See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/328972464

Support for the Safe Use of Zinc Oxide Nanoparticle Sunscreens: Lack of Skin
Penetration or Cellular Toxicity after Repeated Application in Volunteers

Article  in  Journal of Investigative Dermatology · November 2018

DOI: 10.1016/j.jid.2018.08.024

CITATIONS READS

102 727

8 authors, including:

Yousuf Hussain Mohammed Isha Nanthini Haridass

The University of Queensland Curtin University
63 PUBLICATIONS   1,616 CITATIONS    17 PUBLICATIONS   427 CITATIONS   

SEE PROFILE SEE PROFILE

Washington Yamandu Sanchez

Translational Research Institute
19 PUBLICATIONS   845 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Magnesium and the skin View project

Toxicokinetic View project

All content following this page was uploaded by Michael Roberts on 03 September 2019.

The user has requested enhancement of the downloaded file.

 ORIGINAL ARTICLE See related commentary on pg 277

Support for the Safe Use of Zinc Oxide

Nanoparticle Sunscreens: Lack of Skin
Penetration or Cellular Toxicity after
Repeated Application in Volunteers
Yousuf H. Mohammed1, Amy Holmes2, Isha N. Haridass1,3, Washington Y. Sanchez1, Hauke Studier2,4,
Jeffrey E. Grice1, Heather A.E. Benson3 and Michael S. Roberts1,2
Zinc oxide is a widely used broad-spectrum sunscreen, but concerns have been raised about the safety of its
nanoparticle (NP) form. We studied the safety of repeated application of agglomerated zinc oxide (ZnO) NPs
applied to human volunteers over 5 days by assessing the skin penetration of intact ZnO-NPs and zinc ions and
measuring local skin toxicity. Multiphoton tomography with fluorescence lifetime imaging microscopy was used
to directly visualize ZnO-NP skin penetration and viable epidermal metabolic changes in human volunteers. The
fate of ZnO-NPs was also characterized in excised human skin in vitro. ZnO-NPs accumulated on the skin surface
and within the skin furrows but did not enter or cause cellular toxicity in the viable epidermis. Zinc ion con-
centrations in the viable epidermis of excised human skin were slightly elevated. In conclusion, repeated
application of ZnO-NPs to the skin, as used in global sunscreen products, appears to be safe, with no evidence of
ZnO-NP penetration into the viable epidermis nor toxicity in the underlying viable epidermis. It was associated
with the release and penetration of zinc ions into the skin, but this did not appear to cause local toxicity.
Journal of Investigative Dermatology (2019) 139, 308e315; doi:10.1016/j.jid.2018.08.024

INTRODUCTION 61% in 2014 (Cancer Council Australia, 2017). A key

Repeated long-term exposure to UVR causes skin aging and
increased risk of skin cancer (Seite et al., 2010a, 2010b).
Strategies to protect against the harmful effects of UVR,
including regular use of efficient sunscreens, are strongly
advocated as likely to lead to better health outcomes for in-
dividuals and cost savings for health care systems (Gordon
et al., 2009; Mancuso et al., 2017). Although there is no
direct evidence to show that zinc oxide (ZnO) nanoparticles
(NPs) are effective in preventing human cancers, they are
commonly used in sunscreen products, often in combination
with chemical sunscreens. However, some public advocacy
groups have questioned the safety of nanoparticulate-based
sunscreens, and a 2017 National Sun Protection Survey by
Cancer Council Australia found that only 55% of Australians
believed it was safe to use sunscreen every day, down from
1 Toxicology Section, Office of Scientific Evaluation,
Therapeutics Research Centre, The University of Queensland Diamantina
Institute, Translational Research Institute, Brisbane, Australia; 2School of
Pharmacy and Medical Sciences, University of South Australia, Sansom
Institute, City East Campus, Adelaide, Australia; 3School of Pharmacy and
Biomedical Sciences, Curtin Health Innovation Research Institute, Curtin
University of Technology, Perth, Western Australia; and 4Becker and Hickl,
Berlin, Germany
Correspondence: Michael Roberts, Level 5, Translational Research Institute,
37 Kent Street, Woolloongabba, Queensland 4102, Australia. E-mail:
m.roberts@uq.edu.au
Abbreviations: CI, confidence interval; FAD, flavin adenine dinucleotide;
FLIM, fluorescence lifetime imaging microscopy; MPT, multiphoton tomog-
raphy; NAD(P)H, reduced nicotinamide adenine dinucleotide phosphate;
NP, nanoparticle; SHG, second harmonic generation; ZnO, zinc oxide
Received 5 June 2018; revised 8 August 2018; accepted 23 August 2018;
corrected proof published online 15 November 2018
concern is whether avoiding the use of such sunscreens could
increase the incidence of skin cancer.
This article addresses the human safety of ZnO-NP, a
broad-spectrum sunscreen that blocks exposure to UVA
(320e400 nm) and UVB (290e320 nm) radiation and can
contribute to a high sun protection factor rating when
formulated optimally. Engineered ZnO-NPs are transparent
sunscreen products that provide high UVR protection and
cosmetic elegance (Nohynek et al., 2007) and reduce the
reliance on chemical sunscreen agents that can be absorbed
systemically (Gonzalez et al., 2006; Hayden et al., 1997).
A key unresolved issue with ZnO-NPs is their local safety,
with suggestions that they may penetrate the stratum corneum
barrier, gain access to viable epidermis cells, and cause
potentially toxic responses (European Commission Scientific
Committee on Consumer Safety, 2012; Senjen, 2012;
Therapeutics Goods Administration of Australia, 2017).
Toxicity concerns are largely based on cell culture studies
showing that ZnO-NPs can induce cytotoxicity, oxidative
stress, and DNA and metabolic damage when applied for less
than 6 hours to human liver cells (HepG2), epidermal cells
(A431), keratinocytes (NCTC2544), and fibroblasts (Akhtar
et al., 2012; Alarifi et al., 2014; Kocbek et al., 2010; Sharma
et al., 2009). The only study of repeated topical application
of 68ZnO-NP sunscreen to human volunteers found 68Zn ions
in the blood and urine (Gulson et al., 2010, 2012), although the
penetration of intact ZnO-NPs was not proven.
At this time, no studies have directly assessed the skin
penetration and local safety of ZnO-NP after repeated topical

