cosmetics

Article
Biocellulose Masks as Delivery Systems: A Novel
Methodological Approach to Assure Quality
and Safety
Paola Perugini 1,2, *, Mariella Bleve 2 , Fabiola Cortinovis 3 and Antonio Colpani 4
1 Department of Drug Sciences, University of Pavia, 27100 Pavia, Italy
2 Etichub s.r.l, Academic Spin-Off of Pavia University, 27100 Pavia, Italy; mariella.bleve@etichub.it
3 ASST Papa Giovanni XXIII, 24127 Bergamo, Italy; fabyfa74@gmail.com
4 iBeauty, Brembate di Sopra, 24030 Bergamo, Italy; antonio.colpani@i-beauty.it
* Correspondence: paola.perugini@unipv.it; Tel.: +39-038-298-7174




Received: 22 July 2018; Accepted: 31 October 2018; Published: 9 November 2018


Abstract: Bacterial cellulose (BC) has become of great interest in recent years, as a delivery system in
several areas of application, including food, drugs, and cosmetics, thanks to its exclusive advantages,
such as high biocompatibility, water holding capacity, and good gas permeability. The novel
approach of the authors has led to a protocol for checking the quality and safety of bacterial cellulose
matrices in the manufacture of cosmetic masks. Two non-destructive techniques, near-infrared
spectroscopy (NIR) and multiple light scattering (MLS), were used to verify different parameters
affecting the quality of BC sheets, allowing cellulose masks to be checked over time. NIR spectroscopy
allowed for discovering changes in the water content, depending on filling/packaging procedures,
like flat-folding. Multiple light scattering was used to ascertain the stability of solutions in contact
with masks. From a clinical standpoint, the cutaneous tolerability of biocellulose masks, and their
effect on skin parameters, were evaluated through some specific “in vivo” tests. Also, a safety
evaluation during application was conducted through different studies: a short-term one after single
application, and a long-term one upon continued use.

Keywords: bacterial cellulose; tolerability evaluation; stability; NIR spectroscopy; multiple light
scattering; skin parameters

1. Introduction
Cellulose is one of the most abundant biodegradable materials in nature. Currently, cellulose
can be produced in different ways, all by natural synthesis procedures, plant photosynthesis,
microbial synthesis, and synthetic methods [1].
Microbial cellulose (or bacterial cellulose, BC) is produced by various species of bacteria, such as
Gluconacetobacter xylinus, in the presence of oxygen and glucose. Since 1886, Gluconacetobacter xylinus
has been used as the model microorganism for different bases and applied studies on cellulose.
Microbial cellulose is an exopolysaccharide, a long chain polysaccharide consisting of sugars units,
or sugar derivatives, joined by different links (e.g., 1–2 and 1–4). The molecular formula of bacterial
cellulose (C6 H10 O5 ) is the same as that of plant cellulose, but their physical and chemical features are
different [2].
Typically, bacterial cellulose exhibits a basic structure made of microfibrils composed of glucan
chains interlocked by hydrogen bonds, so that a crystalline pattern (microfibrils) is obtained.
BC possesses a higher purity, crystallinity index, and degree of polymerization in comparison to
plant cellulose. Its fibrils are about 100 times thinner (diameter 20–100 nm), such that a very high

Cosmetics 2018, 5, 66; doi:10.3390/cosmetics5040066 www.mdpi.com/journal/cosmetics

Cosmetics 2018, 5, 66 2 of 20

surface area per unit mass is demonstrated. This feature, combined with its highly hydrophilic nature,
results in a very high liquid-loading capacity. Its water-holding capacity (WHC) is over 100 times
(by mass) higher than plant cellulose, and the WHC is 100–200 times its dry weight [1–3].
The hydrogen bonds between its fibrillar units stabilize the whole structure, and define many of its
mechanical properties. Tensile strength, maximum elongation, and elastic modulus, that characterize
BC, depend on its own uniform ultrafine-fiber network structure, and the high planar orientation of
the ribbon-like fibers, when compressed into sheets, results in good chemical stability [3].
Bacterial cellulose shows great biocompatibility, not only because of its non-toxic effects
on biological systems but, also, by eliciting an appropriate host response to ensure satisfactory
performance during a specific application. Petersen and Gatenholm have pointed out that
biocompatibility of BC for tissue engineering applications can be can be due to structure similarities
with extracellular matrix components, such as collagen. In fact, collagen and BC nanofibers have
similar diameters (around 100 nm), and are extracellularly assembled from precursor molecules into
polymer chains [4].
Due to its unique structural properties, BC has become of interest in different fields, such as
in the textile industry, high quality paper production, food, pharmaceutical and medical devices,
electronics, and acoustics [1,2]. In particular, the biomedical field exploits microbial cellulose as a
natural, porous, nontoxic material in tissue-like products for both wound care and the regeneration of
damaged or diseased organs. Due to its unique nanostructure and properties, BC is a natural candidate
for numerous medical and tissue-engineered applications, such as wound dressings, bone tissue
engineering and bone grafting, and cardiovascular applications [3–5]. In wound healing systems,
BC has great potential because it is biocompatible, adherent, elastic, and transparent, resulting in
good permeability and water absorption capacity. Also, it maintains a moist environment in wounds,
absorbing exudates, and shows low solubility and resistance to degradation [1].
Biocompatibility allows for classification of BC among sustainable/renewable sources,
thus suitable when developing value-added cosmetic products [6].
It has been claimed that, due to its high water-holding capacity and good gas permeability,
BC can be an appropriate carrier for delivering active ingredients to the skin, including moisturizers,
whitening ingredients, and anti-wrinkling agents.
In recent years, studies have been developed for the use of BC as a substrate to deliver hydrophilic
cosmetic compounds, in order to obtain masks with exfoliating and brightening effect, purifying hot
clay masks, and anti-wrinkle patches [7–12].
A patented BC facial mask was fabricated with holes for the eyes, mouth, and nose. The author
claimed that such a mask may be suitable for repeated or prolonged use for skin beautifying purposes,
skin nutrition, and moisturizing and cosmetic effects [11]. An American study describes a method for
manufacturing masks from bacterial biocellulose by directly adding extracts of ginseng (known for
its antioxidant and anti-inflammatory properties) to the culture broth; results show overall user
satisfaction in terms of moist feel and skin elasticity in women aged 30+ years [11].
However, available information about the use of BCs in the cosmetic industry and, also,
clinical results, are rather scarce [12].
The production of a safe and stable formulation is one of the more important issues to be addressed
by cosmetic and pharmaceutical companies.
Stability evaluation is always crucial for colloidal systems, like emulsions and suspensions.
They are intrinsically unstable, and particle migration (sedimentation or creaming) combined with
particle aggregation might lead to partial or total separation, or the emulsion breaking. Characterizing
these destructive processes and monitoring of stability are, therefore, essential. This is usually
obtained through Aging tests, accelerating destabilization to detect potentially unstable formulations.
Products are stored under specific conditions (temperature, humidity, light), then organoleptic
properties, pH, and viscosity are analyzed. Besides ordinary visual inspections, non-destructive
instrumental techniques (NIR and MLS) have been used to investigate the expected stability.
Cosmetics 2018, 5, 66 3 of 20
Cosmetics 2018, 5, x FOR PEER REVIEW 3 of 20

The aim
The aimofofthis
thispaper
paperhashasbeen
beentotoestablish
establisha aprotocol
protocolfor
for the
the manufacture
manufacture of of
BCBC face
face masks,
masks, andand
to
to identify appropriate methods for quality assessment of the finished product, establishing
identify appropriate methods for quality assessment of the finished product, establishing criteria to criteria
to evaluate
evaluate their
their shelf-life,
shelf-life, as well.
as well.
Finally, specific “in vivo” tests have
Finally, specific “in vivo” tests have been
been designed
designed in in order
order to
to test
test such
such aspects
aspects as as tolerability,
tolerability,
safety, and efficacy of cosmetic formulations, and their effect on many different skin parameters
safety, and efficacy of cosmetic formulations, and their effect on many different skin parameters upon upon
application using
application using aa BC
BC mask
mask delivery
delivery system.
system.

