Influence of vehicles on antioxidant efficacy in hair

Article abstract of DOI:10.1039/c5ra26815j

The UV radiation of sunlight is known to damage human hair, causing fibre degradation. Gallic acid (GA) was used as an active principle because of its antioxidant properties, which confer protection against free radicals. Encapsulation technologies, such

Full text of DOI:10.1039/c5ra26815j

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Influence of vehicles on antioxidant efficacy in hair
*
L. Rubio, C. Alonso, M. Martı, V. Martınez and L. Coderch
´
´
Cite this: RSC Adv., 2016, 6, 15929
The UV radiation of sunlight is known to damage human hair, causing fibre degradation. Gallic acid (GA) was
used as an active principle because of its antioxidant properties, which confer protection against free
radicals. Encapsulation technologies, such as polymer-based (micro- and nano-spheres, capsules) and
lipid-based systems (liposomes, emulsions), have been used to enhance and prolong the effectiveness of
active ingredients. In this study, GA loaded in PCL microspheres (GA-Micro) and GA encapsulated in PC
liposomes (GA-Lipo) were prepared to study their effectiveness as antioxidants when applied to human
hair compared with treatment with free GA. The antioxidant effects of these structures were measured
by the Bradford colorimetric assay and the thiobarbituric acid (TBA) assay and fluorescence spectroscopy
using 2,7-dichlorodihydrofluorescein (DCFH) as the fluorescent probe. The penetration of GA into hair
depending on the vehicle was also evaluated. Higher penetration of GA was found in hair fibres treated
with GA encapsulated vehicles. The results of the antioxidant studies showed that the efficacy of GA
increased when it was encapsulated in liposomes and microspheres, as well as when the studies were
performed immediately after antioxidant treatment. Moreover, GA that was loaded into microspheres
retained its antioxidant efficacy for one and two months after treatment, indicating that the
microspheres could preserve the antioxidant effect of GA. Moreover, the spectrofluorometric method
was found to be the best method to evaluate the efficacy of GA embedded in different vehicles.
Received 15th December 2015
Accepted 2nd February 2016
DOI: 10.1039/c5ra26815j
www.rsc.org/advances
active ingredients. Encapsulation technologies can be briey
categorized into polymer-based (micro- and nano-spheres,
Introduction
capsules) and lipid-based systems (liposomes, emulsions).
The encapsulation in suitable carrier systems, i.e., lipo-
somes, has been shown to enhance the penetration capacity of
the active ingredients into hair and protect the antioxidants
against oxidation.9,10 They have received considerable attention
because of their ability to trap both hydrophilic and hydro-
phobic active ingredients. However, the lipid-based systems
have problems related to their long-term stability.11,12 The
polymer-based systems are generally more stable than the lipid-
based systems.13 The efficacy of an antioxidant in protecting
hair is usually evaluated directly by measuring the strength,
shine and colour maintenance aer UV irradiation.14 Colori-
metric methods based on protein degradation and lipid per-
oxidation have been used to evaluate the effectiveness of
antioxidants.4 Because lipid and protein fractions play impor-
tant roles in the hair structure, protein degradation and lipid
peroxidation methods have been chosen. Additionally, a new
method based on uorescence spectroscopy has been devel-
oped. The Bradford colourimetric assay has been used to follow
protein degradation; this method can quantify solubilized
proteins and peptides.6 Lipoperoxide formation due to the
degradation of hair lipids by UV radiation was evaluated by
thiobarbituric acid (TBA) assay.6 Additionally, the antioxidant
efficacy of GA encapsulated in different vehicles and applied to
hair was studied by uorescence spectroscopy using
2,7-dichlorodihydrouorescein (DCFH) as the uorescent
The UV components of sunlight are known to damage human
hair, causing bre degradation. UV-B attacks the melanin
pigments and the protein fractions (keratin)1 of hair, and UV-A
produces free radical/reactive oxygen species (ROS) through the
interactions of endogenous photosensitizers. The lipid and
protein fractions play a major role in the structure and integrity
of the hair bre, protecting it against external agents. Therefore,
damage to the protein and lipid fractions2,3 can help us to
evaluate the decomposition of the hair bre.4
Gallic acid (GA) has been used as an active ingredient
because of its anti-inammatory, antifungal and antiviral
properties and because of the antioxidant protection it provides
against free radicals.5,6 Its benecial effects have led to its
incorporation into the different vehicles that are applied to hair.
The efficacy of antioxidants depends on several parameters.
The antioxidant should be an efficient radical scavenger.
