Efficacy of Transdermal Magnesium Ascorbyl Phosphate Delivery after
Ultrasound Treatment with Microbubbles in Gel-Type Surrounding Medium
in Mice
Ai-Ho Liao, Ying-Jui Lu, Chi-Ray Hung
PII:
DOI:
Reference:

S0928-4931(15)30673-1
doi: 10.1016/j.msec.2015.12.058
MSC 6051

To appear in:

Materials Science & Engineering C

Received date:
Revised date:
Accepted date:

13 April 2015
7 July 2015
28 December 2015

Please cite this article as: Ai-Ho Liao, Ying-Jui Lu, Chi-Ray Hung, Ecacy of Transdermal Magnesium Ascorbyl Phosphate Delivery after Ultrasound Treatment with Microbubbles in Gel-Type Surrounding Medium in Mice, Materials Science & Engineering C
(2015), doi: 10.1016/j.msec.2015.12.058

This is a PDF le of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its nal form. Please note that during the production process
errors may be discovered which could aect the content, and all legal disclaimers that
apply to the journal pertain.

ACCEPTED MANUSCRIPT
Efficacy of Transdermal Magnesium Ascorbyl Phosphate Delivery

SC
R

IP

Surrounding Medium in Mice

after Ultrasound Treatment with Microbubbles in Gel-Type

Ai-Ho Liao1,2,*, Ying-Jui Lu1, and Chi-Ray Hung1

Graduate Institute of Biomedical Engineering, National Taiwan University of

NU

MA

Science and Technology, Taipei 10607, Taiwan

Department of medical engineering, National Defense Medical Center, Taipei 11490,

TE

Taiwan

CE
P

*Corresponding author for reprint requests:

Name: Ai-Ho Liao, Ph.D.

AC

Address: Graduate Institute of Biomedical Engineering

National Taiwan University of Science and Technology,
TR-916, #43, Sec. 4, Keelung Rd,
Taipei, 10607, Taiwan
Phone:

+886-2-27303742

Fax:

+886-2-27303742

E-mail: aiho@mail.ntust.edu.tw

ACCEPTED MANUSCRIPT
ABSTRACT

Liquid microemulsions appropriate for topical application were obtained by

IP

increasing their viscosity through the addition of thickening agents. The present study

SC
R

first assessed the usefulness of ultrasound (US) plus US contrast agent, microbubbles
(MBs), in agarose gel for enhancing transdermal drug delivery. The effect of US plus

NU

MBs in agarose gel on the penetration of the skin by magnesium ascorbyl phosphate

MA

(MAP) was explored both in vitro and in vivo. In the in vitro experiments, the
stability of MBs was investigated by examining the penetration of MAP by the model

TE

drug, Evans blue, in two media: an agarose phantom and pig skin. The penetration

CE
P

depth in the agarose phantom and pig skin increased by 40% and 195%, respectively,
when treated with US plus MBs in 0.1% agarose solution combined with MAP

AC

(UMB1), and by 48% and 206%, respectively, when treated with US plus MBs in
0.15% agarose solution and MAP (UMB2). The skin-whitening effects in C57BL/6J
mice in the UMB1 and UMB2 groups over a 4-week experimental period were
significantly increased by 63% and 70%, respectively, in the fourth week. The
findings of this study suggest that the survival of MBs with US is affected by the
viscosity of the surrounding medium, and that in mice, treatment with US plus MBs in
a suitable agarose gel can increase skin permeability and enhance transdermal MAP
delivery.

ACCEPTED MANUSCRIPT
Keywords:

Microbubbles,

Agarose

gel,

Ultrasound,

Transdermal

AC

CE
P

TE

MA

NU

SC
R

IP

Magnesium ascorbyl phosphate, Melanin

delivery,

ACCEPTED MANUSCRIPT

1. INTRODUCTION

IP

The feasibility of controlled cavitation at high frequency for transdermal drug

SC
R

delivery (TDD) using gas-filled microbubbles (MBs) as ultrasound (US) contrast

agents was explored through in vivo experiments in a rat model [1]. Drug or

NU

cosmetics solutions were applied on the animal skin after all visible bubbles had

MA

disappeared, but it was found that liquid microemulsions were more appropriate for
topical application. MB behavior in an US field was recently investigated through

TE

numerical simulation in an effort to enhance high-intensity focused US therapies.

Both the viscosity and shear elasticity of the medium surrounding the MBs reduced

CE
P

the attenuation of US propagation through the MB mixture [2]. Our previous study
demonstrates the penetration depth, concentration, and efficiency of transdermal

AC

-arbutin delivery during 4 weeks after US treatment with MBs solution in mice [3].
The present study explored the feasibility of a gel-type MB compound combined with
US for enhancing the penetration of transdermal magnesium ascorbyl phosphate
(MAP) delivery in vivo.
US-achieved sonophoresis is known to increase skin permeability, but the
fundamental mechanism underlying this effect is still unclear [4]. Shock waves and
microjets generated during inertial cavitations are thought to be responsible for

ACCEPTED MANUSCRIPT
transdermal permeability enhancement, with microjets exerting a significantly greater

effect than shock waves [5]. Such cavitation occurs at various sites, such as the

IP

coupling medium, the skin surface, and the skin tissue [4]. The viscosity, surface

SC
R

tension, density, acoustic impedance, and other bulk and interfacial properties of the
coupling medium play an important role in the enhancement of skin permeability [6].

