Food and Chemical Toxicology 46 (2008) 2776–2781

Contents lists available at ScienceDirect

Food and Chemical Toxicology

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

Low cytotoxicity of creams and lotions formulated with Buriti oil

(Mauritia flexuosa) assessed by the neutral red release test
Cinthia Fernanda Zanatta a,b, Vanessa Ugartondo b, Montserrat Mitjans b,
Pedro A. Rocha-Filho a, María Pilar Vinardell b,*
a
Departament de Ciências Farmacêuticas, Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo, Brazil
b
Departament de Fisiologia, Facultat de Farmácia, Universitat de Barcelona, Av. Joan XXIII s/n, 08028 Barcelona, Spain

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

Article history: The aim of this work was to evaluate possible cytotoxic effects of topical creams and lotions produced
Received 28 December 2007 with Buriti oil and commercial surfactants on human keratinocytes HaCat and 3T3 embryonic mouse
Accepted 7 May 2008 fibroblast cultures. We also aimed to assess the cytotoxicity of the surfactants used to produce the emul-
sions. The neutral red release (NRR) assay was performed as an in vitro method to evaluate the cytotox-
icity of the emulsions in HaCat and 3T3 cell lines and predict potential skin irritation. The Buriti oil
Keywords: emulsions presented low cytotoxicity to the cells at high concentrations and the addition of Vitamin E
Buriti oil
increased cell viability. Among the surfactant tested, UnitolÒ CE 200F proved to be the most cytotoxic,
Emulsion
Cytotoxicity
presenting an IC50 significantly lower than the others. Emulsions formulated with Buriti oil and commer-
Skin irritation cial surfactants could be non irritant to the skin due to their low cytotoxicity, especially when enhanced
Cell culture with vitamin E. When emulsified with Buriti oil, water and Brij 72, Unitol CE200F showed less cytotoxic
effects than when tested alone.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction identify chemicals which might induce adverse skin reactions.

The use of natural ingredients in cosmetics is steadily increas-
ing. As a result, formulators are being offered a host of newly de-
rived lipids, most from renewable plants sources (Rieger, 1994).
The Amazon rain forest is very rich in oily plants, representing
great economic potential for the region. High levels of phytochemi-
cals, especially anti-oxidants are commonly found in many of these
plants. One of such plants is Buriti palm tree (Mauritia flexuosa),
which oil is rich in carotenoids and has been frequently used in
cosmetic production.
Carotenoids are known to be powerful anti-oxidants and their
ability to quench singlet oxygen has been studied (Conn et al.,
1991; Chen et al., 2007) and reviewed extensively (Edge et al.,
1997; Stahl and Sies, 2007; Ghersetich et al., 2007). Oxygen reac-
tive species, such as superoxide radical, peroxide radical and hy-
droxyl radical, which can be generated by cytotoxic compounds
are potentially damaging to cells through initiation of lipid perox-
idation. Consequently, there has been great interest in the anti-oxi-
dant properties of carotenoids with regard to human health
(Cantrell et al., 2003; Rao and Rao, 2007).
During the development of new topic formulations, skin irrita-
tion potential is investigated prior to human exposure, in order to
* Corresponding author. Tel.: +34 934024505; fax: +34 934035901.
E-mail address: mpvinardellmh@ub.edu (M.P. Vinardell).
However, the surfactant present in personal care products are
presumed to be responsible for the majority of these reactions
(Barany et al., 1999; Orton and Wilkinson, 2004; Ferrati et al.,
2005).
Ethical, legal and financial motives have banned the in vivo
method for testing cosmetic ingredients and formulations (Pauw-
els and Rogiers, 2007; Rogiers and Pauwels, 2006). Over the past
few years, several in vitro test methods have been suggested as
valid substitutes of the irritation tests and recently reviewed
(Vinardell and Mitjans, 2007).
To this end, a broad range of cell and tissue culture systems has
been develop for assessing the irritation and phototoxic potential
of chemicals, among them human keratinocytes and fibroblasts
cultures (Dijoux et al., 2006). Previous studies have suggested that
cultured normal human keratinocytes may be predictive of irri-
tancy caused by several surfactants (Benavides et al., 2004; Korting
et al., 1994a,b) and high correlation has been demonstrated with
the in vivo human skin data of irritation potential (Osborne and
Perkins, 1994). Although some authors argue that keratinocytes
(either primary cultures or cell permanent lines) are less sensitive
than fibroblasts to irritation (Maier et al., 1991), for certain specific
purposes, is preferable to use keratinocytes, since in vivo they are
first cells to be exposed to cosmetics. Moreover, this model was
considered a helpful tool to predict irritancy, despite the lack of
100% accuracy (Wilhelm et al., 2001).

