Revised_Small-20_Kanaras-20manuscript

DOI: 10.1002/ ((please add manuscript number))
Article type: Full Paper,
Interactions of Skin with Gold Nanoparticles of Different Surface Charge, Shape and
Functionality.
Rute Fernandes, Neil R. Smyth, Otto L. Muskens, Simone Nitti, Amelie Heuer-Jungemann
Michael R. Ardern-Jones, and Antonios G. Kanaras*
Ms. Rute Fernandes, Ms. Amelie Heuer-Jungemann, Prof. Otto L. Muskens, Dr. Antonios G.
Kanaras
Institute of Life Sciences, Physics and Astronomy, Faculty of Applied and Physical Sciences,
University of Southampton, Southampton, SO171BJ, UK
E-mail:
Mr. Simone Nitti
Istituto Italiano di Technologia, Via Morego 30, 16163 Genova, Italy
Dr. Neil R. Smyth, Faculty of Natural and Environmental Sciences, University of
Southampton, SO17 1BJ, UK
Dr. Michael R. Ardern-Jones
Faculty of Medicine, Southampton General Hospital, University of Southampton,
Southampton, SO17 1BJ, UK
Keywords: gold nanoparticles, peptides, functionality, skin penetration
Abstract
We investigate the interactions between skin and colloidal gold nanoparticles of different
physicochemical characteristics. By systematically varying the charge, shape and
functionality of gold nanoparticles, we assess the nanoparticle penetration through the
different skin layers. The penetration is evaluated both qualitatively and quantitatively using a
variety of complementary techniques. Inductively coupled plasma optical emission
spectrometry (ICP-OES) is used to quantify the total number of particles penetrated into the
skin structure. Transmission electron microscopy (TEM) and two photon photoluminescence
microscopy (TPPL) on skin cross sections provide a direct visualization of nanoparticle
migration within the different skin substructures. Our studies reveal that gold nanoparticles
functionalized with cell penetrating peptides (CPPs) TAT and R 7 are found in the skin in larger
quantities than polyethylene glycol functionalized nanoparticles and are able to enter deep

1


into the skin structure. The systematic studies presented in this work can be of strong interest
for new developments in transdermal administration of drugs and therapy.
1. Introduction
The utilization of nanoparticles in biomedicine holds potential for important developments in
drug delivery, imaging, diagnosis and therapy.[1-4] This is further fueled by significant
advancements in nanoparticle chemical synthesis and surface functionalization which in many
cases allow pre-designing the properties of the functional nanomaterial. [5-12] Having available
a rich library of nanomaterials, one of the biggest challenges is to understand the nanoparticle
behavior when introduced to biological structures.[13-14] Currently many studies focus on how
the morphology, charge and ligand capping of nanoparticles influence their cellular fate. [15-19]
For example, in earlier work it was shown how the physicochemical characteristics of gold
nanoparticles impact on the number of particles taken up or exocytosed by endothelial cells. [2021]

It was also recently discussed how functional nanoparticles can prevent or accelerate the

organization of endothelial cells to create blood vessels. [22] Understanding the basic rules that
govern nanoparticle-cell interactions at the subcellular level is of critical importance for the
development of new biomedical applications exploiting inorganic nanoparticles. However,
equally important and not yet well-understood, is how the physicochemical characteristics of
nanoparticles influence their interactions with complex tissues, which extend at larger scales
than the cells.
A typical example of such a complex tissue structure is the skin. Understanding how the
morphology, charge and function of nanoparticles influences their penetration through the
different layers of the skin would lead to better design of nanoparticles and the development
of new transdermal drug delivery methods. On the other hand, nanoparticle design rules to
minimize skin penetration would be of great interest to the cosmetic industry and health and

safety regulations in an industrial environment.
2


Many studies have recently targeted the subject of nanoparticle-skin interactions. [23-28] For
example, Sonavane et al.[29] studied the penetration of 15nm, 102nm and 198nm citrate-coated
gold nanoparticles through rat skin using Franz diffusion cells. Their TEM, EDS and ICP
analyses showed that the penetration of gold nanoparticles through rat skin is sizeindependent. however the 15nm gold nanoparticles showed higher permeation compared to
the larger particles. It is noteworthy to say here that Franz diffusion cell set-ups, although very
commonly used for skin experiments, result in the skin being exposed to excessive pressure
and shear stress, which can affect the penetration. Krishnan et al. [30] and Filon et al.[31] induced
the penetration of citrate-coated gold nanoparticles through human skin by dermaportation
using a pulsed electromagnetic field and by dermabrasion, respectively. Krishnan et al.[30]
concluded that 10 nm gold nanoparticles do not penetrate intact human skin; however the
stratum corneum penetration was enhanced by the pulsed electromagnetic field. On the other
hand, Filon et al.[31] reported on the penetration of 12.9 nm gold nanoparticles into both the
epidermis and dermis of human skin. Their ICP results further showed that a significantly
higher gold amount was found in damaged skin compared to intact skin. Huang and coworkers showed that 5nm PVP-coated gold nanoparticles are skin permeable. [32] They
attributed the permeability to the nano-bio interaction with skin lipids and the consequent
induction of transient and reversible openings on the stratum corneum. Furthermore, when
they applied a mixture of gold nanoparticles and protein drugs, both were able to penetrate the
skin barrier and migrate into the deep layers. Labouta and co-workers published a number of
studies on the penetration of gold nanoparticles with human skin. [33-38] Their data showed that
15nm citrate-coated gold nanoparticles in aqueous solution tended to aggregate on the
superficial stratum corneum after 24 h exposure, while 6nm dodecanethiol-coated gold
nanoparticles in toluene penetrated through the stratum corneum and into viable epidermal
layers of human skin. In another study by the same group