NU

The most common type of drug delivered through the skin using high-frequency

MA

sonophoresis is anti-inflammatory medication for joint and muscle pain, with a

recently increasing shift in interest from topical steroids to nonsteroidal

TE

anti-inflammatory drugs (NSAIDs), including diclofenac, ibuprofen, ketoprofen,

CE
P

ketorolac, and piroxicam [6]. Moreover, oral NSAIDs generally cause gastrointestinal
side effects, including nausea, heartburn, gastrointestinal ulcers, and nonspecific

AC

colitis [7]. Therefore, the combination of topical NSAID therapy and US is promising;
however, the combination of gel-type MBs plus US for TDD is necessary for such
therapy.

Vitamin C (L-ascorbic acid) is a water-soluble vitamin synthesized from

D-glucose

in the mammalian liver, or in the kidneys in some vertebrates [8]. It can

inhibit melanogenesis, promote collagen biosynthesis, and prevent the formation of

free radicals in the skin [9-13]. It has been considered an interesting ingredient of
cosmetic skin-care products, but formulating products containing vitamin C is

ACCEPTED MANUSCRIPT
impractical because it is readily soluble in water and is extremely unstable [9].

Therefore, vitamin C is chemically modified by esterification of the hydroxyl group

IP

with long-chain organic or inorganic acids. The derivative, magnesium ascorbyl

SC
R

phosphate (MAP), an inorganic water-soluble acid ester, was formulated to overcome

the instability of vitamin C [14]. MAP is a very stable derivative of vitamin C that

NU

may be easily used in various types of cosmetic product. MAP decreases

MA

melanogenesis by interacting with copper at the active site of tyrosinase and by

reducing dopaquinone by blocking dihydrochinindol-2-carboxyl acid oxidation [15].

TE

Shaikh and Mashood (2014) determined the effectiveness of treating melasma with a

CE
P

combination of topical 5% MAP and pulsed fluorescent light, and found that
combining 5% MAP with pulsed fluorescent light is an effective, well-tolerated, and

AC

safe method of treating refractory melasma in Asian patients [16].

The use of US to enhance the delivery of related drugs via the skin surface has
been widely studied. However, the penetration of the stratum corneum lipid bilayer by
water-soluble drugs is a major challenge. The effect of MBs in combination with US
on drug penetration through the skin was evaluated in the present study. In order to
facilitate the smearing of MBs on the skin surface, they were formulated in a gel-type
composition. Since the viscosity of the agarose increases with concentration. In this
study, the efficiency of US plus MBs in different concentrations of agarose gel for

ACCEPTED MANUSCRIPT
enhancement of skin permeability has been demonstrated. The properties of liquid and

gel-type MBs used for US enhanced transdermal MAP delivery were demonstrated

IP

both in vitro and in vivo. US combined with gel-type MBs was found to induce

SC
R

cavitation and thus enhance TDD.

2.1 Production of albumin-shelled MBs

NU

2. MATERIALS AND METHODS

MA

Albumin-shelled MBs were prepared according to the procedure used in our

previous studies [17, 18]. Briefly, albumin-shelled MBs were generated by sonication

TE

in 10 ml of a solution containing 140 mg of albumin (Octapharma, Vienna, Austria)

CE
P

and perfluorocarbon gas in physiological saline (pH 7.4, 0.9% sodium chloride) using
a sonicator (Branson Ultrasonics, Danbury, CT, USA) for 2 min. The number of

AC

perfluorocarbon-filled albumin MBs in the solution was measured using the

MultiSizer III device (Beckman Coulter, Fullerton, CA, USA) with a 30-m-aperture
probe and measurement boundaries of 0.620 m. The size distribution in the
suspension was measured based on dynamic light scattering (Zetasizer Nano, ZS90,
Malvern, UK). The albumin-shelled MBs had a mean diameter of 1.2 m and a mean
concentration of 2.3109 bubbles/ml. The albumin-shelled MBs were centrifuged
(1200 rpm for 1 minute, 128.6g), and half of the physiological saline was removed.
The final concentration of MBs used in this study was 4.640.05109 bubbles/ml
7

ACCEPTED MANUSCRIPT
(meanSD) for a size range of 0.82 m, and their mean diameter was 1195 nm.

2.2 Preparation of agarose-gel-type MBs

IP

Two amounts of agarose powder (0.1 and 0.15 mg; FB-HA0604, FocusBio,

SC
R

Burgos, Spain) were dissolved in 99.9 and 99.85 ml of phosphate-buffered saline,

respectively, and heated to boiling. The solution was fixed on a rotary shaker (50 rpm;

NU

Shaker RS-01, TKS, New Taipei City, Taiwan) for 20 min. The agarose solution was

MA

then placed into the sonication tank (Delta ultrasonic cleaner d150h, Delta, Hsin-Chu

City, Taiwan) to remove the excess gas, and finally stored at 4C in a refrigerator.

TE

Before the experiments were performed, the gel-type MBs were prepared in either

CE
P

0.1% or 0.15% agarose solution at the following four concentrations: 0.23107,

4.6107, 11.5107, and 23107 bubbles/ml. The albumin MBs and gel-type MBs were

AC

filtered with a 5 m syringe filter (Critical Process Filtration, Inc., Nashua, NH, USA)
and then hardened by 0.002% glutaraldehyde (Nippon Shiyaku Co., Tokyo, Japan).
The morphology of the hardened albumin MBs was studied using scanning electron
microscopy (SEM) (Quanta 3D FEG, FEI, ORR, USA). The samples were prepared
for SEM by coating with platinum. SEM images were recorded on a system at an
accelerating voltage of 30 kV.