0278-6915/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.fct.2008.05.001
 C.F. Zanatta et al. / Food and Chemical Toxicology 46 (2008) 2776–2781
Recently the Scientific Advisory Committee of European Centre
for the Validation of Alternative Methods (ECVAM) has today an-
nounced the validation of new ‘‘in vitro” tests, which take an
important step towards ending the practice of using rabbits in
skin-irritancy testing. The two tests for skin irritation validated
have been developed by industry and private companies and are
commercial kits which use in vitro cell culture to evaluate the po-
tential skin-irritancy of chemicals, by providing a realistic repre-
sentation of the properties of human skin to identify irritant and
non-irritant chemicals. Nevertheless, the validation of these two
methods, it is important to consider other potential methods more
economic and simple for pre-screening of new formulations.
Considering the belief that natural lipids and edible substances
are safer for topical applications, the aim of this study was to eval-
uate the possible cytotoxic effects of topical creams and lotions
produced with Buriti oil and commercial surfactants on a simple
method based on human keratinocytes HaCat and 3T3 embryonic
mouse fibroblast monolayers cultures, using the neutral red re-
lease assay. We also studied the influence of emulsification process
in the products cytotoxicity and the cytotoxic effects of the surfac-
tants used to formulate these emulsions and then correlated the
results.
2. Materials and methods
2.1. Materials
Dulbecco’s Modified Eagle’s Medium (DMEM), Hepes buffer, penicillin
(10,000 U/ml)–streptomycin (10,000 lg/ml) mixture and fetal bovine serum (FBS)
were purchased from Bio-Whittaker (Verviers, Belgium). The 75 cm2 flasks and
96-well plates were obtained from TPP (Trasadingen, Switzerland).
For the emulsions preparation, Buriti oil was supplied by Croda (Brazil), Unitol
CE50Ò (Ceteareth-5), Unitol CE200FÒ (Ceteareth-20), Span 80Ò (sorbitan monooleate),
Ultroil R400Ò (PEG-40 castor oil) by Oxiteno (Brazil), Brij 72Ò (Steareth-2) was ob-
tained from Beraca Sabará Ingredients (Brazil) and sodium poliacrylate from ISP
Corp (USA). The vitamin E acetate and, D-panthenol from Sigma.
2.2. Products preparation
Five different emulsions were evaluated, two O/W macroemulsions containing
liquid crystals (29R and 51LC), one simple O/W emulsion (51S), one W/O/W multi-
ple emulsion (37M) and one O/W nanoemulsion (37N). Each system was produced
with and without an active ingredient at concentration of 1%. The vitamin E acetate
(VE) was used to produce the 29R.VE, 51S.VE, 51LC.VE and 37N.VE; and the D-pan-
thenol (P) to produce the 37M.P and the 37N.P. The concentrations of the different
components are described in Table 1.
The O/W macroemulsions (51S, 51S.VE, 51LC, 51LC.VE, 29R and 29R.VE) and the
multiple emulsions (37M and 37M.P) were prepared using the emulsification by
phase inversion (EPI) method (Santos et al., 2005), while the nanoemulsions (37N
and 37N.P and 37N.VE) were obtained by the phase inversion temperature method
(Tadros et al., 2004).
All emulsions were produced under constant stirring (Mechanic Mixer Fisatom
Mod. 713 D) at 600 rpm until they reached room temperature (25 ± 2 °C), except the
simple O/W macroemulsions (51S and 51S.VE) which were emulsified using a high-
pressure homogenizer Ultra TurraxÒ (IKA–25FKR) in order to produce finer
droplets.
Table 1
Concentrations of the emulsions components
Formulation number Oil (%) Water (%) Surfactant system (%)
Total Surfactant A Surfactant B
29Ra 10 80 10 4.53 5.47
51Sb 5 80 15 12.10 2.90
51LCb 5 80 15 12.10 2.90
37Mc 5 90 5 2.30 2.70
37Nc 5 90 5 2.30 2.70
a
Surfactant system composed by surfactant A: BrijÒ 72 and surfactant B: UnitolÒ
CE50and 0.1% of sodium poliacrylate.
b
Surfactant system composed by surfactant A: BrijÒ 72 and surfactant B: UnitolÒ
CE200F .
c
Surfactant system composed by surfactant A: SpanÒ 80 and surfactant B: Ult-
roilÒ R400.
2777
The surfactants were heated to 75 ± 2 °C and poured into distilled water at the
same temperature, under constant stirring at 600 rpm until they reached room
temperature (25 ± 2 °C). The mixtures of surfactants used to produce the emulsions
were prepared at the same percentage (Table 1) for further evaluation.
2.3. Culture of HaCaT and 3T3 cell line
The spontaneously immortalized human keratinocyte cell line HaCaT and the
mouse embryonic fibroblast cell line 3T3 were grown in DMEM medium (4.5 g/l
glucose) supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 10 mM
Hepes buffer and 1% penicillin (10.000 U/ml) – streptomycin (10.000 lg/ml) mix-
ture at 37 °C. 5% CO2. Both cell lines were routinely cultured into 75 cm2 culture
flasks.
When the cells were approximately 80% confluent, they were harvested with
trypsin/EDTA and seeded at a density of 5  104 cells/ml into the central 60 wells
of 96-well plates and then incubated for 48 h at 37 °C, 5% CO2.
2.4. Products dilution
For the cell treatment emulsions and surfactants were diluted in PBS (Invittosx
protocol 1992). Emulsions were prepared in a range of concentrations from 31 mg/
ml to 1000 mg/ml, while the surfactants Span 80, Ultroil R400 and their respective
mixture were diluted into concentrations ranging from 12.5 mg/ml to 200 mg/ml.
The Ceteareth-20 was prepared in a range from 0.5 mg/ml to 50 mg/ml.
2.5. NRR assay
The NRR assay was performed by treating the cultured cells with the emulsions
and surfactants for 10 min and releasing the neutral red (NR) dye from preloaded
cells. This served as a toxicity marker, as described previously (Korting et al.,
1994a,b), with some minor modifications
After removal of the culture medium, the neutral red dye (75 lg/ml de DMEM
10%) was added to the HaCaT and 3T3 cells cultured in the 96-well plates and the
plates were incubated for 3 hours at 37 °C. 5% CO2. Then, the neutral red was aspi-
rated and 150 ll of fresh medium (DMEM 10%) was added to the cells. PBS was then
used to wash the cells to prepare them for the treatment with the products.
The cells were exposed to serial dilutions of the emulsions and surfactants for
10 min. Controls containing culture medium and PBS only, were included in each
plate and were independent for each of the different emulsion and surfactants
tested. After treatment, the products were aspirated and the wells washed twice
with PBS and a solution containing 50% ethanol absolute–1% acetic acid in distilled
water was added to extract the dye. After 10 min on a microtitre-plate shaker, the
absorbance of neutral red was measured at a wavelength of 550 nm in a Bio-Rad
550 microplate reader.
2.6. Statistical analysis
The cytotoxicity of each emulsion and surfactant was expressed as a percentage
of viability compared with control wells (the mean optical density of untreated cells
was set to 100% viability), calculated from the concentration–response curves by
linear regression analysis. NRR assay results are expressed as the percentage of up-
take of neutral red dye by the lysosomes of the viable cells.
The experiments were performed at least three times using three wells for each
surfactant concentration tested. To evaluate the differences in the cytotoxicity of
emulsions and surfactants, variance analyses of were conducted using one-way
ANOVA (Software Origin 6.0) with significance level of 95%.
3. Results and discussion
3.1. Emulsion cytotoxicity
The results of the NRR carried out in keratinocytes and fibro-
blasts, were obtained at different concentrations of emulsions.
The results are expressed as cellular viability (Tables 2 and 3) since
the IC50 could not be estimated due to the low cytotoxicity of the
emulsions.
The emulsions produced with Buriti oil showed in general, to be
non-toxic to 3T3 and HaCat monolayers at concentrations from
31.25 to 1000 mg/ml. Only 37N, 37M and 37M.P showed low
viability for HaCaT cells at the highest concentration tested
(1000 mg/ml). However, the concentrations used to determine
their irritation potential was extremely high in comparison to
those suggested in different works (Invittox protocol 1992, Get-
tings et al., 1996) for the NRR assay, as lower concentrations did
not provide expressive results.
2778 C.F. Zanatta et al. / Food and Chemical Toxicology 46 (2008) 2776–2781

