Hindawi

Oxidative Medicine and Cellular Longevity

Volume 2017, Article ID 5205471, 17 pages
https://doi.org/10.1155/2017/5205471

Research Article
Curcumin Protects Skin against UVB-Induced Cytotoxicity via the
Keap1-Nrf2 Pathway: The Use of a Microemulsion Delivery System

Maya Ben Yehuda Greenwald,1,2,3,4 Marina Frušić-Zlotkin,1 Yoram Soroka,1

Shmuel Ben Sasson,4 Ronit Bitton,5,6 Havazelet Bianco-Peled,2,3 and Ron Kohen1
1
The Institute for Drug Research, School of Pharmacy, The Hebrew University of Jerusalem, 9112100 Jerusalem, Israel
2
Department of Chemical Engineering, Technion-Israel Institute of Technology, Technion City, 3200003 Haifa, Israel
3
The Russell Berrie Nanotechnology Institute and Technion-Israel Institute of Technology, 32000 Haifa, Israel
4
Department of Developmental Biology and Cancer Research, The Hebrew University Medical School, Ein-Karem Campus,
9112100 Jerusalem, Israel
5
Department of Chemical Engineering, Ben-Gurion University of the Negev, 8410501 Beer-Sheva, Israel
6
Ilse Katz Institute for Nanoscale Science and Technology, 8410501 Beer-Sheva, Israel

Correspondence should be addressed to Ron Kohen; [email protected]

Received 24 October 2016; Revised 19 March 2017; Accepted 16 April 2017; Published 5 July 2017

Academic Editor: Luciano Saso

Copyright © 2017 Maya Ben Yehuda Greenwald et al. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work
is properly cited.

Curcumin was found to be beneficial in treating several skin pathologies and diseases, providing antioxidant protection due to its
reducing properties and its electrophilic properties (the ability to activate the Nrf2 pathway and induce phase II cytoprotective
enzymes). Nevertheless, clinical applications of curcumin are being hampered by its insufficient solubility, chemical instability,
and poor absorption, leading to low efficacy in preventing skin pathologies. These limitations can be overcome by using a
nanotechnology-based delivery system. Here, we elucidated the possibility of using curcumin encapsulated in a microemulsion
preserving its unique chemical structure. We also examined whether curcumin microemulsion would reduce UVB-induced
toxicity in skin. A significant curcumin concentration was found in the human skin dermis following topical application of a
curcumin microemulsion. Moreover, curcumin microemulsion enhanced the reduction of UV-induced cytotoxicity in epidermal
cells, paving the way for other incorporated electrophiles in encapsulated form protecting skin against stress-related diseases.

1. Introduction (Nrf2), which is responsible for the induction of a variety of

cytoprotective genes [3]. Regulated by the Keap1 metallopro-
The concept that antioxidants can protect cells and organs tein, Nrf2 is capable of inducing a large number of genes
against oxidative stress has been established in numerous encoding antioxidant enzymes and genes enabling homeosta-
basic and clinical studies [1]. Nevertheless, nowadays, it has sis and controlling processes involved in the pathology of
become evident that antioxidants of low molecular weight many diseases (e.g., immune and inflammatory responses,
cannot protect the living organism against continuous tissue remodeling and fibrosis, carcinogenesis, and metastasis)
stress and sometimes can even be deleterious [2]. Oxidants [4, 5]. Nrf2 plays a vital and crucial role in the maintenance of
(electrophiles), on the other hand, were recently shown to skin homeostasis and repair and regeneration in various
be compounds capable of inducing cellular-protecting disease states of the skin [6]. However, acute and chronic
enzymes such as the phase II enzymes when provided in mod- Nrf2 activation in a healthy epidermis resulted in a negative
erate concentrations. One of the basic factors activated when effect on skin integrity [6]. Endogenous Nrf2 has the ability
an electrophile is present is the transcription factor nuclear to protect skin against UV irradiation [6]. Nrf2 is also capable
factor (erythroid-derived 2)-like 2, an NF-E2-related factor 2 of decreasing symptoms of skin photoaging (e.g., wrinkle
2 Oxidative Medicine and Cellular Longevity

formation, loss of skin flexibility) [6]. The pharmacological treating various skin pathologies and disorders and Nrf2
activation of Nrf2 was proven to provide protection against involvement is summarized in Table 1.
various toxic compounds responsible for a reduction in skin However, the pharmacokinetics of curcumin are unsatis-
toxicity [6]. The role of Nrf2 in the prevention of skin factory due to its chemical instability, scarce solubility in
carcinogenesis has been demonstrated in various research aqueous solutions, deficient absorption, rapid metabolism,
models [6]. Nrf2 is a key element in the prevention of chemi- and systemic elimination [8, 20]. Therefore, curcumin suffers
cally induced tumor formation and promotion [6]. Moreover, from poor bioavailability and its clinical application is
Nrf2 activation reduced solar-simulated UV radiation tumor restricted [8, 20]. Moreover, no double-blinded, placebo-
formation in hairless mice [6]. Nrf2 also demonstrated its controlled clinical trial of curcumin has been successful
essentiality in the healing process of full-thickness wounds [21]. A reasonable approach to overcome these limitations
and in the recovery and repair of an epidermal barrier could be to encapsulate curcumin into delivery systems of
defect [6]. There are compelling evidences demonstrating different characteristics [22, 23]. In addition, a topical deliv-
Nrf2 activation as a promising strategy for the treatment ery system for local administration of curcumin mat results
of atopic dermatitis, psoriasis, and epidermal blistering in an increase in curcumin bioavailability [21]. There are
diseases (e.g., Hailey-Hailey disease) [6]. Nrf2 activation compelling evidences supporting this approach. It was shown
in vitiligo vulgaris pigment disorder was investigated as a that topical application of curcumin exhibited a more
potential strategy to prevent the development of the disor- pronounced effect on wound healing compared to its oral
der and treatment [6]. It was also suggested that activation administration due to a superior accessibility of curcumin
of Nrf2 is important for the treatment of patients suffering at the wound site [10]. One of the leading vehicles for dermal
from allergic skin inflammation (e.g., allergic contact drug delivery is microemulsions [24]. Microemulsions are
dermatitis) [6]. isotropic colloidal nano-formulations, composed of water,
Curcumin (1,7-bis(4-hydroxy 3-methoxy phenyl)-1,6- oil, and surfactants [25]. These vehicles are thermodynami-
heptadiene-3,5-dione) is a natural polyphenol from the pow- cally stable and form almost spontaneously (without any
dered rhizome of the medicinal plant Curcuma longa (also energy input) to a transparent or slightly opalescent formula-
known as turmeric) [7]. It is an amphipathic molecule with tion of low viscosity [25]. The use of microemulsions offers
polar-central and flanking regions that are separated by a many advantages including enhancement of drug solubility,
lipophilic methine segment [8]. Curcumin contains seven protection of labile drugs, controlled drug release, augmenta-
chemical functional groups (see curcumin chemical structure tion in the rate and extent of absorption, and a decrease in
in the Supplementary Data (Figure S1) available online at patient side effects [24]. In addition, it has been shown
https://doi.org/10.1155/2017/5205471) [8]. Among others, that microemulsions significantly increase bioavailability
curcumin contains phenolic groups and thus can act as a compared with classical delivery systems such as emulsions,
reducing antioxidant and directly scavenge oxygen-centered gels, and solutions [24].
reactive intermediates [8, 9]. Curcumin also displays oxidant Incorporating curcumin into a microemulsion may
activity partly due to its Michael acceptor functionalities. As improve its water solubility and bioavailability and hence
such, curcumin is capable of inducing the activation of the lead to better efficacy [21]. Until now, little work has focused
Keap1-Nrf2-EpRE pathway [9]. The unique chemical attri- on topical microemulsion delivery systems containing
butes of curcumin (e.g., log P ensuring curcumin’s accessibility curcumin aimed at treating skin conditions [26–29]. Nano-
to its molecular targets, the capacity to undergo H-bonding formulations of curcumin might potentially improve the
and hydrophobic interactions, and activity as a Michael infiltration of curcumin into cutaneous cells [10]. Indeed,
acceptor) are responsible for curcumin’s pleiotropic biologi- studies to date support this claim [26–29]; Lin et al. devel-
cal activity [8]. These include curcumin’s bifunctional oped curcumin encapsulated in an oil-in-water microemul-
antioxidant properties, anti-inflammatory activity, antican- sion system and investigated its phase diagram and stability
cer effects, wound healing, and antimicrobial effects [8–14]. [29]. In vitro skin permeation assays have demonstrated
Therefore, curcumin was suggested for the treatment of time-dependent increases in permeated curcumin in stable
various disorders like cancer and proinflammatory chronic microemulsion formulations. Enhanced skin permeability
diseases [8–14]. Skin, being an interface between the of curcumin encapsulated in microemulsions was also
environment and the body, suffers from chronic oxidative reported by Liu and Chang [28]. The vehicle composition
stress resulting from exposure to environmental toxicants significantly influenced curcumin solubility and skin
including chemical and physical pollutants, ionization, and permeability [28]. Teichmann et al. incorporated curcumin
UV radiation [15]. The resulting oxidative stress in skin in an oil-in-water microemulsion and in an amphiphilic
may be involved in the pathogenesis of a number of skin dis- cream [30]. A deeper part of the stratum corneum was
orders including some types of cutaneous malignancy and accessible, and significantly smaller amounts of curcumin
photosensitivity diseases [15]. Curcumin, due to its pleiotro- were found on the skin surface following microemulsion
pic behavior, was found to be beneficial in treating several application. Furthermore, curcumin was detected in hair
skin pathology disorders and diseases (e.g., psoriasis, sclero- follicles, indicating that the microemulsion penetrated into
derma, and skin cancer) [8–14, 16–19]. Moreover, drug the complete follicular infundibula [30]. Liu and Huang
development studies were carried out where curcumin demonstrated that the antimicrobial activity of curcumin-
analogues were designed and synthesized due to curcumin’s loaded myristic acid microemulsions against the skin path-
antiangiogenic activities [13]. The role of curcumin in ogen Staphylococcus epidermidis was 12 times higher than
Oxidative Medicine and Cellular Longevity 3

Table 1: Curcumin’s role in treating various skin pathologies and disorders and the interconnectedness with Nrf2 [8–14, 16–19].