2.3 In vitro high-frequency US imaging of MBs and gel type MBs

High-frequency US imaging was performed to evaluate the survival of MBs in
8

ACCEPTED MANUSCRIPT
agarose gel during sonication using a US animal-imaging system (Prospect, S-Sharp

Corporation, New Taipei City, Taiwan) at a frequency of 40 MHz using a transducer

IP

with a diameter of 7 mm and a fixed focus at 12 mm. A 2% agarose square column

SC
R

10x20x20-mm3 phantom was constructed with a 2220-mm3 chamber at its center to

load the 4107 bubbles/ml MBs or gel-type MBs. According to our previous study, the

NU

enhancement of the penetration depth was greatest for 2-W/cm2 US and the condition

MA

was used either for the in vitro skin penetration or for the animal treatments in this
study [3]. During the US imaging, the loaded phantom was sonicated using the

TE

1-MHz US transducer of the sonoporation gene transfection system (ST 2000V,

CE
P

NepaGene, Ichikawa, Japan) at an acoustic intensity of 2 W/cm2 for 1 min. The duty
cycle was set at 50%, the PRF was set at 250 msec and a 6-mm-diameter transducer

AC

was used. The region of interest was drawn over the entire MB-loaded chamber in a
two-dimensional imaging plane by an operator, the dynamic range was set at 50 dB
and the average pre- and postsonication image intensities were measured in B-mode
images.

2.4 Measurement of penetration depth in agarose phantoms

The model drug, Evans blue (0.1 mg; molecular weight=960.81 Da; E2129,
Sigma-Aldrich, St. Louis, MO, USA), was dissolved in 10 ml of physiological saline
(0.9% sodium chloride) and then stirred for 1 hour at 4C. Disk-shaped 0.3% agarose
9

ACCEPTED MANUSCRIPT
phantoms were constructed with a diameter of 1.2 cm and a height of 5 mm (encircled

with US gel to prevent leakage); the round area of each phantom was loaded with

IP

Evans blue or MBs. The probe of the sonoporation gene transfection system (ST

SC
R

2000V, NepaGene) was placed 5 mm from the top of the phantom. After adding
500 l of the 0.1% agarose gel, 0.15% agarose gel, or the MBs at four different

NU

concentrations (0.23107, 4.6107, 11.5107, and 23107 bubbles/ml) in 0.1% or

MA

0.15% agarose gel, the area was sonicated using the 1-MHz US transducer of the
sonoporation system at an acoustic intensity of 2 W/cm2 for 1 min. The duty cycle

TE

was set at 50% and a 1.2-cm-diameter transducer was used. The change in

CE
P

temperature during the US sonication was only 0.3C, as measured by a thermometer

(Optris LS, Optris, Berlin, Germany). The MBs were subsequently removed from the

AC

surface and the area was washed three times for 1 min with physiological saline. The
Evans-blue solution was then injected into the same area on the phantom and the
sample left for 5 min to allow penetration. The Evans blue was then removed and the
area was washed three times for 1 min with physiological saline. Sections of the
phantom were cut at a thickness of 2 mm and prepared for light-microscopy
evaluation.
The penetration depths of the Evans blue were measured using MATLAB (The
MathWorks, Natick, MA, USA). At first, the light-microscopy images were converted

10

ACCEPTED MANUSCRIPT
into grayscale images and image histogram-based binarization was performed [19].

For histogram-based binarization, the peak-and-valley thresholding of target was

IP

performed based on the histogram for many experiments [19]. Then, the boundary

SC
R

was detected by Sobel-operator-based edge detection [20]. At this step, the same
threshold (100-230) was used when processing all of the images. Finally, the area of

NU

the penetration region was measured and the area was divided by the length of the

MA

x-axis of the image to get the mean penetration depth (y-axis) of the Evans blue.

2.5 Measurement of penetration depth in pig skin

TE

Fresh porcine skin was obtained from a local slaughterhouse. The protocol for

CE
P

the experiments involving pig skin was similar to that described for the agarose
phantom. The treatment area on the 2-mm-thick pig skin was sonicated by the 1-MHz

AC

US transducer of the sonoporation system at an acoustic intensity of 2 W/cm2 for

1 min after adding 500 l of MBs. The MBs were removed and the area was washed
three times for 1 min with physiological saline. The Evans-blue solution was then
applied to the same area on the pig skin for 5 min. The Evans blue was subsequently
removed from the surface and the area was washed three times for 1 min with
physiological saline. The treated areas of pig skin were cut off and then embedded
with Surgipath FSC 22 optimal-cutting-temperature solution (Leica Microsystems,
Buffalo Grove, IL, USA) on round specimen disks with a diameter of 2.2 cm. The
11

ACCEPTED MANUSCRIPT
embedded samples were placed on the 25C freezing stage of a cryostat (Microm

HM550 series, Thermo, Braunschweig, Germany) for about 30 min. Transverse

IP

sectioning was performed at a slice thickness of 5 m. Sections were attached to

SC
R

microscopy slides, air-dried at room temperature, and then mounted for microscopy
examination. The average penetration depth was measured by sampling 10 depths on

MA

NU

each slide (15 slices per experiment and each experiment repeated 4 times).