Table 2
Cellular viability, in percentage, of 3T3 cells after exposure to the Buriti oil emulsions and the neutral red release assay

Concentration Cellular viability (%)a

(mg/ml)
Macroemulsion Nanoemulsion Multiple emulsion
29R 29.VE 51LC 51LC.VE 51S 51S.VE 37N 37N.P 37N.VE 37M 37M.P
31.25 103.6 ± 1.5 97.1 ± 1.9b 109.0 ± 7.2 101.7 ± 4.4 82.7 ± 14.5 114.0 ± 2.2b 98.6 ± 4.4 94.5 ± 6.2 107.8 ± 2.4b,c 102.3 ± 2.6 97.3 ± 5.2
62.5 103.9 ± 1.2 105.3 ± 1.2 106.1 ± 4.2 104.7 ± 2.7 98.9 ± 1.5 107.4 ± 1.8b 105.3 ± 4.3 96.4 ± 2.7b 102.9 ± 0.4c 102.3 ± 3.1 92.3 ± 4.5b
125 103.6 ± 1.4 104.8 ± 0.5 105.5 ± 9.7 102.2 ± 3.6 98.5 ± 0.5 107.1 ± 7.4 96.8 ± 4.9 96.2 ± 3.5 106.4 ± 1.2b 98.1 ± 2.8 94.0 ± 5.1
250 99.9 ± 1.9 102.8 ± 3.4 103.5 ± 6.8 103.9 ± 2.2 95.8 ± 2.7 107.8 ± 2.4b 104.0 ± 8.8 88.3 ± 0.7b 104.8 ± 3.9c 106.6 ± 0.5 94.9 ± 4.6b
500 97.8 ± 4.0 104.0 ± 2.7 102.1 ± 1.7 103.9 ± 4.2 114.6 ± 6.5 115.1 ± 5.0 95.4 ± 6.3 92.9 ± 1.7 102.7 ± 3.6c 96.3 ± 3.6 94.0 ± 5.6
750 107.9 ± 3.5 109.6 ± 11.7 110.2 ± 4.1 105.8 ± 4.4 109.2 ± 19.9 99.3 ± 2.8 99.2 ± 4.8 91.7 ± 1.5 99.1 ± 2.2c 95.7 ± 2.2 91.0 ± 0.9b
1000 90.8 ± 3.8 113.8 ± 5.9b 104.1 ± 9.6 107.4 ± 4.5 132.7 ± 17.3 132.0 ± 22.2 85.1 ± 2.6 87.0 ± 2.9 85.7 ± 2.3 83.3 ± 4.0 87.8 ± 2.4
a
Mean of triplicate.
b
Indicates significant difference in comparison to the emulsion without active ingredient, at significance level of 5%.
c
Indicates significant difference in comparison to the emulsion with panthenol, at significance level of 5%.

