan et al., 2000). The largest chemical shift
changes (Table 1) were detected for the protons of the
Table 1
1H NMR chemical shift changes (ꢀδ, ppm) for PBSA in the presence of
cyclodextrins
Protons
ꢀδa
HP-␤-CD
RM-␤-CD
H4
H6
H7
−0.029
−0.017
−0.011
−0.057
−0.041
−0.034
−0.014
−0.008
−0.065
−0.044
H2 , H6
H3 , H4 , H5
a
ꢀδ = δwith cyclodextrin − δPBSA alone
.

246
S. Scalia et al. / European Journal of Pharmaceutical Sciences 22 (2004) 241–249
data, the systems with a 1:1 molar proportion displayed
distinct PBSA peaks at 8.5, 17, 20.3, 26.2◦ (diffraction
patterns not shown). Hence, the X-ray diffraction studies
suggest that complete inclusion of the sunscreen into the
cyclodextrin cavity is achieved when a molar excess (10-
fold) of the host is used. This result is not surprising given
the low values for the equilibrium complexation constants
(91.9 M−1 for RM-␤-CD and 78.6 M−1 for HP-␤-CD).
3.3. Stability studies
Since PBSA has a very low solubility both in water and
oil, it must be neutralized completely with a suitable base in
order to be used as a water-soluble UV-B filter. Moreover,
to prevent crystallization, the pH of the finished sun-care
preparation must be adjusted to values above 7.0 (Johncock,
1999). This is a disadvantage because the topical appli-
cation of such a product can alter the skin physiological
acidic pH, causing irritation (Schneider and Thor, 1991). In
addition, it is important to ensure that no compound with an
acidic reaction is used in the sunscreen formulation since
this could lead to precipitation of the water-insoluble free
acid form of PBSA with loss of efficacy (Johncock, 1999).
In order to evaluate the effect of cyclodextrin complexation
on the chemical–physical stability of PBSA, an aging study
was performed using a cream (oil-in-water emulsion) as a
medium. This vehicle was selected as a model formulation
since it represents the most commonly used type of sun-
screen preparation (Siemer, 1991). In addition, the cream pH
was adjusted to 4.0 because it is under acidic conditions that
product stability is reduced via precipitation. Free PBSA (as
sodium salt) or its complexed form was incorporated (0.5%,
w/w) into the cream and subjected to the stability-indicating
assay. We have reported above that for efficient complexa-
tion of PBSA a 10-times molar excess of cyclodextrin is re-
quired. Accordingly, the complexes in a molar ratio of 1:10
(PBSA:cyclodextrin) were used in the subsequent studies
The higher proportions of cyclodextrins should also reduce
the competitive displacement of the guest molecule from the
host cavity by the various emulsion excipients (Scalia et al.,
1999; Scalia et al., 2002). A number of studies have demon-
strated that an excess amount of cyclodextrin can lead
to decreased drug absorption through skin (Loftsson and
Masson, 2001 and references therein). Accordingly, the ex-
cess of cyclodextrin present in the test formulations could
reduce the percutaneous penetration of PBSA, thereby en-
hancing its retention at the skin surface where the sunscreen
action is highly desirable (Treffel and Gabard, 1996).
Fig. 4. Plots of PBSA concentration vs. time in its formulations
with or without cyclodextrin. (ᮀ) PBSA; (ꢀ) PBSA/RM-␤-CD; (᭺)
PBSA/HP-␤-CD. Values are mean S.D. of at least three experiments.
content was recovered after 6 months in the PBSA/RM-
␤-CD preparation (Fig. 4). Although this result indicated
a reduction in the loss of PBSA, the differences observed
between the samples without or with RM-␤-CD were not
statistically significant (P > 0.1, unpaired t-test), with the
exception of the values measured at day 180 (P < 0.01,
unpaired t-test). However, the formulation containing the
PBSA/HP-␤-CD complex retained 99.1% of the original
sunscreen concentration after the test period (Fig. 4), in-
dicating a higher effectiveness of HP-␤-CD as compared
to RM-␤-CD in enhancing the stability of the sunscreen
agent at acidic pH. Therefore, the efficacy of the examined
complexes in long-term stability studies was not related to
the corresponding complex formation constants (91.9 M−1
for RM-␤-CD and 78.6 M−1 for HP-␤-CD).