308 Journal of Investigative Dermatology (2019), Volume 139 ª 2018 The Authors. Published by Elsevier, Inc. on behalf of the Society for Investigative Dermatology.
 YH Mohammed et al.
Support for Safety of Nanoparticle Sunscreens

application to humans in vivo. The published studies have amplitude-weighted lifetimes (sm) (Figures 1c and 2c) of
been limited to healthy volunteers (Filipe et al., 2009; Leite- reduced nicotinamide adenine dinucleotide phosphate
Silva et al., 2013, 2016a, 2016b; Lin et al., 2011; Zvyagin (NAD(P)H) and flavin adenine dinucleotide (FAD) from the
et al., 2008) or psoriatic patients (Prow et al., 2011) after a viable epidermis at various strata between untreated/vehicle
single application and in vitro studies with excised human controls and repeated ZnO-NP treatments. In addition, no
skin (Baroli et al., 2007; Cross et al., 2007; Holmes et al., morphological changes in the viable epidermis keratinocytes
2016; Zvyagin et al., 2008). These have indicated that skin associated with cellular toxicity or apoptosis (i.e., cell
penetration after a single topical application of ZnO-NP is shrinkage, blebbing) were observed.
limited to the superficial layers of the stratum corneum, with
Concentration profile of ZnO-NP within the skin after
minimal or no penetration into the viable epidermis. How-
repeated topical application
ever, these single-application reports have been criticized as
Representative pseudocolored MPT-FILM images of the stra-
being nonrepresentative of actual consumer use (Senjen,
tum granulosum after repeated hourly and daily treatments of
2012). This concern is exacerbated by the potential of
ZnO-NP are shown in Figures 3a and 4a, respectively: the top
ZnO-NP lipophilic coatings to promote skin penetration
row shows average amplitude-weighted lifetime, the middle
(Osmond-McLeod et al., 2013; Smijs and Pavel, 2011) and
row displays only the ZnO-NP SHG signal (pseudocolored to
the observations of penetration of other nanomaterials into
indicate concentration), and the bottom row is an overlay of
human (Tinkle et al., 2003) and animal (Monteiro-Riviere
the two (with autofluorescence in gray).
et al., 2011) skin.
The ZnO-NPs were detected with SHG/hyper Rayleigh
We assessed the presence of intact ZnO-NP in skin layers
scattering optical phenomena, which can be detected as an
by multiphoton tomography (MPT), with detection of a
ultrafast fluorescence lifetime (almost instantaneous) emis-
second harmonic generation (SHG) signal. SHG occurs in
sion at half the excitation wavelength. This SHG signal
specific nonlinear materials such as ZnO-NP when two
appeared to be localized in the skin furrows at concentrations
identical excitation photons interact with the matrix and
comparable to the administered dose, with no evidence of a
combine to form a single emission photon with exactly twice
ZnO-NP SHG signal within the viable epidermis. Figures 3
the frequency and half the wavelength (Darvin et al., 2012).
and 4b display line intensity plots of the ZnO-NP SHG
Changes in skin properties due to the presence of exogenous
signal corresponding to each treatment group. The data show
ZnO-NPs were evaluated by the addition of fluorescence
signal intensity peaks corresponding to the furrows, due to
lifetime imaging microscopy (FLIM) to multiphoton tomog-
accumulation of both uncoated and coated ZnO-NPs. Within
raphy (MPT). In this technique, fluorophores, including
the viable epidermis, ZnO-NP signal intensities in treated
endogenous fluorophores in the skin, are excited by laser
groups are comparable to the nonspecific background levels
irradiation, and the time taken for them to return to the
in the untreated and vehicle control groups. Scatterplots of
ground state is measured as the fluorescence lifetime, typi-
ZnO-NP concentrations after hourly and daily repeated
cally measured in picoseconds to nanoseconds. Changes in
topical application are displayed in Figures 3d and 4d,
fluorescence lifetimes can be indicative of interactions with
respectively.
exogenous materials, cellular degeneration, or cellular
For repeated hourly topical application (Figure 3d), the
toxicity resulting from exposure (Sanchez et al., 2015). To
mean concentrations measured in the viable epidermis from
further explore the fate of ZnO-NP on the skin surface,
representative regions of interest of the five volunteers were
parallel studies were carried out in excised human skin using
0.48% (95% confidence interval [CI] ¼ e0.6 to 1.57) and
a labile zinc-specific probe, ZinPyr-1, to detect zinc ions in
0.71% (95% CI ¼ e1.0 to 2.4) of the topical dose for un-
the skin.
coated and coated ZnO-NPs, respectively. These levels were
RESULTS not significantly different from nonspecific background un-
The uncoated and coated ZnO-NPs had a mean size
stan- treated (0.06%, 95% CI ¼ 0.03e0.1) and vehicle (0.08%,
dard error of the mean of 74.0
3.1 nm and 65.0 nm
3.4 95% CI ¼ 0e0.15) controls.
nm, respectively. Scanning electron cryomicroscopy showed After repeated daily topical application (Figure 4d), the
that the ZnO-NPs were agglomerated in the topical product. mean levels of uncoated and coated ZnO-NPs in the viable
epidermis measured from the representative regions of in-
Localization and distribution of uncoated and coated
terest of five volunteers were 0.1% (95% CI ¼ 0.07e0.14)
ZnO-NPs in human skin in vivo after repeated topical
and 0.08% (95% CI ¼ 0.06e0.09), respectively. Again, the
application
levels of ZnO-NPs in the viable epidermis after daily
Representative multiphoton tomography multispectral images
repeated treatment were not significantly different from the
of in vivo human skin after repeated hourly (over 6 hours) and
nonspecific background signals measured in the untreated
daily (over 5 days) topical application of uncoated and
(0.09%, 95% CI ¼ 0.06e0.1) or vehicle (0.08%, 95% CI ¼
coated ZnO-NPs are shown in Figures 1a and 2a, respec-
0.04e0.12) controls.
tively. The lack of a ZnO-NP SHG signal (red) at any depth in
In summary, hourly or daily repeated topical treatments of
the viable epidermis (pseudocolored green-blue) shows that
ZnO-NPs did not lead to significantly elevated ZnO-NP
ZnO-NPs were not present in the viable epidermis after either
penetration into the viable epidermis.
of these treatment regimes. Rather, the ZnO-NP were local-
ized on the stratum corneum surface and in the skin furrows. Labile zinc distribution in ex vivo human skin samples
There were no statistically significant differences in the ZinPyr-1 is a bright fluorescent probe based on the fluores-
free-to-bound ratios (ɑ1:ɑ2) (Figures 1b and 2b) and average cein structure that is cell permeable and fluoresces in the