Table 3
Cellular viability, in percentage, of HaCat cells after exposure to the Buriti oil emulsions and the neutral red release assay

Concentration Cellular viability (%)a

(mg/ml)
Macroemulsion Nanoemulsion Multiple emulsion
29R 29.VE 51LC 51LC.VE 51S 51S.VE 37N 37N.P 37N.VE 37M 37M.P
31.25 110.7 ± 0.3 120.8 ± 5.9b 104.2 ± 2.9 105.8 ± 3.0 111.5 ± 4.6 107.5 ± 5.8 103.5 ± 3.5 95.3 ± 1.2b 107.4 ± 14.8 97.5 ± 4.1b 94.8 ± 3.0b
62.5 110.4 ± 8.1 120.3 ± 8.5 106.1 ± 4.2 106.5 ± 3.9 115.2 ± 6.4 106.8 ± 6.0 98.4 ± 3.8 98.6 ± 3.2 111.4 ± 12.2 97.2 ± 5.0b 96.9 ± 2.7b
125 108.0 ± 4.8 111.9 ± 11.3 103.4 ± 6.5 101.6 ± 3.8 110.4 ± 6.6 104.0 ± 5.0 96.9 ± 1.2 96.4 ± 3.8 111.4 ± 1.1 94.6 ± 4.1b 94.1 ± 0.5b
250 105.2 ± 5.4 109.5 ± 7.4 95.2 ± 4.6 101.4 ± 8.1 107.6 ± 3.8 105.7 ± 7.1 93.0 ± 3.4 95.3 ± 6.1 105.1 ± 5.6b 88.9 ± 1.9b 94.3 ± 2.5c
500 102.0 ± 8.3 108.9 ± 11.7 98.9 ± 1.7 114.6 ± 0.8b 118.4 ± 4.5 116.8 ± 3.8 89.1 ± 1.9 89.7 ± 2.2 107.2 ± 5.9b,c 86.2 ± 3.6b 86.5 ± 5.4b
750 99.5 ± 6.1 104.3 ± 9.7 98.4 ± 2.5 124.1 ± 20.7 111.2 ± 2.7 120.1 ± 16.4 89.2 ± 3.3 90.1 ± 1.1 105.3 ± 7.8c 85.4 ± 4.9b 88.6 ± 1.7b
1000 92.8 ± 4.1 99.8 ± 11.3 98.9 ± 1.3 153.5 ± 5.5b 100.5 ± 4.8 148.9 ± 15.2b 69.3 ± 4.5 81.4 ± 5.4b 94.2 ± 7.5b 67.3 ± 4.7b 75.2 ± 4.5b
a
Mean of triplicate.
b
Indicates significant difference in comparison to the emulsion without active ingredient, at significance level of 5%.
c
Indicates significant difference in comparison to the emulsion with panthenol, at significance level of 5%.

According to literature (Maier et al., 1991) the 3T3 cell line is
considered to be more sensitive to irritants and sensitizers than
the HaCat keratinocytes, although the emulsions applied in both
lines did not show great differences. Cell viability was higher for
the HaCat when exposed to the 29R and 29R.VE in comparison to
the 3T3, at concentrations from 31.25 to 500 mg/ml, but no signif-
icant differences were found. The same response was observed
when exposed to 51LC.VE and 51S.VE, however at higher concen-
trations (750 and 1000 mg/ml), cell viability was significantly
different.
On the other hand, 51LC, 51S, 37N, and 37M and showed to be
more cytotoxic to HaCat than to the 3T3 fibroblasts and viability
was significantly different at higher concentrations (750 and
1000 mg/ml).
The emulsions 37N and 37M, which are formulated with the
same pair of surfactants in the same proportions, produced at all
the concentrations tested a significant decrease in cellular viability
of HaCaT cells when compared to the other macroemulsions tested
(29R, 51S and 51LC). This effect could be related to the nature of
the emulsion (nanoemulsion type) beyond the pair of surfactants,
but to our knowledge, there are no studies related to cytotoxicity
of nanoemulsions to confirm this assumption. However, it is
important to highlight that the concentration range used to per-
form this assay was greatly superior to those normally employed
in cosmetic safety tests. Therefore, even in the light of these viabil-
ity values, and having no cytotoxic effects at these higher concen-
tration ranges we could assume no risk of skin irritation. When we
compared the cytotoxic effects of emulsions containing active
ingredient with those that did not, we observed that, the emulsions
produced with vitamin E acetate presented increased cellular via-
bility in all the samples analyzed, and were more effective in HaCat
than in the 3T3. Significant differences in viability values were ob-
served at different concentrations between the formulations 29R
and 29R.VE (1000 mg/ml), and 51S and 51S.VE in fibroblasts 3T3
at low concentrations (31.25 to 250 mg/ml), and between the
51LC and 51LC.VE, 51S and 51.SVE and 37N and 37N.VE in kerati-
nocytes HaCat.
These results could be explained by the synergic anti-oxidant
effect of the vitamin E acetate associated to the carotenoids present
in the Buriti (M. flexuosa) oil. This association may protect the cell
membrane against oxidative stress caused by the non-ionic surfac-
tants present in the emulsions. According to the literature, the
exposure of the cellular membrane to non-ionic surfactants such
as Polysorbate 80 and Cremophor EL, promote a decrease in their
glutathione levels leading to an increase of the hydrogen peroxide
cytotoxicity, which cause oxidative stress (Hirama et al., 2004;
Iwase et al., 2004; Tatsuishi et al., 2005). Previous studies have
shown, that beta-carotene, the major carotenoid in Buriti oil (Gar-
cía-Quiroz et al., 2003) combined with vitamin E are highly effec-
tive against oxy-radicals, offering an additional protection against
cell damage (Böhm et al., 1998). Vitamin E acetate is a free radical
scavenger and can reduce DNA damage and keratinocyte death
(Mcvean and Liebler, 1997). In addition, it can enhance stratum
corneum hydration and reduce skin roughness (Mayer et al., 1993).
For the results of cytotoxicity, the emulsions containing D-pan-
thenol (37N.P and 37M.P), showed in HaCat similar viability values
when compared to cell response to the formulations 37N and 37M
without this compound. Only when the cells were treated with a
concentration of 1000 mg/ml of the 37N.P, a significant increase
in cellular viability was observed. The cellular viability of 3T3
fibroblasts exposed to the formulations 37N.P and 37M.P, at con-
centrations of 62.5, 250 and 750 mg/ml, was significantly lower
compared to 37N and 37M. This response could be explained by
the lower resistance of 3T3 to irritants in comparison to the
 C.F. Zanatta et al. / Food and Chemical Toxicology 46 (2008) 2776–2781 2779