2. Materials
2. Materials and
and Methods
Methods

2.1.
2.1. Materials
The
The bacterial
bacterial cellulose
cellulose sheets (BCS) used in this study were obtained from two manufacturers;
sheets
sheets were
were supplied
supplied inin stacks
stacks (20
(20 sheets)
sheets) soaked
soaked with
with aa preserved
preserved water
water solution
solution (phenoxyethanol
(phenoxyethanol
0.075%,
0.075%, iodopropynyl
iodopropynylbutylcarbamate
butylcarbamate 0.0015%).
0.0015%).BothBoth
manufacturers stated stated
manufacturers that BCthat
had BC
beenhad
obtained
been
by incubation of Gluconacetobacter xylinus in a sterile broth at 30 ◦ C, for a period of up to 120 h, at a
obtained by incubation of Gluconacetobacter xylinus in a sterile broth at 30 °C, for a period of up to 120
pH
h, at= a6.5–7.0, in staticin
pH = 6.5–7.0, conditions. After partially
static conditions. de-watering
After partially the gel, the
de-watering theBC sheet
gel, the had been die-cut
BC sheet to
had been
obtain
die-cutato mask shape,
obtain withshape,
a mask holes with
for eyes,
holesnose, and mouth.
for eyes, nose, and mouth.
The
The diameter
diameter of thethe fibers
fibers is
is about
about 40 40 nm,
nm, so
so that
that the
the ultrafine
ultrafine net
net of
of BC
BC has
has aa very
very smooth
smooth
network
network of of microfibrils,
microfibrils, perfect
perfect to
tofit
fitskin
skin(Figure
(Figure1).1).

Figure 1. Fiber
Figure 1. Fiber structure
structure of
of bacterial
bacterial cellulose
cellulose (BC)
(BC) observed
observed using
using scanning
scanning electron
electron microscopy
microscopy at
at
10,000 × magnification.
10,000× magnification.
2.2. Preparation of Face Masks
2.2. Preparation of Face Masks
In this work, placebo masks (PM) were prepared, starting from BCS in the following way:
In this work, placebo masks (PM) were prepared, starting from BCS in the following way:
bacterial cellulose sheets (BCS, batch 1603, Supplier 2) were folded, embedded with 15 mL of solution
bacterial cellulose sheets (BCS, batch 1603, Supplier 2) were folded, embedded with 15 mL of solution
placebo (P) containing water 95.7%, butylene glycol 3%, phenoxyethanol 0.9%, ethylhexylglycerin
placebo (P) containing water 95.7%, butylene glycol 3%, phenoxyethanol 0.9%, ethylhexylglycerin
0.1%, and xanthan gum 0.3%.
0.1%, and xanthan gum 0.3%.
The folded masks were inserted into Polyethylene terephthalate (PET)/aluminum/ Polyethylene
The folded masks were inserted into Polyethylene terephthalate (PET)/aluminum/ Polyethylene
(PE) sachets, thermo-sealed, and stored for 1 week at room temperature before analysis.
(PE) sachets, thermo-sealed, and stored for 1 week at room temperature before analysis.
Figure 2 shows unfolded face placebo masks, as an example.
Figure 2 shows unfolded face placebo masks, as an example.
Cosmetics 2018, 5, 66 4 of 20
Cosmetics 2018, 5, x FOR PEER REVIEW 4 of 20

Figure 2. left: unfolded face placebo mask (PM); right: pattern of microfibrils network.

Table 11 shows
Table shows all
all samples
samples investigated
investigated in
in this
this study.
study.

Table
Table 1. Bacterialcellulose
1. Bacterial cellulosesheets
sheets (BCS)
(BCS) and
and placebo
placebo mask
mask (PM)
(PM) samples
samples evaluated
evaluated in theinstudy.
the study.
Both
Both sheets and masks are face-shaped.
sheets and masks are face-shaped.
Batch Mask
Batch MaskType
Type Supplier
Supplier
MLA
MLA (BCS)
(BCS) 11
MLB
MLB (BCS)
(BCS) 11
MLC (BCS) 1
MLC
MLD
(BCS)
(BCS)
11
MLD
1603 (BCS)
(BCS) 12
1603
1604 (BCS)
(BCS) 22
1605
1604 (BCS)
(BCS) 22
BN337 (PM) 2
1605 (BCS) 2
BN252 (PM) 2
BN337
BN067 (PM)
(PM) 22
BN252 (PM) 2
BN067 (PM) 2
The “in vitro” analyses were performed on

- The “in vitro”

bacterial analyses
cellulose were
sheets performed on
(BCS),
-- placebo masks
bacterial (PM);
cellulose sheets (BCS),
-- solutionmasks
placebo P before and after squeezing the masks.
(PM);
- solution P before and after squeezing the masks.
2.3. Stability Testing
2.3. Stability
For theseTesting
purposes, two non-destructive techniques, near-infrared spectroscopy (MicroNIR™
Pro Spectrometer, JDSU, Milpitas,
For these purposes, CA, USA) and
two non-destructive multiple near-infrared
techniques, light scattering (MLS, Turbiscan
spectroscopy Tower® ,
(MicroNIR™
Formulaction,
Pro Toulouse,
Spectrometer, JDSU, France),
Milpitas,were
CA,used.
USA) and multiple light scattering (MLS, Turbiscan Tower®,
Formulaction, Toulouse, France), were used.
2.3.1. Near-Infrared (NIR) Analysis
2.3.1.In
Near-Infrared (NIR)
recent years, NIR Analysis has gained wide acceptance within the pharmaceutical industry
spectroscopy
for raw
In material testing,NIR
recent years, product quality control,
spectroscopy and process
has gained wide monitoring.
acceptance The growing
within pharmaceutical
the pharmaceutical
industry for raw material testing, product quality control, and process monitoring. Theanalytical
interest in NIR spectroscopy is probably a direct result of its major advantages over other growing
techniques, namely,
pharmaceutical an easy
interest in NIR sample preparation
spectroscopy withouta any
is probably pretreatments,
direct the use
result of its major of fiber optic
advantages over
probes for measurement of samples far from the spectrometer, and the capacity to
other analytical techniques, namely, an easy sample preparation without any pretreatments, the use predict chemical
andfiber
of physical
opticsample
probesparameters from one
for measurement of single
samplesspectrum
far from[13,14].
the spectrometer, and the capacity to
The combined use of near-infrared (NIR) spectroscopy
predict chemical and physical sample parameters from one single andspectrum
multivariate spectral processing
[13,14].
chemometric techniques
The combined use ofhas enabled the(NIR)
near-infrared development of effective
spectroscopy methods for
and multivariate establishing
spectral the
processing
chemometric techniques has enabled the development of effective methods for establishing the
Cosmetics 2018, 5, 66 5 of 20

Cosmetics 2018, 5, x FOR PEER REVIEW 5 of 20

composition of complex samples with acceptable levels of analytical properties, such as accuracy,
composition of complex samples with acceptable levels of analytical properties, such as accuracy,
precision, and throughput [15]. In this work, we applied principal component analysis (PCA) and
precision, and throughput [15]. In this work, we applied principal component analysis (PCA) and
global indices (viz., the hydroxyl value), obtained from the NIR spectrum of the sample, to define
global indices (viz., the hydroxyl value), obtained from the NIR spectrum of the sample, to define the
the limit of acceptability. The models, thus obtained, are accurate enough for use in quality control
limit of acceptability. The models, thus obtained, are accurate enough for use in quality control
analyses of cosmetic preparations, and provide an effective alternative to existing conventional global
analyses of cosmetic preparations, and provide an effective alternative to existing conventional global
methods [16].
methods [16].
NIR spectroscopy is used as non-destructive mask characterization for two different purposes:
NIR spectroscopy is used as non-destructive mask characterization for two different purposes:
X To test the homogeneity of the cellulose sheets, investigating the overall quality of the biocellulose
 To test the homogeneity of the cellulose sheets, investigating the overall quality of the
item, and assess/validate the BCS supplier;
biocellulose item, and assess/validate the BCS supplier;
X To check the manufacturing process of face masks, and use spectroscopy as a quality control tool
 To check the manufacturing process of face masks, and use spectroscopy as a quality control tool
over time.
over time.