To investigate the influence of the cyclodextrin systems
on the photochemical behaviour of PBSA, photolysis exper-
iments were also performed on the formulations submitted
to the foregoing aging study. The creams were exposed
for 2 h to the solar simulator, the applied UV-B energy
being equivalent to 20 minimal erythemal dose (MED)
which is considered representative of daily solar emission
(Tarras-Wahlberg et al., 1999). Following irradiation, 9.1%
of the sunscreen content was lost in the preparation contain-
ing PBSA alone (Table 2), in good agreement with previous
studies (Berset et al., 1996). The photo-induced decompo-
sition of the sunscreen agent was not significantly affected
(Table 2) by complexation with RM-␤-CD (8.0% degrada-
tion). Conversely, under the same experimental conditions,
a statistically significant reduction of the extent of pho-
todegradation to 3.9% was attained in the cream containing
the PBSA/HP-␤-CD system (Table 2). Hence, the photo-
stabilization effects of the examined cyclodextrins correlate
with their activity in the aging study. Moreover, photolysis
The emulsions containing the sodium salt of the free
sunscreen agent or its cyclodextrin complex were analysed
for PBSA over 6 months at room temperature and in the
dark and the data generated are depicted in Fig. 4. The
plot for the cream containing the uncomplexed sunscreen
agent showed the PBSA level falling below 81.0% of the
initial concentration within 6 months (Fig. 4). Under the
same storage conditions, 89.9% of the UV filter initial

S. Scalia et al. / European Journal of Pharmaceutical Sciences 22 (2004) 241–249
247
Table 2
Comparative photodegradation data for free and complexed PBSA in
cream preparations, after 2 h irradiation with the solar simulator
Sample
%PBSA lossa
Pb
Free PBSA
PBSA/RM-␤-CD
PBSA/HP-␤-CD
9.1
8.0
3.9
2.2
1.5
1.7
>0.1
<0.001
a
Each value is the mean S.D. of eight determinations.
P-values (unpaired t-test) vs. free PBSA.
b
experiments performed after 6 months storage of the cream
samples at room temperature and in the dark, demonstrated
that the sunscreen photostability enhancement achieved by
HP-␤-CD complexation was retained during the above time
interval (the percentage loss of PBSA upon irradiation was
4.2 1.0% in the formulation containing the complexed UV
filter as compared to 10.8 1.1% for the preparation con-
taining free PBSA). The in vitro determination of the sun
protection factor of the creams containing the uncomplexed
PBSA or its complex with HP-␤-CD showed that complex-
ation had no significant influence on the photoprotective
capacity of the examined sunscreen preparations (the SPF
values ranged from 2.5 to 2.8; P > 0.05, unpaired t-test). In
addition, UV spectrophotometric analysis of PBSA and its
HP-␤-CD complex showed that the shape of the spectrum
and the degree of UV absorption of the sunscreen agent
were not affected by complexation (spectra not shown).
Fig. 5. EPR spin-trapping spectrum of the adduct (DMPO-OH•) obtained
after irradiation of buffered (pH 7.4) aqueous solutions containing DMPO
(100 mM) in the presence of PBSA (10 mM). aN = aH = 14.9 G.
DMPO and OH• radicals (Buettner, 1987). When the irradi-
ation was carried out in the absence of PBSA, no EPR signal
was detected. The observation of the spectrum of the adduct
DMPO-OH• is a clear indication that a photosensitisation
process from PBSA to O2 occurred (Inbaraj et al., 2002).
