Colloids and Surfaces B: Biointerfaces 75 (2010) 356–362

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces

journal homepage: www.elsevier.com/locate/colsurfb

Transdermal delivery of anticancer drug caffeine from water-in-oil

nanoemulsions
Faiyaz Shakeel ∗ , Wafa Ramadan
Department of Pharmaceutics, Faculty of Pharmacy, Al-Arab Medical University, Benghazi 5341, Libya

a r t i c l e i n f o a b s t r a c t

Article history: Recently caffeine has been investigated for the treatment of various types of cancers upon oral admin-
Received 16 June 2009 istration. There is also some evidence that dermally applied caffeine can protect the skin from skin
Received in revised form 2 August 2009 cancer caused by sun exposure. Therefore nanoemulsion formulation of caffeine for transdermal drug
Accepted 9 September 2009
delivery was developed and evaluated in the present investigation. Different w/o nanoemulsion formula-
Available online 13 September 2009
tions of caffeine were prepared by oil phase titration method. Thermodynamically stable nanoemulsions
were characterized for morphology, droplet size, viscosity and refractive index. The in vitro skin per-
Keywords:
meation studies were performed on Franz diffusion cell using rat skin as permeation membrane. The
Caffeine
Nanoemulsions
in vitro skin permeation profile of optimized formulation was compared with aqueous solution of
Transdermal delivery caffeine. Significant increase in permeability parameters was observed in nanoemulsion formulations
Phase diagrams (P < 0.05) as compared to aqueous solution of caffeine. The steady-state flux (Jss ) and permeability coeffi-
cient (Kp ) for optimized nanoemulsion formulation (C12) were found to be 147.55 ± 8.21 ␮g/cm2 /h and
1.475 × 10−2 ± 0.031 × 10−2 cm/h, respectively. Enhancement ratio (Er ) was found to be 17.37 in opti-
mized formulation C12 compared with other formulations. Overall these results suggested that w/o
nanoemulsions are good carriers for transdermal delivery of caffeine.
© 2009 Elsevier B.V. All rights reserved.

1. Introduction storage stability, lower preparation cost, good production feasi-

Chemically caffeine is 1,3,7-trimethylhanthine and found in tea
leaves, coffee, cocoa, guarana and kola nuts [1]. Recently some
reports indicated potential anticancer effects of caffeine upon oral
administration [2–4]. There are also some reports that indicated
that orally applied caffeine could protect the skin from ultraviolet
(UV) light induced skin cancer [5,6]. Transdermally applied caffeine
could be suitable for local as well as for systemic treatment of skin
cancer. Therefore transdermal w/o nanoemulsion formulation was
designed and evaluated in the present investigation.
Transdermal drug delivery represents the successful and inno-
vative area of research in drug delivery and known to enhance
therapeutic efficacy, bioavailability and to avoid any adverse effects
[7,8]. Different approaches such as chemical enhancers [9], ion-
tophoresis [10], sonophoresis [11], transdermal gels [12] and
nanoemulsions [7,8,13–15] have been used to enhance skin per-
meation and bioavailability of drugs. There has been increased
interest during recent years in the use of nano or microemulsions
that could modify drug permeation through the skin [7,8,13–15].
Nanoemulsion is one of the most promising technique for trans-
dermal delivery of drugs and present many advantages like higher
∗ Corresponding author. Tel.: +218 913899028.
E-mail address: faiyazs@fastmail.fm (F. Shakeel).
bility, thermodynamic stability, absence of organic solvents and
no need of intensive sonication [14–16]. Nanoemulsions are ther-
modynamically stable transparent or translucent dispersions of
oil and water stabilized by an interfacial film of surfactant usu-
ally in combination with cosurfactant having the droplet size less
than 100 nm [17–19]. Nanoemulsions have also shown improved
transdermal permeation of many drugs over the conventional
topical formulations such as emulsions [20] and gels [13–15].
Aqueous phase titration or spontaneous emulsification method
has been successfully investigated for the preparation of oil-
in-water (o/w) nanoemulsions of many lipophilic drugs in our
previous articles [13–15,17–19]. Caffeine is hydrophilic drug and
aqueous phase titration method is not suitable for preparation
of water-in-oil (w/o) nanoemulsions of caffeine. Therefore differ-
ent w/o nanoemulsions of caffeine were prepared using oil phase
titration method. Oil phase titration method has not been used
extensively for the preparation of nanoemulsions. Moreover trans-
dermal or dermal delivery of caffeine has not been investigated
using nanoemulsion technique. Therefore the aim of this arti-
cle was to investigate the potential of w/o nanoemulsion system
for transdermal delivery of anticancer drug caffeine using non-
irritant, pharmaceutically acceptable ingredients without using
additional chemical enhancers as components of nanoemulsions
themselves act as permeation enhancers. All these chemicals are
nonirritant, nontoxic and safe to the skin and fall under generally