Both sheets
Both sheets and
and masks
masks toto be
be analyzed
analyzed were
were extracted
extracted from
from the
the sachet
sachet and
and completely
completely unfolded,
unfolded,
stretched, and
stretched, and laid
laid on
on the
the work
work surface,
surface, looking
looking forfor visible
visible stains or folds.
stains or Then, three
folds. Then, three specific
specific areas
areas
to be scanned were selected (the same for all performed analyses) with reference to actual
to be scanned were selected (the same for all performed analyses) with reference to actual face areas: face areas:
upper section
upper section (corresponding
(corresponding to to forehead),
forehead), median
median section
section (cheek),
(cheek), and
and lower
lower section
section (chin). For each
(chin). For each
area, the
area, thepoints
pointsanalyzed
analyzed areare
split between
split betweenthe right and the
the right andleft,
theasleft,
shownas in the Figure
shown in the3. Figure
The right/left
3. The
name is referred to the viewer’s point of view, with the sample lying on the work surface.
right/left name is referred to the viewer’s point of view, with the sample lying on the work surface.

1 1 2 2

4 4 3 3

6 6 5 5

Figure 3. Different
Figure 3. points for
Different points for NIR
NIR analysis.
analysis. Each
Each sheet
sheet has
has been
been analyzed
analyzed in
in 66 different
different areas,
areas, with
with 22
replicas for each area.

least 33samples
At least samplesfromfromeacheach batch
batch andand 3 different
3 different batches
batches of sheets
of sheets and placebo
and placebo masksmasks
were
were analyzed.
analyzed.
For the NIR spectroscopy, the scanning wavelength was in the range range of
of 950–1650
950–1650 nm.
nm. In this
region, each constituent of a complex organic mixture has unique absorption properties, due to the
stretching and bending vibrations in molecular bonds [13]. [13].
this analysis,
For this analysis,the
theMicroNIR™
MicroNIR™ ProPro Spectrometer,
Spectrometer, (JDSU,
(JDSU, Milpitas,
Milpitas, CA, CA,
USA)USA) was used,
was used, with
parameters reported in Table 2. This technique presents some advantages with respect to
with parameters reported in Table 2. This technique presents some advantages with respect to other
analytical techniques, for example, its ability to record the spectra of solid and liquid samplessamples without
prior manipulation. Furthermore, it is simple, rapid, and cost-effective
cost-effective [14,15].
[14,15].
Samples werewereplaced
placedonona anon-reflective
non-reflective support,
support, with
with a fixed
a fixed andand constant
constant distance
distance (3 mm)
(3 mm) from
from the acquisition
the acquisition window. window. Three replicates
Three replicates for each for eachwere
sample sample were
always always
used, used, and
and analyses analyses
performed
at 21 ◦ C ± 2at◦ C
performed 21and
°C ±70%
2 °CRH.
and 70% RH.
All the analyzed samples were white, and of uniform appearanceappearance (visual
(visual inspection).
inspection).
Cosmetics 2018, 5, 66 6 of 20

Table 2. MicroNIR instrument parameters.

Illumination Source Two Integrated Vacuum Tungsten Lamps

Illumination geometry Flood illumination/0 observer
Input aperture dimensions 2.5 × 3.0 mm
Sample working plane 0–15 mm from window, 3 mm optimal distance
Detector 128 pixel InGaAs photodiode array
Pixel Size/Pitch 30 µm × 250 µm/50 µm
Wavelength range 950–1650 nm
Pixel to pixel interval 6.2 nm
Spectral bandwidth (FWHM) <1.25% of center wavelength (1% typical)
Spectral in-band LVF blocking >4 OD average

Data were evaluated using the Unscrambler® X software, version 10.4 (Camo software AS, Oslo,
Norway). The spectra were pretreated by using standard normal variate (SNV) followed by the first
derivative with Savitzky–Golay smoothing. Principal component analysis (PCA), on pretreated spectra,
were performed on the obtained spectra.

2.3.2. Multiple Light Scattering (MLS) Analysis

MLS technique allows evaluation of optical characteristics for several formulations,
from suspensions or emulsions to nanosystem or microsystem gels [17–19]. With the multiple
light scattering technique, a macroscopic fingerprint of the sample at a given time is obtained,
providing useful information about changes in droplet size, distribution, and appearance of a
creaming layer. On a smaller scale, MLS also offers a new approach to predict migration phenomena,
like sedimentation and creaming (for emulsion and suspension), and size variation phenomena
(like flocculation and aggregation). MLS provides a major advantage over current ageing tests. It is
possible (i) to measure the stability of both opaque and concentrated colloidal dispersions with a
single instrument; (ii) to detect instability phenomena much earlier and easier than is usually possible
through a visual inspection [20].
Other advantages represented by this technique are (i) a short evaluation time when compared to
those of usual stability protocols (hours/days instead of weeks/months), (ii) a small sample required
(20–30 g); (iii) no dilution nor sample preparation; the instrument works on the product as such.
Destabilization phenomena by multiple light scattering were investigated by Turbiscan Tower®
(Formulaction—France). Solutions were transferred in the specific glass vials and they were scanned,
each 40 µm, by a pulsed near-infrared light source (λ = 880 nm). The instrument is equipped with
two synchronous detectors: the transmission detector receives light through the sample at an angle
of 0 degrees from the incident beam; the backscattering detector receives the light scattered by the
sample at an angle of 135 degrees from the incident beam. Three different sequences of analyses were
established: six hours at 20 ◦ C, six hours at 4 ◦ C and, again, six hours at 20 ◦ C, in order to simulate a
thermal shock.
For each cycle of analysis, 181 scans were acquired every 25 s. All samples, before analysis,
were kept at room temperature (at 25 ◦ C).
Solutions obtained by squeezing placebo masks, from three different batches manufactured over
a period of 12 months, were analyzed using the MLS technique, in order to evaluate the possible
interaction between the cellulose matrix and the solution used to soak the biopolymer.
The solutions obtained by mask squeezing and, then, analyzed as reported below:

- P-M 1 (from PM batch BN337, 1 year old),

- P-M 2 (from PM batch BN067, six months old),
- P-M 3 (from PM batch BN252, just produced)

Two parameters were evaluated and compared: transmission variations (∆T), and Turbiscan
stability index (TSI). Only transmission variations over ±10% were considered significant. TSI sums
Cosmetics 2018, 5, 66 7 of 20

all the variations detected in the sample in terms of size and/or distribution. The higher is the TSI,
the worse the stability. Visual evaluation (by image capture) and pH measurements (by pH-meter
827 pH Lab/Metrohm, Italy) were performed before and after Turbiscan cycles.

2.4. In Vivo Testing

2.4.1. Study Design

In order to investigate the safety of the biocellulose masks applied to skin, different in vivo studies
have been performed on placebo masks. In particular, short-term effects after single application,
and long-term effects after one month of use, were investigated.
In the short-term study, the effect of biocellulose masks was evaluated after one application
by measuring the water content before the application (t0), and 2 h (t2), 4 h (t4), and 6 h (t6) after
application. A site treated with water was used as reference.
In the long-term study, the safety and effect of biocellulose masks was evaluated after 1 month of
application 3 times a week.
Skin parameters investigated in these studies were

â Erythema index
â Water content of the stratum corneum
â Skin barrier integrity

All “in vivo” protocol studies have been carried out according to the Helsinki declaration
(Ethical Principles for Medical Research Involving Human Subjects). For both studies, 89 healthy
female volunteers were recruited according to the following general inclusion criteria:

- good general health;

- absence of cutaneous diseases;
- people that do not show cutaneous lesions or other lesions in the interested area that could
interfere with the study evaluation;
- people that do not present a history of hypersensitivity to the common components of the
cosmetic formulations;
- women that are not pregnant nor breastfeeding;
- people that agreed not to undergo other treatments for the entire duration of the test in the
treated area;
- subjects that signed the informed consent form and following the common exclusion criteria:
- subjects that did not meet the inclusion criteria above;
- subjects under pharmacological therapy;
- participation in similar study by at least 60 days;
- subjects with no known allergies to any components of the product.