[34]

the penetration of four model

gold nanoparticles (15nm citrate-coated in water, 6nm dodecanethiol-coated in toluene, 6nm
3


lecithin-coated in water and 15nm cetrimide-coated in toluene through human skin was
investigated using multiphoton microscopy. They found that the correct skin exposure time
(>6h) was crucial in order to have a significant penetration extent for studying the effect of
the different physicochemical, formulation and environmental factors.
Due to the complexity of the skin penetration experiments and in some cases the lack of a
systematic experimental approach, often the conclusions of different studies have been
contradictory as to whether nanoparticles do or do not penetrate the skin. [39-40] Experimental
parameters such as skin type and condition (e.g. intact skin, and skin chemically or
mechanically treated to enhance penetration), skin surface application area, exposure time,
skin maintenance during the experiment, application vehicle (e.g. solvent, emulsion)
concentration and type of nanoparticles (chemical composition, size, shape and functionality)
play a key role in the evaluation of nanoparticle penetration through the skin and should be
carefully considered in order to obtain reliable conclusions.
In this paper we report a systematic study of the interactions of gold nanoparticles with exvivo mouse and human skin. Utilizing complementary characterization techniques we evaluate
for the first time how the penetration of gold nanoparticles through skin is influenced by the
charge, morphology and function of the nanoparticles. For these experiments, we employed
gold nanospheres and nanorods of a well-defined size distribution and a well-understood
pegylated coating as well as spherical gold nanoparticles containing cell penetration peptides.

2. Results and Discussion
2.1. Skin Structure
In general, the skin structure consists of three major layers, the epidermis (which is the top
layer of the skin), dermis, and the hypodermis (see a schematic illustration and detailed
explanation in supporting information and Figure S1).

4


In our experiments we utilized two types of skin: Human skin explants from the breast area
of an adult and skin from the back of neonatal mice. The choice of skin substrates is very
important and can influence the level of penetration. Although the main skin structure is
universal in mammalians, there are variations to the skin characteristics related to the species
and the part of the body where skin is coming from. For example, mouse skin has a markedly
thinner epidermis than human skin, but in adults has a greater number of hair follicles. The
age of the subject is also critical with the skin having a higher possibility to be scarred in aged
organisms. All these parameters must be taken into account when conclusions are drawn and
they also determine the design of the experiment as it will be discussed later on. The human
skin used in our experiments has a low density of hair follicles while the skin from the newly
born mouse is in the early phases of hair follicle development.
Another parameter, which plays a critical role and it is not always taken into account is the
integrity of the skin during culture. When ex-vivo experiments are conducted, it is critical that
the skin remains in a good condition during the course of the experiment. Cultured skin can
start to degenerate relatively fast, which can falsify the experimental observations. Thus, prior
to any experiments it is important to identify the maximum incubation period for which the
skin sample retains its structure. For this purpose the skin morphology was monitored for 24h
using histological experiments. In our experiments, the human skin retained its integrity for a
maximum of 24h and the mouse skin for a maximum of 6h as shown in skin cross sections in
Figure 1. For comparison, a figure where the skin has lost its integrity with signs of cellular
degeneration is shown in supporting information (Figure S2).
2.2. Nanoparticles
Seven different types of gold nanoparticles were employed to study their penetration through
skin. Gold nanoparticles can be easily synthesized with different morphologies in narrow
distributions and their surface chemistry is well-established. We systematically tuned three
major nanoparticle characteristics namely: charge, shape and function. For this purpose, we

5


synthesized gold nanospheres (NSs) with a narrow size distribution (15 nm ±1) following the
Turkevich method [57] and we functionalized them with thiol containing polyethyelene glycols
with either a terminal amine (positively charged particles) or a carboxylic group (negatively
charged particles). High molecular weight polyethyelene glycols (5000 Da) were chosen
based on their widely accepted biocompatibility.
Gold nanorods (NRs) with an aspect ratio 2.8±0.5 were prepared following well-established
protocols.[41] For consistency, these nanorods were coated with the same types of polyethylene
glycols as the gold nanospheres. To make sure that both nanorods and nanospheres had a
compact layer of polyethylene glycols on their surface we performed a multi-step coating as
previously reported for similar systems.[42]
To assign a particular functionality to the particles we coated gold nanospheres with three
different types of peptides. These are the CALNN, CALNNTat, and CALNNR 7. The first
batch of particles was coated only with CALNN peptide, the second with a mixed monolayer
of CALNN/CALNNTat and the third with a mixed monolayer of CALNN/CALNNR 7. While
CALNN is a well-studied peptide, [43] which facilitates the stabilization of gold nanoparticles,
Tat and R7 have been reported to have important penetration properties. [44-46] Tat has been
successfully applied for intracellular delivery of a broad variety of cargoes including various
nanosized carriers as liposomes,[47] micelles,[48] and nanoparticles,[49] whilst drugs conjugated
to R7 have been shown to cross along the skin barrier. [50] Schematic illustrations of the
different types of particles and their physicochemical characteristics are shown in the
supporting information (Figure S3 and Figure S4).
2.3. Experimental configuration
The configuration of the experimental set up is very important. On one hand it must be
appropriate to preserve the skin viability and structure, while on the other hand the area where
the particle sample is applied must be limited to avoid leakage of nanoparticles outside of the
application domain. Two different configurations, employed in our experiments, are shown in
6