0927-7765/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.colsurfb.2009.09.010
 F. Shakeel, W. Ramadan / Colloids and Surfaces B: Biointerfaces 75 (2010) 356–362
regarded as safe (GRAS) category and are pharmaceutically accept-
able.
2. Materials and methods
2.1. Materials
Caffeine was extracted from tea leaves using sublimation
method. Caprylic/capric triglyceride polyethylene glycol-4 com-
plex (Labrafac), caprylo caproyl macrogol-8-glyceride (Labrasol),
jojoba oil, oleoyl macroglycerides EP (Labrafil), Lauroglycol-
90, Lauroglycol-FCC and diethylene glycol monoethyl ether
(Transcutol-HP) were kind gift samples from Gattefossé (France).
Isopropyl alcohol (IPA), glycerol triacetate (Triacetin) and olive oil
were purchased from E-Merck, Germany. Polyoxy-35-castor oil
(Cremophor-EL), Tween-80 and Tween-85 were purchased form
Sigma–Aldrich, USA. All other chemicals used in the study were of
analytical reagent (AR) grade.
2.2. Screening of oils and water
The solubility of caffeine in various oils (Triacetin, Labrafac,
olive oil, Labrafil, Lauroglycol-90, Lauroglycol-FCC and jojoba oil)
was determined by dissolving excess amount of caffeine in 2 ml
of each of the selected oils in 5 ml capacity stoppered vials sepa-
rately. Excess amount of caffeine was added to each 5 ml capacity
stoppered vial and mixed for 10 min using a vortex mixer. The mix-
ture vials were then kept at 37 ± 1.0 ◦ C in an isothermal memmert
shaker bath (Memmert, Germany) for 72 h to get equilibrium. The
equilibrated samples were removed from shaker and centrifuged
at 3000 rpm for 15 min. The supernatant was taken and filtered
through a 0.45 ␮m membrane filter. The concentration of caffeine
was determined in each oil and distilled water by UV spectropho-
tometer at the wavelength of 273 nm [21].
2.3. Screening of surfactants
Six types of surfactants were tried for w/o nanoemulsion formu-
lation, which included Labrasol, Cremophor EL, Tween 80, Tween
85, Transcutol-HP and Plurol oleique. In water, 2.5 ml of 15% (w/w)
of surfactant solution was prepared and 4 ␮l of oil was added with
vigorous vortexing. If a one-phase clear solution was obtained, the
addition of the oil was repeated until the solution became cloudy.
2.4. Screening of cosurfactants
Transcutol-HP was combined with five types of solubilizers as
cosurfactants, namely, ethanol, isopropyl alcohol (IPA), n-butanol,
PEG 400 and propylene glycol. At a fixed Smix ratio of 1:1, the
pseudoternary phase diagrams were constructed. Twelve different
combinations in different weight ratios of oil and Smix , 1:9, 1:8,
1:7, 1:6, 1:5, 1:4, 1:3, 1:2, 1:1, 6:4 (1:0.7), 7:3 (1:0.43), and 9:1,
were taken so that maximum ratios were covered to delineate the
boundaries of phases precisely formed in the phase diagrams.
2.5. Effect of surfactant and cosurfactant mass ratio on
nanoemulsion formation
Surfactant (Transcutol-HP) was blended with cosurfactant (IPA)
in the weight ratios of 3:1, 2:1, 1:1, 1:0, 1:2, and 1:3. These Smix
ratios were chosen in decreasing concentration of surfactant with
respect to cosurfactant and increasing concentration of cosurfac-
tant with respect to surfactant for detailed study of the phase
diagrams. Oil phase titration method was used for the construc-
tion of the pseudoternary phase diagrams, which involves stepwise
357
addition of oil phase to each weight ratio of water and surfac-