2.4.2. Instrumental Evaluation

The instrumental evaluation was carried out at the beginning of each study, and at predetermined
times, depending on the aim of the study. In all cases, the mask was applied 3 times a week, on the
cleaned face, and 20 min after it was removed.
All measurements were made in an air-conditioned room with controlled temperature and
humidity (T 22 ◦ C, RH 70 ± 5%); subjects were preconditioned in such room for at least 15 min before
the measurements.
The instruments used in the evaluation of the skin properties, such as water content, skin barrier
integrity, and the erythema index, involve contact between the skin and a series of probes that do not
cause discomfort, pain, or damage the skin.
Cosmetics 2018, 5, 66 8 of 20

The water content of the stratum corneum, reflecting the hydrating effect of the product,
was evaluated by Corneometer CM 825 (Cutometer MPA580, Courage & Khazaka, Cologne, Germany).
Corneometry is a technology used to assess the hydration of the outer layer of the epidermis:
the stratum corneum [21]. Since skin is a dielectric medium, all variations in hydration result in
corresponding changes in the skin capacity [22]. The device used in our trial was equipped with a
Cosmetics 2018, 5, x FOR PEER REVIEW 8 of 20
49 mm2 surface probe that allows precise measurements in 1 s within a 10–20 µm depth range in the
stratum corneum. The
corresponding parameters
changes were
in the skin expressed
capacity as an
[22]. The arbitrary
device used inscore scale
our trial (0–100
was A.U.).
equipped with a 49
The
mmassessment
2 surface probe of skin barrierprecise
that allows integrity, and the possible
measurements occlusive
in 1 s within a 10–20effect caused
µm depth by the
range use of
in the
stratumobject
the product corneum. Thestudy,
of this parameters
havewere
beenexpressed
performed as an
byarbitrary score scale
a Tewameter (0–100
TM 300 A.U.).
(Cutometer MPA580,
Courage &The assessment
Khazaka, of skin Germany)
Cologne, barrier integrity, andTransepidermal
[22,23]. the possible occlusive
watereffect
loss caused
(TEWL)byis the use of in
assessed
the product
2 object of this study, have been performed by a Tewameter TM 300
terms of g/m h by a skin evaporimeter made of a small cylindrical open chamber (1 cm in diameter,(Cutometer MPA580,
Courage & Khazaka, Cologne, Germany) [22,23]. Transepidermal water loss (TEWL) is assessed in
2 cm in height) with a couple of hygrometric sensors connected to a microprocessor plugged into a
terms of g/m2h by a skin evaporimeter made of a small cylindrical open chamber (1 cm in diameter,
computer workstation. The device allows recording of TEWL values (ranging from 0 to 90 g/m2 h),
2 cm in height) with a couple of hygrometric sensors connected to a microprocessor plugged into a
as wellcomputer
as the relative humidity
workstation. (ranging
The device from
allows 0 to 100%)
recording and values
of TEWL the probe temperature.
(ranging from 0 to 90 g/m2h), as
The
well as the relative humidity (ranging from 0 to 100%) and the probe temperature. (Mexameter MX 18,
erythema index (EI) was assessed using a reflectance spectrophotometer
Courage &The Khazaka,
erythema Cologne, Germany).
index (EI) was assessed using a reflectance spectrophotometer (Mexameter MX
18, Courage & Khazaka, Cologne, Germany)
2.4.3. Statistical Analysis
2.4.3. Statistical Analysis
All data obtained were processed as both a descriptive statistical analysis, and as a paired t-test.
All data
A significance levelobtained werechosen,
of 5% was processed
so as both a descriptive
changes statistical
were considered analysis, and
statistically as a paired
significant fort-test.
p < 0.05.
A significance level of 5% was chosen, so changes were considered statistically significant for p < 0.05.
3. Results and Discussion
3. Results and Discussion
3.1. Quality Check by NIR
3.1. Quality Check by NIR
3.1.1. Quality Test on Bacterial Cellulose Sheets (BCS)
3.1.1. Quality Test on Bacterial Cellulose Sheets (BCS)
The studies performed on bacterial cellulose sheets (BCS) had different purposes: the first was
The studies performed on bacterial cellulose sheets (BCS) had different purposes: the first was
to evaluate the homogeneity of the biopolymer in all checked areas of the sheet; the second was to
to evaluate the homogeneity of the biopolymer in all checked areas of the sheet; the second was to
ascertain the reproducibility
ascertain among
the reproducibility sheets
among from
sheets fromthe
thesame
samebatch and from
batch and fromdifferent
different batches.
batches.
To assess the quality and the homogeneity of the biopolymer, 28 bacterial cellulose
To assess the quality and the homogeneity of the biopolymer, 28 bacterial cellulose sheets coming
sheets
from the same supplier were analyzed; 3 areas for each matrix were analyzed in duplicate.
coming from the same supplier were analyzed; 3 areas for each matrix were analyzed in duplicate.
Data obtained fromfrom
Data obtained NIRNIR
analysis were
analysis pretreated
were pretreatedand
and then evaluatedusing
then evaluated using principal
principal component
component
analysis
analysis (PCA).(PCA). Figure
Figure 4 shows
4 shows thethe results
results obtainedon
obtained onBCS
BCS coming
coming from
fromSupplier
Supplier 1. 1.

(A)
Figure 4. Cont.
Cosmetics 2018, 5, 66 9 of 20
Cosmetics 2018, 5, x FOR PEER REVIEW 9 of 20

(B)

(C)

Figure 4. Cont.
Cosmetics 2018, 5, 66 10 of 20
Cosmetics 2018,
Cosmetics 2018,5, 5,
x xFOR
FORPEER
PEERREVIEW
REVIEW 10 of1020of 20