Figure 2 for the human and mouse skin (see also supporting information Figure S5). For the
thicker human skin, the skin was placed on a microplate well at the top of the medium at 37 oC
while in the case of the thinner mouse skin, it was placed in a transwell insert.
In these experiments aqueous droplets of the different types of nanoparticles were applied to
the skin surface within the area of an O-ring. In order to assess to which extent the different
types of nanoparticles diffuse through the skin we used high initial concentrations of colloidal
particles (see experimental section). The nanoparticles were incubated with the skin for the
maximum possible time that the skin retained its structure (Here, 24h for the human skin and
6h for the mouse skin). The selection of the longest incubation time, in which skin maintained
its integrity, and the highest nanoparticle concentration applied, allowed us to draw
conclusions for the maximum nanoparticle penetration through skin, under the chosen
experimental conditions.
2.4. Characterization
A powerful quantitative method for evaluating the penetration of the different types of
nanoparticles through skin is the inductively coupled plasma optical emission spectometry
(ICP-OES), where the skin is dissolved, decomposed and ionized and then the amount of gold
content is calculated from the photoemission spectra. Prior to this type of analysis the gold
nanoparticles were incubated with the skin and then the remaining nanoparticle droplet was
removed. The skin was washed several times and it was subsequently tape-stripped six times
to ensure that any nanoparticles attached to the skin surface and the uppermost layers of the
stratum corneum were removed. Thus, the ICP-OES results refer mainly to the amount of
nanoparticles present in the dermis and viable epidermis. ICP-OES results were obtained from
three independent experiments for each type of nanoparticles. Figure 3 shows the percentage
of nanoparticles found in the skin for nanospheres (NSs) and nanorods (NRs) of opposite
charge (Fig. 3a), and for the three types of peptide-coated nanoparticles in comparison to the
non-functional pegylated nanospheres [(fig. 3b)-see also supporting info Figure S6]. As can
7

be seen there is a clear overall trend that positively charged pegylated nanoparticles are found
in the skin in higher numbers (2-6 times) than their negatively charged counterparts. This
observation is in agreement with recent studies of liposomes. There, it was shown that
cationic liposomes penetrate the skin more efficiently than anionic ones. [51] The enhanced skin
permeation of cationic liposomes was attributed to the “Donnan exclusion effect” and is
related to the more efficient interaction of cationic particles with the negatively charged skin
cells.
The second observation is that the percentage of NRs found in the skin is higher than for
NSs, especially in mouse skin. Several studies have demonstrated an effect of nanoparticle
geometry on uptake in cells, related to the radius curvature. [21],[52-54] Our results hint at a similar
possible contribution of particle geometry on transport through tissue. Thirdly, there is an
overall trend that peptide coated NSs are found in skin in larger numbers (up to 10 times more
in some cases) compared to pegylated NSs. Although ICP-OES measurements are
quantitative, they do not give information related to the location of nanoparticles in the skin
structure.
Therefore to qualitatively assess the spatial distribution of nanoparticles found inside the
skin, we employed transmission electron microscopy and two photon photoluminescence
microscopy. These imaging based methods allow identification of areas of the skin where
nanoparticles can be found and together with ICP-OES offer more reliable conclusions as to
the interactions of particles with the skin. To perform these techniques skin was thin-sectioned
after the interaction with the nanoparticles.
For TEM observations, the skin was cut to sections of 90nm thickness. While these sections
are very thin and it is not expected to see a large number of particles we can extract
information about the nanoparticle location within the sectioned domain. It is worth noting
here that the sectioning of the skin is a quite delicate process and needs to be performed
carefully to avoid contamination of the deeper layers of the skin with nanoparticles found in
8

the surface of the skin. For this reason, the epidermis was orientated perpendicular to the
blade when cutting ultra-thin sections. Figure 4 shows different TEM images of skin thinsections containing nanoparticles (more images are provided in the supporting information
Figure S7 and S8). Nanoparticles were found in all layers of the skin, including the deeper
layers i.e.

epidermis and dermis. However, information regarding the localization of

nanoparticles in certain compartments of the skin cannot be extracted from TEM because of
the very small dimensions of the tissue sections.
Compared to TEM sectioning, two photon photoluminescence microscopy allows
characterization of thicker (14 μm) and larger (250 μm) sections of skin. While it does not
provide detailed information on the number of particles in skin, TPPL gives a global overview
of the distribution of densities of particles with micrometer resolution. For our experiment, a
dedicated TPPL set up was built utilizing a femtosecond pulsed laser excitation at a
wavelength of 515 nm. Compared to most TPPL studies using near-infrared laser excitation,
our setup is well matched to the localized surface plasmon resonance (LSPR) of individual
gold NSs.
Figure 5 shows optical and TPPL images of sectioned mouse skin incubated with the
nanoparticles that contain the penetrating peptides Tat and R7 (see also supporting
information Figure S9). These types of particles were chosen due to their particular
functionality and because they have been shown in our ICP-OES studies to penetrate in higher
numbers through the skin. Thus, there is a particular interest about their spatial distribution
within the skin structure. The dark blue background outlining the tissue corresponds to a small
scattering background which was not completely suppressed by the fluorescence filter. The
bright spots indicate the presence of the nanoparticles, which penetrate deep in the dermis in
both cases. Organization of >10 nanoparticles into domains is shown, indicating that
macroscale localization takes place in certain areas of the dermis. Some of these domains are
also visible as characteristic darker regions in the bright field images and indicate that large
9

densities of particles were collected into specific domains. Similar clustering of nanoparticles
was found in varying amounts in sections incubated with the different types of nanoparticles,
as is shown in the supporting information Figure S9. We emphasize that the TPPL maps
represent relatively thin sections of tissue and therefore only provide qualitative information
on nanoparticle localization and not on their quantitative penetration levels, as is obtained
from ICPMS.
The increased penetration of Tat and R7 functionalized nanoparticles is attributed to the key
role of Tat and R7 cell penetrating peptides. This is further indicated in Figure 4D which
depicts peptide-coated particles inside cellular compartments. As reported elsewhere, cell
penetrating peptides can penetrate through the non-viable and viable part of the skin. [55] Both
samples were not tape-stripped and for this reason high densities of nanoparticles are also
observed at the surface of the skin. While at this stage no conclusive answer can be given
regarding the macroscale organization of nanoparticles in the deeper layers of the skin, it is
well possible that these are gathered in cells, specially aimed at harvesting infiltrating
specimens. Indeed, studies of intradermally-injected quantum dots have indicated evidence of
translocation of nanoparticles to adjacent lymph nodes via skin macrophages and dendritic
(Langerhans) cells.[56] Such translocation mechanisms could explain the observation of
concentrated nanoparticle clustering deep inside the dermis layer.
Our studies show that nanoparticle penetration through skin is dependent on the individual
physicochemical characteristics of the nanoparticles and these parameters must be considered
in future studies especially when transdermal drug delivery is chosen as an administration
tool. While the skin explant experiments show a systematic trend which qualitatively confirms
the functionality of advanced nanoparticle coatings, we emphasize that nanoparticle-skin
interactions under in-vivo conditions may differ in the amounts of nanoparticle penetration
and dynamical processes mediated by the living body.