tants, and then mixing the components with the help of vortex
mixer at 25 ◦ C. The nanoemulsion phase was identified as the region
in the phase diagram where clear, easily flowable, and transpar-
ent formulations were obtained based on the visual observation.
Twelve different combinations in different weight ratios of water
and Smix , 1:9, 1:8, 1:7, 1:6, 1:5, 1:4, 1:3, 1:2, 1:1, 6:4 (1:0.7), 112
7:3 (1:0.43), and 9:1, were taken. One axis of the pseudo-three-
component phase diagram represented the aqueous phase, the
other represented the oil phase, and the third represented a mixture
of surfactant and cosurfactant at a fixed weight ratio (Smix ).
2.6. Preparation of caffeine nanoemulsion formulations
From each phase diagram constructed, different formulae were
selected from nanoemulsion region for incorporation of drug into
the aqueous phase. 1% (w/w) of caffeine was dissolved in aqueous
phase (distilled water) of all selected nanoemulsion formulation.
Then selected quantity of Smix was added in aqueous solution of
drug and stirred for 2 min. The oil phase (Lauroglycol-90) was added
slowly with continuous stirring. Selected formulations were sub-
jected to different dispersion stability tests.
2.7. Dispersion stability studies
In order to overcome the problem of metastable and unstable
formulations, dispersion stability tests were performed. Selected
formulations were subjected for three dispersion stability tests
like centrifugation, heating & cooling cycles and freeze thaw cycle
test using procedure as described in our previously published arti-
cles [17–19]. The formulations which showed no phase separation,
creaming, coalescence or phase inversion upon these tests were
selected for characterization.
2.8. Characterization of nanoemulsions
Morphology and shape of the drug loaded nanoemulsion
(C12) were studied using transmission electron microscopy [TEM]
(Philips CM-10, USA).
Droplet size distribution of the nanoemulsion was determined
by photon correlation spectroscopy (PCS) using a Zetasizer 1000
HS (Malvern Instruments, UK). These studies were performed at
refractive index of 1.392 because the refractive index for all formu-
lation was within this range. The viscosity and dielectric constant
of the medium were set at 1.05 cp and 79.4, respectively.
The viscosity of the formulations was determined using Brook-
field DV III ultra V6.0 RV cone and plate rheometer (Brookfield
Engineering Laboratories, Inc, Middleboro, MA) using spindle #
CPE40 at 25 ± 0.5 ◦ C.
Refractive index of placebo and drug-loaded formulations was
determined using an Abbes type refractometer (Precision Standard
Testing Equipment Corporation, Germany).
2.9. In vitro skin permeation studies
In vitro skin permeation of caffeine from selected nanoemul-
sions (C1–C12) and aqueous solution of caffeine all containing same
quantity of caffeine (1%, w/w) was performed on Franz diffusion
cell with an effective diffusional area of 2.54 cm2 and 20 ml of
receiver chamber capacity using rat abdominal skin as permeation
membrane. The full thickness rat skin was excised from abdom-
inal region and hairs were removed with the help of an electric
clipper. The skin was prepared properly and stored in the deep
freezer at −21 ◦ C till further use [13–15]. On the day of experiment
the skin was brought to room temperature and mounted between
358 F. Shakeel, W. Ramadan / Colloids and Surfaces B: Biointerfaces 75 (2010) 356–362