2.6 Animal treatments

TE

The melanin content of origanoside was investigated in the C57BL/6J mouse

CE
P

model [21]. Five-week-old mice weighing 2025 g were obtained from Bio Lasco
(Taipei, Taiwan). The experimental protocol was approved by the Institutional Animal

AC

Care and Use Committee of the National Taiwan University Hospital, Taipei, Taiwan.
Animal care complied with institutional guidelines and regulations. Throughout the
experiments, the animals were housed in stainless-steel cages in an air-conditioned
room with the temperature maintained at 2528C and with alternating light and dark
cycles lasting 12 hours each. The animals were acclimatized for 7 days prior to the
experiment. After their hair had been removed, the skin color was measured using a
Chroma Meter (CR-400, Konica Minolta Sensing, Tokyo, Japan). The animals were
exposed to ultraviolet radiation B (UVB) irradiation (G8T5E, Sankyo, Tokyo, Japan)
12

ACCEPTED MANUSCRIPT
to

induce

hyperpigmentation

(total

energy

dose

per

exposure=1 J/cm2,

wavelength=306 nm, three times/week for 2 weeks), and then the skin color was

IP

measured again.

SC
R

The details of the experimental design are shown in Fig. 1. The animals were
divided into the following five groups (n=5 per group, treatment applied three

NU

times/week for 4 weeks): (1) control (no treatment, C), (2) penetrating MAP only (M),

MA

(3) US combined with penetrating MAP (UM), (4) US plus MB contrast agent in
0.1% agarose solution combined with penetrating MAP (UMB1), and (5) US plus MB

TE

contrast agent in 0.15% agarose solution combined with penetrating MAP (UMB2).

CE
P

The change in skin color induced by each of the treatments was assessed at
predetermined times using the Chroma Meter. The luminosity index, L [22], was

AC

calculated on each measurement day before and after treatment.

2.7 Histochemistry
Skin tissue samples (approximately 88 mm2) were taken from the treatment
area immediately after the experiments and stored in 10% formalin solution.
Hematoxylin and eosin (Sigma-Aldrich) staining was applied, and the samples were
analyzed to evaluate the relative melanin content and damages in the skin structure
after US treatments.

13

ACCEPTED MANUSCRIPT
2.8 Statistical analysis

The obtained data were analyzed statistically using Students t-test. A probability

SC
R

IP

value of p<0.05 was considered indicative of a significant difference.

3. RESULTS

NU

3.1 In vitro high-frequency US imaging of gel-type albumin-shelled MBs

SEM images show that the smooth surface of the MBs (Fig. 2(A)), and a large

MA

area of the MBs surface in agarose was found for gels generated by copolymerization

(Fig. 2(B), (C)). US images of 4107 bubbles/ml albumin-shelled MBs, MBs in 0.1%

TE

agarose gel, MBs in 0.15% agarose gel without and with US sonication at 2 W/cm2

CE
P

for 1 min are shown in Fig. 3(A)(C) and 3(D)(F), respectively; the signal intensities
in panels Fig. 3(A)(F) are quantified in Fig. 3(G). The image intensities of the MBs

AC

in these conditions were 28.61.0, 27.50.4, and 27.11.0 dB in (A)(C), respectively,

and 18.22.0, 4.51.0, and 4.91.0 dB in (D)(F), respectively, indicating that the
gel-type albumin-shelled MBs were significantly destroyed after being sonicated with
US [p0.001; Fig. 3(G)].
3.2 Evaluations of the penetration depth in agarose phantoms
Figure 4 quantifies the penetration depths in the four groups when the MBs were
added

at

various

concentrations

(0.23107,

4.6107,

11.5107,

and

23107 bubbles/ml). Both the agarose concentration and the concentration of MBs
14

ACCEPTED MANUSCRIPT
affected the penetration depth. At the MBs concentration of 0.23107 bubbles/ml, the

penetration depth was enhanced significantly more when using 0.1% agarose gel than

IP

0.15% agarose (p<0.05). At the MBs concentration of 4.6107 bubbles/ml, the

SC
R

penetration depth was enhanced significantly more when using 0.15% agarose gel
than 0.1% agarose (p<0.05). The penetration depth was enhanced significantly more

NU

when using 0.1% agarose gel combined with 23107 bubbles/ml and 0.15% agarose

MA

gel combined with 4.6107 bubbles/ml than their control group (p<0.001). Figure 5
[(A)(D) and (E)(H)] shows microscopy images before and after imaging processing

TE

of the agarose phantoms without US, with US (at an acoustic intensity of 2 W/cm2),
US combined with MBs (23107 bubbles/ml) in 0.1% agarose, and US combined with

CE
P

MBs (4.6107 bubbles/ml) in 0.15% agarose when the Evans-blue solution was

AC

allowed to stand for 5 min. In the US-only group there was no significant difference
in the penetration depth (p>0.05) compared to the control condition. However, the
penetration depths of MBs (23107 bubbles/ml) in 0.1% agarose, and US combined
with MBs (4.6107 bubbles/ml) in 0.15% agarose were 44% and 44.7% greater than
in the control

group

(p<0.01).

The penetration depth was

greater for

23107 bubbles/ml MBs in 0.1% agarose (589.344.0 m) and 4.6107 bubbles/ml

MBs in 0.15% agarose (606.824.0 m) than in the control groups (0.1% agarose,
419.420.0 m; 0.15% agarose, 409.249.0 m).