10

3. Conclusion
In this work we investigated the penetration of different types of gold nanoparticles through
human and mouse skin. We evaluated how the charge, shape and functionality of
nanoparticles can critically influence the interaction of the nanoparticles with the different
types of skin. Our systematic studies include quantitative and qualitative characterization of
our samples and concluded that there is an overall trend that positively charged nanoparticles
penetrated the skin in larger numbers (2-6 times) in comparison to their negatively charged
counterparts. Also, it was observed that rod-shape nanoparticles are found in the skin in
higher numbers than spheres, thus the morphology of the particle has to be taken into account
in such studies. Moreover, there is an overall trend that the peptide-coated nanospheres
employed in our experiments penetrate the skin in larger numbers (up to 10 times more in
some cases) in comparison to pegylated nanospheres, highlighting the important role of cell
penetrating peptides. Thin sections (90 nm) of samples under TEM show that the particles are
found both in epidermis and dermis. In thick sections of skin (14 μm) TPPL studies showed
that Tat and R7 containing nanoparticles migrate in large numbers in the deeper layers of the
skin and present a microscale organization.
4. Experimental Section
Materials and methods.
Reagents were purchased from the following suppliers: trisodium citrate, sodium
tetrachloroaurate (III) dehydrate, sodium chloride, L-ascorbic acid, bis (p-sulfonatophenyl) phenyl phosphine dehydrate dipotassium salt (BSPP), hexadecyltrimethylammonium bromide
(CTAB) sodium borohydride, and tween20 were purchased from Sigma-Aldrich.

Silver

nitrate was purchased from Fisher Scientific. Alpha-Amino-omega-mercapto poly (ethylene
glycol) hydrochloride, MW 5000 Dalton (SH-PEG-NH 2) and alpha-Thio-omega-carboxy
poly(ethylene glycol), MW 5000 Dalton (SH-PEG-COOH) were purchased from Iris Biotech
11

GmbH. The peptides CALNN, CALNNTat and CALNNR 7 were purchased from
PeptideSynthetics. Dulbecco's Modified Eagle Medium (DMEM), Roswell Park Memorial
Institute (RPMI) medium, fetal bovine serum (FCS) were purchased from Life Technologies
Ltd. Penicillin-Streptomycin solution and trypsin were purchased from sigma Aldrich.
Osmium (VIII) Oxide was purchased from OXKEM Limited. Acetonitrile was purchased
from Fisher Scientific. 0.45 μm pore size, 25 mm diameter cellulose acetate membrane
syringe filter was purchased from VWR international Lda. 3.05 mm diameter Carbon coated
400 mesh Copper grids and 3.05 mm diameter Palladium coated 200 mesh Copper grids were
purchased from Agar Scientific. 12 well costar® transwell-clear support tissue culture treated
sterile polyester membrane (0.4 um pore size, 12 mm membrane diameter) and cell culture 6well and 96-well plates were purchased from Fisher Scientific. 3M™ Micropore™ Medical
Tape and O-rings (I.D.= 4.47 mm, W=1.76 mm) were purchase from amazon.co.uk.
26x76mm 1mm-1.2mm thick microscope glass slides and 22x22mm n°1 cover glasses were
purchased from VWR International. UV-visible spectra of colloidal gold nanoparticles were
collected using a Cary 300 Bio UV-vis spectrophotometer over the range from 350 to 800 nm.
Electrophoretic light scattering and dynamic light scattering measurements were obtained
with a Malvern Zetasizer Nano E0248. TEM images were obtained with a FEI Technai12
Transmission electron microscope operating at a voltage of 80 kV. Ultrathin sections were cut
on a Reichurt Om-U3 ultramicrotome and the glass knifes were made with a LKB knife
maker. EDAX X-ray spectrometer was used on the EDS analysis (see supporting info). Twophoton photoluminescence (TPPL) were obtained with a TPPL microscope. ICP-OES
measurements were made with an ICAP 6300 duo Spectrophotometer. Milli-Q water was
used in all experiments.
Synthesis of spherical gold nanoparticles.
Sodium citrate stabilized spherical gold nanoparticles were prepared using the Turkevich
method.[57] In detail, a solution of trisodium citrate (19.5 mM, 2.5 ml) was brought to boil and
12


quickly added into a boiling solution of sodium tetrachloroaurate (III) dihydrate (0.5 mM, 25
ml) while stirring vigorously. The colour of the solution changed from pale yellow to
colourless, then to purple and finally to deep red indicating the formation of nanospheres. The

reaction mixture was boiled and stirred for an additional 5 minutes, cooled down to room
temperature while stirring and purified by filtration through a 0.45 μm syringe filter. Citrate
coated gold nanoparticles were capped with BSPP via ligand exchange reaction. Basically,
BSPP (10 mg; MW = 498.6) was added to a solution of citrate stabilized gold nanoparticles
(27.5 ml, 4.32 nM) and the mixture was stirred overnight at room temperature. BSPP-coated
nanospheres were precipitated with sodium chloride (50 mg), purified by one step
centrifugation (5000 rpm, 5 minutes, 22 °C) and redispersed by sonication in 100 μl of MilliQ water.
Synthesis of gold nanorods.
Gold nanorods were synthesized using an optimized seed mediated growth method. 31 In detail,
a seed solution was prepared by mixing CTAB (0.2M, 1 ml) with sodium tetrachloroaurate
(III) dihydrate (5 mM, 1 ml). Then, an ice-cold solution of sodium borohydride (0.01 M, 0.5
ml) was added dropwise to mixture while stirring vigorously. The colour of the solution
changed from dark yellow to colourless and then to light brown, indicating the formation of
the nanospheres. The solution was stirred for 2 min and used immediately after. A growth
solution was prepared by mixing an aqueous solution of CTAB (0.2 M, 14.24 ml) with sodium
tetrachloroaurate (III) dihydrate (5 mM, 2 ml) and silver nitrate (5 mM, 0.18 ml) at 35 °C.
While stirring at 250 rpm, a L-ascorbic acid solution (78.8 mM, 160 μl) was added to the
mixture. The colour of the growth solution changes from dark yellow to colourless. 30
seconds after the L-ascorbic acid addition, 16 μL of the seed solution were injected into the
growth solution. The solution was kept unstirred at 35 °C for 4 hrs. The colour of the solution
changes from colourless to blue greenish over that period. As-made gold nanorods solution