Furthermore, it should be able to penetrate into the hair sha
and interact with the melanin. An adequate carrier system is of
the utmost importance. Encapsulation technologies with many
biodegradable polymers (PCL, PMMA, PLGA) and lipids have
been used to improve the solubility and long-term stability of
the active ingredients in cosmetic formulations.7,8 They have
also been used to enhance and prolong the effectiveness of the
´
˜
Instituto de Quımica Avanzada de Cataluna IQAC-CSIC, Jordi Girona 18-26, 08034
Barcelona, Spain. E-mail: laia.rubio@iqac.csic.es
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 15929–15936 | 15929

View Article Online
RSC Advances
Paper
probe.15,16 The method used therein was a modication of 100 ml of aqueous PVA solution (2%) and was emulsied for an
a method used in cellular systems to study the formation of ROS additional 10 minutes at 24 000 rpm, resulting in a double
in UVA irradiated cells and to evaluate the cellular oxidative emulsion (W1/O/W2). The preparation was carried out at 4 ꢀC.19
stress.17,18
The mixture was maintained with agitation at 400 rpm for 20 h
The purpose of this study was to prepare GA-loaded PCL at room temperature, leading to the evaporation of the solvent
microspheres (GA-Micro) and GA-PC liposomes (GA-Lipo) to and, consequently, the formation of the microspheres.
study their effectiveness as an antioxidant when applied to
human hair compared with free GA. GA-Micro and GA-Lipo were Liposome preparation
characterized by measuring their size distribution (mean
particle size), polydispersity index (PI), Z-potential, and drug
encapsulation efficiency (% EE). GA retention and penetration
into the hair bres were evaluated to determine the inuence of
Liposomes were prepared using the lm hydration method.20
Three grams of Lipoid® S100 (PC), as the lipid, was weighed and
dissolved in 30 ml of chloroform. The mixture was rotary
evaporated into a thin lipid lm, which was rehydrated with 100
vehicle administration. To determine the antioxidant treatment
ml of a 2% aqueous solution of GA producing multilamellar
efficacy of GA encapsulated in PCL-microspheres, PC-liposomes
vesicles (MLV).
and solubilized in ethanol/water solution, the antioxidant-
treated hair bres were subjected to UV irradiation. Photo-
Microspheres and liposomes characterization
degradation of hair was then evaluated using the three assays
The hydrodynamic diameter (HD) and polydispersity index (PI)
mentioned above.
of the aggregates were measured using the Zetasizer nano ZS
(Malvern Instruments, UK). This equipment employs the
Materials and methods
Materials
dynamic light scattering (DLS) technique to determine particle
sizes between 0.6 nm and 6 mm. DLS measures the Brownian
motion of the particles and correlates this phenomenon to their
For the preparation of the microspheres, poly(vinyl alcohol)
particle size.21 To minimize multiple scattering, non-invasive
(PVA; 87–89% hydrolysed, MW 31 000–50 000 Da) was used as
back scatter technology (NIBS) was used. With this tech-
the dispersant, and polycaprolactone (PCL; MW 45 000 Da) was
nology, the scattering is detected at an angle of 173ꢀ. Therefore,
used as the microsphere polymer. Both PVA and PCL were
concentrated samples can be measured without any modica-
supplied by Sigma-Aldrich (Madrid, Spain). Ethanol was ob-
tion in the case of our systems. The measurements were per-
tained from Merck (Darmstadt, Germany); isopropanol, from
formed at room temperature with DTS1060C-Clear disposable
Carlo Erba (Milan, Italy); and dichloromethane, from Panreac
zeta cell, and water was used as the dispersant. The results were
(Barcelona, Spain). For liposome formation, soybean phospha-
reported as the percentage of scattering intensity. Z-potential
tidylcholine (PC) (Lipoid S100; PC > 96%) was purchased from
was also measured. The data were collected and analysed
Lipoid GmbH (Ludwigshafen am Rhein, Germany). Gallic acid
using the DTS (Dispersion Technology Soware) program
provided by Malvern Instruments Ltd.
was supplied by Sigma-Aldrich (Madrid, Spain). For high-
performance liquid chromatography with UV detector (HPLC-
UV) analysis, methanol (HPLC grade) and puried water were
used. Methanol (HPLC grade) was acquired from Merck
Drug encapsulation efficiency (EE%)
To quantify the encapsulated GA, 1 ml of each of the PCL
(Darmstadt, Germany), and water was puried using an ultra-
microsphere and the liposome formulation were precipitated
pure water system, Milli-Q plus 185 (Millipore, Bedford, USA).
and separated from the supernatant by centrifugation at 14 000
Bovine Serum Albumin (BSA), thiobarbituric acid (TBA) and
rpm for 15 min using a Biocen 22 R centrifuge (Orto Alersa,
Spain). Aer separation, the supernatant was retained. The
initial PCL microspheres, liposomes and both supernatants
malondialdehyde (MDA) were acquired from Sigma (St. Louis,
MO, USA). Methanol, hydrochloric acid (HCl; 37%), trichloro-
acetic acid (TCA) and sodium dodecyl sulphate (SDS) were
were diluted in methanol and analysed by HPLC. The encap-
supplied by Merck (Darmstadt, Germany).
Natural Caucasian brown hair tresses (20 cm in length) were
purchased from De Meo Brothers Inc. (New York). Pantene®
dient present in the entire solution, as well as in the superna-
shampoo was supplied by Procter & Gamble Co. (Cincinnati, USA).
sulation efficacy of GA in the microspheres and in the liposomes
was determined by measuring the amount of the active ingre-
tant, obtained from a GA calibration curve.