15

ACCEPTED MANUSCRIPT
3.3 Evaluation of the penetration depth in pig skin

The pig-skin samples in groups UM (US plus water, 0.1% agarose gel or 0.15%

IP

agarose gel, combined with Evans blue), UMB (US plus MB contrast agent in water

SC
R

combined with Evans blue), UMB1 (US plus MB contrast agent in 0.1% agarose gel
combined with Evans blue), and UMB2 (US plus MB contrast agent in 0.15% agarose

NU

gel combined with Evans blue) were cryosectioned for light-microscopy evaluation at

MA

magnifications of 400 (Fig. 6; Primo Star, Zeiss-Jena, Jena, Germany). Figure 7

shows measurements of the penetration depths in the five groups (n=4) for various

TE

concentrations of MBs and agarose. The degree of penetration in either the cuticle or
the epidermis was significantly greater in groups UMB, UMB1, and UMB2 than in

CE
P

group UM (p<0.05). Moreover, the penetration of 23107, 11.5107, and 0.23107

AC

MBs in water was significantly greater than in agarose gel (p<0.05). For MB contrast
agent in agarose gel, the greatest conditions were 11.5107 bubbles/ml MBs in 0.1%
agarose (14.81.1 m) and in 4.6107 bubbles/ml MBs in 0.15% agarose
(14.40.8 m). The greatest conditions were not significantly different from 4.6107
MBs in water (p>0.05). The penetration depths in those groups treated with 0.1% and
0.15%

agarose

were

5.00.6

and

4.60.6 m,

respectively.

When

11.5107 bubbles/ml MBs in 0.1% agarose or 4.6107 bubbles/ml MBs in 0.15%

agarose were added, the penetration depths were 196% and 213% greater than with

16

ACCEPTED MANUSCRIPT
US only in 0.1% and 0.15% agarose, respectively (p<0.05). Therefore, the

enhancement of the penetration depth was dependent upon the proper proportions of

IP

MBs and agarose gel for both the in vitro skin penetration and for the animal

SC
R

treatments. Moreover, there was no significant difference between the penetration

depth with 11.5107 bubbles/ml MBs in 0.1% agarose (14.81.1 m) and

NU

4.6107 bubbles/ml MBs in 0.15% agarose.

MA

3.4 Animal treatments

Figure 8 shows photographs of the mouse skin (Fig. 8AE) and histology images

TE

(Fig. 8FJ) for groups C (Fig. 8A, F), M (Fig. 8B, G), UM (Fig. 8C, H), UMB1

CE
P

(Fig. 8D, I), and UMB2 (Fig. 8E, J) at week 4. Figure 9 plots the brightness values
(i.e., L) to demonstrate the whitening effects of MAP, US, and gel-type MBs on

AC

UV-induced hyperpigmentation over a 4-week period. Fig. 8B, Fig. 8C and Fig. 9
show that the brightness of the skin in groups M and UM had recovered. The skin
brightness was more effectively increased and closer to the original color in groups
UMB1 and UMB2 than in groups C, M, and UM. The skin brightness in the
observation area was more uniform in groups UMB1 and UMB2 than in groups C, M,
and UM. Moreover, no skin damage was evident in any of the US treatment groups.
Histology analysis revealed that both the UMB1 and UMB2 treatments resulted in a
significant decrease in the relative melanin content of the skin. No changes in the skin
17

ACCEPTED MANUSCRIPT
structure or bilayerbilayer interface were observed in any of the treatment groups.

In Fig. 9, the brightness value was around 36.238.4 in each group after UVB

IP

exposure. At week 1 the brightness values in groups UM, UMB1, and UMB2 had

SC
R

increased by 25%, 31%, and 31%, respectively. Significant differences (p<0.05) were
obvious between each US treatment group and non-US treatment group (i.e., C and

NU

M), but not between groups UM, UMB1, and UMB2 (p>0.05). At week 2 the

MA

brightness values in groups M, UM, UMB1, and UMB2 had increased by 35%, 41.8%,
49.6%, and 57.7%, respectively. At week 3 the brightness values in groups UMB1 and

TE

UMB2 had increased by 58.6% and 63.1%, respectively, making them close to the

CE
P

original skin color, while those in groups M and UM had increased by smaller
amounts (by 52.1% and 54.5%, respectively). At week 4 the brightness values in

AC

groups M, UM, UMB1, and UMB2 had reached a plateau, with increases of 57%,
61.4%, 63.3%, and 69.9%, respectively, while that in group C had increased by
31.1%.

4. DISCUSSION
Gel cream containing vitamin C can be applied to the skin in patients to suppress
melanin formation by tyrosinase and melanoma cells therein [23]. Since a solution
containing MBs has a very low viscosity, it cannot be applied directly to the skin. The
usual way of resolving this problem is to add a suitable thickening agent that can
18

ACCEPTED MANUSCRIPT
modify the rheological properties without significantly influencing the other features

of MBs [24, 25]. In the present study, MBs in agarose gel were applied with US to the

IP

skin to enhance MAP delivery. Agar gels have been found useful in the fields of

SC
R

medicine and pharmacy [26], and are commonly used in the food industry as
thickening and gelling agents [26, 27]. Moreover, agarose gel with graphite suspended

NU

in it can be constructed as US phantoms, and mimic the acoustic properties of human

MA

tissue (e.g., speed of sound and average attenuation) [28]. Rozman and Gasperlin
(2007) used gel-like microemulsions as carrier systems for the dermal delivery of