donor and receiver compartment of Franz diffusion cell where stra-
tum corneum side faced the donor compartment and dermal side
faced receiver compartment of cell. Initially the donor compart-
ment was empty and receiver chamber was filled with phosphate
buffer saline (PBS) pH 7.4. The receiver fluid was stirred with a
magnetic rotor at a speed of 100 rpm and the assembled apparatus
was placed in hot air oven and the temperature was maintained at
37 ± 1 ◦ C. After complete stabilization of the skin, 1 g nanoemulsion
formulation (10 mg/g of caffeine) or 1 g of caffeine aqueous solution
(10 mg/g) was placed into each donor compartment. Samples were
withdrawn at regular interval (0.5, 1, 2, 3, 6, 12 and 24 h), filtered
through 0.45 membrane filter and analyzed for drug content by UV
spectrophotometer at the wavelength of 273 nm [21].
The skin permeation profiles of nanoemulsion formulations
were compared with aqueous solution using Dunnett test of one-
way analysis of variance (ANOVA).
2.10. Permeation data analysis
Cumulative amount of caffeine permeated through the skin
(␮g/cm2 ) was plotted as a function of time (t) for each formula-
tion. Rate of drug permeation at steady state (Jss ) was determined
from the slope of the linear portion of graph plotted between cumu-
lative drug permeated and time. Permeability coefficient (Kp ) was
calculated by dividing Jss with initial concentration of drug in donor
cell (Co ) by using the following equation [12]:
Jss
Kp =
Co
Enhancement ratio (Er ) was calculated by dividing the Jss of
respective formulation with Jss of control formulation by using the
equation:
Jss of formulation
Er =
Jss of control
2.11. Histopathological examination of skin specimens
Abdominal skin of Wistar rat was treated with optimized
nanoemulsion (C12) in PBS (pH 7.4). After 24 h, rat was sacrificed
and the skin samples from treated and untreated sites were taken.
These skin samples were stored in 10% formalin solution prepared
in PBS (pH 7.4). The skin samples were cut into different sections
vertically. Each section was dehydrated using ethanol, embedded
in paraffin for fixing and stained with xylene. These samples were
then observed under Carl Zeiss light microscope (Axioskop 40 FL,
Carl Zeiss, Germany)) fitted with canon power shot G3 digital cam-
era and compared with control sample. The light microscopy was
performed at low magnification (40×) level.
Table 1
Solubility of caffeine in various oils and distilled water at 37 ± 1 ◦ C.
Components Solubility mean ± SD (mg/ml)a
Labrafac 4.35 ± 0.24
Olive oil 6.64 ± 0.38
Jojoba oil 7.11 ± 0.93
Labrafil 3.21 ± 0.10
Lauroglycol-90 14.25 ± 1.24
Lauroglycol-FCC 12.62 ± 1.13
Distilled water 51.21 ± 3.14
a
Mean ± SD, n = 3.
for the development of the optimal nanoemulsion formulation of
caffeine. The solubility of caffeine in distilled water was found to be
51.21 ± 3.14 mg/ml (Table 1). Since solubility of caffeine was higher
in aqueous phase as compared to oil phase, w/o nanoemulsions
were developed for transdermal delivery of caffeine.
3.2. Screening of surfactants
The most critical problem related in the development of
nanoemulsion based drug delivery systems is the toxicity of the
surfactants. Large amounts of surfactants may cause skin irrita-
tion when administered transdermally. It is therefore important
to determine the surfactant concentration properly and use the
minimum concentration in the development of nanoemulsion for-
mulation. Another important criterion for surfactant selection is
the hydrophilic lipophilic balance (HLB) value of surfactants. The
right blend of surfactants with proper HLB value leads to the for-
mation of a stable nanoemulsion formulation. After selection of
oil (Lauroglycol-90), surfactant was selected based on the highest
solubilization capacity for the oil phase. In the present study, six
nonionic surfactants (Labrasol, Cremophor EL, Tween 80, Tween
85, Transcutol-HP and Plurol oleique) were chosen for screening. As
Transcutol-HP solubilized the maximum amount of Lauroglycol-90
(2.62%, w/w), it was selected as the surfactant for the development
of suitable w/o nanoemulsion. The solubilization behavior of the
Lauroglycol-90 into six types of surfactant solutions is given in
Fig. 1.
3.3. Screening of cosurfactants
Addition of cosurfactants causes further reduction in the interfa-
cial tension between oil phase and aqueous phase and increase the
fluidity of the interface. Therefore, ethanol, IPA, n-butanol, PEG-400
and propylene glycol were selected as cosurfactants. Nanoemul-
sion area was used as the selection criterion for the evaluation