Skin pathology/disorder Effect of curcumin treatment Nrf2 involvement

Inflammatory reduction via
Inflammatory diseases (e.g., psoriasis, (1) Inhibition of NF-κB transcription factor and reducing
atopic dermatitis, contact dermatitis, acne, the production of TNF-α, IL-1, and interferon-γ +
rosacea, and erythroderma) (2) Scavenging reactive oxygen species
(3) Modulating production of antioxidant enzymes
Scleroderma Antifibrotic effect via suppressing TGF-β +
Protection against disease progression via
(1) An increase in MAPK/ERK phosphorylation and
inhibition of apoptosis
Vitiligo vulgaris +
(2) An increase in total antioxidant capacity and a decrease in
intracellular reactive oxygen species generation
(3) Improving mitochondrial activity
Enhancing effective wound healing in three stages
(a) Inflammation (see above)
(b) Proliferation
(1) Enhancing fibroblast migration, granulation tissue formation,
collagen deposition, and re-epithelialization
(2) Apoptosis in the early stage of wound healing resulting
Wound healing +
in removal of nondesirable inflammatory cells
from the wound site
(c) Remodeling
(1) Enhancing wound closure via the production of TGF-β1 and
fibronectin resulting in increased migration and proliferation
of fibroblasts
Delay the aging process via induction of Keap1-Nrf2-EpRE and
Aging +
phosphatidylinositol 3-kinase/Akt pathways
Anticarcinogenic activity in different stages of cancer
(a) Transformation of normal cells into tumor cells: curcumin
inhibits NF-κB and its target genes like COX-2 and cyclin
D1 and induces apoptosis via activation of caspase-3,
caspase-8, and Fas receptor
Carcinogenesis +
(b) Tumor growth and progression: curcumin inhibits mTOR
signaling resulting in blocking of tumor progression
(c) Tumor promotion: curcumin inhibits 12-o-tetradecanoylphorbol-
(TPA-) induced tumor promotion and TPA-induced tumor markers via
modulation of transmembrane signal transduction via protein kinase

that of curcumin dissolved in dimethyl sulfoxide (DMSO) radicals were encapsulated into a microemulsion delivery
[26]. Yutani et al. assessed the distribution of polyphenols system resulting in enhanced Nrf2 activation, protection
in skin from a di-2-ethylhexyl sodium sulfosuccinate against UVB-induced injury, and relief in inflamed skin con-
(Aerosol OT) microemulsion and detected enhanced dition [33]. While encapsulating these synthetic antioxidants
intradermal delivery [27]. with diverse lipophilicity and ability to shuttle between the
In the present study, we hypothesize that incorporating nitroxide radical, the reduced hydroxylamine, and the
curcumin into a topical microemulsion delivery system will oxidized oxoammonium cation formed by one- and two-
preserve its unique chemical structure allowing it to induce electron transfer reactions [33] (i.e., members of the nitrox-
the Keap1-Nrf2-EpRE pathway more efficiently than the ide family) may be challenging, the case of encapsulating
unformulated curcumin. This hypothesis holds an additional the natural polyphenol curcumin into a microemulsion
rationale; it was shown that chronic and enhanced activation delivery system holds different challenges since curcumin is
of the Keap1-Nrf2-EpRE pathway in the epidermis suffers from prone to oxidative degradation and has low solubility in
several detrimental complications including defects in the aqueous solution [8, 20].
epidermal barrier, inflammation, and induced keratinocyte In order to investigate our hypothesis, curcumin was
hyperproliferation [31]. Thus, precise and temporary activa- incorporated into a microemulsion delivery system and
tion of the Keap1-Nrf2-EpRE in skin is essential [32]. microemulsion nanometric structure was evaluated using
Here, we suggest to expand our prior work demonstrat- dynamic light scattering (DLS), small-angle light scattering
ing the feasibility of encapsulating Nrf2-activating agent into (SAXS), and cryo-transmission electron microscopy (cryo-
a delivery system. We have previously shown that three TEM) measurements. Moreover, the antioxidant activity of
members of the nitroxide family representing synthetic stable curcumin incorporated into the microemulsion was evaluated
4 Oxidative Medicine and Cellular Longevity

in vitro using oxygen radical absorbance capacity assay manipulations (the final curcumin concentration was 1%
(ORAC), luminol-dependent chemiluminescence (LDCL) w/w (27.1 mM)) using a small-angle diffractometer
assay, and 2-diphenyl-lpicrylhydrazyl (DPPH) radical (a Molecular Metrology SAXS system with Cu Kα radiation
assay. The capability of a curcumin-loaded microemulsion from a sealed microfocus tube (MicroMax-002+S), two
to induce the Keap1-Nrf2-EpRE pathway was evaluated Göbel mirrors, and three-pinhole slits; the generator was
in keratinocyte culture and in human skin. Human skin powered at 45 kV and 0.9 mA). Scattering patterns were
organ culture was used to access the reduced UVB- recorded by a 20 × 20 cm two-dimensional position-
induced cytotoxicity resulting from topical application of sensitive wire detector (gas-filled proportional type of
the curcumin-loaded microemulsion. Gabriel design with 200 μm resolution) that was positioned
150 cm behind the sample. Scattered intensity I (q) was
2. Material and Methods recorded in the interval 0.07 < q < 2.7 nm−1, where q is the
scattering vector defined as q = (4π/λ) sin (Ѳ), where 2Ѳ is
∗
Similar material and methods were used in [33]. the scattering angle and λ is the radiation wavelength
(0.1542 nm). Microemulsions were sealed in a thin-walled
2.1. Microemulsion Preparation. Microemulsions were pre- capillary (glass) of about 2 mm diameter and 0.01 mm wall
pared by first mixing the surfactants lauric acid (pKa = 5 3 thickness. Experiments were performed under vacuum at
at room temperature, Sigma-Aldrich, Israel), Span® 20 ambient temperature. Scattering curves were adjusted for
(sorbitan laurate, Sigma-Aldrich, Israel), and Tween® 80 counting time and sample absorption.
(polysorbate 80, Sigma-Aldrich, Israel) with isopropyl myris-
tate (IPM, Sigma-Aldrich, Israel). Upon receiving a transpar- 2.5. Spectrofluorometer Measurements. Curcumin location in
ent blend of surfactants and oil, curcumin (Sigma-Aldrich, the microemulsion was investigated using the fluorescent
Israel) was added to the solution and then mixed until probe method [35], which can sense the microenvironment
completely dissolved. This step was followed by a drop- of the probe from changes in the intensity and wavelength
wise addition of double-distilled water (DDW; pH = 6.8 of the emission peak. Curcumin’s emission properties highly
± 0.2). Solutions were allowed to equilibrate for 24 h to depend on its specific microenvironment; therefore, curcu-
obtain a clear oil-in-water microemulsion. The ratio of min could be used directly as a probe [8, 36]. Curcumin
Tween 80® : Span 20 : lauric acid : IPM : curcumin was was dissolved in different microemulsion components to a
33.3 : 1.6 : 1 : 5 : 1.3 and kept constant throughout the study. final concentration of 0.007% w/w (1.9 μM), and fluorescence
The final concentrations (%w/w) in the microemulsion measurements were obtained using a Jobin Yvon Horiba
were 26.8 : 1.3 : 0.8 : 4 : 1 : 66.1 for Tween 80, Span 20, lauric Fluoromax 4 spectrofluorometer. The excitation source was
acid, isopropyl myristate, curcumin, and water, respec- a xenon arc lamp. The excitation and emission slit widths
tively. tert-Butylhydroquinone (tBHQ) and trolox were were 5 nm. Excitation was set at 450 nm, and emission was
purchased from Sigma-Aldrich, Israel. scanned from 460 nm to 600 nm.
2.2. Dynamic Light Scattering (DLS). DLS measurements on
microemulsions (microemulsions were diluted to 1 : 100 with 2.6. Voltammetric Measurements of Reducing Power. The
DDW; the final curcumin concentration was 0.01% w/w overall reducing power of microemulsions (the final curcu-
(0.27 mM)) were performed using a Zetasizer Nano Series min concentrations were as follows (%w/w): 0 (empty micro-
(Malvern) and analyzed using Zetasizer software. The drop- emulsion), 0.25 (6.8 mM), 0.5 (13.9 mM), 0.75 (20.4 mM),
let diameter was calculated from the diffusion coefficient, and 1 (27.1 mM)) was examined using a cyclic voltammeter
using Stokes-Einstein equation [34]. (Electrochemical Analyzer, CH Instruments, Austin, TX,
USA). Samples were placed in a well with three electrodes:
2.3. Cryogenic Transmission Electron Microscopy (Cryo-TEM). a glassy carbon, with a working electrode of 3.3 mm diameter;
Cryo-TEM specimens were prepared in a controlled environ- an Ag/AgCl reference electrode; and a platinum wire as an
ment box using a vitrification robot (Vitrobot). 60 μL of the auxiliary electrode [37]. Potential was applied to the working
microemulsion (the final curcumin concentration was 1% w/ electrode at a constant rate (100 mV/s) receiving cyclic
w (27.1 mM)) was dropped onto a glow-discharged TEM grid voltammogram. Cyclic voltammogram was composed of
(300-mesh Cu Lacey substrate; Ted Pella Ltd.). Excess was two parameters: the peak potential (Ep(a)), which reflects
automatically blotted with a filter paper, and the specimen the ability to donate electrons, and the anodic current (AC),
was rapidly plunged into liquid ethane and transferred to which correlates with the concentrations of these compounds
liquid nitrogen where it was kept until used. Specimens were [38]. Reducing power was determined from the cyclic
analyzed below −175°C using an FEI Tecnai 12G2 TWIN voltammogram. The working electrode was tested prior to
TEM operated at 120 kV in a low-dose mode and with a each series of measurements, by performing a cyclic
few micrometers under focus to increase phase contrast. voltammogram of 1 mm potassium ferricyanide in PBS.
Images were recorded with a Gatan charge-coupled device
camera (model 794) and examined using Digital Micrograph 2.7. Oxygen Radical Absorbance Capacity Assay (ORAC).
software, Version 3.1. ORAC assay adapted to fluorescein labeling [39] was used to
determine the total antioxidant capacity of curcumin-loaded
2.4. Small-Angle X-ray Scattering (SAXS). SAXS experi- microemulsions (the final curcumin concentrations were as
ments were performed on microemulsions without further follows (%w/w): 0 (empty microemulsion), 0.25 (6.8 mM),
Oxidative Medicine and Cellular Longevity 5