TE

vitamins C and E, and found that these microemulsions offered the best protection for

CE
P

both vitamins and increased their stability compared with that of a basic vitamin
solution [29]. Those authors also found that the addition of thickener significantly

AC

increased the viscosity and changed the behavior of the delivery systems. According
to the results of the present study, although agarose gel did not affect the survival of
MBs before sonication, it did affect their behaviors with US treatments and the
efficacy of drug penetration. This efficacy can be controlled by optimizing the MB
conditions, such as their concentration. According to the results of phantom study
(Fig. 5), the use of optimum concentrations of agarose and MBs0.1% agarose gel
combined with 23107 bubbles/ml and 0.15% agarose gel combined with
4.6107 bubbles/mlsignificantly enhanced the MAP penetration depth. In pig skin,

19

ACCEPTED MANUSCRIPT
although the penetration of 23107, 11.5107, and 0.23107 MB in water was

significantly greater than in agarose gel. It is not easy to encircle with US gel to

IP

prevent leakage and apply to the skin surface in living animals. However, MBs in gel

SC
R

type surrounding medium is still a challenge. Evaluation of the penetration depth in

pig skin revealed optimum agarose and MB concentrations of 0.1% agarose gel

NU

combined with 11.5107 bubbles/ml and 0.15% agarose gel combined with

MA

4.6107 bubbles/ml. These two conditions were not significantly different from
4.6107 MBs in water. Moreover, the penetration in pig skin with 0.15% agarose gel

TE

was more uniform than with 0.1% agarose gel at all concentrations of MBs tested.

CE
P

Thus, 0.15% agarose gel could improve the uniformity of the effects of MBs and US
on the skin surface. This finding is consistent with those of the small-animal

AC

experiments, which demonstrated that the skin brightness in the observation area was
more uniform in groups UMB1 and UMB2 than in the other groups.
In animal experiments, the skin brightness values of the control group (C)
increased very slowly, and those of the MAP group (M) increased relatively slowly in
the first week. However, during the same period, the skin brightness values of the UM,
UMB1, and UMB2 groups increased more clearly. By the second week the rate of
brightness increase in the M group was similar to that in the UM, UMB1, and UMB2
groups. During the 4 weeks of observation, the brightness values of the UMB1 and

20

ACCEPTED MANUSCRIPT
UMB2 groups were similar, and higher than those of the UM group. This may be due

to the cavitation induced by US, which may have improved the permeability of the

IP

lipid bilayer membrane of the stratum corneum for a short time. Thus, combined

SC
R

gel-type MBs and US and US alone can enhance MAP delivery at the early stage.
Combining US with MBs increased the delivered concentration of MAP and did not

NU

influence the time trend of MAP delivery. After US treatment with gel-type MBs in

MA

mice, more MAP passed through the stratum corneum lipid bilayer and directly

affected the tyrosinase activity so as to inhibit the formation of melanin.

TE

5. CONCLUSIONS

CE
P

Although the penetration of some concentrations of MBs in water was

significantly greater than in agarose gel, it is not easy to apply to skin surface in living

AC

animal. The potential of gel-type MBs in combination with US to enhance

transdermal MAP delivery was evaluated in this study using both in vitro and in vivo
methods. It is expected that gel-type MBs containing formulations of different drugs
for TDD will be evaluated in the near future. The present study applied MBs in an
agarose gel preparation to the skin in combination with US to enhance MAP delivery.
The agarose gel positively affected both the behavior of the MBs in combination with
US energy and the efficacy of TDD enhancement. The optimal MB and agarose

21

ACCEPTED MANUSCRIPT
concentrations were determined. The in vivo results demonstrated that combination

treatment with US and MBs in agarose gel (at optimum concentrations of both) can

NU

ACKNOWLEDGMENTS

SC
R

melanogenesis without damaging the skin.

IP

increase skin permeability in mice and thus enhance MAP delivery to inhibit

MA

This work was supported by the Ministry of Science and Technology of the
Republic of China, Taiwan under Contract of MOST 100-2628-E-011-015-MY3 and

CE
P

REFERENCES

TE

103-2221-E-011-004-MY3.

[1] D. Park, H. Ryu, H.S. Kim, Y.S. Kim, K.S. Choi, H. Park, J. Seo, Sonophoresis

AC

using ultrasound contrast agents for transdermal drug delivery: an in vivo

experimental study, Ultrasound Med Biol 38 (2012) 642650.
[2] K. Okita, K. Sugiyama, S. Takagi, Y. Matsumto, Microbubble behavior in an
ultrasound field for high intensity focused ultrasound therapy enhancement, J
Acoust Soc Am 134 (2013) 15761585.
[3] A.H. Liao, W.C. Ma, C.H. Wang, M.K. Yeh, Penetration depth, concentration and
efficiency of transdermal -arbutin delivery after ultrasound treatment with
albumin-shelled microbubbles in mice, Drug Deliv (2015) in press.
22

ACCEPTED MANUSCRIPT
[4] D. Park, H. Park, J. Seo, S. Lee, Sonophoresis in transdermal drug deliverys,

Ultrasonics 54 (2014) 5665.

SC
R

sonophoresis, J Control Release 148 (2010) 204211.

IP

[5] L. Wolloch, J. Kost, The importance of microjet vs shock wave formation in

[6] Y.Z. Zhao, L.N. Du, C.T. Lu, Y.G. Jin, S.P. Ge, Potential and problems in

NU

ultrasound-responsive drug delivery systems, Int J Nanomedicine 8 (2013)

MA

16211633.