The percentage of encapsulation efficiency (EE%) in the
microspheres and in the liposomes is given by:
PCL microsphere preparation
PCL microspheres loaded with GA were prepared using the
solvent evaporation method forming microemulsions (W1/O/W2
double emulsion). A total of 25 ml of a dispersion of 1 g of GA in
monoalcohol (2% isopropanol/98% ethanol) was added drop-
wise to 15 ml of PCL (1 g) in dichloromethane. A simple
EE% ¼ [drug(encapsulated)/drug(total)] ꢁ 100
HPLC analysis
emulsion (W1/O) was generated by mechanical agitation The concentration of GA in the different samples and the EE%
(ULTRA-TURRAXT25, IKA) for 30 min at 24 000 rpm. Later, this were determined by using a VWR-Hitachi LaChrom Elite HPLC
simple emulsion was added to a continuous phase consisting of system (Darmstadt, Germany). The equipment consisted of an
15930 | RSC Adv., 2016, 6, 15929–15936
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
L-2130 pump, an L-2200 autosampler and an L-400 UV detector. uorescence spectroscopy, the time of exposure to UV radiation
The system was operated using Merck EZChrom Elite v 3.1.3 was 2 hours.
soware. Chromatographic separation was performed at room
temperature using a LiChrocart 125-4/Lichrosorb RP-18 (5 mm) Protein degradation test
column (Darmstadt, Germany). The mobile phase was 90%
The Bradford colorimetric assay was used to test protein
methanol/10% water + 0.7% H3PO4 owing at a rate of 1.0 ml
degradation. This assay quanties dissolved proteins and
minꢂ1. A total of 20 ml of sample or calibration standard was
peptides. It is based on the formation of a complex between the
injected into the column and eluted with the mobile phase.
dye, Brilliant Blue G, and the proteins in solution, leading to an
Detection was carried out by monitoring the absorbance at 280
increase in absorption at 595 nm that is proportional to the
nm. The calibration curve exhibited a linear behaviour (>0.99%)
amount of protein in the solution (4). BSA was used as a stan-
dard to calculate the amount of protein.
over the concentration range of 0.104–130 mg mlꢂ1
.
Approximately 100 mg of hair in different swatches (treated
and exposed to UV irradiation or untreated) were weighed.
Dissolution of hair proteins and peptides was carried out in 2
ml of 2% aqueous SDS solution, which was sonicated using
a Labsonic 1510 device (B. Braun, Melsungen, Germany) for 5 h,
at 45 ꢀC. The hair extracts were then diluted down to 0.01% SDS
concentration, and the Bradford colorimetric assay was per-
formed. This assay was performed in triplicate for each hair
swatch. Tests with untreated hair (only washed with shampoo)
were also performed.
Hair treatments
The bre samples were all Caucasian brown hair. Hair swatches
of 2 g weight were used to carry out protein degradation, lip-
operoxidation, uorescence spectroscopy tests and hair pene-
tration evaluation. The hair swatches were initially washed
using 40 mg of Pantene® shampoo, and then rinsed with 4 ꢁ
250 ml of distilled water to remove any traces of cosmetic or
dirt. Hair strands weighing approximately 2 g were treated with
1 ml of each of the different formulations containing 1% GA
(microspheres, liposomes and ethanolic-aqueous solution) or
placebo (vehicles without GA) twice a day for 5 days. Before each
Lipid peroxidation measurements
subsequent application, the hair was washed as in the rst Lipid peroxidation refers, in our case, to an oxidative degrada-
wash. During the experimentation the hair samples were kept in tion of the lipids in hair by UV irradiation. To evaluate lip-
ꢀ
a temperature and humidity-controlled room (23 ꢃ 2 C, 50 ꢃ operoxidation, the thiobarbituric acid reactive substances
5% RH) and preserved to the light.
(TBARS) assay4 was performed. It is based on the reaction of
MDA (one of the products formed by the decomposition of the
lipoperoxided compounds) with TBA to form a coloured
complex (MDA:TBA).
Antioxidant hair penetration
The total GA retained in the hair and the GA that penetrated
into the bre, depending on the vehicle used, were evaluated
using the following procedure. To remove the external GA from
the hair surface, a small strand (approximately 40 mg) of each
hair treated with GA encapsulated in different vehicles were
washed three times with 10 ml of 0.1% SDS with ultrasonication
for 1 min. And then washed three times with 15 ml of water
using the same conditions.22 The samples were then dried at
room temperature.
To determine the total and the internalized amount of GA in
the hair bres, washed and non-washed samples were cut into
snippets of 1–3 mm. To ensure sample homogeneity, at least of
20–30 mg of hair were subjected to the above procedure. To
extract the GA from the hair samples, the washed and non-
washed bres were subjected to ultrasonication for 10 min in
methanol (10 ml). The extracted samples were analysed by
HPLC (as previously described) to determine the total amount
of GA that was retained in the hair and the amount of GA that
penetrated into the hair matrix. This study was performed in
triplicate for each hair sample.
Approximately 500 mg of different hair samples were
weighed. Then, the lipids from each of the different hair sample
were extracted with methanol (500 mg hair/10 ml methanol)
and sonicated for 15 min at room temperature. Next, the hair
extracts were dried under an N2 current and diluted again in 1.5
ml of methanol.