www.jidonline.org 309
 YH Mohammed et al.
Support for Safety of Nanoparticle Sunscreens

Figure 1. Effect of hourly topical

application of uncoated and coated
ZnO-NPs on human skin in vivo for 6
hours. Negative (untreated) and
vehicle controls (CCT) are included.
(a) Representative pseudocolored
multispectral images (lEm ¼ 370e390
nm [red], lEm ¼ 450e515 nm [green],
lEm ¼ 515e620 nm [blue]) of different
skin strata after treatment. Images
were pseudocolored to show cellular
autofluorescence (green/blue) and
ZnO-NP (red). (b, c) Metabolic and
redox-associated changes in FLIM
channel 2 (lEm: 450-515 nm) within
the stratum granulosum, stratum
spinosum and stratum basale
following treatment. (b) Average redox
(a1/a2) ratios obtained following
treatment. (c) Average NAD(P)H/FAD
fluorescence lifetime (sm) obtained
following treatment. Data represent
mean lifetime (ps) or ratio
95% CI
(n ¼ 5). The white scale bar
represents 40 mm. au, arbitrary unit;
CCT, caprylic/capric triglyceride;
Em, emission; FAD, flavin
adenine dinucleotide; FLIM,
fluorescence lifetime imaging
microscopy; NAD(P)H, reduced
nicotinamide adenine dinucleotide
phosphate; NP, nanoparticle; ZnO,
zinc oxide.

presence of labile zinc in biological systems (Walkup et al., DISCUSSION

2000). MPT images of ZinPyr-1estained full-thickness skin
overlaid with the SHG channel corresponding to both ZnO-
NP and collagen show increased labile zinc in the viable
epidermis (Figure 5). White arrows illustrate the presence of
ZnO-NP on the stratum corneum surface in the ZnO-
NPedosed skin and not in control skin, although SHG is also
present because of the collagen fibers of the dermis
(Figure 5aec, eeg). After repeated hourly application of
either coated or uncoated ZnO-NPs, there was an increase in
solubilized zinc species within the stratum corneum and the
viable epidermis compared with the control (Figure 5aec).
The signal intensity for ZinPyr-1 for two selected regions of
interest shows that the solubilized zinc species increased in
the skin after repeated dosing of coated ZnO-NP (P < 0.05,
n ¼ 6) compared with control skin (Figure 5d).
After repeated daily ZnO-NP dosing (Figure 5eeg), an
increase of solubilized zinc species was observed
compared with the control. Figure 5h shows the increased
ZinPyr-1 signal intensity after application of ZnO-NP in
caprylic/capric triglycerides compared with control. It is
evident that repeated daily application of ZnO-NP leads to
an increase in the concentration of zinc species in the skin.
The greatest increase was observed when coated ZnO-NPs
were applied daily, compared with uncoated ZnO-NPs,
although both coated and uncoated ZnO-NPs resulted in
a higher solubilized zinc signal in the viable epidermis; in
all but one case the differences were not statistically
significant.
In summary, repeated application of ZnO-NPs to human skin
in vivo over several days did not result in NP penetration
through the stratum corneum or cause any visible morpho-
logical or redox changes. This finding extends the earlier
observations that single ZnO-NP applications to human skin
in vivo are associated with minimal skin penetration and
local toxicity (Leite-Silva et al., 2013, 2016a, 2016b; Lin
et al., 2011). These findings contrast with those of Bakand
et al. (2017), who suggested that cellular toxicity arises af-
ter ZnO-NPs are applied to human skin fibroblasts in cell
culture. Deng et al. (2015) also suggested that NP-based
sunscreens can penetrate skin and cause local toxicity.
However, these researchers used murine models that poorly
represent the effective barrier property of human skin (Scott
et al., 1986). Deng et al. (2015) argued that NP skin pene-
tration could be minimized by using sunscreen NPs that
adhere to the skin surface, yet the available evidence from
single and repeated dosing studies on human skin shows that
ZnO-NPs do remain on the skin surface.
This study adds to the body of knowledge used by various
health-related regulatory bodies to assess the safety of topi-
cally applied sunscreen NPs (European Commission
Scientific Committee on Consumer Safety, 2012; Senjen,
2012; Toxicology Section, Office of Scientific Evaluation,
Therapeutics Goods Administration of Australia, 2017). In
particular, we addressed the knowledge gap of what happens
when coated and uncoated ZnO-NPs are applied by repeated
topical administration under in-use conditions, using a highly