3. Results and discussion

3.1. Screening of oils and water

Development of nanoemulsion formulation depends on physic-

ochemical properties of drugs. Lipophilic drugs are encapsulated
in oil phase of o/w nanoemulsions, whereas hydrophilic drugs are
encapsulated in aqueous phase of w/o nanoemulsions. Solubil-
ity of the lipophilic drugs in the oil phase and hydrophilic drugs
in aqueous phase is an important criterion for the selection of
oils and water, respectively. Because caffeine is hydrophilic drug,
its solubility in water is more important than oil phase. The sol-
ubility of caffeine in different oils as well as in distilled water
was determined (Table 1). The solubility of caffeine was found to
be highest in Lauroglycol-90 (14.25 ± 1.24 mg/ml) as compared to
other oils. Therefore Lauroglycol-90 was selected as the oil phase Fig. 1. % (w/w) of oil solubilized by different surfactants.
 F. Shakeel, W. Ramadan / Colloids and Surfaces B: Biointerfaces 75 (2010) 356–362
Fig. 2. Pseudoternary phase diagrams of W/O nanoemulsion region of Lauroglycol-90 (oil phase), Transcutol-HP (surfactant) with different cosurfactants indicated in A
(ethanol), B (IPA), C (n-butanol), D (PEG-400) and E (PG) at 1:1 Smix ratio.
of cosurfactants. The nanoemulsion region in the phase diagrams
were compared at a fixed Smix (1:1) ratio, keeping the surfactant
the same but replacing the cosurfactant (Fig. 2). It was found that,
when the chain length was increased from ethanol (Fig. 2A) to iso-
propyl alcohol (Fig. 2B), it increased the area of the existence of
the nanoemulsion. However, with n-butanol (Fig. 2C), a consider-
able decrease in the area was observed. The nanoemulsion zone
was lowest for PEG-400 and PG (Fig. 2D and E). The presence of
the cosurfactant and its type can thus affect the phase behavior
of the nanoemulsion. Based on these results, IPA was found to be
an efficient cosurfactant for its maximum nanoemulsion zone, and
hence, it was selected as the cosurfactant for the development of
w/o nanoemulsion formulation of caffeine.
3.4. Effect of surfactant and cosurfactant mass ratio on
nanoemulsion formation
The existence of nanoemulsion formation zones can be illus-
trated with aid of the pseudoternary phase diagrams [22].
359
Pseudoternary phase diagrams were constructed using Laurogly-
col 90 as the oil phase and Transcutol-HP & IPA as the surfactant
and cosurfactant respectively. Effect of surfactant and cosurfac-
tant mass ratio on nanoemulsion formation was evaluated for the
further optimization of the nanoemulsion system. In Fig. 3, a low-
nanoemulsion area was observed when Transcutol-HP or IPA were
used alone not in combination i.e., at the Smix of ratio 1:0 (Fig. 3A and
B). A w/o nanoemulsion region was found towards the oil-rich apex
of the phase diagrams. The maximum amount of water that was
solubilized, as can be seen in the phase diagram, was around 22%
(w/w) at 52% (w/w) of surfactant (Fig. 3A and B). When cosurfactant
(IPA) was added with surfactant (Transcutol-HP) in equal amounts,
a higher nanoemulsion region was observed, perhaps because of
the further reduction of the interfacial tension and increased flu-
idity of the interface at Smix 1:1 (Fig. 3C). The maximum amount
of water that was solubilized using 1:1 Smix ratio was around 28%
(w/w) at 41% (w/w) of Smix . On further increasing the surfactant
concentration, i.e., at Smix 2:1 (Fig. 3D), the nanoemulsion region
started to decrease again as compared to Smix 1:1. When the sur-
360 F. Shakeel, W. Ramadan / Colloids and Surfaces B: Biointerfaces 75 (2010) 356–362

Fig. 3. Pseudoternary phase diagrams of W/O nanoemulsion region of Lauroglycol-90 (oil phase), Transcutol-HP (surfactant) and isopropyl alcohol (cosurfactant) at different
Smix ratios (1:0 to 3:1).

factant concentration was further increased in the Smix ratio of 3:1
(Fig. 3E), a decrease in the nanoemulsion region was observed again
as compared to Smix 1:1. Because nanoemulsion region was started
decrease in Smix ratio of 2:1 and 3:1, there was no need to go for
Smix ratio of 4:1 or 5:1.
observed using TEM as shown in Fig. 4. The shape of droplets was
found to be spherical. Most of the droplets were of uniform shape
and size.
The mean droplet size of caffeine nanoemulsions was found in
the range of 20.14–105.25 nm. The droplet size increased with the