The results are expressed as malonaldehyde bis(dimethyla-
cetal) equivalents (mM MDA) using a standard curve for the pure
MDA–TBA complex. The calibration curve was obtained by
using different concentrations of MDA (0.5–30 mM). The
absorbance values obtained from the hair samples allowed us to
determine the amount of oxidized lipid in each sample. Here,
the absorbance values from each of the samples were extrapo-
lated using the equation from the calibration curve. The results
were expressed as the number of equivalents of MDA peroxides
formed (in mM concentrations). Ultimately, the amount of MDA
peroxides formed for each mg of hair was found to relate to the
initial amount of hair.
Evaluation of antioxidant efficacy by uorescence
spectroscopy
Exposure to articial UV radiation
To evaluate the antioxidant efficacy of the application on the
Each swatch was divided into two parts (0 and 72 h): one part hair of GA encapsulated in the two different vehicles (GA-Lipo
was not irradiated (0 h), whereas the other part was subjected to and GA-Micro) and as an aqueous-ethanolic solution
UV irradiation for 72 hours on the SUNTEST CPS + (250 W mꢂ2 (GA-Aqu), DCFH was used as the uorescent probe. DCFH is
equivalents to 1.5 J minꢂ1 cmꢂ2). During evaluation by a non-uorescent molecule that in contact with an oxidizing
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 15929–15936 | 15931

View Article Online
RSC Advances
Paper
agent, as free radicals generated by UV-radiation, produces 2,7- the values obtained from the characterization of GA-Micro and
dichlorouorescein (DCF) (Fig. 1). DCF is uorescent GA-Lipo are shown.
compound (lexcitation ¼ 488 nm; lemission ¼ 522 nm) that was The percentage of GA concentration in the three formula-
initially thought to be useful as a specic indicator for the tions are very similar, 1.08 ꢃ 0.09% for GA-Lipo, 1.15 ꢃ 0.08%
quantitation of reactive oxygen species (ROS).
a
for GA-Micro and 1.15 ꢃ 0.02% for GA-Aqu. The EE% was 34.8%
The method used here is based on that is based in the for GA-Micro and 20.9% for GA-Lipo. These values indicated
method described by Venkatachari.23 The DCFH solution was a higher encapsulation of GA in the microspheres than in the
prepared by rst cleaving 20,70-dichlorouorescin diacetate liposomes.
(DCFH-DA) in 0.01 N NaOH solution according to the following
The results obtained by DLS for GA-Lipo and GA-Micro
procedure. Briey, 6 ml of an ethanolic solution of 1 mM DCFH- showed mean particle sizes of 320.75 ꢃ 32.28 nm (PI 0.630)
DA was hydrolysed to DCFH with 24 ml of 0.01 N NaOH solu- and 2096.63 ꢃ 122.78 nm (PI 0.356), respectively. The results
tion. The reaction mixture was placed in the dark for 30 minutes indicated a larger diameter for GA-Micro than for GA-Lipo.
at room temperature, and then neutralized with 60 ml of These results were expected because the size for liposomes
phosphate buffered saline (PBS). The solution was kept on ice in generally ranges from 30 nm to several micrometers,24 whereas
the dark until further use.
that for the microspheres made up with PCL ranges from 1 to 20
Approximately 500 mg of hair subjected to different treat- mm (mostly in the range of 5–10 mm).25 The larger size of the
ments and exposure to UV radiation, or untreated hair, were microspheres was in accordance with the higher % EE of GA in
weighed. A total of 10 ml of the DCFH solution was added to the the microspheres. Moreover, the polydispersity index shows
different hair samples, which were incubated at 37 ꢀC for 1 hour greater homogeneity for GA-Micro (0.356) than for GA-Lipo
and 30 minutes in the dark. Aer incubation, the solutions were (0.630).
collected to measure the DCF formed due to the generation of
The Z-potential values obtained were +8.57 mV in the case of
ROS in the hair when exposed to UV radiation. The DCF was GA-Lipo and ꢂ0.97 mV in the case of GA-Micro. Because the hair
detected with a Shimadzu RF-540 spectrouorometer.
bres are negatively charged,26 GA-Lipo appears to show better
interaction with hair, but the low negative charge on GA-Micro
may not provide a great impediment to interact with the hair
bres either. Moreover, the higher the Z-potential in magnitude
(it does not matter whether the zeta potential is + or ꢂ), the
better the stability of the particles due to repulsion. At lower
values of Z-potential, the repulsion between particles is minor;
therefore, they tend to occulate. Therefore, these results
indicated a higher stability for GA-Lipo compared with GA-
Micro.
In the hair penetration evaluation studies, the total amount
of GA retained (both inside and outside of the hair samples) and
the amount of GA that penetrated into the hair matrix were
quantied. The results obtained are included in the following
table (Table 2). In addition, a graphical representation of the
percentage of GA penetration with respect to the total amount
applied to the hair is plotted in Fig. 2.