310 Journal of Investigative Dermatology (2019), Volume 139


sensitive, in vivo, noninvasive, multiphoton approach to
evaluate NP skin penetration and distribution after realistic
in-use administration of sunscreen products. Repeated hourly
applications simulate in-use conditions of outdoor activities
such as sports or swimming, and repeated daily application
rates simulate standard everyday activities. The approach also
enabled potential local skin toxicity in the various skin layers
of human volunteers to be assessed for changes in skin
morphology and redox state, an early marker of apoptosis
change. Finally, we corroborated our in vivo findings with
those of in vitro studies, in which the same ZnO-NP formu-
lations were repeatedly applied to excised human skin and
the penetrations of solid ZnO-NP and labile zinc species
were quantified.
Our repeated application results extend the earlier prop-
osition that single-dose administration of ZnO-NP (Leite-Silva
et al., 2013, 2016a, 2016b; Lin et al., 2011) is not associated
with any skin penetration and local toxicity. Our work also
brings into context the work of Gulson et al. (2010, 2012),
who reported small but significantly elevated zinc levels in
the blood and urine of volunteers after repeated applications
of ZnO-NP that were more than 99% enriched with the stable
isotope 68Zn. Localized skin toxicity was not examined in
that work, and the detection methodology (acid digestion
followed by inductively coupled plasma mass spectrometry)
did not permit determination of whether 68Zn had been
absorbed in a particulate or a soluble form. In contrast, in this
work we used the specific chelating agent ZinPyr-1 to map
the deposition of solubilized zinc species in ex vivo human
skin (Holmes et al., 2016) to show that the form of zinc
penetrating through the skin is most likely to be the zinc ion,
Zn2þ. The increase of labile zinc in the viable epidermis
YH Mohammed et al.
Support for Safety of Nanoparticle Sunscreens
Figure 2. Effect of daily application
of uncoated and coated ZnO-NPs on
human skin in vivo for 5 days.
Negative and vehicle controls are
included. (a) Representative
pseudocolored multispectral
images (lEm ¼ 370e390 nm [red],
lEm ¼ 450e515 nm [green], lEm ¼
515e620 nm [blue]) of different
skin strata after treatment. The images
were pseudocolored to show cellular
autofluorescence (green/blue) and
ZnO-NP (red). (b, c) Metabolic- and
redox-associated changes in FLIM
channel 2 (lEm ¼ 450e515 nm) in
the stratum granulosum, stratum
spinosum, and stratum basale after
treatment. (b) Average redox (a1:a2)
ratios obtained after treatment. (c)
Average fluorescence NAD(P)H/FAD
lifetime (sm) obtained after treatment.
Data represent mean lifetime (ps) or
ratio
95% confidence interval
(n ¼ 5). Scale bar ¼ 40 mm. Em,
emission; FAD, flavin
adenine dinucleotide;
FLIM, fluorescence lifetime imaging
microscopy; NAD(P)H, reduced
nicotinamide adenine dinucleotide
phosphate; NP, nanoparticle; ZnO,
zinc oxide.
shown here is also most likely due to permeation of zinc ions
liberated from ZnO-NPs accumulated on the stratum cor-
neum and in the skin furrows, promoted by the hydrolysis of
ZnO-NP to Zn2þ at the normal acidic pH of the skin. The
resulting zinc ions in the viable epidermis would then
redistribute in vivo into the systemic circulation and result in
a nonsignificant increase in zinc ion concentrations. The
redistribution of isotopic zinc ions in the viable epidermis,
liberated from the hydrolysis of intact ZnO-NPs on the skin
surface and zinc ion penetration into the viable epidermis,
explains the low isotopic zinc levels reported in the blood
and urine of human volunteers by Gulson et al. (2010, 2012).
ZnO-NP levels of less than 20 mg/ml were reported to
induce cytotoxicity, oxidative stress, and DNA and meta-
bolic damage in human liver cells (HepG2), human
epidermal cells (A431), human keratinocytes (NCTC2544)
and fibroblasts after less than 6 hours (Akhtar et al., 2012;
Alarifi et al., 2014; Bakand et al., 2017; Kocbek et al.,
2010; Sharma et al., 2009). However, we observed no
comparable changes, such as changes in cellular
morphology or evidence of necrosis or apoptosis, after
repeated topical application of ZnO-NP to in vivo human
skin over 5 days. We conclude that the toxic effects that may
be elicited in isolated cells after exposure to ZnO-NP
in vitro do not occur when they are applied to the skin of
humans under realistic in-use conditions. This is supported
by the findings of O’Keefe et al. (2016) that despite causing
similar levels of in vitro cytotoxicity in human monocytes
and macrophages compared with chemical UV filters, ZnO-
NPs were likely to be at least one order of magnitude less
cytotoxic in vivo, because of their very low rates of skin
penetration.
www.jidonline.org 311
 YH Mohammed et al.
Support for Safety of Nanoparticle Sunscreens

Figure 3. Visualization and quantification of ZnO-NPs in the viable epidermis and furrows after hourly application of uncoated and coated ZnO-NPs for 6 hours.
Negative (untreated) and vehicle control groups are included. (a) First row: representative multiphoton tomography with fluorescence lifetime imaging microscopy
images of the viable epidermis, pseudocolored to indicate fluorescence lifetimes (0e1,500 ps). Second row: pseudocolored spectral images (lEm ¼ 370e390 nm) to
indicate the concentration (mg/ml) of uncoated and coated ZnO-NPs. Third row: merging of multiphoton tomography with fluorescence lifetime imaging microscopy
and spectral images. (b) Line intensity plot of the ZnO-NP signal profile along a dotted line the viable epidermis after treatment. (c) Average (percentage of applied dose)
ZnO-NP quantified in the regions of interest in the viable epidermis (purple box) and (d) in the furrows (green box). Scale bar ¼ 40 mm. CCT, caprylic/capric triglyceride;
NP, nanoparticle; ROI, region of interest; ZnO, zinc oxide.