3.5. Dispersion stability studies

Table 2
Nanoemulsions are thermodynamically and physically stable Composition of selected w/o caffeine nanoemulsion formulations.
systems and are formed at a particular concentration of water,
Code %, w/w of components Smix ratio Smix :water ratio
surfactant and oil. It is the thermostability which differentiates
*
nano or micro emulsion from emulsions that have kinetic stability Water Smix Oil

and will eventually phase separate [17–19]. Thus, the formula- C1 5 40 55 1:1 8.00
tions were tested for different dispersion stability tests. Only those C2 10 40 50 1:1 4.00
C3 15 40 45 1:1 2.66
formulations, which showed no phase separation, creaming, crack-
C4 20 40 40 1:1 2.00
ing, coalescence or phase inversion upon these stress tests, were C5 5 40 55 2:1 8.00
selected for further studies. The compositions of these formulations C6 10 40 50 2:1 4.00
are given in Table 2. C7 15 40 45 2:1 2.66
C8 20 40 40 2:1 2.00
C9 5 40 55 3:1 8.00
3.6. Characterization of nanoemulsions C10 10 40 50 3:1 4.00
C11 15 40 45 3:1 2.66
C12 20 40 40 3:1 2.00
Morphology of nanoemulsion C12 was determined using TEM
*
experiments. A “positive” image of nanoemulsion (C12) was Aqueous phase (water) containing 1% (w/w) of caffeine.
 F. Shakeel, W. Ramadan / Colloids and Surfaces B: Biointerfaces 75 (2010) 356–362 361

Fig. 4. Transmission electron microscopic positive image of formulation C12.

Fig. 5. In vitro skin permeation profile of caffeine from different W/O nanoemulsions
(C1–C12) and aqueous solution of caffeine (control).
increase in concentration of oil (Lauroglycol-90) in the formula-
tions (Table 3). The droplet size of the formulation C12, containing
20% of distilled water and 40% of oil was lowest (20.14 ± 2.12 nm) that the nanoemulsion formulations were not only physically sta-
as compared to other formulations. When ratio of Smix to water was ble but also chemically stable and remained isotropic in nature,
increased, it was found that droplet size was significantly increased thus having no interactions between nanoemulsion components
(P < 0.05) as shown in Table 3. The droplet size was lowest when and drug.
concentration of Smix was two times of water (C12). All the for-
mulations had droplets in the nanometer range with low values 3.7. In vitro skin permeation studies
of polydispersity index. The polydispersity values of all formula-
tions were very low (0.105–0.177) which indicated the uniformity In vitro skin permeation studies were performed to compare
of droplets within the formulation. The polydispersity index was the permeation of caffeine from 12 different nanoemulsion for-
lowest for formulation C12 (0.105). mulations (C1 to C12) and aqueous solution of caffeine, all having
The viscosity of nanoemulsion formulations was found in the same quantity (1%, w/w) of caffeine. In vitro skin permeation pro-
range of 15.31–54.64 cps. Formulation C12 had least viscosity as file of all nanoemulsions was significant as compared to aqueous
compared to other formulations (15.31 cp) as shown in Table 3. solution of caffeine (control) as shown in Fig. 5 (P < 0.05). The in