(D)
Figure 4. NIR spectra andmathematical
mathematical elaboration
(D) of of
thethe
6 areas analyzed for each
Figure 4. NIR spectra and elaboration 6 areas analyzed forofeach
28 sheets:
of 28 R,
sheets:
right;
Figure L,
4. left
NIR (A) NIR
spectra spectra
and data expressed
mathematical as a mean
elaboration value
of the 6(all sheets,
areas all
analyzedanalyzed
R, right; L, left (A) NIR spectra data expressed as a mean value (all sheets, all analyzed areas), (B) PCAfor areas),
each of 28(B) PCA
sheets: R,
plot
plotright;
and L, and
(C)left(C) loadings
(A) NIRplot
loadings plot
spectrafor PC1,
data (D)
for PC1, (D) loadings
expressed
loadings plot
as aplot
meanfor PC2
forvalue for
PC2 for BCS
(allBCS coming
sheets, from Supplier
all analyzed
coming 1.
areas), (B)
from Supplier 1. PCA
plot and (C) loadings plot for PC1, (D) loadings plot for PC2 for BCS coming from Supplier 1.
The Theresultsresults highlighteda agreat
highlighted great variability
variability among among thethe
different zoneszones
different analyzed in the sheets.
analyzed in theAs sheets.
shown in
TheinresultsFigure 4B (PCA
highlighted plot), some
a greatsome changes
variability in
among the biopolymer
thebiopolymer structure
different zones of
analyzedall analyzed
in the areas As
sheets.
As shown Figure 4B (PCA plot), changes in the structure of all analyzed areas
occurred, though to a different extent. This greater variability (83%) is expressed along the PC1 axis,
shown
occurred, in Figure
though to 4B (PCA plot),
a different some
extent. changes
This greaterinvariability
the biopolymer (83%) structure
is expressed of all analyzed
along the PC1areas
axis,
as shown in the PCA plot. This is confirmed by the loadings plot specific for PC1 (Figure 4C), in
occurred,
as shown though to a different extent. This greater variability (83%) is expressed along the PC1 axis,
whichinathe PCA plot.peak
characteristic Thisatis 1676
confirmed by the loadings
cm−1, corresponding to OHplotandspecific for PC1 peak,
CH overtone (Figure 4C),toinbewhich
seems
as shown in the
a characteristic peak PCA at plot. This
1676 cm −1is
, confirmed by the
corresponding loadings plot specific forpeak,
PC1 (Figure 4C),
bein
the responsible for this variation. This peak can be related to the cellulose structure. Also,tothe
to OH and CH overtone seems the
which
responsible a characteristic
for this peak
variation. at
This1676
peak cm −1, corresponding to OH and CH overtone peak, seems to be
can be related to the cellulose structure. Also, the characteristic
characteristic peak of water around 1450 cm , in the loading plot specific for component 2 in the PCA
−1
the responsible for this variation. −1 , inThis
peak of water
plot, indicates around 1450 cmdistribution
that a different the peak
loading
of
can be related to the cellulose structure. Also, the
plot specific
water occurs for component
among cellulose zones. The2reproducibility
in the PCA plot,
characteristic
among peak
different of water around
batchesdistribution 1450
of BCS was tested cm −1, in the loading plot specific for component 2 in the PCA
by analyzing at least 3cellulose
different batches
indicates that a different of water occurs among zones. for The each supplier;
reproducibility
plot,
5 indicates
matrices for that
each a batch
different
were distribution
evaluated oftriplicate.
in water occurs among cellulose zones. The reproducibility
among different batches of BCS was tested by analyzing at least 3 different batches for each supplier;
among different are batches of BCS was testedofbymatrices
analyzing at least 3 different batches for each supplier;
5 matricesHereafter
for each batch reported
werePCA analyses
evaluated in triplicate. coming from the two suppliers (Figure 5).
5 matrices for each batch were evaluated in triplicate.
Hereafter are reported PCA analyses of matrices coming from the two suppliers (Figure 5).
Hereafter are reported PCA analyses of matrices coming from the two suppliers (Figure 5).

(A)

(A)
Figure 5. Cont.
Cosmetics 2018, 5, 66 11 of 20
Cosmetics 2018, 5, x FOR PEER REVIEW 11 of 20

(B)
Figure PCA
5. 5.
Figure PCAplots
plotsofofNIR
NIRspectra
spectra of
of BCS
BCS different batches coming
different batches comingfrom
from(A)
(A)Supplier
Supplier 1 and
1 and (B)(B)
Supplier
Supplier 2. 2.

PCAPCAplots indicate
plots that that
indicate the variability of batches
the variability comingcoming
of batches from Supplier 2 depends
from Supplier 2 quite exclusively
depends quite
on exclusively on only (PC1
only one variable one variable
of about(PC1
85%),of about 85%),aprobably
probably differentadistribution
different distribution
of waterof water matrices;
among among
on matrices; on the
the contrary, contrary,
at least two at least twoare
variables variables
involved areininvolved in the variability
the variability among
among sheets sheets
from from 1.
Supplier
Supplier
In fact, 1. In
in this fact,PC1
case, in this case,63%.
is only PC1 For
is only
this63%. For only
reason, this reason, only
matrices matrices
coming coming
from from2 Supplier
Supplier have been
2 have
further been furtherand
investigated investigated and used
used to prepare to prepare
placebo masks.placebo masks.
Finally, in order to understand the cause of the variability found
Finally, in order to understand the cause of the variability foundanalyzing
analyzingBC BCsheets
sheetsfrom
fromthethe
same batch, 4 selected sheets from each of three batches (Supplier 2) were evaluated,
same batch, 4 selected sheets from each of three batches (Supplier 2) were evaluated, picked from the picked from the
original
original packages
packages asas follows:
follows:
1604_1: first sheet on top of the stack
1604_1: first sheet on top of the stack
1604_2 &1604_3: sheets from the center of the stack
1604_2 & 1604_3:
1604_4: last sheetsheets
fromfrom the center of the stack
the bottom.
1604_4: last sheet from the bottom.
Hereafter, PCA and loading graphs are reported.
Hereafter, PCA and loading graphs are reported.
The results show very clearly that the first one sheet from the top (1604_1) is different from the
others (Figure 6A). The loading plot (Figure 6B) indicates that the same peaks, at around 1676 cm−1
and 1450 cm−1 , are responsible for this difference. This is probably due to a non-homogeneous
distribution of the aqueous solution used for soaking BC sheets, also provoking a different stretching
of the cellulose molecule OH, responsible for the hydrogen bond with water. This finding reveals that
a suitable storage protocol must be applied by manufacturers, before releasing every batch product,
to allow a thorough soaking process.

3.1.2. Quality Testing on Placebo Masks (PM)

The maximum wavelength of all recorded spectra values was 1453.2 cm–1 , for placebo masks
manufactured with BC Sheets (from Supplier 2), as shown in Figure 7A,B. This wavelength has
eventually been used to check the quality and homogeneity of samples. A correlation coefficient (95%)
between each spectrum was taken as a threshold for the acceptance of batches, as suggested by the
European Medicines Evaluation Agency (EMEA) notes of guidance for NIR applications in industrial
processes [24]. Samples were considered to conform if they had a maximum absorption within the
range mean ±95%. All samples tested showed acceptable values, according to this principle.
(A)
 original packages as follows:
1604_1: first sheet on top of the stack
1604_2 &1604_3: sheets from the center of the stack
1604_4: last sheet from the bottom.
Cosmetics 2018, 5, 66 12 of 20
Hereafter, PCA and loading graphs are reported.

Cosmetics 2018, 5, x FOR PEER REVIEW 12 of 20

(A)

(B)
Mathematical elaboration
Figure6.6.Mathematical
Figure elaborationofof
NIR spectra:
NIR (A) (A)
spectra: PCAPCA
plot and
plot(B)
andloadings plot for plot
(B) loadings PC1 for
forBCS
PC1 for
coming from Supplier 2.
BCS coming from Supplier 2.

The results show very clearly that the first one sheet from the top (1604_1) is different from the
others (Figure 6A). The loading plot (Figure 6B) indicates that the same peaks, at around 1676 cm−1
and 1450 cm−1, are responsible for this difference. This is probably due to a non-homogeneous
distribution of the aqueous solution used for soaking BC sheets, also provoking a different stretching
of the cellulose molecule OH, responsible for the hydrogen bond with water. This finding reveals
that a suitable storage protocol must be applied by manufacturers, before releasing every batch
product, to allow a thorough soaking process.

3.1.2. Quality Testing on Placebo Masks (PM)

The maximum wavelength of all recorded spectra values was 1453.2 cm–1, for placebo masks
manufactured with BC Sheets (from Supplier 2), as shown in Figure 7A,B. This wavelength has
eventually been used to check the quality and homogeneity of samples. A correlation coefficient
(95%) between each spectrum was taken as a threshold for the acceptance of batches, as suggested by
the European Medicines Evaluation Agency (EMEA) notes of guidance for NIR applications in
Cosmetics 2018, 5, 66 13 of 20
Cosmetics 2018, 5, x FOR PEER REVIEW 13 of 20

(A)

(B)
Figure7.7. NIR
Figure NIR spectra
spectra and
and mathematical
mathematical elaboration
elaboration of placebo
placebo masks
masks (PM)
(PM) produced
produced using
using BCS
BCS
comingfrom
coming fromSupplier
Supplier2:2:(A)
(A)NIR
NIRspectra
spectra and
and (B)
(B) loadings
loadings plot for PC1.