It is important to note the health care importance of our
findings. The benefits of providing regular and ongoing skin
protection from UVR exposure are well documented (Green
et al., 2011; Hughes et al., 2013; Thompson et al., 1993; van
der Pols et al., 2006), yet chemical sunscreens have been
shown to be systemically absorbed (Gonzalez et al., 2006). A
real concern is that the conflicting views provided to the
public regarding sunscreen and NP safety may be inhibiting
sunscreen use. Indeed, the recent National Sun Protection
Survey by the peak Australian nongovernment cancer control
organization, Cancer Council Australia, recorded a decline in
the past 3 years from 61% to 55% of Australians who believe
it is safe to use sunscreen every day (Cancer Council
Australia, 2017). This study supports the stated position of
Cancer Council Australia that on the current weight of evi-
dence, ZnO-NPs remain on the surface and outer layers of
the stratum corneum and do not reach viable cells. The po-
tential risk of increased skin cancer due to a reduction in
sunscreen use is far greater than could conceivably result
from NP toxicity (Slevin, 2012). The significance of our study,
in the context of a need for confidence in sun protection
measures such as the use of aesthetically acceptable and
effective ZnO-NP products, is evident in the estimated
13,941 new cases of and 1,839 deaths from melanoma in
Australia in 2017 (Cancer_Australia, 2018).
MATERIALS AND METHODS
Uncoated (Z-COTE; BASF, Ludwigshafen, Germany) and coated
(Z-COTE HP1 ZnO-NP, manufacturer estimated particle size < 200
nm) were supplied as a dry white powder. Z-COTE HP1 is composed
of 98% ZnO and 2% triethoxycaprylylsilane, a silicone derivative-
based hydrophobic coating, whereas Z-COTE is composed of
100% ZnO. These NPs are commonly used in commercial sunscreen
products. ZnO-NPs were dispersed in caprylic/capric triglyceride
(10% weight/weight) for application to the skin. ZnO-NPs were
characterized using a JEOL (Munich, Germany) scanning electron
microscope (JSM7100F) and JEOL 1011 transmission electron mi-
croscope, as described previously (Leite-Silva et al., 2016a).
Repeated topical application of ZnO-NP
Five healthy volunteers (ages 20e30 years) with undamaged skin
and no active skin disease gave written informed patient consent
(University of Queensland Human Research Ethics Committee
approval number 2007/197-2008001342) and abstained from using
sunscreen or cosmetic skin product use for 24 hours before the
experiment.
Treatments of the vehicle (caprylic/capric triglyceride) control and
uncoated and coated ZnO-NPs were topically applied at a rate of 2
mg/cm2 (based on the standard protocol for determination of the sun
protection factor of sunscreen products) over 4-cm2 areas on each
volunteer’s volar forearm repeatedly over two time periods: one
application every hour to the same site for 6 hours and one appli-
cation per day to the same site for 5 days. The hourly and daily
applications were performed at comparable sites on opposite arms.
MPT-FLIM images were acquired at the end of the hourly treatment
on the first day and approximately 6 hours after the final treatment
on day 5.
MPT with fluorescence lifetime imaging
A multiphoton tomograph DermaInspect (JenLab, Jena, Germany)
equipped with a tunable titanium sapphire femtosecond laser Mai

312 Journal of Investigative Dermatology (2019), Volume 139

 YH Mohammed et al.
Support for Safety of Nanoparticle Sunscreens

Figure 4. Visualization and quantification of ZnO-NPs in the viable epidermis and furrows after daily application of uncoated and coated ZnO-NP for 5 days.
Negative (untreated) and vehicle controls are included. (a) First row: representative multiphoton tomography with fluorescence lifetime imaging microscopy
images of the viable epidermis, pseudocolored to indicate fluorescence lifetimes (0e1,500 ps). Second row: Pseudocolored spectral images (lEm ¼ 370e390
nm) to indicate the concentration (mg/ml) of uncoated and coated ZnO-NP. Third row: Merging of multiphoton tomography with fluorescence lifetime imaging
microscopy and spectral images. (b) Line intensity plot of the ZnO-NP signal profile along a dotted line the viable epidermis after treatment. (c) Average
(percentage of applied dose) ZnO-NP quantified in the regions of interest in the viable epidermis (purple box) and (d) in the furrows (green box). Scale bar ¼ 40
mm. CCT, caprylic/capric triglyceride; Em, emission; NP, nanoparticle; ROI, region of interest; SHG, second harmonic generation; ZnO, zinc oxide.

Tai (Spectra Physics, Mountain View, CA) and a time-correlated
single photon-counting module SPC-830 (Becker & Hickl, Berlin,
Germany) was used for imaging. Time-correlated single photon-
counting was performed by building up a photon distribution
based on the arrival times of the detected photons for each pixel in
the scanned area, constructing a decay curve over 256 time
channels within a 10-ns interval (less than the 13-ns pulse width)
between two laser pulses. Three Becker & Hickl photo multiplier
detectors PMC-100 were integrated into the MPT system, enabling
splitting of the fluorescence emission into three wavelength ranges.
The bandpass filters were 400/100 nm, 482/65 nm, and 567/105 nm.
To spectrally select the ZnO-NP SHG signal (and, to a degree, hyper

Figure 5. Hourly (aed) and daily (eeh) topical application of uncoated and coated ZnO-NPs on ex vivo human skin. Representative overlay images of
ZinPyr-1 fluorescence (gray; lEx ¼ 488 nm, lEm ¼ 520e560 nm) and second harmonic generation (pink; lEx ¼ 800 nm; lEm ¼ 405/10 nm) for skin treated with
(a, e) vehicle control, (b, f) coated ZnO-NP (10% weight/weight in caprylic/capric triglycerides), and (c, g) uncoated ZnO-NPs (10% weight/weight in caprylic/
capric triglyceride). White arrows indicate ZnO-NP on the stratum corneum surface. Scale bar ¼ 20 mm. (d, h) ZinPyr-1 signal intensities for the stratum
corneum and viable epidermis after hourly and daily applications, respectively. Data represent the mean fluorescence
standard error of the mean (n ¼ 6).
**P < 0.05. AU, arbitrary unit; DE, dermis; Em, emission; Ex, excitation; NP, nanoparticle; SC, stratum corneum; VE, viable epidermis; ZnO, zinc oxide.