[7] K.D. Rainsford, Profile and mechanisms of gastrointestinal and other side effects

TE

of nonsteroidal anti-inflammatory drugs (NSAIDs), Am J Med 107 (1999)

[8]

D.C.

CE
P

27S35S.

Mahan,

A.J.

Lepine,

K.

Dabrowski,

Efficacy

of

AC

magnesium-L-ascorbyl-2-phosphate as a vitamin C source for weanling and

growing-finishing swine, J Anim Sci 72 (1994) 23542361.
[9] R. Austria, A. Semenzato, A. Bettero, Stability of vitamin C derivatives in solution
and topical formulations, J Pharm Biomed Anal 15 (1997) 795801.
[10] R. Crippa, V. Horak, G. Prota, P. Svoronos, L. Wolfram, Chemistry of melanins,
1th ed., Academic Press, New York, 1989.
[11] S. Murad, D. Grove, K.A. Lindberg, G. Reynolds, A. Sivarajah, S.R. Pinnell,
Regulation of collagen synthesis by ascorbic acid, Proc Natl Acad Sci USA 78

23

ACCEPTED MANUSCRIPT
(1981) 28792882.

[12] P.T. Pugliese, C.B. Lampley, Biochemical assessment of the anti-aging effects of

IP

cosmetic products, J Appl Cosmetol 3 (1985) 129138.

SC
R

[13] J.M. Wilmott, M.C. Duggan, A.P. Znaiden, Sunscreens: Developments,

Evaluation, and Regulatory Aspects. 1th ed., Marcel Dekker, New York, 1990.

NU

[14] A.I. Segall, M.A. Moyano, Stability of vitamin C derivatives in topical

MA

formulations containing lipoic acid, vitamins A and E, Int J Cosmet Sci 30 (2008)
453458.

TE

[15] A. Slominski, D.J. Tobin, S. Shibahara, J. Wortsman, Melanin pigmentation in

CE
P

mammalian skin and its hormonal regulation, Physiol Rev 84 (2004) 11551228.
[16] Z.I. Shaikh , A.A. Mashood, Treatment of refractory melasma with combination

AC

of topical 5% magnesium ascorbyl phosphate and fluorescent pulsed light in

Asian patients, Int J Dermatol 53 (2014) 9399.
[17] A.H. Liao, H.L. Liu, C.H. Su, M.Y. Hua, H.W. Yang, Y.T. Weng, P.H. Hsu, S.M.
Huang,

S.Y.

Wu,

H.E.

Wang,

T.C.

Yen,

P.C.

Li,

Paramagnetic

perfluorocarbon-filled albumin-(Gd-DTPA) microbubbles for the induction of

focused-ultrasound-induced blood-brain barrier opening and concurrent MR and
ultrasound imaging, Phys Med Biol 57 (2012) 27872802.
[18] A.H. Liao, C.C. Shen, C.H. Cheng, H.C. Chuang, C.H. Wang, C.H. Lin,

24

ACCEPTED MANUSCRIPT
Measurements of lifetime and attenuation properties of ultrasound/magnetic

resonance multimodality molecular probe, J Med Biol Eng 33 (2013) 285292.

IP

[19] M. Sezgin, B. Sankur, Survey over image thresholding techniques and

SC
R

quantitative performance evaluation, J Electron Imaging 13 (2004) 146165.

[20] H. Ahammer, T.T. DeVaney, The influence of edge detection algorithms on the

NU

estimation of the fractal dimension of binary digital images, Chaos 14 (2004)

MA

183188.

[21] M. Ishikawa, I. Kawase, F. Ishii, Glycine inhibits melanogenesis in vitro and

TE

causes hypopigmentation in vivo, Biol Pharm Bull, 30 (2007) 20312036.

CE
P

[22] Y.H. Tsai, K.F. Lee, Y.B. Huang, C.T. Huang, P.C. Wu, In vitro permeation and in
vivo whitening effect of topical hesperetin microemulsion delivery system. Int J

AC

Pharm 388 (2010) 257262.

[23] K. Kameyama, C. Sakai, S. Kondoh, K. Yonemoto, S. Nishiyama, M. Tagawa, T.
Murata, T. Ohnuma, J. Quigley, A. Dorsky, D. Bucks, K. Blanock, Inhibitory
effect of magnesium L-ascorbyl-2-phosphate (VC-PMG) on melanogenesis in
vitro and in vivo, J Am Acad Dermatol 34 (1996) 2933.
[24] H. Osmann-Gardabbou, J. Pelletier, S. Sfar-Gandoura, M.C. Martini,
Thickening of hydrophilic/lipophilic and lipophilic/hydrophilic microemulsions I.
Comparative study of the gelling agent influence on the physical characteristics

25

ACCEPTED MANUSCRIPT
of the H/L and L/H microemulsions, STP Pharma Sci 10 (2000) 220223.

[25] P. Spiclin, M. Homar, A. Zupancic-Valant, M. Gasperlin, Sodium ascorbyl

IP

phosphate in topical microemulsions, Int J Pharm 256 (2003) 6573.

SC
R

[26] M. Watase, K. Nishinari, A.H. Clark, S.B. Ross-Murphy, Differential scanning

calorimetry, rheology, x-ray, and NMR of very concentrated agarose gels,

NU

Macromolecules 22 (1989) 11961201.