keratinocytes HaCat. Furthermore, the D-panthenol did not prove
to be effective against oxidative damage caused by surfactants in
this assay. Slyshenkov et al. in 2004 reported that the pantothenic
acid and its derivates were very efficient against cell damage pro-
duced by oxygen free radicals, by increasing the levels of glutathi-
one (Slyshenkov et al., 2004), a thiol-containing compound that is
the most abundant cell anti-oxidant (Lu, 1999; Dickinson et al.,
2003). However, the protection mechanism induced by D-panthe-
nol involves the biosynthesis of glutathione, which a time-consum-
ing process and it is possible that 10-min exposure was not enough
time to allow this reaction to happen.
The cytotoxicity of emulsions containing the same components
in the same proportions, but prepared by different emulsification
method, were compared. We aimed to discern whether the way
in which the surfactants are disposed on the droplets’ interface
and thus the skin’s exposure to them, can alter their cytotoxic
effects. Comparing nanoemulsions and multiple emulsions, no sig-
nificant differences were found in viability values obtained for
emulsions prepared by different emulsification method (described
in Methods section), although this would influence the size and
morphology of the droplets of each type of emulsion.
For macroemulsions, 51LC presented similar viability values in
fibroblasts 3T3 and lower in keratinocytes HaCat compared to
51S. Interestingly, 51S.VE showed a tendency to be less cytotoxic
3T3, probably due to the faster liberation of the Vitamin E from
its droplets when compared to the 51LC.VE, which emulsion con-
tain liquid crystals evolving the droplets. This structures form a
barrier to the diffusion of the active and thus delaying its release
and action (Brinon et al., 1999). Nevertheless, no significant differ-
ences were observed as cytotoxic effects were more dependent on
concentration and surfactant type than on the way they are dis-
posed on the droplets interface.
3.2. Surfactants cytotoxicity
HaCat and 3T3 cell lines have proven to be viable for the study
of surfactant cytotoxicity, as an alternative to in vivo studies on
skin irritation (Benavides et al., 2004; Sánchez et al., 2004. Some
surfactants employed in emulsions attainment did not undergo
cytotoxicity evaluation due to their high viscosity, as they
surfactants tested and are presented as dose–response curves
(Fig. 1). The release of the neutral red by the lysosomes increased
with the increase of surfactant concentration. The IC50 values
obtained (Table 4) revealed that the SpanÒ 80 was the least cyto-
toxic surfactant analyzed, showing lower viability values in 3T3
than in HaCat, although no significant differences were observed.
UltroilÒ R400 presented significant lower IC50 values when com-
pared to SpanÒ 80 and was also significantly more toxic to 3T3
than to HaCat.
The SpanÒ 80 and UltroilÒ R400 mixture used in emulsions’
attainment, showed lower viability values in comparison to SpanÒ
80 samples and higher values compared to UltroilÒ R400 samples.
The cytotoxicity of the mixture was higher in 3T3 than in HaCat
cells, but significant differences were found at the highest concen-
tration analyzed. Based on the percentage of SpanÒ 80 (44%) and
UltroilÒ R400 (66%) used in their association to produce the emul-
sions (37N, 37N.P, 37N.VE, 37M and 37M.P) and the IC50 obtained
for each surfactant, we have calculated a prediction of IC50 value
for the surfactant mixture, in order to compare with the experi-
mental values attained. For 3T3 fibroblasts, the IC50 value
predicted was 206 mg/ml and the value experimentally obtained
was 164.8 mg/ml, these values suggested that the 3T3 could be
more sensitive to the SpanÒ 80 and UltroilÒ R400 mixture than
to these surfactants isolated. On the other hand, for HaCat querati-
nocytes, the cytotoxicity results showed an IC50 value of 262 mg/
ml, significantly higher than the value (204 mg/ml) empirically cal-
culated, suggesting that their mixture were less cytotoxic than the
pure surfactants.
This could be explained by a possible interaction between the
surfactants that would promote a synergic effect on the cytotoxic-
ity, more evident in 3T3 cells due to their lower resistance to
chemicals in comparison to HaCat cells.
UnitolÒ CE200F was the most cytotoxic surfactant tested and
significantly more toxic to 3T3 than to HaCat. We also observed
that the cellular viability values of the fibroblasts and keratino-
Table 4
Values of IC50 and IC80 for the surfactants in 3T3 fibroblasts and HaCat keratinocytes
Surfactant IC50 (mg/ml) IC80 (mg/ml)