The results of the hair penetration studies indicated that the
hair treated with the aqueous solution of GA (GA-Aqu) retained
more GA than the hairs treated with the other formulations,
especially, the differences are signicant with hair treated with
liposomes. The microspheres also signicantly helped to retain
GA to a greater extent than the liposomes. This is in contrast
with the Z-potential results that indicated the possibility of
a better interaction between GA-Lipo and the hair bres.
However, it is important to note that a higher amount of GA was
found inside the bres when the two vehicles, liposomes and
microspheres, were applied. Statistically differences are found
between penetration of GA vehiculized in microspheres and
solubilized with ethanolic-aqueous solution.
Statistical analysis
The results obtained in the different experiments are expressed
as the mean ꢃ standard deviation (SD) for three determina-
tions. As regards to the statistical signicance of the results
obtained for each vehicle used and experiment carried out,
analysis of variance (ANOVA) and a Kruskal–Wallis test have
been applied. The soware used was the Statgraphics® plus 5. A
p value of less than 0.1 was considered signicant.
Results and discussion
Three formulations were prepared with the aim to obtain 1% GA
encapsulated in PC liposomes and PCL microspheres, and
solubilized in ethanol/water solution (1 : 1). Following the
experimental steps described previously, the three formula-
tions, GA-Lipo, GA-Micro and GA-Aqu, were obtained. The GA
concentrations were determined for all formulations by HPLC
analysis. Similar concentrations of approximately 1.1% GA were
obtained for all the three samples (see Table 1). The GA
formulations were characterized using DLS and HPLC to
determine particle size distribution, zeta potential and drug
encapsulation efficiency (EE%) (Table 1). In the following table,
To evaluate the efficacy of the GA encapsulated in the
different vehicles as an antioxidant, three types of tests were
carried out: protein degradation, lipid peroxidation and uo-
rescence spectroscopy. Efficacy was evaluated at different times
aer treatment. This was planned to determine the possible
Fig. 1 DCF formation by oxidation of DCFH.
15932 | RSC Adv., 2016, 6, 15929–15936
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Table 1 Average values (ꢃSD) from the characterization of GA-Lipo,
It can be pointed out the antioxidant efficacy of all formu-
lations containing GA evaluated by the three methodologies.
Moreover, the results show that GA-Lipo and GA-Micro provide
better protection against UV radiation than free GA when eval-
uated by protein solubilization and uorescence spectroscopy.
The higher antioxidant protection observed for the GA-
encapsulated formulations can be attributed to the higher
amount of the antioxidant present inside the bres due to the
penetration promoted by the liposomes and microspheres. In
contrast, the lipid peroxidation values indicated that better
protection was provided by GA-Aqu. However, the different
values obtained for the non-irradiated samples indicated that
lipid peroxidation could affect not only the small lipid fraction
in the hair but also the lipid components of the liposomes and
microspheres, which can otherwise mask the lipid peroxidation
of the components of the hair bre. Therefore, the lipid per-
oxidation test was discarded and not used for further
evaluation.
GA-Micro and GA-Aqu
Conc
(%)
D
(nm)
Z-Pot
(mV)
% EE
(%)
Sample
GA-Lipo
PI
1.08 ꢃ 0.09 320.75 ꢃ 32.28
0.630 8.57
20.9
mA-Micro 1.15 ꢃ 0.08 2096.63 ꢃ 122.78 0.356 ꢂ0.97 34.8
GA-Aqu
1.15 ꢃ 0.02
—
—
—
—
Protein degradation and uorescence spectroscopy studies
of GA-Lipo and GA-Micro one and two months aer application
to hair were performed to evaluate the stability of GA and the
antioxidant efficacy of GA in liposomes and microspheres over
time. Free GA was not evaluated over time because it demon-
strated poor results in terms of its antioxidant efficacy just
before the application. A comparison of the results obtained
from the protein degradation and spectrouorometry tests for
GA-Lipo- and GA-Micro-treated and untreated hair samples
immediately aer application and one and two months aer
application are presented in the following table (Table 4):
The graphical representations of the values obtained from
the protein degradation and uorescence spectroscopy tests
immediately aer the application of GA to hair and aer one
and two months are presented in Fig. 4.
The results obtained from the protein degradation and
uorescence spectroscopy tests, which were carried out imme-
diately aer GA application and one and two months later,
showed a similar trend. When GA was loaded into the lipo-
somes (GA-Lipo), most of the antioxidant effect achieved
immediately aer applications was lost with time; there was less
antioxidant effect aer 1 month and 2 months. However, the
effectiveness of GA-Micro was clearly maintained over time.
This difference in the antioxidant efficacy of GA between
encapsulation in liposomes and in microspheres (especially
statistically signicant in the spectrouorometric results) can
be attributed to the difference in their EE%. GA-Micro has
a higher % EE than GA-Lipo. Liposomes are known to have
lower stability over time,11,12 whereas polymer based
Fig. 2 Graphical representation of the GA retained and the GA that
penetrated with respect to the total amount of GA in the three
formulations that were applied to the hair. *p > 0.1.
preservation of the antioxidant effect of GA due to
encapsulation.
