This preview shows page 1. Sign up to view the full content.
Unformatted text preview: 1988 Biol. Pharm. Bull. 33(12) 1988—1993 (2010) Regular Article Vol. 33, No. 12 Macromolecular Delivery into Skin Using a Hollow Microneedle
Nanthida WONGLERTNIRANT,a,b Hiroaki TODO,b Praneet OPANASOPIT,a Tanasait NGAWHIRUNPAT,a and
Faculty of Pharmacy, Silpakorn University; Nakhon Pathom 73000, Thailand: and b Faculty of Pharmaceutical Sciences,
Josai University; 1–1 Keyakidai, Sakado, Saitama 350–0295, Japan.
Received June 23, 2010; accepted September 6, 2010; published online September 17, 2010 The objective of the present study was to obtain information to develop an effective delivery device regarding a sophisticated hollow microneedle array-patch system. Thus, the potential of hollow microneedles was investigated for enhancing the transdermal delivery of hydrophilic large molecular compounds, and the effect of variable parameters on drug release behavior was determined from skin. Fluorescein isothiocyanate (FITC)dextrans (4.3 kDa), FD-4, was used as the main model compound, and it was successfully loaded into the lower
epidermis as well as the superﬁcial dermis of the skin in hairless rats by a hollow microneedle. The higher the
volume of FD-4 solution injected, the faster the FD-4 release rate from skin. In addition, release rate tended to
increase when FD-4 was administered dividedly by multiple injections. These release proﬁles of FD-4 were
expressed by Fick’s law of diffusion. Furthermore, a combination of the formulation strategy and hollow microneedle-assisted delivery was useful for controlling the drug release rate from skin. Release proﬁles from drugloaded skin were also compared by changing the molecular weights of model compounds. The larger molecular
size of compounds caused a lower release rate from skin. These results suggest the utilization of hollow microneedle to enhance transdermal delivery of large molecular compounds and provide useful information for designing an effective hollow microneedle system.
Key words hollow microneedle; drug release; injection; transdermal delivery; large molecular compound Recently much attention has been paid to a new approach
for enhancing transdermal drug delivery employing needles
in micron scale, termed microneedles. Such a technique
combines the concepts of the user-friendly delivery of transdermal patches and the broad effectiveness of hypodermic injections. Several studies have established the increasing efﬁcacy of microneedle-assisted transdermal delivery for a variety of compounds,1—4) especially for high molecular weight
hydrophilic compounds.5—7) Microneedles are also expected
to be safe, because they are minimally invasive and do not
typically cause bleeding or any severe pain at the injection
Although microneedles assist in penetrating the stratum
corneum (SC) barrier and offer several microchannels facilitating drug transport across the skin, the release of macromolecular drugs at the desired therapeutic rate might not be
achieved by only (enhanced) passive diffusion from the
patch-based drug reservoir due to the small diffusional area
produced by needles.9) It is necessary to further improve the
transdermal delivery of drugs of large molecular weight by
contributing push pressure to propel the drug toward the
skin, similar to conventional topical injection. Wu et al. reported that the combination of solid microneedle pretreatment and subsequent iontophoresis signiﬁcantly enhanced
FD ﬂux compared with microneedle pretreatment alone or
It is difﬁcult, however, to determine the penetration-enhancing effect as well as the parameters affecting the delivery efﬁciency produced by each enhancing technique and
also by each individual microneedle. Yoshida et al. carried
out a study to investigate the dermatopharmacokinetics
(DPK) and systemic drug disposition after topical (intracutaneous (i.c.)) injection of sodium salicylate and concluded
that the injection volume was one of the factors which can be
utilized to control the drug migration rate from the injection
∗ To whom correspondence should be addressed. e-mail: email@example.com site.10) Al-Qallaf et al. found that different surface areas of
the hollow microneedle array-patch system affected the
blood concentration of human growth hormone (hGH).11)
Moreover, information on the effect of several factors related
to the delivery efﬁciency of hollow microneedles, for instance, various needle parameters, including injection conditions (e.g. length of microneedle/injection depth, needle
numbers, distance between each needle, hollow size, pressure
of injection, etc.), physicochemical properties of drugs (e.g.
molecular weight, lipophilicity, etc.), and formulations must
be considered for designing the optimum device.
In this study, we therefore concentrated only on the inﬂuence of variables related to the hollow microneedle system
on in vitro drug release behavior from skin. A single 33gauge hypodermic needle, where the diameter was almost the
same to the arrayed microneedles recently reported,2,7,9,12,13)
was used as a single type hollow microneedle in order to reduce the effect of needle parameters. The effects of injection
volume and number of injections on in vitro drug release
from skin were assessed. Moreover, the effects of formulations and molecular size were investigated.
MATERIALS AND METHODS
Materials Fluorescein isothiocyanate (FITC)-dextrans
(FD-4; average molecular weight, 4.3 kDa) and cholesterol
(CH) were purchased from Sigma Aldrich (St. Louis, MO,
U.S.A.). Calcein sodium (CAL; molecular weight, 623 Da)
was provided by Tokyo Chemical Industry (Tokyo, Japan).