MA

[27] M. Djabourov, A.H. Clark, D.W. Rowlands, S.B. Ross-Murphy, Small angle
X-ray scattering characterization of agarose sols and gels, Macromolecules, 22

TE

(1989) 180188.

CE
P

[28] R.O. Bude, R.S. Adler, An easily made, low-cost, tissue-like ultrasound phantom
material, J Clin Ultrasound 23 (1995) 271273.

AC

[29] B. Rozman, M. Gasperlin, Stability of vitamins C and E in topical

microemulsions for combined antioxidant therapy, Drug Deliv 14 (2007)
235245.

26

ACCEPTED MANUSCRIPT
FIGURE CAPTIONS

Fig. 1. Experimental design and protocol. The skin color was measured before and

IP

after the animals were exposed to UVB irradiation. The change in skin color induced

SC
R

by each of the treatments was assessed on each recording (R) day.

Fig. 2. SEM micrograph of the (A) MBs , (B) MBs in 0.1% agarose, and (C) MBs in

NU

0.15% agarose.

MA

Fig. 3. In vitro US images of albumin-shelled MBs in physiological saline (A, D),

0.1% agarose (B, E), and 0.15% agarose (C, F) before and after sonication at 2 W/cm2

TE

for 1 min. The signal intensities in panels AF are quantified in panel G. The signal

CE
P

intensity of MBs decreased more significantly in 0.1% and 0.15% agarose than in
physiological saline after sonication (*p0.001, **p0.0001). Data in panel G are

AC

mean and SD values.

Fig. 4. Quantification of the penetration depths of Evans blue in agarose phantom in

the four experimental groups at an acoustic intensity of 2 W/cm2 when the MBs were
added

at

various

concentrations

(0.23107,

4.6107,

11.5107,

and

23107 bubbles/ml). 0.1% agarose gel combined with 23 107 bubbles/ml and 0.15%
agarose gel combined with 4.6 107 bubbles/ml enhanced the penetration depth
significantly than other groups (*p0.05, **p0.001).

27

ACCEPTED MANUSCRIPT
Fig. 5. Light-microscopy images before (AD) and after (EH) imaging processing of

the agarose phantoms without US (control), with US (at an acoustic intensity of

IP

2 W/cm2), US combined with MBs (23107 bubbles/ml) in 0.1% agarose, and US

SC
R

combined with MBs (4.6107 bubbles/ml) in 0.15% agarose. The penetration depths
of Evans blue are quantified in panel I. The penetration depth was greater for

NU

23107/ml MBs in 0.1% agarose and 4.6107/ml MBs in 0.15% agarose than in the

MA

control and US groups (*p0.01).

Fig. 6. Light-microscope evaluation of pig-skin samples in groups UM (upper row),

TE

UMB1 (middle row), and UMB2 (bottom row) for MBs at various concentrations.

CE
P

Fig. 7. Quantification of the penetration depths of Evans blue shown in Fig. 5. The
penetration of 23107, 11.5107, and 0.23107 MBs in water was significantly greater

AC

than in agarose gel (p<0.05). The degree of penetration was greatest with
11.5107 bubbles/ml MBs in 0.1% agarose (14.81.1 m) and in 4.6107 bubbles/ml
MBs in 0.15% agarose (14.40.8 m) (*p0.05). These conditions were not
significantly different from 4.6107 MBs in water (**p>0.05).

Fig. 8. Photographs (AE) and histology images (FJ) of mouse skin from groups C,
M, UM, UMB1, and UMB2 at week 4 of treatment.

Fig. 9. Skin-whitening effects of MAP on UV-induced hyperpigmentation in groups C,

28

ACCEPTED MANUSCRIPT

AC

CE
P

TE

MA

NU

SC
R

IP

M, UM, UMB1, and UMB2 over the 4-week posttreatment period.

29

MA

NU

SC
R

IP

ACCEPTED MANUSCRIPT

AC

CE
P

TE

Fig. 1

30

AC

CE
P

TE

MA

NU

Fig. 2

SC
R

IP

ACCEPTED MANUSCRIPT

31

AC

CE
P

TE

MA

NU

SC
R

IP

ACCEPTED MANUSCRIPT

Fig. 3

32

Fig. 4

AC

CE
P

TE

MA

NU

SC
R

IP

ACCEPTED MANUSCRIPT

33

AC

CE
P

TE

MA

NU

SC
R

IP

ACCEPTED MANUSCRIPT

Fig. 5

34

Fig. 6

AC

CE
P

TE

MA

NU

SC
R

IP

ACCEPTED MANUSCRIPT

35

Fig. 7

AC

CE
P

TE

MA

NU

SC
R

IP

ACCEPTED MANUSCRIPT

36

NU

SC
R

IP

ACCEPTED MANUSCRIPT

AC

CE
P

TE

MA

Fig. 8

37

Fig. 9

AC

CE
P

TE

MA

NU

SC
R

IP

ACCEPTED MANUSCRIPT

38

ACCEPTED MANUSCRIPT

AC

CE
P

TE

MA

NU

SC
R

IP

Graphical abstract

39

ACCEPTED MANUSCRIPT

AC

CE
P

TE

MA

NU

SC
R

IP

Highlights
1. the survival of MBs with US is affected by the viscosity of the surrounding medium
2. treatment with US plus MBs in an optimum agarose gel increases skin permeability
3. skin penetration of some concentrations of MBs in water was greater than in gel
4. MBs in gel is more easy to apply to skin surface in living animal

40