13


was purified by two steps of centrifugation (8500 rpm, 20 min, 22 °C) and redispersed in 5 ml
of Milli-Q water.
Surface functionalization of gold nanospheres with PEG-containing molecules.
A freshly prepared SH-PEG-COOH or SH-PEG-NH2 aqueous solution (5 mg/ml, 200 μl, MW
= 5000 Da) was added to a solution of BSPP coated gold nanospheres (5 nM, 10 ml), while

stirring. The mixture was incubated for 2 hours at room temperature while shaking and then
overnight at 4 °C. Functionalised gold nanoparticles were purified by three steps of
centrifugation (16400 rpm, 15 minutes, 10°C) and redispersed by sonication in 100 μl of
Milli-Q water.
Surface functionalization of gold nanorods with PEG-containing molecules.
A freshly prepared SH-PEG-COOH or SH-PEG-NH2 aqueous solution (0.5 mg/ml, 2 ml, MW
= 5000 Da) was added to a solution of gold nanorods (as-prepared, 2 ml), while stirring at 500
rpm. The mixture was sonicated for 30 seconds and kept overnight at 500 rpm at room
temperature. After, the solution was centrifuged (8500 rpm, 16 minutes, 22°C) and
redispersed in a freshly prepared SH-PEG-COOH or SH-PEG-NH 2 aqueous solution (0.25
mg/ml, 4 ml, MW = 5000 Da). The mixture was incubated for 4 hours at room temperature
while shaking at 500 rpm. Capped gold nanorods were purified by two steps of
centrifugation/decantation (8500 rpm, 16 minutes, 22°C) and redispersed by sonication in 100
μl of Milli-Q water. The concentration of gold nanorods was calculated by ICP-OES.
Surface functionalization of gold nanospheres with CALNN peptide and CALNN peptide
derivatives.
The surface of spherical gold nanoparticles was functionalized with CALNN, 99% CALNN 1% CALNNTat and 99% CALNN - 1 % CALNNR 7 peptides via the thiol group of the
cysteine. In all the cases a 5000:1 peptide to nanoparticle molar ratio was used. Briefly, three
solutions of BSPP coated gold nanospheres (5 nM, 5 ml) were injected with the following
aqueous solutions, respectively: CALNN (0.5 mg/ml, 133.4 μl), CALNN – CALNNTat
14


mixture [CALNN (0.5 mg/ml, 132.0 μl) + CALNNTat (0.1 mg/ml, 26.7 μl)] and CALNN CALNNR7 mixture (CALNN (0.5 mg/ml, 132.0 μl) + CALNNR7 (0.1 mg/ml, 20.35 μl))
whilst shaking at 500 rpm. In order to avoid unspecific biding, an aqueous solution of tween
20 (1 wt%, 150 μl) was added to the solution. After 4 hours shaking at room temperature, the
reaction mixture was centrifuged (16400 rpm, 10 min, 22°C) and redispersed in 5 ml of water
with CALNN (0.5 mg/ml, 133.4 μl), CALNNTat (CALNN (0.5 mg/ml, 132.0 μl) +
CALNNTat (0.1 mg/ml, 26.7 μl) and CALNNTat (CALNN (0.5 mg/ml, 132.0 μl) +
CALNNR7 (0.1 mg/ml, 20.35 μl). An aqueous solution of tween 20 (1wt%, 150 μl) was then

added to the solution. The solutions were incubated overnight at room temperature while
shaking at 500 rpm and purified from the excess peptide by 3 centrifugation steps (16400
rpm, 15 min) and redispersed in 100 μl of phosphate buffer (0.01M, pH 7.2).
Human skin preparation.
Human full thickness skin was obtained from surgical resection after receiving written
consent with approval by the Southampton and South West Hampshire Research Ethics
Committee in adherence to Helsinki Guidelines (NRES 07/Q1704/59). The experiments were
carried on immediately after skin was excised. The subcutaneous fat was carefully removed
with a blade and the skin was washed with PBS. For the entire study skin samples from four
different donors were utilized. The skin was equally divided for experiments into pieces of
~133mg each. Skin samples from 2 donors were used in the histological studies; 2-4 sections
were investigated for each sample.
Mouse skin preparation.
The dorsal skin from newborn hairless mice was excised. C57/BL6 mice were bred at the
University of Southampton (Certificate of designation number 70/2906). Tissue was removed
from day old mice after euthanasia by a schedule one method following the UK Animals
(Scientific Procedures) Act 1986. Briefly, a cut was made at the base of the mouse torso with
sterile scissors. The scissors were then used to gently peel away the skin toward the frontal
15


part of the mouse. Mouse skin was then washed with PBS buffer and used in further
experiments. For the entire study, the skin of 22 mice was used. The skin was equally divided
for experiments into pieces of ~19mg each. Skin samples from 6 mice were used in the
integrity studies; 2-4 sections were investigated for each sample.