0.5 (13.9 mM), 0.75 (20.4 mM), and 1 (27.1 mM)). Analysis was atmosphere of 5% CO2. Cells were subcultured twice weekly
performed using 2, 2′-azobis(2-amidinopropane) dihydrochlor- at a 1 : 10 ratio using a trypsin-EDTA (0.05%) solution (Bio-
ide (AAPH) as a peroxyl generator. This assay is a kinetic assay logical Industries, Beit Haemek, Israel) as a detaching agent.
which measures the loss of fluorescein fluorescence over time
due to peroxyl radical formed by AAPH, enabling evaluation 2.11. Human Skin Organ Culture. Human skin was obtained
of antioxidant protection. Measurements were performed on with informed consent from 20- to 60-year-old healthy
a Fluostar Galaxy plate reader (BMG, Offenburg, Germany) women, who had gone through breast or abdomen reduction.
equilibrated at 37°C, with excitation and emission set up at Testing was performed according to the Declaration of
485 nm and 520 nm, respectively. Trolox was used as a calibra- Helsinki and approved by the Hadassah University Hospital
tion standard. Reagents were prepared in phosphate buffer Ethics Committee, #0639-12-HMO. Skin was cut into pieces
(pH 7.4). 40 μL samples were pipetted into a 96-well plate. of approximately 0.5 × 0.5 cm and cultured, dermal side
Fluorescein was added to a final concentration of 96 nM. down and epidermal side up, in 35 mm diameter petri dishes
ORAC fluorescence was read every 2 min for 70 min. Oxidation containing DMEM (Dulbecco’s Modified Eagle’s Medium,
resulting from peroxyl radical started immediately following Biological Industries, Beit Haemek, Israel) at 37°C, under
AAPH addition. Total antioxidant capacity was calculated by 5% CO2. 4 μL of curcumin-loaded microemulsion (the final
measuring the area below the kinetic curve [39]. curcumin concentration was 1%w/w (27.1 mM)) was applied
to the air-exposed epidermis 24 h before irradiation as
2.8. Quantification of Oxidant-Scavenging Abilities (OSA) by described below. The samples were incubated for another
a Luminol-Dependent Chemiluminescence (LDCL) Assay. A 24 h for apoptosis determination. The epidermis was
highly sensitive luminol-dependent chemiluminescence- separated from the dermis by 1 min heating in phosphate-
inducing cocktail [40] was employed to quantify the OSA of buffered saline (PBS) at 56°C and apoptosis was examined.
microemulsions (the final curcumin concentrations were as
2.12. Dermal Absorption of Curcumin: An Ex Vivo Model
follows (%w/w): 0 (empty microemulsion), 0.25 (6.8 mM),
Using Human Skin Organ Culture. Microemulsion penetra-
0.5 (13.9 mM), 0.75 (20.4 mM), and 1 (27.1 mM)). Briefly, the
tion was investigated using Franz-type diffusion cells
following were added into 850 μL of Hanks’ balanced salt
(PermeGear Inc., Hellertown, PA, USA) with a diffusion area
solution (HBSS) (pH 7.4): 10 μL of luminol (1 mM), H2O2
of 1 cm2 and an acceptor compartment of 8 mL containing
(100 mM), sodium selenite (IV) (2 mM), and CoCl2·6H2O
fetal bovine serum and PBS (pH 7.4) (1 : 9, v/v). Skin was
(II) (1 mM). This cocktail produces an immediate wave
mounted on Franz-type diffusion cells, epidermal side up,
of light due to peroxide and hydroxyl radical. Light
and dermal side facing the receptor compartment. Diffusion
quenching by microemulsions indicates the degree of their
cells were kept at 32°C. 100 μL of different treatments (the
oxidant-scavenging ability.
final curcumin concentration was 1%w/w (27.1 mM)) was
Light quenching was measured as counts per minute by a
applied to the mounted skin. Following 24 h incubation, skin
Lumac 2500 Luminometer (Landgraaf, The Netherlands).
was removed and washed three times using a cotton cloth
containing ethanol and the viable epidermis was separated
2.9. Quantification of Oxidant-Scavenging Abilities (OSA) by
from the dermis. Separation of the full epidermis from the
the 2-diphenylpicrylhydrazyl (DPPH) Radical Assay. Modi-
dermis was achieved by heat shock treatment; skin was
fied DPPH assay [41] was used to determine the oxidant-
placed for 30 seconds at 55–60°C followed by 1 min at 4°C,
scavenging ability of curcumin-loaded microemulsion (the
both in PBS. Curcumin was extracted from the separated
final curcumin concentrations were as follows (%w/w): 0
layers with DMSO. The extraction was performed by
(empty microemulsion), 0.25 (6.8 mM), 0.5 (13.9 mM), 0.75
incubation in a shaker (60 ×g) until all curcumin were
(20.4 mM), and 1 (27.1 mM)). 2,2-diphenylpicrylhydrazyl
released (24 h). Finally, 100 μL from the receptor fluids
(DPPH) free radical was used as a probe; upon reduction, this
was collected. Curcumin existence in skin layers and in
stable, purple, free radical changed its color to a yellow diphe-
the acceptor compartment was determined by measuring
nylpicryl hydrazine. Briefly, 10 μL of microemulsions was
fluorescence excitation at 485/40 nm and emission at
mixed with 20 μL of a DPPH solution (10 mM in absolute
528/20 nm, using a BioTek microplate reader (BioTek
methanol). One minute later, 800 μL of absolute methanol
Instruments Inc., Winooski, VT).
was added. The reaction mixtures were centrifuged at
425 ×g for 2 min, and the change in absorption at 517 nm 2.13. Skin Exposure to UVB Irradiation. Prior to irradiation,
using a Whittaker microplate reader 2001 was determined. culture medium was removed and skin was washed with
Oxidant-scavenging ability is expressed in terms of micro- PBS to remove all traces of treatments. PBS was added to cover
mole equivalents of trolox per 100 grams of sample. the dermis, and the sample was irradiated with a UVB source
(VL-6.M lamp, emission spectrum 280–350 nm, emission
2.10. Cell Culture. Immortalized human keratinocytes, peak 312 nm, filter size 145 × 48 mm; Vilber Lourmat, Torcy,
HaCaT cells [42], were grown in Dulbecco’s Modified Eagle’s France) at 300 mJ/cm2. Immediately following irradiation,
Medium (DMEM, Biological Industries, Beit Haemek, Israel) PBS was replaced by human skin organ culture medium (see
containing 4.5 g/L D-glucose and supplemented with 10% above) and skin was incubated for an additional 24 h.
fetal bovine serum, 1 mM L-glutamine, 100 U/mL penicillin,
and 100 U/mL streptomycin in DMEM. The cultures were 2.14. Apoptosis Determination by Caspase-3 Activity Assay.
maintained in an incubator at 37°C in a humidified The epidermis was incubated in 100 μL PBS containing
6 Oxidative Medicine and Cellular Longevity