InInconclusion,
conclusion,NIRNIRspectroscopy
spectroscopyhas proven successful,
has proven whenever
successful, used to check
whenever usedthetohomogeneity
check the
ofhomogeneity
placebo masks of by monitoring
placebo masks theby water absorption
monitoring peak. absorption
the water Probably, the same
peak. approachthe
Probably, could
samebe
used to check
approach the be
could quality
used ofto BC masks
check the containing hydrophilic
quality of BC (water-soluble)
masks containing actives;
hydrophilic in such cases,
(water-soluble)
the presence
actives; of substances
in such dissolved
cases, the presence ofin water willdissolved
substances change the waterwill
in water absorption peak,
change the proportionally
water absorption
with
peak,the concentrationwith
proportionally of the
theactives.
concentration of the actives.
Changes
Changesarearealways
alwaysobserved,
observed, probably
probably due due to
to the
the folding
folding process
process and/or the positioning
and/or the positioningofof
sachetsinside
sachets insidethe
theouter
outerpackaging
packaging whenwhen stored;
stored; however, these changes
changes seem
seem toto be
be not
not significant
significant
enoughtotoaffect
enough affectthe
themoisture
moisturedistribution
distributionthrough
throughthetheentire
entireBCBCsheet.
sheet.There
Therewere
wereno noareas
areasshowing
showinga
a significant
significant decrease
decrease in in water
water content(graph
content (graphwould
wouldshow
showsuch suchaa behavior,
behavior, with PCAs seenseenasaspoints
points
distributedatataaconsiderable
distributed considerabledistance
distancefrom
fromthethe average
average value).
value).
Cosmetics 2018, 5, 66 14 of 20
Cosmetics 2018, 5, x FOR PEER REVIEW 14 of 20

The loading plots were extremely

extremely similar
similar among
amongmasks
masksfrom
fromthe
thesame
sametype:
type:this
thisalso
alsoshows
showsthat
that
variability
variability in the three different areas of the mask
mask (forehead,
(forehead, cheek,
cheek,chin)
chin)follows
followsthe
thesame
samepattern.
pattern.

3.2.
3.2. Stability Evaluation by
Stability Evaluation by Multiple
Multiple Light
Light Scattering
Scattering
Several variables could influence
variables could influence the
the stability
stability of
of solutions
solutions embedding
embeddingmasks,masks,such
suchasasthethe
interaction between substrate (cellulose matrix) and solution ingredients
interaction between substrate (cellulose matrix) and solution ingredients (rheology modifiers,(rheology modifiers,
preservatives).
preservatives). InIn this
this study,
study, placebo
placebo solutions
solutionsfrom
frommasks
masksbelonging
belongingtotothree
threedifferent
differentbatches
batches
(obtained
(obtained as described in Section
Section 2.3.2)
2.3.2) were
werecharacterized,
characterized,ininorder
ordertotopredict
predictsample
samplestability
stabilitydue
duetoto
possible
possible interaction between biopolymer
biopolymer matrix
matrixand
andsolution
solutionconstituents.
constituents.TheTheresults
resultsshowed
showeda large
a large
difference
difference in terms of physical
physical stability
stabilityof
ofsamples,
samples,directly
directlyrelated
relatedtotothe
themasks’
masks’ date
date ofof manufacture.
manufacture.
Figure
Figure 88shows
showstransmission
transmission profile (∆T)
profile of batch
(ΔT) P-MP-M
of batch 1, the1,oldest batchbatch
the oldest manufactured (twelve
manufactured months
(twelve
monthsthe
before before the analysis).
analysis).

Transmissionprofiles
Figure8.8.Transmission
Figure profilesofofP-M
P-M1.1.

The
The transmission profile revealed
transmission profile revealedthat
thatsome
someaggregation
aggregationoccurred
occurredon onthe
thesurface
surfaceofof the
the sample,
sample,
though
though notnot detectable
detectable byby visual
visual inspection.
inspection.This
Thiscould
couldclearly
clearlybe
beseen
seenasasthe
theresult
resultofofananinteraction
interaction
over
over time between the cellulose polymer and the placebo solution, probably leading to an extractionof
time between the cellulose polymer and the placebo solution, probably leading to an extraction
some
of somematrix molecules.
matrix molecules.This
Thishypothesis
hypothesis was confirmed
was confirmedbyby
a pH
a pHmeasurement
measurement of of
thethe
solution. AA
solution. pH
value of 6.77 for P-M 1 solution, was higher than those observed for P-M 2 and P-M 3,
pH value of 6.77 for P-M 1 solution, was higher than those observed for P-M 2 and P-M 3, two batchestwo batches that
maintained theirtheir
that maintained pH value (5.12(5.12
pH value and and
4.94 4.94
respectively).
respectively).
In
In Figure
Figure9 transmission
9 transmission profiles (∆T) (ΔT)
profiles of batches P-M 2 and
of batches P-MP-M 3 are P-M
2 and reported,
3 areandreported,
measurements
and
were made at different temperatures (20 ◦ C, then 4 ◦ C and, finally, 20 ◦ C again).
measurements were made at different temperatures (20 °C, then 4 °C and, finally, 20 °C again)
As it is possible to observe in Figure 9. After one cycle at 20 ◦ C, that the solutions showed an
irregular transmission profile in the range ±5%, probably due to the air bubbles and light foam,
formed when masks were squeezed to extract the sample solution.
No significant negative peak, corresponding to possible sedimentation, was observed. For each
sample, comparable profiles were obtained after the second and third analysis (4 ◦ C and
20 ◦ C): transmission variations were in an acceptable range, ±10%. As reported in the literature,
the transmission variation permits breaking phenomena to be detected at a very early stage in
non-absorbing concentrated dispersions. In particular, the instrument detects these phenomena
at least 4 times earlier than the naked eye for dilute emulsions (about 5%), and more than 20 times
earlier for concentrated ones (>20%) [20].
Cosmetics 2018, 5, 66 15 of 20

Cosmetics 2018, 5, x FOR PEER REVIEW 15 of 20

(A)

(B)

Figure 9. Cont.
 Cosmetics 2018, 5, 66 16 of 20
Cosmetics 2018, 5, x FOR PEER REVIEW 16 of 20

-4%

(C)
Figure 9. Transmission
Figure 9. Transmission profiles
profilesofofP-M
P-M22andandP-M
P-M 3:
3: (A)
(A) after one
one cycle
cycleat 20◦°C;
at20 (B)after
C; (B) afterone
onecycle
cycleatat
◦
4 (C)
C; (C) after ◦
4 °C; after thethe second
second cycleatat2020°C.C.
cycle