Histological analysis of un-treated human skin.
The cleaned human skin was then evenly cut into smaller pieces and placed in a 6-well
microplate and RPMI media (RPMI + 5% Penicillin/Streptimycin + 10% Foetal Calf Serum;
600 ul per well) was added in order to immerse the dermis and leave the epidermis at the airmedium interface. Samples were then incubated for 12, 24 and 48 hours at 37°C in 5%

CO2/air. After incubation time, the skin was minced into ~ 3-4 mm3 portions and fixed with
10% formaldehyde. The samples were then embedded in paraffin, hematoxylin and eosin
stained, cut into 0.5 µm sections and observed by light microscopy.
Histological analysis of un-treated mouse skin.
The cleaned mouse skin was evenly cut into smaller pieces and placed in 12 mm transwell
insert with DMEM media (DMEM + 5% Penicillin/Streptomycin + 10% Foetal Calf Serum;
400 μl per well) in the basolateral chamber. Samples were then incubated for 6 and 24 hours at
37°C in 5% CO2/air. After incubation time, the skin was minced into ~ 3-4 mm3 portions and
fixed with 10% formaldehyde. The samples were then embedded in paraffin, hematoxylin and
eosin stained, cut into 0.5 µm sections and observed by light microscopy.
Assessment of human skin penetration.
The cleaned human skin was then evenly cut into smaller pieces and rubber O-rings (I.D. =
4.47 mm; W = 1.76 mm) were clamped and sealed with vaseline on the top of each piece of
skin. The skin pieces were then placed in a 6-well microplate and RPMI media (RPMI + 5%
Penicillin/Streptomycin + 10% Foetal Calf Serum; 600 ul per well) was added in order to
immerse the dermis and leave the epidermis at the air-medium interface. Each O-ring was
16


filled with 20 μl of gold NPs solution (100 nM for gold nanospheres and 14 nM for gold
nanorods). (One of the skin pieces was used as a control where the skin was not incubated
with nanoparticles and exposed to the same conditions as the NP-treated skin pieces). The
skin was then incubated for 24 hours at 37°C in 5% CO 2/air. After incubation, the gold NPs
solution that remained in the ring was collected and the skin was washed 2 times with 20 µl of
Milli-Q water. The ring was carefully removed from the skin and the skin underneath the ring
was collected for characterization.
Assessment of mouse skin penetration.
The cleaned mouse skin was evenly cut into smaller pieces and rubber O-rings (I.D. = 4.47
mm; W = 1.76 mm) were clamped and sealed with vaseline on the top of each piece of skin.
The skin pieces were then placed in 12 mm transwell insert with DMEM media (DMEM +

5% Penicillin/Streptomycin + 10% Foetal Calf Serum; 400 μl per well) in the basolateral
chamber. Each O-ring was filled with 20 μl of gold NPs solution (100 nM for gold
nanospheres and 14 nM for gold nanorods). (As in the case of human skin, one of the skin
pieces was used as a control where the skin was not incubated with nanoparticles and exposed
to the same conditions as the NP-treated skin pieces). The skin was then incubated for 24
hours (human skin) 6 hours (mouse skin) at 37°C in 5% CO 2/air. After incubation, the gold
NPs solution that remained in the ring was collected and the skin was washed 2 times with 20
µl of Milli-Q water. The ring was carefully removed from the skin and the skin underneath the
ring was collected for characterization.
TEM and EDX of gold nanoparticle-treated skin.
After the 24 hours gold (for human skin) or 6 hours (for mouse skin) of gold nanoparticles
treatment, the skin underneath the O-ring was minced into ~ 3-4 mm3 pieces. For electron
microscopy, tissue specimens were fixed overnight in a solution containing glutaraldehyde
(3%) and formaldehyde (4%) in PIPES buffered (0.1 M, pH 7.2). Tissues were then washed
twice in PIPES (0.1 M, pH 7.2) and postfixed for 1 hour in a solution containing osmium
17


tetraoxide buffered (2%) with PIPES (0.1 M, pH 7.2). Next, tissues were washed twice in
PIPES buffer (0.1 M, pH 7.2) and treated with uranyl acetate for 20 min. Tissues were
dehydrated in an ascending ethanol series [30% (10 minutes), 50% (10 minutes), 75% (10
minutes), 95% (10 minutes) and 100% ethanol (twice 20 minutes)] and embedded in
acetonitrile for 10 min, 50:50 acetonitrile:Spurr resin overnight and Spurr resin for 6 hours.
Samples were then polymerized in fresh Spurr resin for 16 hours at 60°C. Specimens were
subsequently thin-sectioned into 90 nm slices, mounted on 400 mesh copper/palladium 3.05
mm grids and observed on a FEI Technai12 TEM, operating at 80 kV. EDX analysis of the
TEM grids was performed to evaluate the gold content of the samples.
ICP-OES for the human and mouse skin treated with nanoparticles.
After the skin samples were tape-stripped sic times for each case, they were dissolved in aqua
regia (10%). For the human skin, a microwave digester was used to facilitate the process.

Once the samples were digested, Milli-Q water was used to bring the final volume to 10 ml.
Finally, ICAP 6300 duo Spectrophotometer was used to measure the amount of gold present
in the skin specimens in ppm.
Two-photon microscopy of gold nanoparticle-treated mouse skin.
After the mouse skin was treated for 6 hours with gold NP the skin samples were fixed for 24
hours in a 10% buffered formalin solution at room temperature, using 10 times the volume of
fixative to the volume of specimen. Tissues were dehydrated in an ascending ethanol series
(50% (1 hour), 70% (1 hour), 90% (1 hour), 100% (twice 1 hour) and 100% ethanol (twice 2
hours)) and then placed in Xylene (three times 1 hour and 30 minutes each) and then in
paraffin (1 hour and 30 min followed by three times 1 hour each). Finally samples were
embedded into a block al let to set for 1 hour in a freezer plate. Specimens were subsequently
thin-sectioned into 14 μm slices, mounted in cover slips and let to dry overnight. Sections
were then dewaxed in xylene (twice 10 minutes), 100% ethanol (twice 5 minutes), 70%
ethanol (twice 5 minutes) and water (5 minutes). Slides were then dried at room temperature
18


and coverslips were mounted on the top of the glass slides with Mowiol. Samples were
imaged on a two-photon photoluminescence microscope. Illumination was done at 515nm
using the second-harmonic of a 1030nm femtosecond laser (Pharos, Light Conversion)
producing short pulses of 100 fs duration at 76 MHz repetition frequency. Emitted light was
spectrally filtered using a bandpass filter between 500nm-510nm to detect the anti-Stokes
fluorescence emission mediated by the surface plasmon resonance of the gold nanospheres.
Emission from a single gold nanosphere corresponds to approx. 100 counts in the image.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.
Acknowledgements
The University of Southampton, Royal Society and EPSRC are gratefully acknowledged for
partial support of this project. A.G.K. would also like to thank the EU COST actions MP1202,