2.5 μM Ac-DEVD-AMC as a substrate, with 0.02% Triton X- Microemulsion ingredients also have a pivotal role in the
100 and 10 mM DTT, at 37°C in a 96-well plate [43]. beneficial dermal delivery; surfactants and cosurfactants are
Fluorescence of the released coumarin derivative was often penetration enhancers resulting in the decrease in the
measured at 390/435 nm, using a Fluostar-BMG spectrofluo- diffusional barrier of the stratum corneum [48]. Moreover,
rometer (Offenburg, Germany). Caspase-3 activity was microemulsions may have a beneficial hydration effect on
calculated over 40 min in a linear range from the fluorescence the stratum corneum, influencing permeation ability [49].
versus time slope. Results were normalized relative to the Therefore, o/w microemulsions were designed. Ingredients
control group. were carefully chosen for their biocompatibility and lack of
toxicity, and the usage of alcohol as a cosurfactant due to tox-
2.15. Viability Measurements through Mitochondrial Assay. icity and irritancy issues [50] was denied. Nonionic surfac-
Cytotoxicity of treated cell culture (HaCaT cells) was evaluated tants were selected due to their activity as solubilizing
by the MTT method described elsewhere [44]. Treatments agents and their effects on the skin barrier function [25].
(empty microemulsion, curcumin-loaded microemulsion, The stabilization of the microemulsion was achieved using
and curcumin dissolved in DMSO) according to dilutions in a mixture of surfactants with different HLB values.
increasing curcumin concentrations (0–3 μM) were added to Figure 1(a) demonstrates the ability to form an empty micro-
24-microwell plates containing cell cultures of 30,000 cells/ emulsion formulation. Next, curcumin incorporation in the
mL. After 24 h, cell survival was evaluated by measuring the empty microemulsion formulation without disrupting its
absorbance at 540 nm, using a Whittaker microplate reader phase consistency was tested. As mentioned above, increas-
2001. The percentage of cell survival was normalized relative ing curcumin’s solubility would enhance its dermal delivery.
to the control group. Curcumin which is highly insoluble in water was solubilized
in the microemulsion (Figure 1(b)). Microemulsions demon-
2.16. Keap1-Nrf2-EpRE Pathway Activation. Real-time PCR
strated stability; visual evaluation following accelerated con-
of Nrf2 and enzyme expression after treatments (microemul-
ditions (40 ± 3°C) in darkness for 24 months revealed a
sions, free curcumin, and catalase; Sigma-Aldrich, Israel) were
transparent and isotropic behavior. Microemulsion particle
measured in cell culture. Subconfluent cells were treated and
size was similar to that of the freshly prepared samples with
harvested at the desired times after treatment (see below).
the same monomodal size distribution pattern.
In the case of catalase treatments, catalase (300 U/mL) was
coadministered simultaneously with the other treatments.
3.2. Reduction of Cytotoxicity Using Curcumin-Loaded
Total RNA from cell culture was extracted according to tri-
Microemulsions Compared to Free Curcumin in Keratinocyte.
reagent protocol (Sigma). Reverse transcription was
Curcumin has poor solubility in water, yet good solubility in
performed as previously described [45]. Aliquots of cDNA
dimethyl sulfoxide (DMSO) and chloroform [8]. Due to the
culture were subjected to real-time PCR using PerfeCTa
low aqueous solubility of curcumin, some researchers dissolve
SYBR Green SuperMix, Low ROX (Quanta Biosciences
it in base medium; however, this approach does not address the
Inc.), Stratagene real-time PCR machine, and oligonucleotide
alkaline decomposition of curcumin: degradation products
sets (see oligonucleotide sequence in the Supplementary
including ferulic acid and feruloylmethane [8]. Therefore,
Data). In all cases, the samples were normalized relative to
through all of this study, curcumin dissolved in DMSO was
GAPDH expression.
used as a control. DMSO curcumin solutions are termed in
2.17. Statistical Analysis. Experiments were performed inde- the following as free curcumin.
pendently at least three times. For oxidant-scavenging ability Cytotoxicity of an empty microemulsion, curcumin-
assays, each experiment included three repetitions (n = 3). loaded microemulsion, and free curcumin in DMSO in
For organ culture experiments, experiments were performed immortal human keratinocyte cells (HaCaT) was measured
with three different donors. Each independent experiment by the MTT assay. Figure 2 shows cell viability (%) follow-
included four repetitions, with four skin pieces being proc- ing 24 h treatment. Axis x represents the treatment
essed in parallel. Data were expressed as mean ± standard concentration in the cell culture (%). As can be seen, the
errors of the mean (SEM) or standard deviation of the viability of HaCaT cells exposed to an empty microemul-
mean (STDEV) as specified. Statistical significance of sion at a concentration of 0.11% (v/v) or less was greater
differences was determined using one-way ANOVA, than 80%. Introduction of cells into a microemulsion at
followed by Kruskal-Wallis test. The significance threshold a concentration of 0.15% (v/v) significantly reduced cell
was set at P < 0 05. viability (~50%). This decline in cell viability could be
derived from the surfactants composing the microemul-
3. Results and Discussion sion. It has been shown that all nonionic surfactants are
capable of causing cell damage due to destruction of the
3.1. Design of Curcumin-Loaded Microemulsion. The usage of cell membrane and its solubilization, in a concentration-
microemulsions for dermal delivery offers several advan- dependent manner [51]. In particular, it has been shown
tages. Few mechanisms of activity were suggested in order that Tween 80 can cause cellular damage [52]. Consistent
to elucidate micreomulsion penetration ability. High solubi- with our results (see above), it has also been shown that
lization capacity of the drug in the microemulsion may Tween 80 used at a concentration above 0.03% (v/v)
increase its activity towards the skin by raising the drug gra- reduced cell viability [52]. However, incorporation of cur-
dient across the skin [46] and may favor skin partition [47]. cumin into the microemulsion mitigated this cytotoxicity
Oxidative Medicine and Cellular Longevity 7

(a) (b)

Figure 1: Images of (a) clear empty microemulsion and (b) curcumin-loaded microemulsion.

120
Cell viability by MTT (% of control)

100

80

⁎
60 ⁎

⁎ ⁎
40

20

0
0.02 0.04 0.08 0.11 0.15
Treatment concentration (%)

Curcumin-loaded microemulsion
Curcumin dissolved in DMSO
Empty microemulsion
DMSO

Figure 2: HaCaT cell viability as measured by MTT (in percentage) after 24 h treatment. Cell viability was expressed as the percentage of the
untreated control (dashed line) as the function of treatment concentration. Average values are presented in the figure with standard deviation of
the mean (∗ P < 0 05).

resulting in cell survival. Introducing free curcumin (cur- 3.3. Structural Investigation. One of the main challenges in
cumin dissolved in DMSO) demonstrates cytotoxicity at incorporating curcumin into a microemulsion is avoiding
a concentration of 0.08% (v/v), indicating the curcumin- the interruption of microemulsion structure [53]. Cryo-
loaded microemulsion’s preference in terms of cytotoxicity. TEM micrographs of the empty microemulsion and
8 Oxidative Medicine and Cellular Longevity

Table 2: Diameters (averaged by volume) of empty microemulsion Table 3: Best-fit parameters for the core and shell model (Eq. 1-2),
and microemulsion containing curcumin, as measured by dynamic with 95% confidence bounds of the fit. Rc is the radius of the
light scattering. Standard deviation is specified (N = 6). core, Rs is the radius of the droplet, and σ is the standard deviation
of Rc (N = 4).
Diameter (nm)
Empty microemulsion 6.75 ± 2.9 Empty Microemulsion containing
Parameter
microemulsion curcumin
Microemulsion containing curcumin 9.33 ± 3.6
Rc (nm) 3.85 ± 0.04 3.87 ± 0.07
Rs (nm) 5.39 ± 0.07 5.17 ± 0.04
curcumin-loaded microemulsions are presented in the Shell density
36.34 ± 1.24 48.19 ± 2.08
Supplementary Data (Figure S2). Both microemulsions dem- (el/nm3)
onstrate a level of order of the closely packed droplets with a σ 0.57 ± 0.02 0.61 ± 0.02
diameter of about 10 nm. The dimension of microemulsion’s
spherical droplet was further evaluated using dynamic light
scattering (DLS) measurements. As can be seen in Table 2, depend on its specific microenvironment (e.g., polar and
there is no significant difference between the size of the nonpolar solvents) [8, 36]. Therefore, curcumin can be used
droplets in the empty microemulsion and that in the as a probe and directly monitor the polarity of its surround-
curcumin-loaded microemulsion. Additional nanostructure ings instead of using a probe, pointing out its site in the
information was obtained using small-angle X-ray scattering. microemulsion. The fluorescence curves of curcumin in dif-
SAXS plots are presented in the Supplementary Data (Figures ferent solvents (background of the corresponding solvent
S3), demonstrating a broad peak at q ≈ 0.07 A−1, which was subtracted) are presented in the Supplementary Data
corresponds to a structure with a dimension of about 9 nm (Figure S4). The wavelength of the peak is dependent on
according to Bragg’s law, in agreement with DLS measure- the solvent; the peak in DDW (524 nm) shifts in IPM (463-
ments and cryo-TEM images. Further analysis was done by 464 nm). The peak of curcumin in the microemulsion is at
fitting the core and shell model, which is frequently used to a wavelength of 509 nm, similar to the peak of curcumin in
describe micelles and microemulsions [53]. In the case of Tween 80 (504 nm), suggesting that the microenvironment
oil-in-water microemulsions, the core is the hydrophobic of curcumin is alike in both the microemulsion and Tween
component and the surfactants comprise the shell. Table 3 80 and that curcumin is located in the Tween 80 layer of
reveals the best-fit parameters for the core and shell model the droplets. This is consistent with other studies that show
(see Supplementary Data, Eq. 1-2), with 95% confidence that the drug is in the interface of microemulsion [53]. In
bounds of the fit. The oil component (IPM) in the microe- addition, this data are also in agreement with SAXS data
mulsion has a strong influence on the microemulsion’s presented in Table 3 supporting curcumin’s location in the
formation and stability (data is not shown). Thus, density surfactant shell.
of the core was calculated from the oil properties (IPM)
and kept constant. The shell in this model is composed
from the surfactant mixture. The incorporation of curcu- 3.4. Curcumin-Loaded Microemulsions Maintain Oxidant-
min into the microemulsion might affect the shell density Scavenging Ability In Vitro. Maintaining the oxidant-
of the microemulsion. Data in Table 3 demonstrate no scavenging ability of curcumin loaded in microemulsions is
significant difference in the core and shell radius between crucial for its utilization. Therefore, antioxidant capacity
the empty microemulsion and the microemulsion containing was evaluated by a variety of methods on curcumin-loaded
curcumin. However, the shell density of the curcumin-loaded microemulsions in five increasing curcumin concentrations
microemulsion is higher than the shell density of the (%w/w): 0 (empty microemulsion), 0.25 (6.8 mM), 0.5
empty microemulsion, hinting at curcumin’s location in (13.9 mM), 0.75 (20.4 mM), and 1 (27.1 mM). Oxygen radical
the microemulsion. absorbance capacity (ORAC) assay measures the degree of
The complete model used in this study was the core and inhibition of peroxyl radical-induced oxidation by the com-
shell, with a normal size distribution of the core. Small pounds of interest, expressed in trolox equivalents (y-axis).
deviations of the model from the experimental data could Figure 3(a) demonstrates the protection of the curcumin-
originate from the droplet not being ideally spherical or from loaded microemulsions against the free radical. As can be
the nature of the shell, which is not constant in density due to seen, the microemulsion with an increased curcumin
radial concentration gradients. concentration demonstrates a linear trend with ORAC values
Curcumin’s location inside the microemulsion and its expressed in trolox equivalents (ORAC = 17.451c − 3.9076,
interaction with the other ingredients seem to be of major where c is the curcumin concentration in mM, coefficient of
importance since its mobility in the vehicle can be affected determination (R2 = 0.99)). Free curcumin (curcumin dis-
and may influence its delivery [54]. Therefore, the location solved in DMSO) in increasing concentrations also demon-
of curcumin within the microemulsion was examined using strates a linear trend with ORAC value (ORAC = 19.463c
the fluorescent probe method. This method can detect the − 16.702, R2 = 0.99). Thus, curcumin-loaded microemulsions
microenvironment near a substance and is commonly used demonstrate improved protection against peroxyl radicals
for revealing phase changes and structure of microemulsions relative to trolox (~17.5 times more). Similar behavior is
and micelles [35]. Curcumin’s emission properties highly observed for free curcumin (~19.5).
Oxidative Medicine and Cellular Longevity 9