Because proteins constitute of a major part of hair (88%),
and UV radiation leads to loss of protein,27 a protein degrada-
tion test is a good way to study the efficacy of an antioxidant
against damage due to UV radiation. Although it is not easy to
evaluate lipid modication because of the low amount of lipids
in the hair bres (2.3%),28 a lipid peroxidation test is also an
efficient method to study the damage produced by UV radiation,
and, hence, the efficiency of an antioxidant applied to hair.
Fluorescence spectroscopy using DCFH as the uorescent probe
is another tool for evaluating the generation of ROS due to UV
irradiation in hair.
The results of the three tests immediately aer the applica-
tion of GA vehiculized in different vehicles to the hair samples,
irradiated or not, are shown in Table 3. The graphical repre-
sentation of the percentage of oxidation increase in the hair
samples, treated with the different formulations of GA and
submitted to UV radiation, which was evaluated using the
aforementioned techniques, are exhibited in Fig. 3. The
percentages are relative to 100% oxidation of untreated hair
samples.
Table 2 Average values (ꢃSD) obtained in the evaluation of hair penetration
Total amount GA
retained
(mg GA per mg hair)
Internal amount GA
penetrated
(mg GA per mg hair)
Total applied
% GA retained
of total applied
% penetrated
% penetrated
Sample
(mg GA per mg hair)
of amount retained
of total applied
GA-Lipo
mA-Micro
GA-Aqu
5.40
5.75
5.75
0.44 ꢃ 0.14
1.28 ꢃ 0.55
1.70 ꢃ 0.57
8.11 ꢃ 2.53
22.28 ꢃ 9.48
29.55 ꢃ 9.93
0.08 ꢃ 0.05
0.10 ꢃ 0.05
0.03 ꢃ 0.02
22.74 ꢃ 19.42
9.58 ꢃ 6.04
2.18 ꢃ 1.23
1.57 ꢃ 0.85
1.77 ꢃ 0.83
0.57 ꢃ 0.26
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 15929–15936 | 15933

View Article Online
RSC Advances
Paper
Table 3 Average values (ꢃSD) obtained in the protein degradation, lipid peroxidation and fluorescence spectroscopy tests carried out imme-
diately after GA application to the hair and after 0 h (non-irradiated), 2 h or 72 h of UV irradiation
GA-Lipo
GA-Micro
GA-Aqu
Untreated
Protein degradation [mg protein per mg
hair]
0 h UV
72 h UV
% increase
Relative to 100% untreated
0 h UV
2 h UV
% increase
Relative to 100% untreated
36.50 ꢃ 5.51
38.91 ꢃ 3.49
7.28 ꢃ 6.61
38.91 ꢃ 3.23
40.99 ꢃ 9.21
4.67 ꢃ 16.85
9.12 ꢃ 29.83
18.18 ꢃ 3.39
29.23 ꢃ 3.49
66.20 ꢃ 44.22
55.34 ꢃ 38.38
1.26 ꢃ 0.73
41.59 ꢃ 6.86
46.76 ꢃ 5.00
13.27 ꢃ 9.24
20.96 ꢃ 10.89
17.65 ꢃ 0.80
32.99 ꢃ 3.07
33.69 ꢃ 6.57
53.64 ꢃ 9.52
59.85 ꢃ 11.86
100 ꢃ 0.00
11.52 ꢃ 10.75
22.08 ꢃ 2.98
30.91 ꢃ 7.54
39.90 ꢃ 30.03
31.59 ꢃ 21.46
3.48 ꢃ 2.05
Fluorescence spectroscopy (intensity
(u.a.))
45.75 ꢃ 3.19
102.27 ꢃ 1.26
89.90 ꢃ 16.42 124.21 ꢃ 15.16
71.82 ꢃ 23.13
2.27 ꢃ 0.95
3.38 ꢃ 1.72
100 ꢃ 0.00
0.41 ꢃ 0.36
3.38 ꢃ 2.42
Lipid peroxidation [mM MDA per mg hair] 0 h UV
72 h UV
% increase
Relative to 100% untreated
8.54 ꢃ 2.87
4.01 ꢃ 0.94
193.29 ꢃ 153.92 271.97 ꢃ 139.69 45.85 ꢃ 27.57 795.20 ꢃ 144.35
22.45 ꢃ 16.16
33.16 ꢃ 13.94
6.02 ꢃ 3.85
100 ꢃ 0.00
microspheres are usually more stable. Moreover, the slightly
higher penetration of GA into the hair bres when it is encap-
sulated in microspheres can also support its greater antioxidant
efficacy over time. These results were indicative of better
protection of GA against degradation when it is encapsulated in
microspheres than in liposomes at least two months aer the
application of GA formulations to hair samples. Therefore,
microspheres are a better candidate to encapsulate GA for
application to hair over time.
In this work, three assays were performed to determine the
antioxidant properties of GA encapsulated in different ways. The
TBARS assay is a well-recognized and established method for
quantifying lipid peroxidation of a substrate subjected to UV
radiation. Unfortunately, in our case, there was interference with
Fig. 3 Comparison of % increase in oxidation from the different
techniques studied immediately after the application of GA formula-
tions, relative to 100% oxidation of the untreated hair samples.