MP1005, TD1003 and TD1004 for networking opportunities associated with this work. R.F.
thanks the School of Physics and Astronomy for a mayflower studentship. NRS was funded
by MRC (grant number G0501515).
Received: ((will be filled in by the editorial staff))
Revised: ((will be filled in by the editorial staff))
Published online: ((will be filled in by the editorial staff))
[1]

G. Chen, H. Qiu, P. N. Prasad, X. Chen, Chem. Rev. 2014, 114, 5161-5214.

[2]

J. A. Webb, R. Bardhan, Nanoscale 2014, 6, 2502-2530.

[3]

T. Vo-Dinh, A. M. Fales, G. D. Griffin, C. G. Khoury, Y. Liu, H. Ngo, S. J. Norton, J.

K. Register, H. N. Wang, H. Yuan, Nanoscale 2013, 5, 10127-10140.

19


[4]

A. Heuer-Jungemann, P. Harimech, T. Brown, A. G. Kanaras, Nanoscale 2013, 5,

9503-9510.
[5]

A. Comin, L. Manna, Chem. Soc. Rev. 2014, 43, 3957-3975.

[6]

D. Astruc, P. Zhao, N. Li, Angew. Chem. Int. Ed. 2014, 53, 1756-1789.

[7]

Z. Luo, K. Zheng, J. Xie, Chem. Commun. 2014, 5143-5155.

[8]

L. Y. T. Chou, K. Zagorovsky, W. C. W. Chan, Nature Nanotechnology 2014, 9, 148-

155.
[9]

A. Heuer-Jungemann, R. Kirkwood, A. El-Sagheer, T. Brown, A. G. Kanaras,

Nanoscale 2013, 5, 7209-7212.
[10]

B. Pelaz, V. Grazu, A. Ibarra, C. Magen, P. del Pino, J. M. de la Fuente,

Langmuir 2012, 28, 8965-8970.
[11]

Z. Lu, Y. Yin, Chem. Soc. Rev. 2012, 41, 6874-6887.

[12]

R. A. Sperling, W. J. Parak, Phil. Trans. R. Soc. A 2010, 368, 1333-1383.

[13]

M. Mahmoudi, J. Meng, X. Xue, X. J. Liang, M. Rahman, C. Pfeiffer, R. Hartmann, P.

R. Gil, B. Pelaz, W. J. Parak, P. del Pino, S. Carregal-Romero, A. G. Kanaras, S. T. Selvan,
Biotechnology Advances 2014, 32, 679-692.
[14]

S. T. Kim, K. Saha, C. Kim, V. M. Rotello, Acc. Chem. Res. 2013, 46, 681–691.

[15]

C. D. Walkey, J. B. Olsen, F. Song, R. Liu, H. Guo, D. W. H. Olsen, Y. Cohen, A.

Emili, W. C. W. Chan, ACS Nano 2014, 8, 2439-2455.
[16]

J. A. Yang, S. E. Lohse, C. J. Murphy, Small 2014, 10, 1642-1651.

[17]

P. P. Pillai, S. Huda, B. Kowalczyk, B. A. Grzybowski, J. Am. Chem. Soc. 2013, 135,

6392-6395.
[18]

D. Huhn, K. Kantner, C. Geidel, S. Brandholt, I. De Cock, S. J. H. Soenen, P. R. Gil, J.

M. Montenegro, K. Braeckmans, K. Mullen, G. U. Nienhaus, M. Klapper, W. J. Parak, ACS
Nano 2013, 7, 3253-3263.
[19]

F. Zhao, Y. Zhao, Y. Liu, X. Chang, C. Chen, Y. Zhao, Small 2011, 7, 1322-1337.
20


[20]

D. Bartczak, S. Nitti, T. M. Millar, A. G. Kanaras, Nanoscale 2012, 4, 4470-4472.

[21]

D. Bartczak, O. L. Muskens, S. Nitti, T. M. Millar, T. Sanchez-Elsner, A. G. Kanaras,

Small 2012, 8, 122-130.
[22]

D. Bartczak, O. L. Muskens, T. Sanchez-Elsner, A. G. Kanaras, T. M. Millar, ACS

Nano 2013, 7, 5628-5636.
[23]

D. Zheng, D. A. Giljohann, D. L. Chen, M. D. Massich, X. Q. Wang, H. Iordanov, C.

A. Mirkin, A. S. Paller, PNAs, 2012, 109, 11975-11980.
[24]

L. A. DeLouise, Journal of Investigative Dermatology 2012, 132, 964-975.

[25]

F. Rancan, Q. Gao, C. Graf, S. Troppens, S. Hadam, S. Hackbarth, C. Kembuan, U.

Blume-Peytavi, E. Ruhl, J. Lademann, A. Vogt, ACS Nano 2012, 6, 6829-6842.
[26]

C. Pegoraro, S. MacNeil, G. Battaglia, Nanoscale 2012, 4, 1881-1894.

[27]

D. Papakostas, F. Rancan, W. Sterry, U. Blume-Peytavi, A. Vogt, Arch. Dermatol. Res.

2011, 303, 533-550.
[28]

B. Baroli, Journal of pharmaceutical sciences 2010, 99 (1), 21-50.

[29]

G. Sonavane, K. Tomoda, A. Sano, H. Ohshima, H. Terada, K. Makino, Colloids and

surfaces. B, Biointerfaces 2008, 65 (1), 1-10.
[30]

G. Krishnan, J. Edwards, Y. Chen, H.A. Benson, Journal of pharmaceutical sciences

2010, 99 (6), 2724-31.