The LDCL assay is based on the ability of an antioxidant homeostasis maintenance is the transcription factor Nrf2, a
agent to quench the luminescence generated by a “cocktail of central key target for skin protection and cancer prevention
oxidants.” Figure 3(b) shows that while the luminescence [31]. As mentioned above, curcumin is capable of activating
induced by the “cocktail” is kept steady at a high level for the Keap1-Nrf2-EpRE pathway [9]. Therefore, the effects of
2.5 min, the addition of curcumin-loaded microemulsions microemulsions on the activation of the Keap1-Nrf2-EpRE
dramatically affects the luminescence observed. The sharp pathway were examined using real-time PCR. The mRNA
and steady decline in luminescence due to the consumption expression of a few phase II enzymes was examined: (catalase
of the bulk of oxidants generated yields a curve of light (EC 1.11.1.6), glutathione S-transferase (EC 2.5.1.18), super-
emission. Figure 3(d) demonstrates similar behavior for free oxide dismutase (EC 1.15.1.1), glutathione reductase (EC
curcumin (curcumin dissolved in DMSO). From calculating 1.8.1.7), NAD(P)H dehydrogenase [quinone] 1 (EC 1.6.5.2),
the area under the curve for a curcumin-loaded microemul- and glutamate-cysteine ligase (EC 6.3.2.2) [5]. Although the
sion and free curcumin, it can be concluded that the relative mRNA expression of most of these enzymes was
scavenging ability of a curcumin-loaded microemulsion is not significantly affected, the relative mRNA expression
not significantly different than that of free curcumin. of HO-1, a known phase II enzyme, was significantly
Using cyclic voltammetry, the oxidation potentials of a induced. HO-1 regulates the level of intracellular heme
curcumin-loaded microemulsion and free curcumin were by catalyzing the oxidative degradation of heme to biliverdin,
measured. Two oxidation potentials were observed corre- iron, and carbon monoxide, resulting in cytoprotective,
sponding to two electron-donating centers in the curcumin antiapoptotic, and anti-inflammatory effects on various
molecule. Table 4 summarizes the oxidation potentials of a experimental models [57]. HO-1 levels are associated with
curcumin-loaded microemulsion and free curcumin. the proliferating epidermis [58].
Figure 3(c) shows the anodic current at oxidation potential Figure 4 demonstrates relative mRNA expression of
of 407 mV and 473 mV for free curcumin and for HO-1 6, 12, and 24 h after treatments. As can be seen,
curcumin-loaded microemulsions, respectively. As expected, the most significant relative mRNA expression increase
an increase in curcumin concentration resulted in an occurs following 6 h treatment with a microemulsion
increased anodic current both for the curcumin-loaded containing curcumin. Empty microemulsion and free cur-
microemulsions and for free curcumin. The anodic current cumin also exhibit activation of the Keap1-Nrf2-EpRE
drop for a curcumin-loaded microemulsion in the highest pathway. As can be seen, following 6 h treatment, microe-
curcumin concentration might be explained by other mulsion containing curcumin has a synergistic effect. The
processes involved apart from curcumin diffusion (e.g., observation that the empty micreomulsion is capable of
interaction with surfactants and oils). This observation is activating the Keap1-Nrf2-EpRE pathway can be explained
consistent with other works [55]. by the nanodroplets composing it. It was shown that fibers
The DPPH assay measures the hydrogen atom (or one and particles may activate the Keap1-Nrf2-EpRE pathway
electron)-donating activity and hence evaluates the antioxi- via production of reactive oxygen species [43], and we
dant activity due to free radical scavenging. Expressed as assume that the microemulsion nanodroplets operate
trolox equivalents, Figure 3(e) shows that free curcumin similarly. Alternatively, the oxidation status of the micreo-
and the curcumin-loaded microemulsions display similar mulsion might generate reactive oxygen species capable of
antioxidant activity. activating the pathway.
The redox assay presented here indicates that the oxygen- Overall, our results demonstrate the advantage of
scavenging ability of curcumin-loaded microemulsions is curcumin-loaded microemulsions over free curcumin.
similar to that of free curcumin. However, taking into Microemulsion containing curcumin enhanced the Keap1-
consideration that curcumin dissolved in DMSO showed Nrf2-EpRE pathway in an epidermal cell culture with a
cytotoxicity (even in low concentrations), curcumin-loaded 180% increase over free curcumin. It is worth noting that
microemulsions may provide improved protection against tert-butylhydroquinone (tBHQ), a synthetic electrophile
free radicals without raising cytotoxicity issues. These known for its ability in activating the Keap1-Nrf2-EpRE
experiments demonstrate preservation of curcumin phenolic pathway in epidermal cell culture [59], induced the relative
group’s activity. Thus, curcumin-loaded microemulsions mRNA expression of HO-1 following 6 h treatment
can scavenge directly and potently oxygen-centered (50 μM) similar to the curcumin-loaded microemulsions
reactive intermediates. (3.5 ± 0.7 fold change). Treatment with DMSO had no effect.

3.5. Microemulsions Containing Curcumin Enhance the 3.6. Microemulsion Containing Curcumin Induced Activation
Activation of the Keap1-Nrf2-EpRE Pathway in Keratinocyte. of the Keap1-Nrf2-EpRE Pathway in Keratinocyte: Mechanism
Cellular redox homeostasis guarantees a suitable cell of Action. Polyphenols in general and curcumin in particular,
response to a variety of exogenous or endogenous stimuli under in vitro conditions, in the presence of oxygen and metal
[56]. Upon disrupting this gentle balance, reactive oxygen ions, may exhibit pro-oxidant activity [60]. Polyphenols
species which can activate proliferative and cell-survival sig- can undergo autoxidation involving oxygen consumption
naling [56] can alter apoptotic pathways that may be involved generating O2·−, hydrogen peroxide (H2O2), semiquinones,
in the pathogenesis of a number of skin disorders including and quinones [61]. Hydrogen peroxide (H2O2) production
photosensitive diseases and some types of cutaneous malig- by polyphenols in culture media was well demonstrated
nancy [15]. One of the central players involved in the redox [61]. H2O2 is an important mild oxidant capable of
10 Oxidative Medicine and Cellular Longevity

600

ORAC value (equivalent trolox, mM)

400

200

0
0 5 10 15 20 25 30
Curcumin concentration (mM)
Curcumin in DMSO
Microemulsion containing curcumin
(a)
2.50 E+05

2.00 E+05
LDCL (cpm)

1.50 E+05

1.00 E+05

5.00 E+04

0.00 E+00
0 20 40 60 80 100 120 140 160
Time (sec)
Me 27.1 mM curcumin Me 6.8 mM curcumin
Me 20.4 mM curcumin Empty Me
Me 13.9 mM curcumin

2.50 E+05

2.00 E+05
LDCL (cpm)

1.50 E+05

1.00 E+05

5.00 E+04

0.00 E+00
0 20 40 60 80 100 120 140 160
Time (sec)
DMSO 27.1 mM curcumin DMSO 6.5 mM curcumin
DMSO 20.4 mM curcumin DMSO
DMSO 13.9 mM curcumin
(b)

Figure 3: Continued.
Oxidative Medicine and Cellular Longevity 11

‒12

‒10

‒8

Ip (
휇A) ‒6

‒4

‒2
0 5 10 15 20 25 30
0
Curcumin concentration (mM)

A nodic peak current (microemulsion containing curcumin)

A nodic peak current (curcumin in DMSO)
(c)
50
Antioxidant activity (equivalent trolox, mM)

45
40
35
30
25
20
15
10
5
0
0 5 10 15 20 25 30
Curcumin concentration (mM)
Microemulsion containing curcumin
Curcumin in DMSO
(d)

Figure 3: (a) ORAC value (equivalent trolox, mM) versus curcumin concentration for (■) curcumin-loaded microemulsion and (♦)
curcumin dissolved in DMSO, (b) LDCL generated by “cocktail,” for curcumin-loaded microemulsion (b) and curcumin dissolved in
DMSO (d) containing in increasing concentration (-) 27.1 mM curcumin, (♦) 20.4 mM curcumin, (▲) 13.9 mM curcumin, (×) 6.8 mM
curcumin, and (■) microemulsion without curcumin (microemulsion and DMSO, respectively). Statistical analysis indicated significantly
higher antioxidant activity indicated by a decrease in LDCL (P < 0 01). (c) Typical anodic peak as measured by cyclic voltammetry for (♦)
curcumin dissolved in DMSO (1st oxidation potential, 407 mV) and (■) curcumin-loaded microemulsion (1st oxidation potential,
473 mV) in increasing curcumin concentrations. (e) Antioxidant activity of (♦) curcumin dissolved in DMSO and (■) curcumin-loaded
microemulsion as measured by the DPPH assay and expressed in trolox equivalents.

Table 4: Oxidation potential of microemulsion containing curcumin reacting with cysteines and therefore is capable of inducing
and curcumin in DMSO is measured by cyclic voltammetry. several transcription factors involved in cell replication,
regulation of metabolism, apoptosis, and necrosis [62]. It
1st oxidation 2nd oxidation is worth noting that H2O2 is electronically neutral and
potential (mV) potential (mV)
can freely diffuse through cellular membranes [63].
Microemulsion An important question regarding the mechanism of
473 701
containing curcumin activity by which the curcumin-loaded microemulsion
Curcumin dissolved operates in keratinocyte is whether curcumin’s phenolic
407 702
in DMSO groups preserve their activity following incorporation into
microemulsions and moreover whether curcumin retains its
pro-oxidative activity. It has been shown that curcumin
12 Oxidative Medicine and Cellular Longevity

4.5

4.0 ⁎⁎

Relative mRNA expression (HO‒1)

3.5

3.0 ⁎

2.5

2.0

1.5

1.0

0.5

0.0
6 hours 12 hours 24 hours

Control (untreated)
Empty microemulsion
Curcumin-loaded microemulsion
Curcumin dissolved in DMSO

Figure 4: Activation of the Keap1-Nrf2-EpRE pathway following 6 h, 12 h, and 24 h treatments. mRNA expression determined by real-time
PCR. GAPDH mRNA expression was used for normalization, and the basal mRNA normalized expression was considered 1. Average values
are presented in the figure with standard deviation of the mean (∗ P < 0 05, ∗∗ P < 0 01).