Table 4 Values obtained in the protein degradation and fluorescence spectroscopy tests carried out immediately after GA application to hair and
one and two months thereafter
GA-Lipo
GA-Micro
Non treat
Protein degradation [mg prot per mg hair] Just applied
1 month aer
0 h UV
72 h UV
% increase
Relative to 100% non treat
0 h UV
36.50 ꢃ 5.51
38.91 ꢃ 3.49
7.28 ꢃ 6.61
38.91 ꢃ 3.23
40.99 ꢃ 9.21
4.67 ꢃ 16.85
9.12 ꢃ 29.83
42.96 ꢃ 18.30
46.88 ꢃ 9.41
16.84 ꢃ 30.64
24.08 ꢃ 22.04
45.63 ꢃ 2.13
45.28 ꢃ 8.76
ꢂ3.42 ꢃ 17.76
33.69 ꢃ 6.57
53.64 ꢃ 9.52
59.85 ꢃ 11.86
100 ꢃ 0.00
11.52 ꢃ 10.75
33.72 ꢃ 4.93
47.35 ꢃ 3.49
39.04 ꢃ 18.84
44.05 ꢃ 3.30
57.85 ꢃ 19.29
29.98 ꢃ 34.56
100 ꢃ 0.00
72 h UV
% increase
Relative to 100% non treat 146.08 ꢃ 100.54
2 months aer 0 h UV
42.91 ꢃ 6.32
45.12 ꢃ 5.37
5.37 ꢃ 3.02
35.40 ꢃ 16.89
43.54 ꢃ 10.46
37.03 ꢃ 47.31
72 h UV
% increase
Relative to 100% non treat
0 h UV
2 h UV
% increase
Relative to 100% non treat
0 h UV
2 h UV
% increase
30.28 ꢃ 37.82
22.08 ꢃ 2.98
30.91 ꢃ 7.54
39.90 ꢃ 30.03
31.59 ꢃ 21.46
22.56 ꢃ 4.52
78.27 ꢃ 23.68
243.27 ꢃ 36.14
83.22 ꢃ 26.56
18.91 ꢃ 3.90
56.39 ꢃ 6.55
200.97 ꢃ 27.39
79.90 ꢃ 28.24
ꢂ55.83 ꢃ 104.69 100.00
Fluorescence spectroscopy (intensity
(u.a.))
Just applied
18.18 ꢃ 3.39
29.23 ꢃ 3.49
66.20 ꢃ 44.22
55.34 ꢃ 38.38
20.32 ꢃ 0.90
27.05 ꢃ 2.50
33.50 ꢃ 18.22
10.78 ꢃ 4.18
17.81 ꢃ 2.93
23.67 ꢃ 6.70
31.62 ꢃ 15.97
11.69 ꢃ 3.50
45.75 ꢃ 3.19
102.27 ꢃ 1.26
124.21 ꢃ 15.16
100 ꢃ 0.00
1 month aer
26.07 ꢃ 8.15
102.32 ꢃ 18.97
300.72 ꢃ 52.57
100 ꢃ 0.00
Relative to 100% non treat
2 months aer 0 h UV
21.08 ꢃ 2.17
75.63 ꢃ 4.43
261.81 ꢃ 58.27
100 ꢃ 0.00
2 h UV
% increase
Relative to 100% non treat
15934 | RSC Adv., 2016, 6, 15929–15936
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
References
1 A. C. Nogueira, L. E. Dicelio and I. Joekes, Photochem.
Photobiol. Sci., 2006, 5, 165–169.
2 E. Hoting, M. Zimmerman and B. Hilterhaus, J. Soc. Cosmet.
Chem., 1995, 46, 85–99.
3 E. Hoting and M. Zimmerman, J. Soc. Cosmet. Chem., 1996,
47, 201–211.
4 E. Fernandez, C. Barba, C. Alonso, M. Marti, J. L. Parra and
L. Coderch, J. Photochem. Photobiol., B, 2012, 106, 101–106.
5 B. Badhani, N. Sharma and R. Kakkar, RSC Adv., 2015, 5,
27540–27557.
6 Y. Yilmaz and R. T. Toledo, J. Agric. Food Chem., 2004, 52,
255–260.
7 P. Perugini, I. Genta, F. Pavanetto, B. Conti, S. Scalia and
A. Baruffini, Int. J. Pharm., 2000, 196, 51–61.
8 M. H. Lee, S. G. Oh, S. K. Moon and S. Y. Bae, J. Colloid
Interface Sci., 2001, 240, 83–89.