The adduct can be expected to derive from direct trapping
(Beauchamp a•n−d Fridovich, 1970), very fast decomposition
of DMPO-O2
formed via an electron transfer process
(Finkelstein et al., 1979) or singlet oxygen generated via an
energy transfer reaction from photogenerated PBSA triplet
state to O2 (Jones and Wilson, 1980). The increase of the
DMPO-OH• adduct EPR signal intensity during PBSA irra-
diation is illustrated in Fig. 6. Under the same experimental
conditions, no EPR signal was detected (see Fig. 6) in the
solution containing PBSA complexed with HP-␤-CD (1:10,
guest:host). The results shown in Fig. 6 clearly demonstrated
that the photo-induced production of oxygen-centered
radicals by PBSA is completely inhibited by inclusion
complexation of the sunscreen agent with HP-␤-CD. This
3.4. EPR studies
Recent reports in the literature (Stevenson and Davies,
1999; Inbaraj et al., 2002) have demonstrated that when
exposed to solar UV radiation, PBSA generates a variety
of free radicals and active oxygen species that are capable
of causing in vitro damages in single- and double-stranded
DNA. Accordingly, the next aspect of the present investiga-
tion was to examine the influence of the cyclodextrin com-
plex microenvironment on the photosensitising properties of
the sunscreen agent. Aqueous solutions (pH 7.4) containing
PBSA free or complexed with HP-␤-CD were illuminated
for 25 min with simulated sunlight and the photo-induced
free radicals were characterized by the EPR spin-trapping
technique, using DMPO as spin-trap. Radical species (R•)
generated during sunscreen irradiation were monitored by
detection of the more stable paramagnetic adduct formed ac-
cording to the following equation (Janzen and Haire, 1990):
On exposure to UV light of the sample containing uncom-
plexed PBSA and the spin-trap, a spectrum was obtained
consisting of four lines with a 1:2:2:1 intensity ratio and with
hyperfine splitting constants aN = aH = 14.9 G (Fig. 5)
which can be ascribed to the formation of the adduct between
Fig. 6. EPR signal intensity at a fixed field position of the paramagnetic
adduct (DMPO-OH•) during irradiation of buffered (pH 7.4) aqueous
solutions containing DMPO (100 mM) in the presence of PBSA (10 mM).
(ᮀ) Free PBSA; (ꢀ; 1:10) PBSA/HP-␤-CD; (᭺; 1:1) PBSA/HP-␤-CD.
Each point is the mean S.D. of at least three experiments.

248
S. Scalia et al. / European Journal of Pharmaceutical Sciences 22 (2004) 241–249
Gasparro, F.P., Mitchnick, M., Nash, J.F., 1998. A review of sunscreen
safety and efficacy. Photochem. Photobiol. 68, 243–256.
Green, A., Williams, G., Neale, R., Hart, V., Leslie, D., Parsons, P., Marks,
G.C., Gaffney, P., Battistutta, D., Frost, C., Lang, C., Russell, A.,
1999. Daily sunscreen application and betacarotene supplementation
in prevention of basal-cell and squamous-cell carcinomas of the skin:
a randomised controlled trial. Lancet 354, 723–729.
Hayden, C.G., Roberts, M.S., Benson, H.A.E., 1998. Sunscreens: are
Australian getting the good oil? Aust. NZ. J. Med. 28, 639–646.
Hedges, A.R., 1998. Industrial applications of cyclodextrins. Chem. Rev.
98, 2035–2044.
Higuchi, T., Connors, K.A., 1965. Phase-solubility techniques. Adv. Anal.
Chem. Instrum. 4, 117–212.