4.0 ⁎
Relative mRNA expression (HO-1)

3.0 ⁎

⁎
2.0 ⁎

1.0

⁎⁎
0.0
Control Control + Cat Empty Empty ME cont. Cur ME cont. Cur + Cur in DMSO Cur in DMSO +
microemulsion microemulsion + Cat Cat
Cat
Treatments

Figure 5: Activation of the Keap1-Nrf2-EpRE pathway following treatment with microemulsions (empty ME or ME containing curcumin)
and free curcumin in the absence or presence of catalase (Cat, 300 U/mL). mRNA expression determined by real-time PCR. GAPDH mRNA
expression was used for normalization, and the basal mRNA normalized expression was considered 1. Average values are presented in the
figure with standard deviation of the mean (∗ P < 0 05, ∗∗ P < 0 01).

generated extracellular H2O2 in cell growth medium during introduction of catalase (300 U/mL) to microemulsions
autoxidation [60]. Therefore, it can be speculated that activa- decreased the relative mRNA expression of HO-1 indicating
tion of the Keap1-NRF2-EpRE system is partially mediated lower activation of the Keap1-Nrf2-EpRE pathway. Catalase
by extracellular H2O2 production by curcumin [9]. In order addition to the empty microemulsion decreased the relative
to test this hypothesis, microemulsions and free curcumin were mRNA expression of HO-1 from ~1.75-fold to ~0.95-fold,
applied to keratinocyte in the presence of the enzyme catalase. similar to that to the control group. This decrease
H2O2 is decomposed by catalase to water and oxygen [60]. demonstrated H2O2 production by the microemulsion and
Therefore, H2O2 involvement is expected to be abrogated involvement in activation of the Keap1-Nrf2-EpRE
following catalase addition. As can be seen in Figure 5, pathway under experimental conditions. Curcumin-loaded
Oxidative Medicine and Cellular Longevity 13

microemulsions increased the relative mRNA expression of curcumin-loaded microemulsion was applied to human skin
HO-1 by ~3.65-fold. As can be seen, following catalase organ culture in Franz-type diffusion cells in order to
addition, the induction was lowered to ~2.57-fold indicat- perform and evaluate microemulsion’s penetration ability.
ing that ~30% of this microemulsion’s activity resulted Curcumin was analyzed separately in the epidermis and
from H2O2 involvement and the other ~70% is related dermis. Figure 6 demonstrates extracted curcumin (μg/cm2)
to microemulsion penetration ability and curcumin’s from the epidermis (Figure 6(a) and dermis (Figure 6(b)) of
pro-oxidative activity. human skin following topical applications. As can be seen
It can be speculated that a few factors contribute to HO-1 in Figure 6(a), a significant elevation in curcumin compared
induction: microemulsion’s skin penetration, H2O2 involve- to the control group (untreated skin) was observed only for
ment in curcumin phenolic groups, and the pro-oxidant application of free curcumin. However, Figure 6(b), which
activity of curcumin. Although we realize that this summa- demonstrates curcumin’s quantity in the human skin dermis,
tion is not perfectly accurate, it is interesting to evaluate reveals that curcumin-loaded microemulsions and free cur-
and quantify the significance of each of these contributing cumin (curcumin dissolved in DMSO) both penetrated the
factors. Therefore, an assumption that the different contribu- dermis by a significantly similar and elevated quantity com-
tions are additives was made and a rough estimation was pared to the control group (untreated skin). The observation
obtained. Free curcumin treatment induced relative mRNA that free curcumin was found in the epidermis is consistent
expression of HO-1 to increase by ~1.9-fold; following with DMSO’s skin adsorption enhancement properties [68].
catalase addition, the induction decreased into ~1.65-fold, DMSO, a polar and aprotic molecule, is one of the most
indicating ~12.2% H2O2 involvement and ~87.8% curcumin efficient transdermal delivery agents [69]. However, due to
pro-oxidant activity. By comparing the free curcumin follow- its side effects (including erythema, scaling, and contact
ing catalase addition and the microemulsion containing urticaria) and its potential toxicity, DMSO is rarely used as
curcumin following catalase addition, H2O2 extracellular a transdermal delivery agent [69]. The ability of microemul-
involvement is eliminated and emphasizes the microemul- sions to penetrate skin may be attributed to the use of pene-
sion’s contribution to the means of penetration. Both tration enhancers in the formulation, for example, isopropyl
treatments resulted in the induction of relative mRNA myristate, Tween 80, and Span 20 [25]. The observation that
expression of HO-1 by ~1.65- and ~2.57-fold, respectively, curcumin-loaded microemulsions penetrated the skin and
caused by the penetration ability and curcumin’s pro- reached the dermis in a similar quantity as free curcumin
oxidative activity. Since curcumin’s pro-oxidative activity is without any cytotoxicity highlights microemulsion superior-
exactly the same, it can be concluded that the penetration ity. It is worth mentioning that a similar level of fluorescence
ability of the microemulsion in comparison to that in DMSO was observed in the untreated skin (control group) and in the
is higher by ~156% in keratinocyte. Another interesting empty microemulsion and DMSO treatments, which can be
result is the sharp decrease in relative mRNA expression explicated by the basal levels of skin autofluorescence [44].
of HO-1 in the control group with catalase addition,
indicating H2O2 involvement in the basal state. It was 3.8. Reduction of UVB Cytotoxicity Using Curcumin-Loaded
shown that reactive oxygen species is formed in the Microemulsions in Human Skin Organ Culture. Skin expo-
cellular medium [64]; catalase addition can deplete reactive sure to environmental stressors (e.g., UVB) may cause injury
oxygen species production and therefore decrease phase II to epidermal cells through enhanced production of reactive
detoxification enzyme expression. oxygen species, thus leading to a variety of skin pathologies
[15]. One of the approaches that was suggested to enable skin
3.7. Evaluation of Dermal Absorption of Curcumin-Loaded protection was the use of various nontoxic antioxidants
Microemulsions in an Ex Vivo Model Using Human Skin which displayed efficacy in cell culture systems and animal
Organ Culture. The skin constitutes a barrier between the models [15]. However, absolute efficacy in humans was not
body and the environment [15]. It preserves homeostasis by well demonstrated [15]. Topical application of Keap1-Nrf2-
avoiding water loss via evaporation and protects against the EpRE-inducing agents may present a protective strategy to
environment by preventing penetration of exogenous reduce UVB-induced skin injury. Indeed, it has been shown
substances [15]. Skin layers which are continuously renewed that UVB-induced damage to skin cells can be efficiently
enable efficient protection against the penetration of external limited by Keap1-Nrf2-EpRE-inducing agents [6]. A
substances, especially thanks to the stratum corneum [65]. curcumin-loaded microemulsion was applied to human skin
The outermost stratum corneum layer, despite its thickness organ culture in order to perform and evaluate the microemul-
of only 15–20 μm [25], regulates the barrier properties of sion’s ability to impede UVB-induced cell toxicity in the
the skin by regulating the fluxes of chemicals and water epidermis via Keap1-Nrf2-EpRE activation. Elevated HO-1
between the environment and the organism [66]. Moreover, levels following 24 h incubation with the curcumin-loaded
the hermetic barrier of the stratum corneum makes topical microemulsion were observed using immunohistochemical
application challenging in spite of the large available staining (indication for Keap1-Nrf2-EpRE pathway activa-
surface area, relative low enzymatic degradation, and long tion) as presented in the Supplementary Data (Figure S5).
application time [67]. Following treatment and 24 h incubation, skin was irradiated
A prerequisite for the success of a dermatological drug, and apoptosis was then monitored by caspase-3 activity
primarily, is its ability to penetrate through or into the skin assay. UVB irradiation caused a ~17-fold increase in
in sufficient quantities to achieve the desired effect. A caspase-3 activity indicating an increase in epidermal cell
14 Oxidative Medicine and Cellular Longevity

1800
1600
1400
⁎⁎⁎

Curcumin (
휇g/cm2)
1200
1000
800
600
400
200
0
Treatments
Control (untreated) Curcumin dissolved in DMSO
Empty microemulsion DMSO
Curcumin-loaded microemulsion
(a)
1800
⁎⁎⁎ ⁎⁎⁎
1600
1400
Curcumin (
휇g/cm2)

1200
1000
800
600
400
200
0
Treatments

Control (untreated) Curcumin dissolved in DMSO

Empty microemulsion DMSO
Curcumin-loaded microemulsion
(b)

Figure 6: Dermal absorption evaluation of curcumin-loaded microemulsion in (a) the human skin epidermis and in (b) human skin dermis.
Average values are presented in the figure with standard deviation of the mean (∗∗∗ P < 0 001).
Epidermal cell apoptosis by caspase-3 activity (fold

90 ⁎
change relative to nonirradiated control)