AsThese data confirmed

it is possible that in
to observe thermal
Figureshock did not
9. After onecause
cycleaggregation or flocculation
at 20 °C, that phenomena
the solutions showed an
for both batches (P-M 2 and P-M 3), their respective values being very similar.
irregular transmission profile in the range ±5%, probably due to the air bubbles and light foam,
Table 3 shows the Turbiscan stability index (TSI), at different times, for all samples tested.
formed when masks were squeezed to extract the sample solution.
No significant negative peak, corresponding to possible sedimentation, was observed. For each
Table 3. Turbiscan stability index (TSI)—solutions squeezed out of the masks.
sample, comparable profiles were obtained after the second and third analysis (4 °C and 20 °C):
transmission variations were inBatch 50
an acceptable 300 range,
1h 3h
±10%. 6 h reported in the literature, the
As
transmission variation permits breaking
P-M 1 0.7 phenomena
2.3 3.4 to be
6.1 detected
7.2 at a very early stage in non-
P-M 2 0.4 1.7 2.5 4.5 5.9
absorbing concentrated dispersions. In particular, the instrument detects these phenomena at least 4
P-M 3 0.2 1.5 1.7 3.1 3.9
times earlier than the naked eye for dilute emulsions (about 5%), and more than 20 times earlier for
concentrated ones (>20%) [20].
It seems
These data obvious
confirmed thatthat
batch P-M 3,shock
thermal the last
didone
notmanufactured, showed
cause aggregation the lowest values
or flocculation at all
phenomena
forreadings. As a (P-M
both batches rule, TSI index
2 and P-Mincreases
3), theirwith ageing values
respective of batches,
beingindicating that stability is worsening
very similar.
over time.
Table 3 shows the Turbiscan stability index (TSI), at different times, for all samples tested.
3.3. In Vivo Testing
Table 3. Turbiscan stability index (TSI)—solutions squeezed out of the masks.
The in vivo tests aimed at verifying the tolerability of bacterial cellulose masks when applied onto
Batch
the skin, allowing a novel approach 5′ release/delivery
for the 30′ 1 h 3 hof actives,
6h particularly small molecules
and hydrophilic substances. P-M 1 0.7 2.3 3.4 6.1 7.2
Some studies were conducted P-M to2 evaluate
0.4 1.7the2.5effect4.5 5.9 membranes on the skin [25–28].
of this
Amnuaikit et al. 2011 concludedP-M that3cellulose
0.2 1.5 1.7obtained
masks 3.1 3.9from A. xylinum could be used as a
natural cosmetic product, in order to increase moisture uptake in the skin. However, no other skin
It seems obvious
characteristics that batch
were deemed P-M 3,significantly
to change the last one[25].
manufactured, showed the lowest values at all
readings. As a rule, TSI index increases with ageing of batches, indicating that stability is worsening
over time.
 Some studies were conducted to evaluate the effect of this membranes on the skin [25–28].
Amnuaikit et al. 2011 concluded that cellulose masks obtained from A. xylinum could be used as a
natural cosmetic product, in order to increase moisture uptake in the skin. However, no other skin
characteristics were deemed to change significantly [25].
Results obtained from this work highlighted that a noticeable hydration effect is attributable
Cosmetics 2018, 5, 66
to
17 of 20
placebo masks (M). Upon a single application, hydration becomes statistically significant after two
hours (p = 0.0292*). Control area C (not treated) and W area (treated with water) were used for
Results obtained from this work highlighted that a noticeable hydration effect is attributable
comparison.
to placebo
Resultsmasks (M). Upon
are reported a single
in Figure 10A,application, hydration
showing that becomes
the cellulose statistically
matrix significant
can transfer water toafter
the
two hoursepidermal
outermost (p = 0.0292*). Control
layers; area does
this water C (not
nottreated) andasWquickly
evaporate area (treated with water)
as it happens when were used
free water
forapplied
is comparison.
onto the skin.
Results are reported
Biocellulose in Figure
masks seem to be a10A, showing
promising that the
delivery cellulose
system matrix can
for releasing transfer
active water toonto
ingredients the
outermost
the epidermal layers;
skin. Transepidermal this loss
water water does not
(TEWL) evaporate
values as quickly
confirmed that as
BCitmasks
happens when
were free
well water is
tolerated,
applied
and onto the skin.
the integrity of the barrier function was not affected (10B).

60
Water content (A.U.)

40

20

0
M 2H
M 4H

W H

W 2H
W 4H

C H

C 2H
C 4H
6H
M T0

W _T0

C 0
_T
6

6
_

_T
_T
_T
_T
_T
_T

_T
_T
_T
M

(A)
15
TEWL (g/m2h)

10

0
M 4H

W H
M 2H

W 2H
W 4H

C H

C 2H
C 4H
6H
M T0

W _T0

C 0
_T
6

6
_

_T
_T
_T
_T
_T
_T

_T
_T
_T
M

(B)
Figure 10.
Figure (A)Water
10.(A) Watercontent
content
andand(B)(B) transepidermal
transepidermal (TEWL)
(TEWL) datadata obtained
obtained in short
in short time analysis
time analysis after
after one
only onlyapplication
one application
on theon the following
following sites:
sites: M M (placebo
(placebo mask); mask);
W (areaWtreated
(area treated with water);
with water); C (not
C (not treated
treated areaasused
area used as control).
control).

Biocellulose
A monthlongmasks seem
clinical to be a promising
evaluation delivery system
has been performed as partfor
of releasing active
the present ingredients
study; onto
it has been of
the skin.
the Transepidermal
utmost importance to water loss (TEWL)
investigate values confirmed
the interactions that BC
between masks were
biocellulose maskswell and
tolerated,
skin
and the integrity
parameters. In a of the barrier
previous skinfunction was notstudy,
compatibility affected (Figure 10B).
Almeida et al. evaluated the skin irritating
A monthlong clinical evaluation has been performed
potential of biocellulose membranes through an occlusion test:as partconclusions
of the present study;
were thatitthis
has material
been of the
is
utmost
very importance
well tolerated to investigate the interactions between biocellulose masks and skin parameters. In a
[26].
previous skin compatibility study, Almeida et al. evaluated the skin irritating potential of biocellulose
membranes through an occlusion test: conclusions were that this material is very well tolerated [26].
Results obtained at the end of the long-term study confirmed the good tolerability of biocellulose
masks. The continued use of the masks did not cause any irritating effects (the change in EI between
T0 and T1 is of no importance, accounting for 0.02%, as shown in Figure 11A) while TEWL values
were statistically significant, but the values remains within the physiological range (medium TEWL
value at T1 is 10.71), (Figure 11B).
masks. The continued use of the masks did not cause any irritating effects (the change in EI between
T0 and T1 is of no importance, accounting for 0.02%, as shown in Figure 11A) while TEWL values
were statistically significant, but the values remains within the physiological range (medium TEWL
value at T1 is 10.71), (Figure 11B).
As 2018,
Cosmetics concerning
5, 66 water content, the results showed that PM did not significantly increase the
18 of 20
hydration of stratum corneum (Figure 11C).

500

400

300

E.I
200

100

T0

T1
(A)
20

15
TEWL (g/m2h)

10

0
T0

T1

(B)

80
Arbitrary Unit (A.U.)

60

40

20

0
T0

T1

(C)
Figure 11. (A)
(A) Erythema
Erythema index
index (B)
(B) TEWL,
TEWL, and
and (C)
(C) water
water content
content data
data obtained
obtained from
from 1 month of
continuous application of PM.

As concerning water content, the results showed that PM did not significantly increase the
4. Conclusions
hydration of stratum corneum (Figure 11C).
Bacterial cellulose is used, among several applications, for the manufacture of cosmetic masks,
with beautifying purposes, to promote skin nutrition and moisture, plus other desirable cosmetics
4. Conclusions
effects.
Bacterial cellulose is used, among several applications, for the manufacture of cosmetic
masks, with beautifying purposes, to promote skin nutrition and moisture, plus other desirable
cosmetics effects.
In order to develop stable, safe, and effective commercial products, it is essential to establish a
protocol to check their quality and stability. So far, no methods, nor complete protocols for quality
assurance of BC mask, have been published.
Cosmetics 2018, 5, 66 19 of 20

In this study, NIR and MLS techniques were applied to bacterial cellulose sheets and masks.
These fast and reproducible techniques proved to be suitable methods for quality and stability
testing. NIR proved to be able to detect changes in water and ingredients distribution in different
areas of the mask, and to ascertain the influence of folding as well. Multiple light scattering was
successful to evaluate stability, as well as possible interactions between the formulation and the
cellulose matrix, leading to a more precise definition of the expiry date (shelf-life) attributable to these
new delivery system.
If one refers to the tested products, the most likely shelf-life period for stable and safe commercially
available BC masks would be 6 months.
Erythema index values and TEWL values, obtained during in vivo tests, highlighted the great
tolerability of BC masks: skin parameters were not altered upon continued use, no occlusive effect was
reported, nor had the skin barrier function been affected.
To summarize, bacterial cellulose may be a promising substrate to develop a new class of cosmetics,
able to deliver a large number of hydrophilic/water-soluble substances onto the skin.