www.jidonline.org 313
 YH Mohammed et al.
Support for Safety of Nanoparticle Sunscreens
Rayleigh scattering), an additional bandpass filter of 380/20 nm was
used in the first detection channel. The excitation wavelength was
740 nm, and the laser power 31 mW in the sample plane. Volunteer
skin was imaged at four depths: the stratum corneum (w5e10 mm),
stratum granulosum (w15e20 mm), stratum spinosum (25e30 mm),
and stratum basale (35e40 mm). Keratinocyte morphology was used
to ensure the correct imaging depth of these layers. FLIM data were
then recorded as described previously (Leite-Silva et al., 2016a)
using 256 time channels. MPT-FLIM images were analyzed with
Becker & Hickl SPCImage 5.2 software as color-coded FLIM images.
The extent of ZnO-NP penetration was quantified using a standard
curve constructed by measuring the photon count of the SHG signal
with a 27  27-pixel square area from known concentrations (0.05, 0.1,
0.5, 1, 5, 10, 50 and 100 mg/ml) of uncoated and coated ZnO-NPs
suspended in caprylic/capric triglyceride on a glass slide. In the
volunteer experiments, regions of interest with identical dimensions
were selected from the viable epidermis and furrows of in vivo MPT-
FLIM images of human skin after the repeated topical application of
ZnO-NP.
Cellular toxicity in various strata of the viable epidermis after
topical treatments was assessed by measuring average amplitude-
weighted lifetimes (sm) and ratios of the free-to-bound amplitude
coefficients (ɑ1:ɑ2) of NAD(P)H and FAD. The average lifetime, sm,
consists of a short component, s1, associated with free NAD(P)H and
FAD, and a long component, s2, associated with bound NAD(P)H
and FAD. The amplitude coefficients ɑ1 and ɑ2 are in turn associated
with these short and long lifetimes, and the ratio ɑ1:ɑ2 indicates the
proportion of these fluorophores in free and bound states. A
reduction in ɑ1:ɑ2 signifies decreased binding and an inhibition of
oxidative phosphorylation that is inversely proportional to the
metabolic rate of the cells in the analyzed part of the image. These
parameters are used as rapid and sensitive indicators of oxidative
stress, toxicity, viability, and the redox state of the cells (Deka et al.,
2013; Heikal, 2010; Sanchez et al., 2010). Fluorescence lifetimes of
the endogenous fluorophores NAD(P)H and FAD in skin are unique,
and a change in the microenvironment of the fluorophore as a result
of exogenous or proinflammatory processes can be easily detected
as a deviation in the average lifetimes (Sanchez et al., 2010).
Furthermore, inspection of MPT-FLIM images allowed visual
morphological examination of keratinocytes, and a comparison of
the cell and nucleus size was performed to provide an estimate of
cellular health.
In vitro Franz diffusion cell study using ex vivo human skin
Written informed patient consent was obtained from a female patient
(age ¼ 36 years) to donate skin after elective abdominoplasty (Queen
Elizabeth Hospital Protocol-2009208). Subcutaneous fat was
removed by blunt dissection, and the skin was mounted in a Franz-
type diffusion cell. After barrier integrity check by transepithelial
electrical resistance (Cross and Roberts, 2008), 0.1% sodium azide
solution was placed in the receptor, and dose formulations were
applied at 2 mg/cm2 to the skin (replicating in vivo dosing). Franz cells
were unoccluded, placed in a water bath at 35
1 C, and constantly
agitated by a magnetic flea. At completion, excess formulation was
removed with a moist swab, and the skin surface was blotted dry. Skin
samples were then snap frozen (e80  C), embedded in optimal cutting
temperature media, cryosectioned to 20 mm (CM1950 cryostat; Leica,
Wetzlar, Germany), placed on a poly-L-lysine coated microscope
slide, and kept frozen on dry ice until staining. A 10-mmol/L aqueous
solution of ZinPyr-1, a selective fluorescent probe for labile zinc, was
314 Journal of Investigative Dermatology (2019), Volume 139
applied to skin sections (10 ml) and left at ambient temperature in the
dark for 15 minutes.
Labile zinc distribution within ex vivo human skin samples
Imaging of ex vivo skin cryosections stained with ZinPyr-1 (n ¼ 6)
was conducted on a Zeiss LSM 710 multiphoton microscope
equipped with an argon gas laser (488 nm) and a tunable titanium
sapphire laser (Mai Tai, Spectra Physics). ZinPyr-1 was detected at
excitation wavelength 488 nm and emission wavelengths 520560
nm (laser power of 1.2 mW and 500 gain). SHG was detected at 800
nm using a bandpass filter of 405/10 nm at a constant laser power of
9 mW and 700 gain.
Statistics
The data presented in Figures 1e4 (n ¼ 5) and Figure 5 (n ¼ 6) are
expressed as mean with 95% CI. Statistical differences between the
means were evaluated using a Kruskal-Wallis test with post hoc
Dunn multiple comparisons test. Statistical differences between the
ZinPyr-1 intensities were assessed using one-way analysis of vari-
ance with post hoc Bonferroni multiple comparison test. All analyses
were conducted using GraphPad Prism (La Jolla, CA), with a P value
of less than 0.05 defined as statistically significant.
CONFLICT OF INTEREST
The authors state no conflict of interest.
ACKNOWLEDGMENTS
MSR acknowledges support from the National Health and Medical Research
Council of Australia for grants APP1107356 and APP1055176. We
acknowledge Elizabeth Ryan, Carole Pilorget, Anne-Laure Penloup, and
Wolfgang Becker for providing technical support.
REFERENCES
Akhtar MJ, Ahamed M, Kumar S, Khan MM, Ahmad J, Alrokayan SA. Zinc
oxide nanoparticles selectively induce apoptosis in human cancer cells
through reactive oxygen species. Int J Nanomed 2012;7:845e57.
Alarifi S, Ali D, Alakhtani S, Al Suhaibani ES, Al-Qahtani AA. Reactive oxygen
species-mediated DNA damage and apoptosis in human skin epidermal
cells after exposure to nickel nanoparticles. Biol Trace Elem Res 2014;157:
84e93.
Bakand S, Hayes A, Dechsakulthorn F. Nanotoxicity and safety evaluation of
nanoparticles in sunscreen products in vitro. Paper presented at: Austral-
asian College of Toxicology and Risk Assessment 10th Annual Scientific
Meeting. 27e29 September 2017; Canberra, Australia.
Baroli B, Ennas MG, Loffredo F, Isola M, Pinna R, Lopez-Quintela MA.
Penetration of metallic nanoparticles in human full-thickness skin. J Invest
Dermatol 2007;127:1701e12.
Cancer_Australia. Melanoma of the Skin Statistics, https://melanoma.
canceraustralia.gov.au/statistics; 2018 (accessed 7 March 2018).
Cancer Council Australia. Almost Half of Australians Confused About Sunscreen,
http://www.cancer.org.au/news/media-releases/almost-half-of-australians-
confused-about-sunscreen.html; 2017 (accessed 27 October 2017).
Cross SE, Innes B, Roberts MS, Tsuzuki T, Robertson TA, McCormick P.
Human skin penetration of sunscreen nanoparticles: in-vitro assessment of
a novel micronized zinc oxide formulation. Skin Pharmacol Physiol
2007;20:148e54.
Cross SE, Roberts MS. Use of in vitro human skin membranes to model and
predict the effect of changing blood flow on the flux and retention of
topically applied solutes. J Pharm Sci 2008;97:3442e50.
Darvin ME, König K, Kellner-Hoefer M, Breunig HG, Werncke W,
Meinke MC, et al. Safety assessment by multiphoton fluorescence/second
harmonic generation/hyper-Rayleigh scattering tomography of ZnO nano-
particles used in cosmetic products. Skin Pharmacol Physiol 2012;25:
219e26.
Deka G, Wu WW, Kao FJ. In vivo wound healing diagnosis with second
harmonic and fluorescence lifetime imaging. J Biomed Opt 2013;18(6):
061222.