Inbaraj, J.J., Bilski, P., Chignell, C.F., 2002. Photophysical and photo-
chemical studies of 2-phenylbenzimidazole and UVB sunscreen 2-
phenylbenzimidazole-5-sulfonic acid. Photochem. Photobiol. 75, 107–
116.
remarkable effect can be traced to hindered interaction of the
photoactivated complexed PBSA with oxygen or to a reduc-
tion of the life-time of the photogenerated sunscreen triplet
state. Additional measurements were performed on sample
containing the complex in a 1:1 (sunscreen:cyclodextrin)
molar ratio. On UV illumination of this system, the para-
magnetic adduct between DMPO and OH• was detected,
though its formation rate was significantly lower as com-
pared to free PBSA (see Fig. 6). These data indicated that
the influence of HP-␤-CD on the sunscreen photosensitiz-
ing activity is much greater for the complex with a 1:10
(PBSA:cyclodextrin) molar proportion, in line with the
powder X-ray results on the efficiency of the inclusion
process.
Janzen, E.G., Haire, D.L., 1990. Two decades of spin trapping. Adv. Free
Radic. Chem. 1, 253–295.
Jones, R.L., Wilson, W.D., 1980. Singlet oxygen and spin trapping with
nitrones. J. Am. Chem. Soc. 102, 7778–7779.
4. Conclusions
Johncock, W., 1999. Sunscreen interactions in formulations. Cosmet Toil.
114, 75–82.
Loftsson, T., Brewster, M.E., 1996. Pharmaceutical applications of cy-
clodextrins. 1. Drug solubilization and stabilization. J. Pharm. Sci. 85,
1017–1025.
Loftsson, T., 1998. Increasing the cyclodextrin complexation of drugs
and drug bioavailability through addition of water-soluble polymers.
Pharmazie 53, 733–740.
Loftsson, T., Masson, M., 2001. Cyclodextrins in topical drug formula-
tions: theory and practice. Int. J. Pharm. 225, 15–30.
Maier, H., Schauberger, G., Brunnhofer, K., Hönigsmann, H., 2001.
Change of ultraviolet absorbance of sunscreens by exposure to solar-
simulated radiation. J. Invest. Dermatol. 117, 256–262.
National Institute of Health, 1989. NIH Consens statement online. Sun-
light, Ultraviolet Radiat. Skin 7, 1–29
The results described in this study demonstrate that the
complexation of PBSA with HP-␤-CD extends the stability
of the sunscreen agent to the acidic pH conditions optimal
for topical formulations, such as the sun-care products.
Moreover, the inclusion of PBSA into the HP-␤-CD cavity
represents an effective strategy to enhance the sunscreen
photostability as well as to suppress the light-induced
production of free radicals by the UV filter. Hence, the
PBSA/HP-␤-CD system should also minimize the damage
inflicted by the sunscreen to DNA following illumination
with simulated sunlight.
Piel, G., Pirotte, B., Delneuville, I., Neven, P., Llabres, G., Delarge, J.,
Delattre, L., 1997. Study of the influence of both cyclodextrins and
l-lysine on the aqueous solubility of nimesulide isolation and char-
acterization of nimesulide-l-lysine-cyclodextrin complexes. J. Pharm.
Sci. 86, 475–479.
Rajewski, R.A., Stella, V.J., 1996. Pharmaceutical applications of cy-
clodextrins. 2. In vivo drug delivery. J. Pharm. Sci. 85, 1142–1169.
Scalia, S., Villani, S., Casolari, A., 1999. Inclusion complexation
of the sunscreen agent 2-ethylhexyl-p-dimethylaminobenzoate with
hydroxypropyl-␤-cyclodextrin: effect on photostability. J. Pharm. Phar-
macol. 51, 1367–1374.
Scalia, S., Casolari, A., Iaconinoto, A., Simeoni, S., 2002. Compara-
tive studies of the influence of cyclodextrins on the stability of the
sunscreen agent 2-ethylhexyl-p-methoxycinnamate. J. Pharm. Biomed.
Anal. 20, 1181–1189.
Schauder, S., Ippen, H., 1997. Contact and photocontact sensitivity to
sunscreens. Contact Dermatitis 37, 221–232.
References
Beauchamp, C., Fridovich, I., 1970. A mechanism for the production of
ethylene from methional. The generation of the hydroxyl radical by
xanthine oxidase. J. Biol. Chem. 245, 4641–4646.