80
70
60
50
40
30
20
10 ⁎
0
Control Empty ME ME cont. Cur Cur in DMSO DMSO

Figure 7: Organ cultures were treated with microemulsions for 24 h and then irradiated with UVB at 300 mJ/cm2, and cell apoptosis was
evaluated by caspase-3 activity assay 24 h after irradiation. Data were normalized on the basis of untreated (control), nonirradiated skin
(equal to 1). Average values are presented in the figure with standard deviation of the mean (∗ P < 0 05).
Oxidative Medicine and Cellular Longevity 15

apoptosis. On the other hand, the prior application of a Acknowledgments

curcumin-loaded microemulsion reversed this trend, with
only ~2.7-fold increase in caspase-3 activity in the same The authors thank Yael Levi-Kalisman for her assistance in
conditions (Figure 7). The previous application of free curcu- obtaining cryo-TEM images. This work was funded by the
min or an empty microemulsion did not affect caspase-3 David and Ines Myers Fund of Cleveland, Ohio, USA.
activity significantly. However, DMSO, as expected, increased
capsase-3 activity by ~78-fold. These results demonstrate an
intense effect of a curcumin-loaded microemulsion to restrain References
UV-induced cytotoxicity in epidermal cells. [1] R. Rn and C. Bosco, “Oxidative stress and protective effects of
It is worth noting that curcumin’s preventive effect polyphenols: comparative studies in human and rodent kidney.
against UVB-induced damage in skin might also be the A review,” Comparative Biochemistry and Physiology Part C:
consequence of molecular events such as downregulation of Toxicology & Pharmacology, vol. 142, no. 3, pp. 317–327, 2006.
cell proliferative controls, involving thymine dimer, apopto- [2] S.-K. Myung, W. Ju, B. Cho et al., “Efficacy of vitamin and anti-
sis, transcription factor NF-κB, and inflammatory responses oxidant supplements in prevention of cardiovascular disease:
or upregulation of p53, and these different contributions systematic review and meta-analysis of randomised controlled
need to be further revealed [70]. trials,” British Medical Journal, vol. 346, p. f10, 2013.
[3] G. Tsuji, M. Takahara, H. Uchi et al., “Identification of
4. Conclusions ketoconazole as an AHR-NRF2 activator in cultured human
keratinocytes: the basis of its anti-inflammatory effect,”
The work presented in this study supports the usage of Journal of Investigative Dermatology, vol. 132, no. 1,
curcumin-loaded microemulsions for treating oxidative pp. 59–68, 2012.
stress-related conditions in skin. The incorporation of cur- [4] B. M. Hybertson, B. Gao, S. K. Bose, and J. M. McCord,
cumin in a microemulsion, from a structural point of view, “Oxidative stress in health and disease: the therapeutic poten-
tial of Nrf2 activation,” Molecular Aspects of Medicine, vol. 32,
resulted in a stable nanometric-size microemulsion composed
no. 4, pp. 234–246, 2011.
of core and shell droplets. Curcumin-loaded microemulsions
[5] G. P. Sykiotis and D. Bohmann, “Stress-activated cap’n’collar
maintained curcumin’s activity as a reactive oxygen species
transcription factors in aging and human disease,” Science
scavenger. Moreover, curcumin-loaded microemulsions Signaling, vol. 3, no. 112, p. re3, 2010.
enabled an efficient Keap1-Nrf2-EpRE pathway activation.
[6] M. Schafer and S. Werner, “NRF2-a regulator of keratinocyte
Curcumin-loaded microemulsions promoted a powerful redox signaling,” Free Radical Biology and Medicine, vol. 88,
effect on the reduction of UV-induced cytotoxicity in Part B, pp. 243–252, 2015.
epidermal cells. This work provided insights regarding [7] A. Goel and B. B. Aggarwal, “Curcumin, the golden spice from
the mechanism of activity in which curcumin-loaded Indian saffron, is a chemosensitizer and radiosensitizer for
microemulsions operate and thus supports our suggested tumors and chemoprotector and radioprotector for normal
strategy for ameliorating skin injuries and damages. organs,” Nutrition and Cancer, vol. 62, no. 7, pp. 919–930, 2010.
[8] M. Heger, R. F. van Golen, M. Broekgaarden, and M. C.
Abbreviations Michel, “The molecular basis for the pharmacokinetics and
pharmacodynamics of curcumin and its metabolites in
ME: Microemulsion relation to cancer,” Pharmacological Reviews, vol. 66, no. 1,
SOD1: Superoxide dismutase pp. 222–307, 2013.
Nrf2: Nuclear factor (erythroid-derived 2)-like 2, [9] A. T. Dinkova-Kostova and P. Talalay, “Direct and indirect
NF-E2-related factor antioxidant properties of inducers of cytoprotective proteins,”
EpRE: Electrophile response element Molecular Nutrition & Food Research, vol. 52, Supplement 1,
HO-1: Heme oxygenase-1 or haem oxygenase-1 pp. S128–SS38, 2008.
GAPDH: Glyceraldehyde 3-phosphate dehydrogenase [10] D. Akbik, M. Ghadiri, W. Chrzanowski, and R. Rohanizadeh,
NQO1: NAD(P)H dehydrogenase [quinone] 1 “Curcumin as a wound healing agent,” Life Sciences, vol. 116,
no. 1, pp. 1–7, 2014.
ROS: Reactive oxygen species
tBHQ: tert-Butylhydroquinone. [11] J. L. Arbiser, N. Klauber, R. Rohan et al., “Curcumin is an
in vivo inhibitor of angiogenesis,” Molecular Medicine, vol. 4,
no. 6, p. 376–383, 1998.
Conflicts of Interest [12] S. S. Bhandarkar and J. L. Arbiser, Curcumin as an Inhibitor of
Angiogenesis. The Molecular Targets and Therapeutic Uses of
The authors report no conflicts of interest. Ron Kohen is Curcumin in Health and Disease, pp. 185–195, Springer,
affiliated with the Bloom Center of Pharmacy and is the Boston, MA, 2007.
incumbent of the Richard and Jean Zarbin Chair in Medical [13] T. P. Robinson, R. B. Hubbard, T. J. Ehlers, J. L. Arbiser, D. J.
Studies at the Hebrew University of Jerusalem. Goldsmith, and J. P. Bowen, “Synthesis and biological evalua-
tion of aromatic enones related to curcumin,” Bioorganic &
Authors’ Contributions Medicinal Chemistry, vol. 13, no. 12, pp. 4007–4013, 2005.
[14] R. L. Thangapazham, S. Sharad, and R. K. Maheshwari, “Skin
The authors alone are responsible for the content and writing regenerative potentials of curcumin,” BioFactors, vol. 39,
of the article. no. 1, pp. 141–149, 2013.
16 Oxidative Medicine and Cellular Longevity

[15] D. R. Bickers and M. Athar, “Oxidative stress in the pathogen- [30] A. Teichmann, S. Heuschkel, U. Jacobi et al., “Comparison of
esis of skin disease,” Journal of Investigative Dermatology, stratum corneum penetration and localization of a lipophilic
vol. 126, no. 12, pp. 2565–2575, 2006. model drug applied in an o/w microemulsion and an
[16] G. Kuttan, K. B. H. Kumar, C. Guruvayoorappan, and R. amphiphilic cream,” European Journal of Pharmaceutics and
Kuttan, Antitumor, Anti-Invasion, and Antimetastatic Effects Biopharmaceutics, vol. 67, no. 3, pp. 699–706, 2007.
of Curcumin. The Molecular Targets and Therapeutic Uses of [31] M. Schafer, H. Farwanah, A. H. Willrodt et al., “Nrf2 links
Curcumin in Health and Disease, pp. 173–184, Springer, epidermal barrier function with antioxidant defense,”
Boston, MA, 2007. EMBO Molecular Medicine, vol. 4, no. 5, pp. 364–379,
[17] V. P. Menon and A. R. Sudheer, Antioxidant and Anti- 2012.
Inflammatory Properties of Curcumin. The Molecular Targets [32] T. A. Beyer, U. Auf dem Keller, S. Braun, M. Schafer, and S.
and Therapeutic Uses of Curcumin in Health and Disease, Werner, “Roles and mechanisms of action of the Nrf2
pp. 105–125, Springer, Boston, MA, 2007. transcription factor in skin morphogenesis, wound repair
[18] K. Song, S. Peng, Z. Sun, H. Li, and R. Yang, “Curcumin sup- and skin cancer,” Cell Death & Differentiation, vol. 14, no. 7,
presses TGF-β signaling by inhibition of TGIF degradation pp. 1250–1254, 2007.
in scleroderma fibroblasts,” Biochemical and Biophysical [33] M. Ben Yehuda Greenwald, M. Frusic-Zlotkin, Y. Soroka et al.,
Research Communications, vol. 411, no. 4, pp. 821–825, 2011. “Nitroxide delivery system for Nrf2 activation and skin
[19] R. L. Thangapazham, A. Sharma, and R. K. Maheshwari, protection,” European Journal of Pharmaceutics and Biophar-
Beneficial Role of Curcumin in Skin Diseases. The Molecular maceutics, vol. 94, pp. 123–134, 2015.
Targets and Therapeutic Uses of Curcumin in Health and [34] A. Shukla and R. H. H. Neubert, “Diffusion behavior of
Disease, pp. 343–357, Springer, Boston, MA, 2007. pharmaceutical O/W microemulsions studied by dynamic
[20] P. Anand, A. B. Kunnumakkara, R. A. Newman, and B. B. light scattering,” Colloid and Polymer Science, vol. 284, no. 5,
Aggarwal, “Bioavailability of curcumin: problems and prom- pp. 568–573, 2006.
ises,” Molecular Pharmaceutics, vol. 4, no. 6, pp. 807–818, 2007. [35] G. B. Behera, B. K. Mishra, P. K. Behera, and M. Panda,
[21] K. M. Nelson, J. L. Dahlin, J. Bisson, J. Graham, G. F. Pauli, and “Fluorescent probes for structural and distance effect studies
M. A. Walters, “The essential medicinal chemistry of curcu- in micelles, reversed micelles and microemulsions,”
min: miniperspective,” Journal of Medicinal Chemistry, Advances in Colloid and Interface Science, vol. 82, no. 1,
vol. 60, no. 5, pp. 1620–1637, 2017. pp. 1–42, 1999.
[22] M. Mouslmani, J. M. Rosenholm, N. Prabhakar, M. [36] D. Patra and C. Barakat, “Synchronous fluorescence spectro-
Peurla, E. Baydoun, and D. Patra, “Curcumin associated scopic study of solvatochromic curcumin dye,” Spectrochimica
poly (allylamine hydrochloride)-phosphate self-assembled Acta Part a: Molecular and Biomolecular Spectroscopy, vol. 79,
hierarchically ordered nanocapsules: size dependent inves- no. 5, pp. 1034–1041, 2011.
tigation on release and DPPH scavenging activity of [37] Y. Abramov, J. G. Schenker, A. Lewin, S. Friedler, B. Nisman,
curcumin,” RSC Advances, vol. 5, no. 24, pp. 18740– and V. Barak, “Endocrinology: plasma inflammatory cytokines
18750, 2015. correlate to the ovarian hyperstimulation syndrome,” Human
[23] D. Patra and F. Sleem, “A new method for pH triggered Reproduction, vol. 11, no. 7, pp. 1381–1386, 1996.
curcumin release by applying poly (l-lysine) mediated nano- [38] R. Kohen and I. Gati, “Skin low molecular weight antioxidants
particle-congregation,” Analytica Chimica Acta, vol. 795, and their role in aging and in oxidative stress,” Toxicology,
pp. 60–68, 2013. vol. 148, no. 2, pp. 149–157, 2000.
[24] R. H. H. Neubert, “Potentials of new nanocarriers for dermal [39] J. R. Cronin, “The biochemistry of alternative medicine:
and transdermal drug delivery,” European Journal of Pharma- comparing antioxidant values with the ORAC method,”
ceutics and Biopharmaceutics, vol. 77, no. 1, pp. 1–2, 2011. Alternative & Complementary Therapies, vol. 10, no. 3,
[25] A. Kogan and N. Garti, “Microemulsions as transdermal drug pp. 167–170, 2004.
delivery vehicles,” Advances in Colloid and Interface Science, [40] I. Ginsburg, M. Sadovnic, M. Oron, and R. Kohen, “Novel
vol. 123, pp. 369–385, 2006. chemiluminescence-inducing cocktails, part II: measurement
[26] C. H. Liu and H. Y. Huang, “Antimicrobial activity of of the anti-oxidant capacity of vitamins, thiols, body fluids,
curcumin-loaded myristic acid microemulsions against Staph- alcoholic beverages and edible oils,” Inflammopharmacology,
ylococcus epidermidis,” Chemical and Pharmaceutical Bulletin, vol. 12, no. 4, pp. 305–320, 2004.
vol. 60, no. 9, pp. 1118–1124, 2012. [41] M. S. Blois, “Antioxidant determinations by the use of a sta-
[27] R. Yutani, S. Y. Morita, R. Teraoka, and S. Kitagawa, “Distribu- ble free radical,” Nature, vol. 181, no. 4617, pp. 1199–1200,
tion of polyphenols and a surfactant component in skin during 1958.
Aerosol OT microemulsion-enhanced intradermal delivery,” [42] P. Boukamp, R. T. Petrussevska, D. Breitkreutz, J. Hornung, A.
Chemical and Pharmaceutical Bulletin, vol. 60, no. 8, Markham, and N. E. Fusenig, “Normal keratinization in a
pp. 989–994, 2012. spontaneously immortalized aneuploid human keratinocyte
[28] C. H. Liu and F. Y. Chang, “Development and characterization cell line,” The Journal of Cell Biology, vol. 106, no. 3,
of eucalyptol microemulsions for topic delivery of curcumin,” pp. 761–771, 1988.
Chemical and Pharmaceutical Bulletin, vol. 59, no. 2, pp. 172– [43] M. Frusic-Zlotkin, R. Pergamentz, B. Michel et al., “The
178, 2011. interaction of pemphigus autoimmunoglobulins with epider-
[29] C. C. Lin, H. Y. Lin, H. C. Chen, M. W. Yu, and M. H. Lee, mal cells: activation of the fas apoptotic pathway and the use
“Stability and characterisation of phospholipid-based of caspase activity for pathogenicity tests of pemphigus
curcumin-encapsulated microemulsions,” Food Chemistry, patients,” Annals of the New York Academy of Sciences,
vol. 116, no. 4, pp. 923–928, 2009. vol. 1050, no. 1, pp. 371–379, 2005.
Oxidative Medicine and Cellular Longevity 17