remained stuck to the cells, thus causing a false-negative result. 3T3 Hacat 3T3 Hacat
The effect of the surfactants on cell membrane integrity was Span 80 Ò
199.5 ± 2.1 237.2 ± 9.4 99.3 ± 14.6 85.9 ± 27.9
measured using the NRR, where the living cells has the ability of UltroilÒ R400 161.9 ± 27.4 197.4 ± 43.6 56.1 ± 14.9 54.4 ± 5.2
take up the neutral red and in case of damage will release it. SpanÒ 80 + UltroilÒ R400 164.8 ± 4.3 325 ± 83.1 69.8 ± 2.3 76 ± 25.8
UnitolÒ CE200F 0.82 ± 0.7 10.7 ± 2.9 0.39 ± 0.04 1.51 ± 1.1
The results of the NRR carried out in HaCat keratinocytes and
3T3 fibroblasts were obtained at different concentrations for the Values are expressed as Mean ± SD of at least three independent experiments.

120
Unitol CE200F Unitol CE200F
120
A Span 80
Ultroil R400
B Span 80
110 110 Ultroil R400
Span 80 and Ultroil R400 Span 80 and Ultroil R400
100 100

90 90
Celular Viability (%)
Celular viability (%)

IC80
80 80 IC80

70 70

60 60

50
IC50 50 IC50
40
40
30
30
20
20
0 25 50 75 100 125 150 175 200
0 25 50 75 100 125 150 175 200
Concentration (mg/ml) Concentration (mg/ml)