The difference in viscosity of the formulations C12 was significant vitro skin permeation profile of C12 was significantly different
from other formulations but it was observed that the viscosity of from other nanoemulsion formulations (P < 0.05). The significant
all formulations was very low which is expected, as one of the difference in caffeine permeation between nanoemulsion formu-
characteristics of the nanoemulsion formulation is lower viscosity lations and aqueous solution of caffeine (control) could be due to
[15]. the mean size of internal phase droplets, which was different in
The mean values of refractive index of drug loaded and placebo different nanoemulsions. The droplet size of formulation C12 was
formulations were found in the range of 1.392 to 1.402. When significantly lowest than other formulations (P < 0.05). Moreover,
the values of refractive index were compared with the placebo, polydispersity index and viscosity of formulation C12 was also sig-
it was found that there were no significant changes in the val- nificantly lower (P < 0.05).
ues containing the drug (P ≥ 0.05). The values of refractive index
of all formulations were similar to the refractive index of oil phase 3.8. Permeation data analysis
(Lauroglycol-90) which indicated that developed nanoemulsions
were of w/o type. From values of refractive index it was concluded The graph between cumulative caffeine permeated and time
was plotted for each nanoemulsion formulation (Fig. 5). Per-
meability parameters like Jss , Kp and Er were significantly
Table 3 increased in nanoemulsions as compared to aqueous solution
Mean droplet size, polydispersity index and viscosity of the nanoemulsion
of caffeine (P < 0.05) as shown in Table 4. Aqueous solution of
formulations.
caffeine was used as control formulation for the determination
Formulation code Droplet size Polydispersity Viscosity mean of permeability parameters. The permeability parameters were
mean ± SD ± SD (cps)a significant for formulation C12 as compared to other formula-
(nm)a
tions (P < 0.05). Er was found to be highest (17.37) in formulation
C1 105.25 ± 13.14 0.124 54.64 ± 4.87 C12 compared with other formulations. The values of Jss and Kp
C2 84.54 ± 10.69 0.126 43.76 ± 3.12
for formulations C12 were found to be 147.55 ± 8.21 ␮g/cm2 /h
C3 55.61 ± 9.65 0.173 34.42 ± 2.51
C4 32.14 ± 5.72 0.177 24.32 ± 2.01 and 1.475 × 10−2 ± 0.031 × 10−2 cm/h, respectively (Table 4). The
C5 98.54 ± 12.59 0.127 50.12 ± 4.65 enhanced permeability parameters of C12 could be due to
C6 75.64 ± 9.11 0.120 37.87 ± 2.97 the presence of permeation enhancers such as Lauroglycol-90,
C7 48.54 ± 8.14 0.167 29.98 ± 2.11
Transcutol-HP and IPA.
C8 25.66 ± 4.02 0.156 20.42 ± 1.95
C9 92.14 ± 10.69 0.116 44.53 ± 4.12
C10 68.54 ± 8.13 0.118 32.65 ± 2.15 3.9. Histopathological examination of skin specimens
C11 42.25 ± 7.24 0.171 23.76 ± 1.94
C12 20.14 ± 2.12 0.105 15.31 ± 1.11 Histopathological examination of treated and untreated (con-
a
Mean ± SD, n = 3. trol) sites of rat skin was performed using Carl Zeiss light
362 F. Shakeel, W. Ramadan / Colloids and Surfaces B: Biointerfaces 75 (2010) 356–362