[31]

F.L Filon, M. Crosera, G. Adami, M Bovenzi, F. Rossi, G. Maina, Nanotoxicology

2011, 5 (4), 493-501.
[32]

Y. Huang, F. Yu, Y.S. Park, J. Wang, M.C. Shin, H.S. Chung, V.C. Yang, Biomaterials

2010, 31 (34), 9086-91.
[33]

H.I. Labouta, D.C. Liu, L.L. Lin, M.K. Butler, J.E. Grice, A.P. Raphael, T. Kraus, L.K.

El-Khordagui, H.P. Soyer, M.S. Roberts, M. Schneider, T.W. Prow, Pharmaceutical research
2011, 28 (11), 2931-44.

21


[34]

H.I. Labouta, L.K. El-Khordagui, T.Kraus, M. Schneider, Nanoscale 2011, 3 (12),

4989-99.
[35]

H.I. Labouta, M. Schneider, Nanomedicine : nanotechnology, biology, and medicine

2013, 9 (1), 39-54.

[36]

H.I. Labouta, M. Hampel, S. Thude, K. Reutlinger, K.H. Kostka, M. Schneider,

Journal of biophotonics 2012, 5 (1), 85-96.
[37]

H.I Labouta, L.K. El-Khordagui, M. Schneider, Skin pharmacology and physiology

2012, 25 (4), 208-18.
[38]

H.I. Labouta, T. Kraus, L.K. El-Khordagui, M. Schneider, International journal of

pharmaceutics 2011, 413 (1-2), 279-82.
[39]

M. Schneider, F. Stracke, S. Hansen, U. H. Schaefer, Dermato-Endocrinology 2009, 1,

197-206.
[40]

B. Marie-Alexandrine, B. Stephanie, C.Yves, Wiley interdisciplinary reviews.

Nanomedicine and nanobiotechnology 2011, 3 (5), 463-478.
[41]

B. Nikoobakht, M. A. El-Sayed, Chem. Mater. 2003, 15, 1957-1962.

[42]

C. Kinnear, H. Dietsch, M. J. D. Clift, B. Rothen-Rutishauser, A. Petri-Fink, Angew.

Chem. Int. Ed.2013, 52, 1934-1938.
[43]

R. Levy, N. T. K. Thanh, R. C. Doty, I. Hussain, R. J. Nichols, D. J. Schiffrin, M.

Brust, D. G. Fernig, J. Am. Chem. Soc. 2004, 126, 10076-10084.
[44]

J. Jo, S. Hong, W. Y. Choi, D. R. Lee, Scientific Reports 2014, 4, 1-7.

[45]

D. M. Copolovici, K. Langel, U. Langel, ACS Nano 2014, 8, 1972-1994.

[46]

F. Madani, S. Lindberg, U. Langel, S. Futaki, A. Graslund, Journal of Biophysics 2011,

2011, 1-10.
[47]

M. M. Fretz, G. Storm, Methods Mol. Biol. 2010, 605, 349-359.

[48]

X. Li, M. Wang, C. Liu, X. Jing, Y. Huang, J. Appl. Polym. Sci. 2013, 130, 4598-4607.

[49]

P. Nativo, I. A. Prior, M. Brust, ACS Nano, 2008, 2, 1639-1644.
22


[50]

J. B. Rothbard, S. Garlingthon, Q. Lin, T. Kirschberg, E. Kreider, L. McGrane, P. A.

Wender, P. A. Khavari, Nature Medicine 2000, 6, 1253-1257.
[51]

S. Shanmugam, C. K. Song, S. Nagayya-Sriraman, R. Baskaran, C.S. Yong, H. G.

Choi, D. D. Kim, J. S. Woo, B. K. Yoo Arch Pharm. Res. 2009, 32, 1067-1075.
[52]

S. Dasgupta, T. Auth, G. Gompper, NanoLett. 2014, 14, 687-693.

[53]

S. D. Perrault, C. Walkey, T. Jennings, H. C. Fischer, W. C. W. Chan, Nano Lett. 2009,

9, 1909-1915.
[54]

B. D. Chithrani, A. A. Ghazani, W. C. W. Chan, Nano Lett. 2006, 6, 662–668.

[55]

P. Desai, R. P. Patlolla, M. Singh, Mol. Mambr. Biol. 2010, 27, 247-259.

[56]

S. Kim, Y. T. Lim, E. G. Soltesz, A. M. De Grand, J. Lee, A. Nakayama, J. A. Parker,

T. Mihaljevic, R. G. Laurence, D. M. Dor, L. H. Cohn, M. G. Bawendi, J. V. Frangioni,
Nature Biotechnology 2004, 22, 93-97.
[57]

J. Turkevich, P. C. Stevenson, J. Hillier, Discuss. Faraday Soc. 1951, 11, 55–75.

Nu

Nu

Figure 1. Cross-sections of viable human (A) and mouse (B) skin under a light microscope
(1) and a transmission electron microscope (2). The human skin and mouse skin used in our
experiments remain intact for 24 h and 6h, respectively (Nu: Nucleus).
23


Figure 2. Schematic illustration and images of the experimental configuration for human (A,
C) and mouse (B, D) skin, respectively. For the thinner mouse skin a transwell insert was
used. The nanoparticle droplet is placed in the o-ring on the top of the skin. In both cases the
skin is remained viable over the time course of the experiment.

24

Figure 3. ICP-OES measurements showing the amount of gold nanoparticles found in the
skin in respect to the initial concentration of nanoparticles applied to the skin surface. In (A)
pegylated gold nanospheres and gold nanorods are compared in respect to their shape and
charge. In (B) the gold nanospheres are compared in respect to their function.

25