Author Contributions: P.P., M.B. and A.C. designed the study. P.P. performed the in vitro experiments,
M.B. performed the in vivo experiments. F.C. performed statistical analysis. All authors analyzed the results and
wrote the manuscript.
Funding: This research received no external funding.
Acknowledgments: The authors acknowledge Silvio Valle for technical support.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Lina, F.; Yue, Z.; Jin, Z.; Guang, Y. Bacterial Cellulose for Skin Repair Materials, Biomedical Engeneering.
In Frontiers and Challenges; Fazel-Rezai, R., Ed.; InTech Publisher: London, UK, 2011; pp. 249–274,
ISBN 978-953-307-309-5.
2. Chawla, P.R.; Bajaj, I.B.; Survase, S.A.; Singhal, R.S. Fermentative Production of Microbial Cellulose.
Food Technol. Biotechnol. 2009, 47, 107–124.
3. Torres, F.G.; Commeaux, S.; Troncoso, O.P. Biocompatibility of Bacterial Cellulose Based Biomaterials.
J. Funct. Biomater. 2012, 3, 864–878. [CrossRef] [PubMed]
4. Petersen, N.; Gatenholm, P. Bacterial cellulose-based materials and medical devices: Current state and
perspectives. Appl. Microbiol. Biotechnol. 2011, 91, 1277–1286. [CrossRef] [PubMed]
5. Czaja, W.K.; Young, D.J.; Kawecki, M.; Brown, R.M., Jr. The future Prospects of Microbial cellulose in
Biomedical Applications. Am. Chem. Soc. 2007, 8, 1–2. [CrossRef] [PubMed]
6. Gallegos, A.M.A.; Carrera, S.H.; Parra, R.; Keshavarz, T.; Iqbal, M.N. Bacterial Cellulose: A sustainable
source to develop value-added products—A review. Bioresources 2016, 11, 5641–5655. [CrossRef]
7. Legendre, J.Y. Assembly Comprising a Substrate Comprising Biocellulose, and a Powdered Cosmetic
Composition to Be Brought into Contact with the Substrate. U.S. Patent 2009/0041815, 12 February 2009.
8. Silva, N.H.C.S.; Drumond, I.; Almeida, I.F.; Costa, P.; Rosado, C.F.; Pascoal Neto, C.; Freire, C.S.R.;
Silvestre, A.J.D. Topical caffeine delivery using biocellulose membranes: A potential innovative system for
cellulite treatment. Cellulose 2014, 21, 665–674. [CrossRef]
9. Hasan, N.; Biak, D.R.A.; Kamarudin, S. Application of Bacterial Cellulose (BC) in Natural Facial Scrub. Int. J.
Adv. Sci. Eng. Inf. Technol. 2012, 2, 1–4. [CrossRef]
10. Ullah, H.; Santos, H.A.; Khan, T. Applications of bacterial cellulose bin food, cosmetics and drug delivery.
Cellulose 2016, 23, 2291–2314. [CrossRef]
11. Lee, C.K.; Hsu, K.C.; Cho, J.C.; Kim, Y.J.; Han, S.H. Cosmetic Bio-Cellulose Mask Pack Sheet and Method for
Manufacturing Same. U.S. Patent 2013/0244977A1, 19 September 2013.
12. Pacheco, G.; de Mello, C.V.; Galdorfini Chiari-Andreo, B.; Borges Isaac, V.L.; Lima Ribeiro, S.J.; Pecoraro, E.;
Trovatti, E. Bacterial cellulose skin masks—Properties and sensory tests. J. Cosmet. Dermatol. 2017, 17,
840–847. [CrossRef] [PubMed]
Cosmetics 2018, 5, 66 20 of 20

13. Reich, G. Near-infrared spectroscopy and imaging: Basic principles and pharmaceutical applications.
Adv. Drug Deliv. Rev. 2005, 57, 1109–1143. [CrossRef] [PubMed]
14. Roggo, Y.; Chalus, P.; Maurer, L.; Lema-Martinez, C.; Edmond, A.; Jent, N. A review of near infrared
spectroscopy and chemometrics in pharmaceutical technologies. J. Pharm. Biomed. Anal. 2007, 44, 683–700.
[CrossRef] [PubMed]
15. U.S. Department of Health and Human Services, Food and Drug Administration. Development and Submission
of Near Infrared Analytical Procedures- Guidance for Industry; FDA Office Publisher: Silver Spring, MD, USA,
2015; pp. 1–24.
16. Blanco, M.; Alcalá, M.; Planells, J.; Mulero, R. Quality control of cosmetic mixtures by NIR spectroscopy.
Anal. Bioanal. Chem. 2007, 389, 1577–1583. [CrossRef] [PubMed]
17. Celia, C.; Trapasso, E.; Cosco, D.; Paolino, D.; Fresta, M. Turbiscan Lab®Expert analysis of the stability of
ethosomes®and ultradeformable liposomes containing a bilayer fluidizing agent. Colloids Surf. B Biointerfaces
2009, 72, 155–160. [CrossRef] [PubMed]
18. Kirilov, P.; Rum, S. Aqueous dispersions of organogel nanoparticles—Potential systems for cosmetic and
dermo-cosmetic applications. Int. J. Cosmet. Sci. 2014, 36, 336–346. [CrossRef] [PubMed]
19. Olejnik, A.; Goscianska, J. Stability determination of the formulations containing hyaluronic acid. Int. J.
Cosmet. Sci. 2015, 37, 401–407. [CrossRef] [PubMed]
20. Mengual, G.; Meunier, I.; Cayre, K.; Puech, P. Snabre, Characterisation of instability of concentrated
dispersions by a new optical analyser: The TURBISCAN MA 1000. Colloids Surf. A Physicochem. Eng. Asp.
1999, 152, 111–123. [CrossRef]
21. Blichman, C.W.; Serup, J. Assessment of skin moisture. Measurement of electrical conductance,
capacitance and transepidermal water loss. Acta Dermato-Venereol. 1988, 68, 284.
22. Lee, C.M.; Maibach, H.I. Bioengineering Analysis of Water Hydration: An Overview. Exog. Dermatol. 2002, 1,
269–275. [CrossRef]
23. Pinnagoda, J.; Tupkek, R.; Agner, T.; Serup, J. Guidelines for transepidermal water loss (TEWL) measurement.
Contact Dermat. 1990, 22, 164–178. [CrossRef]
24. Guideline on the Use of Near Infrared Spectroscopy (NIRS) by the Pharmaceutical Industry and the Data Requirements
for new Submissions and Variations; EMEA/CHMP/CVMP/QWP/17760/2009 Rev2; EMEA Office Publisher,
EMEA Office: London, UK, 20 January 2012.
25. Amnuaikit, T.; Chusuit, T.; Raknam, P.; Boonme, P. Effects of a cellulose mask synthesized by bacterium on
facial skin characteristics and user satisfaction. Med. Devices Evid. Res. 2011, 4, 77–81.
26. Almeida, I.F.; Pereira, T.; Silva, N.H.C.S.; Gomes, F.P.; Silvestre, A.J.D.; Freire, C.S.R.; Sousa Lobo, J.M.;
Costa, P.C. Bacterial cellulose membranes as drug delivery systems: An in vivo skin compatibility study.
Eur. J. Pharma. Biopharm. 2014, 86, 332–336. [CrossRef] [PubMed]
27. Lin, Y.C.; Wey, Y.-C.; Lee, M.-L. Bacterial Cellulose Film and Uses. U.S. Patent 2011/0286948,
24 November 2011.
28. Lati, E.; Daniel, M.; Durand, C.; Sandager, E.; Monstatier, C. Evaluation of the release of active ingredients
from a bio-cellulose mask on support followed by a clinical evaluation of its efficiency in combination with a
serum with a depigmentation activity. In Proceedings of the International Federation of Societies of Cosmetic
Chemists (IFSCC), Paris, France, 27–30 October 2014.

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).