Fig. 6. Photomicrograph of skin sample from (A) control group animal and (B) nanoemulsion (C12) treated animal at low power (40×).

Table 4 Results of histopathological examinations indicated that developed

Permeability parameters of different formulations.
nanoemulsion is safe for transdermal delivery of caffeine.
Formulation codes Jss (␮g/cm2 /h)a Kp (cm/h)a × 10−2 Er Overall these results indicated that w/o nanoemulsions are
Control* 8.49 ± 0.98 0.084 ± 0.001 – good carriers for transdermal delivery of anticancer drug caffeine.
C1 71.17 ± 4.12 0.711 ± 0.014 8.38 However further, in vivo investigations are required to evaluate
C2 91.46 ± 5.61 0.914 ± 0.016 10.77 improved anticancer efficacy of caffeine.
C3 109.90 ± 6.21 1.099 ± 0.021 12.94
C4 124.68 ± 6.81 1.246 ± 0.026 14.68
C5 69.16 ± 3.31 0.691 ± 0.012 8.14
Acknowledgement
C6 98.58 ± 5.92 0.985 ± 0.019 11.61
C7 119.80 ± 6.32 1.198 ± 0.022 7.88 The authors are thankful to Gattefosse, France for providing the
C8 135.37 ± 7.05 1.353 ± 0.028 15.94 gift samples of Lauroglycol-90 and Transcutol-HP.
C9 92.21 ± 5.41 0.922 ± 0.018 10.86
C10 102.36 ± 6.01 1.023 ± 0.024 12.05
C11 123.70 ± 7.11 1.237 ± 0.026 14.57 References
C12 147.55 ± 8.21 1.475 ± 0.031 17.37
[1] D.S. Murray, P.J. Hansen, J. Chem. Edu. 72 (1995) 851.
a
Mean ± SD, n = 3. [2] J.N. Sarkaria, E.C. Busby, R.S. Tibbetts, P. Roos, Y. Taya, L.M. Karnitz, R.T. Abraham,
*
Aqueous solution of caffeine was used as control. Cancer Res. 59 (1999) 4375.
[3] S. Michal, S. Andrzej, Curr. Pharm. Biotechnol. 9 (2008) 325.
[4] K. Tomita, H. Tsuchiya, Cancer Sci. 80 (2008) 83.
microscope. The photomicrographs of untreated or control rat skin [5] Y.P. Lu, Y.R. Lou, Y. Lin, W.J. Shih, M.T. Huang, C.S. Yang, A.H. Conney, Cancer
showed normal skin layers. As shown in Fig. 6A. When the skin was Res. 61 (2001) 5002.
treated with nanoemulsion (C12) for 24 h, definite changes were [6] Y.P. Lu, Y.R. Lou, Q.Y. Peng, J.G. Xie, P. Nghiem, A.H. Conney, Cancer Res. 68
(2008) 2523.
observed in the skin morphology as shown in Fig. 6B that could [7] F. Shakeel, S. Baboota, A. Ahuja, J. Ali, S. Shafiq, J. Nanobiotechnol. 6 (2008) E8.
be due to the action of nanoemulsion on stratum corneum [7,8]. [8] F. Shakeel, S. Baboota, A. Ahuja, J. Ali, S. Shafiq, J. Drug Target 16 (2008) 733.
Dermis does not show any edema or inflammatory cell infiltration [9] R. Walker, E. Smith, Adv. Drug. Deliv. Rev. 18 (1996) 295.
[10] R. Guy, J. Pharm. Pharmacol. 50 (1998) 371.
(Fig. 6B). There were no apparent signs of skin irritation (erythma [11] A. Boucaud, L. Machet, B. Arbeille, M. Machet, M. Sournac, A. Mavon, F. Patat, L.
and edema) observed on skin specimens treated with nanoemul- Vaillant, Int. J. Pharm. 228 (2001) 69.
sion C12 indicating absence of any skin irritation as a consequence [12] S. Baboota, F. Shakeel, K. Kohli, Method Fin. Exp. Clin. Pharmcol. 28 (2006) 109.
[13] F. Shakeel, S. Baboota, A. Ahuja, J. Ali, M. Aqil, S. Shafiq, AAPS Pharm. Sci. Technol.
of nanoemulsion treatment. These results indicated that developed 8 (2007) E104.
nanoemulsion is safe for transdermal delivery of caffeine. [14] F. Shakeel, S. Baboota, A. Ahuja, J. Ali, S. Shafiq, J. Disp. Sci. Technol. 30 (2009)
834.
[15] F. Shakeel, W. Ramadan, M.A. Ahmed, J. Drug Target 17 (2009) 435.
4. Conclusion [16] S. Shafiq, F. Shakeel, AAPS Pharm. Sci. Technol. 9 (2008) 1097.
[17] S. Shafiq, F. Shakeel, S. Talegaonkar, F.J. Ahmad, R.K. Khar, M. Ali, Eur. J. Pharm.
Based on significant in vitro skin permeation profile, highest Biopharm. 66 (2007) 227.
[18] S. Shafiq, F. Shakeel, S. Talegaonkar, F.J. Ahmad, R.K. Khar, M. Ali, J. Biomed.
steady state flux (147.55 ␮g/cm2 /h), highest enhancement ratio
Nanotechnol. 3 (2007) 28.
(17.37), lowest droplet size (20.14 nm), least polydispersity index [19] S. Shafiq, F. Shakeel, S. Talegaonkar, J. Ali, S. Baboota, A. Ahuja, R.K. Khar, M. Ali,
(0.105), lowest viscosity (15.31 cps) and lower concentrations of oil, AAPS Pharm. Sci. Technol. 8 (2007) 28.
[20] M.R. Gasco, M. Gallarate, F. Pattarino, Int. J. Pharm. 69 (1991) 193.
surfactant and cosurfactant, formulation C12 have been optimized
[21] Y. Yamauchi, A. Nakamura, I. Kohno, M. Kitai, K. Hatanaka, T. Tanimoto, Chem.
as nanoemulsion formulation of caffeine containing 1.0% (w/w) of Pharm. Bull. 56 (2008) 185.
caffeine, 20% (w/w) of distilled water, 30% (w/w) of Transcutol-HP, [22] M.J. Lawrence, G.D. Rees, Adv. Drug. Deliv. Rev. 45 (2000) 89.
10% (w/w) of isopropyl alcohol and 40% (w/w) of Lauroglycol-90.