[44] M. Portugal-Cohen, Y. Soroka, M. Frusic-Zlotkin et al., “Skin quinones,” Free Radical Biology and Medicine, vol. 51, no. 12,
organ culture as a model to study oxidative stress, inflamma- pp. 2319–2327, 2011.
tion and structural alterations associated with UVB10 induced [61] L. H. Long, M. V. Clement, and B. Halliwell, “Artifacts in
photodamage,” Experimental Dermatology, vol. 20, no. 9, cell culture: rapid generation of hydrogen peroxide on
pp. 749–755, 2011. addition of (−)-epigallocatechin, (−)-epigallocatechin gallate,
[45] J. Sambrook DWR and D. W. Russell, Molecular Cloning: A (+)-catechin, and quercetin to commonly used cell culture
Laboratory Manual Cold Spring Harbour, Cold Spring media,” Biochemical and Biophysical Research Communica-
Harbour Laboratory Press, New York, 2001. tions, vol. 273, no. 1, pp. 50–53, 2000.
[46] V. Singh, H. Sharma, R. Veerma, A. Javed, and M. Singh, [62] S. G. Rhee, “Cell signaling. H2O2, a necessary evil for cell
“Topical non steroidal anti inflammatory drug (NSAIDs) signaling,” Science, vol. 312, no. 5782, pp. 1882–1883, 2006.
microemulsions: rationale, review and future prospective,” [63] G. P. Bienert, J. K. Schjoerring, and T. P. Jahn, “Membrane trans-
Asian Journal of Pharmaceutics, vol. 7, no. 1, p. 1, 2014. port of hydrogen peroxide,” Biochimica et Biophysica Acta
[47] M. Kreilgaard, “Influence of microemulsions on cutaneous (BBA)-Biomembranes, vol. 1758, no. 8, pp. 994–1003, 2006.
drug delivery,” Advanced Drug Delivery Reviews, vol. 54, [64] A. Balcerczyk, A. Grzelak, S. Kozio, R. Bae, and G. Bartosz,
Supplement 1, pp. S77–S98, 2002. “Formation of reactive oxygen species in cellular media,”
[48] Y. S. Rhee, J. G. Choi, E. S. Park, and S. C. Chi, “Transdermal Current Topics in Biophysics, vol. 29, no. 1-2, pp. 109–111,
delivery of ketoprofen using microemulsions,” International 2005.
Journal of Pharmaceutics, vol. 228, no. 1, pp. 161–170, 2001. [65] M. A. Bolzinger, S. P. Briançon, J. Pelletier, and Y. Chevalier,
[49] M. Changez and M. Varshney, “Aerosol-OT microemulsions “Penetration of drugs through skin, a complex rate-controlling
as transdermal carriers of tetracaine hydrochloride,” Drug membrane,” Current Opinion in Colloid & Interface Science,
Development and Industrial Pharmacy, vol. 26, no. 5, vol. 17, no. 3, pp. 156–165, 2012.
pp. 507–512, 2000. [66] J. Hadgraft and R. H. Guy, Transdermal Drug Delivery:
[50] M. J. Lawrence and G. D. Rees, “Microemulsion-based media Developmental Issues and Research Initiatives, Marcel Dekker
as novel drug delivery systems,” Advanced Drug Delivery Inc, New York, 1989.
Reviews, vol. 45, no. 1, pp. 89–121, 2000. [67] M. R. Prausnitz and R. Langer, “Transdermal drug delivery,”
[51] C. Korzeniewski and D. M. Callewaert, “An enzyme-release Nature Biotechnology, vol. 26, no. 11, pp. 1261–1268, 2008.
assay for natural cytotoxicity,” Journal of Immunological [68] S. G. Elfbaum and K. Laden, “The effect of dimethyl sulfoxide
Methods, vol. 64, no. 3, pp. 313–320, 1983. on percutaneous absorption: a mechanistic study. Part III,”
[52] S. Roohinejad, D. Middendorf, D. J. Burritt, U. Bindrich, D. W. Journal of the Society of Cosmetic Chemists, vol. 19, no. 13,
Everett, and I. Oey, “Capacity of natural β-carotene loaded pp. 841–847, 1968.
microemulsion to protect Caco-2 cells from oxidative damage [69] S. Kwak and M. Lafleur, “Effect of dimethyl sulfoxide on the
caused by exposure to H2O2,” Food Research International, phase behavior of model stratum corneum lipid mixtures,”
vol. 66, pp. 469–477, 2014. Chemistry and Physics of Lipids, vol. 161, no. 1, pp. 11–21, 2009.
[53] A. S. Narang, D. Delmarre, and D. Gao, “Stable drug encapsu- [70] K. D. Tsai, J. C. Lin, S. M. Yang et al., “Curcumin protects
lation in micelles and microemulsions,” International Journal against UVB-Induced skin cancers in SKH-1 hairless mouse:
of Pharmaceutics, vol. 345, no. 1, pp. 9–25, 2007. analysis of early molecular markers in carcinogenesis,”
[54] M. Lapteva and Y. N. Kalia, “Microstructured bicontinuous Evidence-Based Complementary and Alternative Medicine,
phase formulations: their characterization and application in vol. 2012, Article ID 593952, 11 pages, 2012.
dermal and transdermal drug delivery,” Expert Opinion on
Drug Delivery, vol. 10, no. 8, pp. 1043–1059.
[55] G. Chen, Y. Chen, N. Yang, X. Zhu, L. Sun, and G. Li,
“Interaction between curcumin and mimetic biomembrane,”
Science China Life Sciences, vol. 55, no. 6, pp. 527–532, 2013.
[56] D. Trachootham, W. Lu, M. A. Ogasawara, N. R. D. Valle, and
P. Huang, “Redox regulation of cell survival,” Antioxidants &
Redox Signaling, vol. 10, no. 8, pp. 1343–1374, 2008.
[57] Y. M. Kim, H. O. Pae, J. E. Park et al., “Heme oxygenase in the
regulation of vascular biology: from molecular mechanisms to
therapeutic opportunities,” Antioxidants & Redox Signaling,
vol. 14, no. 1, pp. 137–167, 2011.
[58] R. M. Tyrrell, “Solar ultraviolet A radiation: an oxidizing skin
carcinogen that activates heme oxygenase-1,” Antioxidants &
Redox Signaling, vol. 6, no. 5, pp. 835–840, 2004.
[59] F. Lieder, F. Reisen, T. Geppert et al., “Identification of
UV43 protective activators of nuclear factor erythroid-derived
2-related factor 2 (Nrf2) by combining a chemical library
screen with computer-based virtual screening,” Journal of
Biological Chemistry, vol. 287, no. 39, pp. 33001–33013, 2012.
[60] H. Erlank, A. Elmann, R. Kohen, and J. Kanner, “Polyphenols
activate Nrf2 in astrocytes via H2O2, semiquinones, and