Berset, G., Gonzenbach, H., Christ, R., Martin, R., Deflandre, A., Mas-
cotto, R., Jolley, J.D., Lowell, W., Pelzer, R., Stiehm, T., 1996. Pro-
posed protocol for the determination of photostability. Part I. Cosmetic
UV filters. Int. J. Cosmet. Sci. 18, 167–177.
Block, L.H. Medicated topicals. In: Gennaro, A.R., Der Marderosian,
A.H., Hanson, G.R., Medwick, T., Popovich, N.G., Schnaare, R.L.,
Schwartz, J.B., White, H.S. (Eds.), Remington: The Science and Prac-
tice of Pharmacy, 20th ed. Lippincott Williams & Wilkins, Baltimore,
2000, pp. 836, 847
Buettner, G.R., 1987. Spin trapping: ESR parameters of spin adducts.
Free Rad. Biol. Med. 3, 259–303.
Schneider, W., Thor, G., 1991. Cosmetic preparations for skin care. In:
Umbach, W. (Ed.), Cosmetics and Toiletries. Ellis Horwood, New
York, p. 52.
Chan, L.W., Kurup, T.R.R., Muthaiah, A., Thenmozhiyal, J.C., 2000.
Interaction of p-hydroxybenzoic esters with beta-cyclodextrin. Int. J.
Pharm. 195, 71–79.
Serre, I., Cano, J.P., Picot, M.C., Meynadier, J., Meunier, L., 1997. Im-
munosuppression induced by acute solar-simulated ultraviolet expo-
sure in humans: prevention by a sunscreen with a sun protection factor
of 15 and high UVA protection. J. Am. Acad. Dermatol. 37, 187–194.
Siemer, E., 1991. Sunscreening preparations. In: Umbach, W. (Ed.), Cos-
metics and Toiletries. Ellis Horwood, New York, pp. 99–105.
Stevenson, C., Davies, R.J.H., 1999. Photosensitization of guanine-specific
DNA damage by 2-phenylbenzimidazole and the sunscreen agent 2-
phenylbenzimidazole-5-sulphonic acid. Chem. Res. Toxicol. 12, 38–
45.
Diffey, B.L., Robson, J., 1989. A new substrate to measure sunscreen
protection factors throughout the ultraviolet spectrum. J. Soc. Cosm.
Chem. 40, 127–133.
Ducheˆne, D., Wouessidjewe, D., Poelman, M., 1999. Cyclodextrins in
cosmetics. In: Magdassi, S., Touitou, E. (Eds.), Novel Cosmetic De-
livery Systems. Marcel Dekker, New York, pp. 275–293
EEC Directive, 1976. European Economic Community Council Directive
76/768 Annex VII
Finkelstein, E., Rosen, G.M., Raudkman, E.J., Paxton, J., 1979. Spin
trapping of superoxide. Mol. Pharmacol. 16, 676–685.

S. Scalia et al. / European Journal of Pharmaceutical Sciences 22 (2004) 241–249
249
Tarras-Wahlberg, N., Stenhagen, G., Larkö, O., Rosén, A., Wennberg,
A.M., Wennerström, O., 1999. Changes in ultraviolet absorption of
sunscreens after ultraviolet irradiation. J. Invest. Dermatol. 113, 547–
553.
Treffel, P., Gabard, B., 1996. Skin penetration and sun protection factor
of ultra-violet filters from two vehicles. Pharm. Res. 13, 770–774.
US Food and Drug Administration, 1999. Final monograph on sunscreen
drug products. Federal Register 64, 27666–27691.
Ziegler, A., Jonason, A.S., Leffell, D.J., Simon, J.A., Sharma, H.W.,
Kimmelman, J., Remington, L., Jacks, T., Brash, D.E., 1994. Sun-
burn and p53 in the onset of skin cancer. Nature 372, 773–
776.