Fig. 1. Dose–response curves of the cytotoxicity of the surfactants used in emulsions formulation, in 3T3 fibroblasts (A) and HaCat keratinocytes (B). The cellular viability of
50 and 80% used to calculate the concentrations inducing a 50% and 80% of cellular viability (IC50 and IC80) are indicated in the figures.
2780 C.F. Zanatta et al. / Food and Chemical Toxicology 46 (2008) 2776–2781
cytes exposed to UnitolÒ CE200F were significantly lower when
compared to all the others surfactants evaluated.
According to other authors (Preté et al., 2002), the concentra-
tion of membrane solubilization within the non-ionic polyoxyeth-
ylene surfactants, which causes 100% of hemolysis, decreases with
the increase of the hydrophobic chain, reflecting the importance of
hydrophobic interaction in the hemolytic process. Considering that
solubilization involves hydrophilic interactions between the sur-
factant chain and the lipids or proteins of the cell membrane, this
effect could explain the significant differences found between the
IC50 values of UnitolÒ CE200F and the others surfactants tested.
The former possess 18:3 carbon chain and 20 moles of polyoxyeth-
ylene, while SpanÒ 80 contains a smaller hydrophobic portion.
Nevertheless, UltroilÒ R400 possesses a long carbon chain, as it is
a polymer composed of 40 sub-units of ethylene oxide (PEG-40
castor oil), however it is less hydrophobic than UnitolÒ CE200F,
which only contains 20 units of ethylene oxide. These chemical
characteristics are closely related to the hydrophilic–lipophilic bal-
ance (HLB) of the surfactant. This property was correlated with
hemolytic activity, as well as the critical micelle concentration
(CMC), by many authors (Söderlind and Karlsson, 2006; Vinardell
and Infante, 1999), since the presence of micelles is a prerequisite
for surfactant solubilization of substances in aqueous solution.
Taking into account the dilutions made for the NRR assay and
the concentration of the surfactants used to produce the emulsions
(23 mg of SpanÒ 80, 27 mg of UltroilÒ R400 or 32 mg of UnitolÒ
CE200F per gram of emulsion), we concluded that the surfactants
isolated, especially UnitolÒ CE200F, are more cytotoxic to cells than
when emulsified with Buriti oil and water.
At a concentration of 1000 mg of emulsion/ml of PBS, the emul-
sions containing UnitolÒ CE200F showed cellular viability equal to
104.1% and 113.3% for 3T3 cells, while for HaCat cells were 98.9%
and 100.5%, respectively for 51LC and 51S (Table 2 and 3). How-
ever, the concentration of UnitolÒ CE200F present in this emulsion
dilution is 16 mg/ml of PBS, a much higher value than the IC50
found for this surfactant in both cell lines. What suggest that the
Buriti oil was capable to significantly reduce the cytotoxicity of this
surfactant? For SpanÒ 80 and UltroilÒ R400, no significant differ-
ences were found between the cytotoxicity values attained for
emulsions containing 23 mg of SpanÒ 80 and 27 mg of UltroilÒ
R400, and the values obtained for the same concentrations of pure
surfactants and their mixture.
4. Conclusions
According to the results obtained, we conclude that emulsions
formulated with Buriti oil and commercial surfactants seem non
irritant to skin due to their low cytotoxic effects in the 3T3 and
HaCat cell monolayers. The addition of vitamin E not only
decreased the toxicity of the products, but also increased the cellu-
lar viability. The action of panthenol should be studied for a longer
time, in order to examine a possible increase in the anti-oxidant
effect in cells, which would reduce the cellular damage caused
by the surfactants. When emulsified with Buriti oil, water and Brij
72, Unitol CE200F showed less cytotoxic effects than when tested
alone.
Conflict of interest statement
The authors declare that there are no conflicts of interest.
Acknowledgements
This work was funded by CAPES (Brazil) and project 501 (Fun-
dació Bosch i Gimpera). Acknowledgements to CRODA, Oxiteno
and, ISP Corp for material supply. The authors are grateful to Robin
Rycroft for linguistic assistance.
References
Barany, E., Lindberg, M., Loden, M., 1999. Biophysical characterization of skin
damage and recovery after exposure to different surfactants. Contact Dermatitis
40 (1999), 98–103.
Benavides, T., Mitjans, M., Martínez, V., Clapés, P., Infante, M.R., Clothier, R.H.,
Vinardell, M.P., 2004. Assessment of primary eye and skin irritants by in vitro
cytotoxicity and phototoxicity models: an in vitro approach of new arginine-
based surfactant-induced irritation. Toxicology 197, 229–237.
Böhm, F., Edge, R., McGarvey, D.J., Truscott, T.G., 1998. Beta-carotene with vitamins
E and C offers synergistic cell protection against NOx. FEBS Lett. 436, 387–389.
Brinon, L., Geiger, S., Alard, V., Doucet, J., Tranchant, J.F., Couarraze, G., 1999.
Percutaneous absorption of sunscreens from liquid crystalline phases. J. Control.
Release 60, 67–76.
Cantrell, A., McGarvey, D.J., Truscott, T.G., Rancan, F., Böhm, F., 2003. Singlet oxygen
quenching by dietary carotenoids in a model membrane environment. Arch.
Biochem. Biophys. 412, 47–54.
Chen, J., Jarvi, M., Lo, P.C., Stefflova, K., Wilson, B.C., Zheng, G., 2007. Using the
singlet oxygen scavenging property of carotenoid in photodynamic molecular
beacons to minimize photodamage to non-targeted cells. Photochem.
Photobiol. Sci. 6, 1311–1317.
Conn, P.F., Schalch, W., Truscott, T.G., 1991. The singlet oxygen and carotenoid
interaction. J. Photochem. Photobiol. B. 11, 41–47.
Dickinson, D.A., Moellering, D.R., Iles, K.E., Patel, R.P., Levonen, A.L., Wigley, A.,
Darley-Usmar, V.M., Forman, H.J., 2003. Cytoprotection against oxidative stress
and the regulation of glutathione synthesis. Biological Chemistry 384, 527–
537.