Deng Y, Ediriwickrema A, Yang F, Lewis J, Girardi M, Saltzman WM. A sunblock
based on bioadhesive nanoparticles. Nat Mater 2015;14:1278e85.
European Commission Scientific Committee on Consumer Safety. Opinion on
Zinc Oxide (Nano Form), http://ec.europa.eu/health/scientific_committees/
consumer_safety/docs/sccs_o_103.pdf; 2012 (accessed 1 November 2017).
Filipe P, Silva JN, Silva R, Cirne de Castro JL, Marques Gomes M, Alves LC,
et al. Stratum corneum is an effective barrier to TiO2 and ZnO nanoparticle
percutaneous absorption. Skin Pharmacol Physiol 2009;22:266e75.
Gonzalez H, Farbrot A, Larko O, Wennberg AM. Percutaneous absorption of
the sunscreen benzophenone-3 after repeated whole-body applications,
with and without ultraviolet irradiation. Br J Dermatol 2006;154:337e40.
Gordon LG, Scuffham PA, van der Pols JC, McBride P, Williams GM, Green AC.
Regular sunscreen use is a cost-effective approach to skin cancer prevention
in subtropical settings. J Invest Dermatol 2009;129:2766e71.
Green AC, Williams GM, Logan V, Strutton GM. Reduced melanoma after
regular sunscreen use: randomized trial follow-up. J Clin Oncol 2011;29:
257e63.
Gulson B, McCall M, Korsch M, Gomez L, Casey P, Oytam Y, et al. Small
amounts of zinc from zinc oxide particles in sunscreens applied outdoors
are absorbed through human skin. Toxicol Sci 2010;118:140e9.
Gulson B, Wong H, Korsch M, Gomez L, Casey P, McCall M, et al. Com-
parison of dermal absorption of zinc from different sunscreen formulations
and differing UV exposure based on stable isotope tracing. Sci Total En-
viron 2012;420:313e8.
Hayden CG, Roberts MS, Benson HA. Systemic absorption of sunscreen after
topical application. Lancet 1997;350(9081):863e4.
Heikal AA. Intracellular coenzymes as natural biomarkers for metabolic ac-
tivities and mitochondrial anomalies. Biomark Med 2010;4:241e63.
Holmes AM, Song Z, Moghimi HR, Roberts MS. Relative penetration of zinc
oxide and zinc ions into human skin after application of different zinc
oxide formulations. ACS Nano 2016;10:1810e9.
Hughes MC, Williams GM, Baker P, Green AC. Sunscreen and prevention of
skin aging: a randomized trial. Ann Intern Med 2013;158:781e90.
Kocbek P, Teskac K, Kreft ME, Kristl J. Toxicological aspects of long-term
treatment of keratinocytes with ZnO and TiO2 nanoparticles. Small
2010;6:1908e17.
Leite-Silva VR, Lamer ML, Sanchez WY, Liu DC, Sanchez WH, Morrow I,
et al. The effect of formulation on the penetration of coated and uncoated
zinc oxide nanoparticles into the viable epidermis of human skin in vivo.
Eur J Pharm Biopharm 2013;84:297e308.
Leite-Silva VR, Liu DC, Sanchez WY, Studier H, Mohammed YH, Holmes A,
et al. Effect of flexing and massage on in vivo human skin penetration and
toxicity of zinc oxide nanoparticles. Nanomedicine (Lond) 2016a;11:
1193e205.
Leite-Silva VR, Sanchez WY, Studier H, Liu DC, Mohammed YH,
Holmes AM, et al. Human skin penetration and local effects of topical nano
zinc oxide after occlusion and barrier impairment. Eur J Pharm Biopharm
2016b;104:140e7.
Lin LL, Grice JE, Butler MK, Zvyagin AV, Becker W, Robertson TA, et al. Time-
correlated single photon counting for simultaneous monitoring of zinc
oxide nanoparticles and NAD(P)H in intact and barrier-disrupted volunteer
skin. Pharm Res 2011;28:2920e30.
Mancuso JB, Maruthi R, Wang SQ, Lim HW. Sunscreens: an update. Am J Clin
Dermatol 2017;18:643e50.
Monteiro-Riviere NA, Wiench K, Landsiedel R, Schulte S, Inman AO,
Riviere JE. Safety evaluation of sunscreen formulations containing titanium
dioxide and zinc oxide nanoparticles in UVB sunburned skin: an in vitro
and in vivo study. Toxicol Sci 2011;123:264e80.
Nohynek GJ, Lademann J, Ribaud C, Roberts MS. Grey goo on the skin?
Nanotechnology, cosmetic and sunscreen safety. Crit Rev Toxicol 2007;37:
251e77.
View publication stats
YH Mohammed et al.
Support for Safety of Nanoparticle Sunscreens
O’Keefe SJ, Feltis BN, Piva TJ, Turney TW, Wright PF. ZnO nanoparticles and
organic chemical UV-filters are equally well tolerated by human immune
cells. Nanotoxicology 2016;10:1287e96.
Osmond-McLeod MJ, Osmond RI, Oytam Y, McCall MJ, Feltis B, Mackay-
Sim A, et al. Surface coatings of ZnO nanoparticles mitigate differentially a
host of transcriptional, protein and signalling responses in primary human
olfactory cells. Part Fibre Toxicol 2013;10(1):54.
Prow TW, Grice JE, Lin LL, Faye R, Butler M, Becker W, et al. Nanoparticles
and microparticles for skin drug delivery. Adv Drug Deliv Rev 2011;63:
470e91.
Sanchez WY, Pastore MN, Haridass IN, Konig K, Becker W, Roberts MS.
Fluorescence lifetime imaging of the skin. In: Becker W, editor. Advanced
time-correlated single photon counting applications. Basel, Switzerland:
Springer; 2015. p. 457e508.
Sanchez WY, Prow TW, Sanchez WH, Grice JE, Roberts MS. Analysis of
the metabolic deterioration of ex vivo skin from ischemic necrosis
through the imaging of intracellular NAD(P)H by multiphoton tomog-
raphy and fluorescence lifetime imaging microscopy. J Biomed Opt
2010;15(4):046008.
Scott RC, Walker M, Dugard PH. A comparison of the in vitro permeability
properties of human and some laboratory animal skins. Int J Cosmet Sci
1986;8(4):189e94.
Seite S, Christiaens F, Bredoux C, Compan D, Zucchi H, Lombard D,
et al. A broad-spectrum sunscreen prevents cumulative damage from
repeated exposure to sub-erythemal solar ultraviolet radiation repre-
sentative of temperate latitudes. J Eur Acad Dermatol Venereol
2010a;24:219e22.
Seite S, Fourtanier A, Moyal D, Young AR. Photodamage to human skin by
suberythemal exposure to solar ultraviolet radiation can be attenuated by
sunscreens: a review. Br J Dermatol 2010b;163:903e14.
Senjen R. Submission regarding SCCS Opinion on Zinc oxide (nano form), http://
www.ntn.org.au/wp/wp-content/uploads/2012/11/NTA-SCCS-submission-Oct-
2012.pdf; 2012 (accessed 1 November 2017).
Sharma V, Shukla RK, Saxena N, Parmar D, Das M, Dhawan A. DNA
damaging potential of zinc oxide nanoparticles in human epidermal cells.
Toxicol Lett 2009;185:211e8.
Slevin T. How Worried Should We Be About Nanoparticles in Sun-
screen?, http://www.cancer.org.au/news/blog/risks/how-worried-should-
we-be-about-nanoparticles-in-sunscreen.html; 2012 (accessed 24
November 17).
Smijs TG, Pavel S. Titanium dioxide and zinc oxide nanoparticles in sun-
screens: focus on their safety and effectiveness. Nanotechnol Sci Appl
2011;4:95e112.
Thompson SC, Jolley D, Marks R. Reduction of solar keratoses by regular
sunscreen use. N Engl J Med 1993;329:1147e51.
Tinkle SS, Antonini JM, Rich BA, Roberts JR, Salmen R, DePree K, et al. Skin
as a route of exposure and sensitization in chronic beryllium disease. En-
viron Health Perspect 2003;111:1202e8.
Toxicology Section, Office of Scientific Evaluation, Therapeutics Goods
Administration of Australia. Literature Review on the Safety of Titanium
Dioxide and Zinc Oxide Nanoparticles in Sunscreens, https://www.tga.gov.
au/literature-review-safety-titanium-dioxide-and-zinc-oxide-nanoparticles-
sunscreens; 2017 (accessed 1 November 2017).
van der Pols JC, Williams GM, Pandeya N, Logan V, Green AC. Prolonged
prevention of squamous cell carcinoma of the skin by regular sunscreen
use. Cancer Epidemiol Biomark 2006;15:2546e8.
Walkup GK, Burdette SC, Lippard SJ, Tsien RY. A new cell-permeable fluo-
rescent probe for Zn2þ. J Am Chem Soc 2000;122:5644e5.
Zvyagin AV, Zhao X, Gierden A, Sanchez W, Ross JA, Roberts MS. Imaging of
zinc oxide nanoparticle penetration in human skin in vitro and in vivo.
J Biomed Opt 2008;13(6):064031.
www.jidonline.org 315