Fig. 4 Comparison of the % increase in oxidation obtained from the
protein degradation and fluorescence spectroscopy studies immedi-
ately after the application of GA to hair and after one and two months,
related to 100% oxidation of untreated hair samples.
the lipidic matrix of the liposomes, which caused signicant
standard deviations in the results. Other studies were carried out
using the Bradford test; this assay is useful in the evaluation of
protein degradation due to oxidation by UV radiation. As
a consequence, the effectiveness of the antioxidant could be
evaluated. This method is simple and faster than the other
methods used to evaluate protein degradation but is also subject
to interference from the non-protein components of biological
samples.29 In our work, the results showed an signicant stan-
dard deviation indicating that this was not the best method to
evaluate the effectiveness of GA, loaded in different vehicles, as
an antioxidant. The last method used to evaluate the antioxidant
efficacy of GA in different vehicles was a modication of
a method that was used to study the formation of ROS in UVA
irradiated cells30,31 and to evaluate cellular oxidative stress.17,18
This spectrouorometric method is based on the oxidation of
DCFH to DCF in the presence of an oxidizing agent. This method
adapted for our samples is rapid and has a high reproducibility,
as indicated by the low standard deviation in our results.
9 K. Jung, T. Herrling, G. Blume, M. Sacher and
¨
¨
D. Teichmuller, So Journal, 2006, 132, 32–36.
10 T. Herrling, K. Jung and J. Fuchs, Spectrochim. Acta, Part A,
2008, 69, 1429–1435.
11 I. P. Kaur, M. Kapila and R. Agrawal, Ageing Res. Rev., 2007, 6,
271–288.
12 G. M. El Maghraby, B. W. Barry and A. C. Williams, Eur. J.
Pharm. Sci., 2008, 34, 203–222.
13 L. J. Peek, C. R. Middaugh and C. Berkland, Adv. Drug
Delivery Rev., 2008, 60, 915–928.
14 A. C. Santos Nogueira and I. Joekes, J. Photochem. Photobiol.,
B, 2004, 74, 109–117.
15 A. Gomes, E. Fernandes and J. L. Lima, J. Biochem. Biophys.
Methods, 2005, 65, 45–80.
16 C. P. LeBel, H. Ischiropoulos and S. C. Bondy, Chem. Res.
Toxicol., 1992, 5, 227–231.
17 W. Jakubowski and G. Bartosz, Cell Biol. Int., 2000, 24, 757–
760.
18 M. M. Tarpey, D. A. Wink and M. B. Grisham, Am. J. Physiol.:
Regul., Integr. Comp. Physiol., 2004, 286, R431–R444.
˜
´
19 N. Carreras, V. Acuna, M. Martı and M. J. Lis, Colloid Polym.
Conclusions
Sci., 2013, 291, 157–165.
In summary, these results indicate that microspheres could
preserve the antioxidant efficacy of GA over a longer period of time,
which was due in part to their higher penetration ability. More-
over, the spectrouorometric method was found to be the best
method to evaluate the efficacy of GA embedded in the different
vehicles. Future work will be based on the vehiculization of other
antioxidants in microspheres to demonstrate their efficacy along
the time as it was the case of GA. This approach could serve to
verify the use of encapsulated antioxidants to be released into the
hair. The present study opens new prospects for the development
of antioxidant formulations able to increase hair protection.
20 E. Ramon, C. Alonso, L. Coderch, A. de la Maza, O. Lopez,
J. L. Parra and J. Notario, Drug Delivery, 2005, 12, 83–88.
21 H. G. Merkus, in Particle Size Measurements, Springer,
Netherlands, 2009, vol. 17, pp. 299–317.
22 Y. Nakahara, T. Ochiai and R. Kikura, Arch. Toxicol., 1992, 66,
446–449.
23 P. Venkatachari, P. Hopke, B. Grover and D. Eatough, J.
Atmos. Chem., 2005, 50, 49–58.
24 A. Akbarzadeh, R. Rezaei-Sadabady, S. Davaran, S. W. Joo,
N. Zarghami, Y. Hanifehpour, M. Samiei, M. Kouhi and
K. Nejati-Koshki, Nanoscale Res. Lett., 2013, 8, 102.
25 W.-J. Lin, D. R. Flanagan and R. J. Linhardt, Polymer, 1999,
40, 1731–1735.
Acknowledgements
26 B. Bhushan, in Biophysics of Human Hair, Introduction,
Springer-Verlag, Berlin, Heidelberg, 2010, p. 14.
The authors wish to thank the Ministerio de Economia y
Competitividad (Spanish National Project CTQ2013-44998-P)
and the Generalitat de Catalunya (2014 SGR 1325).
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 15929–15936 | 15935

View Article Online
RSC Advances
Paper
27 C. W. M. Yuen, C. W. Kan and S. Y. Cheng, Fibers Polym., 30 C. F. Chignell and R. H. Sik, Free Radical Biol. Med., 2003, 34,
2007, 8, 414–420. 1029–1034.
28 Y. Masukawa, H. Narita and G. Imokawa, J. Cosmet. Sci., 31 C. Rota, C. F. Chignell and R. P. Mason, Free Radical Biol.
2005, 56, 1–16.
Med., 1999, 27, 873–881.
29 N. Kruger, in The Protein Protocols Handbook, ed. J. Walker,
Humana Press, 2009, ch. 4, pp. 17–24, DOI: 10.1007/978-1-
59745-198-7_4.
15936 | RSC Adv., 2016, 6, 15929–15936
This journal is © The Royal Society of Chemistry 2016