Dijoux, N., Guingand, Y., Bourgeois, C., Durand, S., Fromageot, C., Combe, C., Ferret,
P., 2006. Assessment of the phototoxic hazard of some essential oils using
modified 3T3 neutral red uptake assay. Toxicol. In Vitro 20 (2006), 480–489.
Edge, R., McGarvey, D.J., Truscott, T.G., 1997. The carotenoids as anti-oxidants. J.
Photochem. Photobiol. B: Biol. 41, 189–200.
Ferrati, M., Moscato, G., Imbriani, M. 2005. Allergic cutaneous diseases in
hairdressers.
García-Quiroz, A., Moreira, S.G.C., De Morais, A.V., Silva, A.S., Da Rocha, G.N.,
Alcantara, P., 2003. Physical and chemical analysis of dielectric properties and
differential scanning calorimetry techniques on Buriti oil. Instrum. Sci. Technol.
31, 93–101.
Gettings, S.D., Lordo, R.A., Hintze, K.L., Bagley, D.M., Casteton, P.L., Chudkowski, M.,
et al., 1996. The CTFA evaluation of alternatives program: an evaluation of
in vitro alternatives to the Draize primary eye irritation test. (Phase III)
surfactant-based formulations. Food Chem. Toxicol. 34, 79–117.
Ghersetich, I., Troiano, M., De Giorgi, V., Lotti, T., 2007. Receptors in skin ageing and
antiageing agents. Dermatol. Clin. 25, 655–662.
Hirama, S., Tatsuishi, T., Iwase, K., Nakao, H., Umebayashi, C., Nishizaki, Y.,
Kobayashi, M., Ishida, S., Okano, Y., Oyama, Y., 2004. Flow-cytometric analysis
on adverse effects of polysorbate 80 in rat thymocytes. Toxicology 199, 137–
143.
Iwase, K., Oyama, Y., Tatsuishi, T., Yamaguchi, J., Nishimura, Y., Kanada, A.,
Kobayashi, M., Maemura, S., Ishida, Y., Okano, Y., 2004. Cremophor EL
augments the cytotoxicity of hydrogen peroxide in lymphocytes dissociated
from rat thymus glands. Toxicol. Lett. 154, 143–148.
Korting, H.C., Herzinger, T., Hartinger, A., Kerscher, M., Angerpointner, T., Maibach,
H.I., 1994a. Discrimination of the irritancy potential of surfactants in vitro by
two cytotoxicity assays using normal human keratinocytes, HaCaT cells and 3T3
mouse fibroblasts: correlation with in vivo data from a soap chamber assay. J.
Dermatol. Sci. 7, 119–129.
Korting, H.C., Schindler, S., Hartinger, A., Kerscher, M., Angerpointner, T., Maibach,
H.I. 1994b. MTT-assay and neutral red release (NRR)-assay: relative role in the
prediction of the irritancy potential of surfactants.
Lu, S.C., 1999. Regulation of hepatic glutathione synthesis: current concepts and
controversies. FASEB J. 13, 1169–1183.
Maier, K., Schmitt-Landgraf, R., Siegemund, B., 1991. Development of an in vitro test
system with human skin cells for evaluation of phototoxicity. Toxicol. In Vitro 5,
457–461.
Mayer, P., Pittermann, W., Wallat, S., 1993. The effects of vitamin E on the skin
cosmet. Toil 108, 99–109.
Mcvean, M., Liebler, D.C., 1997. Inhibition of UVB induced DNA photodamage in
mouse epidermis by topically applied alpha-tocopherol. Carginogenesis 18,
1617–1622 (Med. Lav. 96, 102–18).
Orton, D.I., Wilkinson, J.D., 2004. Cosmetic allergy: incidence, diagnosis, and
management. Am. J. Clin. Dermatol. 5, 327–337.
Osborne, R., Perkins, M.A., 1994. An approach for development of alternative test
methods based on mechanisms of skin irritation. Food Chem. Toxicol. 32, 133–
142.
Pauwels, M., Rogiers, V., 2007. EU legislations affecting safety data availability of
cosmetic ingredients. Regul. Toxicol. Pharmacol. 49, 308–315.
Preté, P.S.C., Gomes, K., Malheiros, S.V.P., Meirelles, N.C., de Paula, E., 2002.
Solubilization of human erythrocyte membranes by non-ionic surfactants of the
polyoxyethylene alkyl ethers series. Biophys. Chem. 97, 45–54.
Rao, A.V., Rao, L.G., 2007. Carotenoids and human health. Pharmacol. Res. 55, 207–
216.
 C.F. Zanatta et al. / Food and Chemical Toxicology 46 (2008) 2776–2781
Rieger, M.M., 1994. Cosmetic use of selected natural fats and oils. Cosmet. Toiletries
109, 57–68.
Rogiers, V., Pauwels, M., 2006. Good science must be the key factor in the
development and use of alternative methods for safety assessment of cosmetics.
ALTEX 223, 346–352.
Sánchez, L., Mitjans, M., Infante, M.R., Vinardell, M.P., 2004. Assessment of the
potential skin irritation of lysine-derivative anionic surfactants using mouse
fibroblasts and human keratinocytes as an alternative to animal testing. Pharm.
Res. 21, 1637–1641.
Santos, O.D.H., Miotto, J.V., Moraes, J.M., Oliveira, W.P., Rocha-Filho, P.A., 2005.
Attainment of emulsions with liquid crystal from marigold oil using the
required HLB method. J. Dispers. Sci. Technol. 26, 243–249.
Slyshenkov, V.S., Dymkowska, D., Wojtczak, L., 2004. Pantothenic acid and
pantothenol increase biosynthesis of glutathione by boosting cell energetics.
FEBS Lett. 569, 169–172.
Söderlind, E., Karlsson, L., 2006. Haemolytic activity of maltopyranoside surfactants.
Eur. J. Pharm. Biopharm. 62, 254–259.
2781
Stahl, W., Sies, H., 2007. Carotenoids and flavonoids contribute to nutritional
protection against skin damage from sunlight. Mol. Biotechnol. 37, 26–30
(review).
Tadros, T., Izquierdo, P., Esquena, L., Solans, C., 2004. Formation and stability of
nano-emulsions. Adv. Colloid Interface Sci. 20, 303–318.
Tatsuishi, T., Oyama, Y., Iwase, K., Yamaguchi, J., Kobayashi, M., Nishimura, Y.,
Kanada, A., Hirama, H., 2005. Polysorbate 80 increases the susceptibility to
oxidative stress in rat thymocytes. Toxicology 207, 7–14.
Vinardell, M.P., Infante, M.R., 1999. The relationship between the chain length of
non-ionic surfactants and their hemolytic action on human erythrocytes.
Comput. Biochem. Physiol. C 124, 117–120.
Vinardell, M.P., Mitjans, M., 2007. Alternative methods for eye and skin irritation
tests: an overview. J. Pharm. Sci. 97, 46–59.
Wilhelm, K.P., Boèttjer, B., Siegers, C.P., 2001. Quantitative assessment of primary
skin irritants in vitro in a cytotoxicity model: comparison with in vivo human
irritation tests. Brit. J. Dermatol. 145, 709–715.