Deuterium oxide (D2O; 99.9%, NMR grade, molecular
weight, 20.03 Da) was obtained from Wako Pure Chemical
Industries (Osaka, Japan). 1,2-Dimyristoyl-sn-glycero-3phosphocholine (DMPC) was provided by NOF Co., Ltd.
(Tokyo, Japan). Disodium ethylenediamine (EDTA · 2Na)
was purchased from Dojindo Laboratories (Kumamoto,
© 2010 Pharmaceutical Society of Japan December 2010 Japan). Propylene glycol monocaprylic ester (Sefsol-218)
and Polyoxyethylene (60) hydrogenated castor oil (HCO-60)
were supplied by Nikko Chemicals Co., Ltd. (Tokyo, Japan).
All other chemicals were of analytical grade and used without further puriﬁcation.
Preparation of Hollow Microneedles The hollow microneedles, NanopassTM (33-gauge hypodermic needle, i.d.,
0.20 mm), were kindly provided by Terumo Co. (Tokyo,
Japan). A hollow microneedle was manufactured from a microneedle connected to a 27-gauge hypodermic needle (i.d.,
0.22 mm; o.d., 0.40 mm; Terumo Co.).
Experimental Animals Male hairless rats (WBM/ILAHt, 7—9 weeks-old, body weight: 180—250 g) were supplied
either by Life Science Research Center, Josai University
(Sakado, Saitama, Japan) or Ishikawa Experimental Animal
Laboratories (Fukaya, Saitama, Japan). They were housed in
temperature-controlled rooms (25 2 °C) with a 12 h
light–dark cycle (07:00—19:00 h). The rats were allowed
free access to food (M.F. Oriental, Tokyo, Japan) and tap
water for a week before experiments. All animal experiments
were conducted under the guidelines of Josai University.
The upper part of full-thickness skin was carefully excised
from the dorsal region of the rats under anesthesia of intraperitoneally (i.p.) injection of sodium pentobarbital (50 mg/
kg), and excess subcutaneous fat was carefully trimmed off.
The excised skin, 1.0—1.3 mm in thickness, was immediately used for experiments.
Formulations for Injection To examine the effect of
formulations on the drug release behavior from skin after injection by hollow microneedle, three different formulations
were considered: solution, o/w emulsion, and liposome suspension. CAL was used as the model drug in this study.
CAL (1 mM) solution was prepared by adding to 1 mM
EDTA · 2Na in pH 7.4 phosphate-buffered saline (PBS).
Emulsion-based formulation was prepared by adding the
aqueous phase containing CAL and Sefsol-218 (5%, v/v) to
HCO-60 (1%, v/v) in oil phase. Emulsiﬁcation was carried
out using a homogenizer (Polytron® PT3100; Kinematica
Inc., NY, U.S.A.) at 13000 rpm for 5 min. The ﬁnal concentration of CAL in emulsion formulation was 1 mM. Liposome-based formulation was prepared by the reverse phase
evaporation method.14) Brieﬂy, the lipid phase (mixture of
DMPC and CH; 7 : 3, w/w) was dissolved in a chloroform/
isopropyl ether mixture (1 : 1, v/v), and then mixed with the
aqueous solution containing CAL (initial concentration
10 mM) in a sonication bath (5510 J-DTH; Branson Ultrasonics Corp., Danbury, CT, U.S.A.) for 5 min to obtain a waterin-oil emulsion. The organic solvent was evaporated under
reduced pressure. During this process, the material ﬁrst
forms a viscous gel and subsequently becomes an aqueous
suspension. The freeze–thaw cycle was performed and repeated three times. The liposome suspension was then extruded through polycarbonate membranes with pore sizes of
0.4 m m, 0.2 m m, and 0.1 m m, respectively. The extrusion
process was run 5 times per membrane pore size. Liposomes
were separated from non-encapsulated CAL by ultracentrifugation at 289000 g for 10 min and resuspension of the vesicles in buffer (quadruple). The entrapment efﬁciency of liposomes was 2.43%. Size distribution was determined by dynamic light scattering (Malvern Instruments, Southborough,
MA, U.S.A.) and revealed a mean vesicle diameter of 1989 127.3 nm.
In Vitro Skin Release Studies The extent and rate of
drug release from skin were investigated to examine the effect of different four variables (a—d) related to the hollow
microneedle-based delivery system on the drug DPK. Unless
otherwise mentioned, FD-4 (1 mM) was used as the high molecular compound for testing in this study. In vitro release
studies were carried out after the following treatments: (1)
different volumes of drug solution (5, 10, or 20 m l) were administered into the excised dorsal skin by single injection;
(2) different numbers of injections (10 m l single injection,
5 m l two injections, and 2.5 m l four injections) were administered into excised skin. Distance between each injection
point was 0.5 cm; (3) different drug formulations (solution,
o/w emulsion, or liposome suspension) at 10 m l were injected
into excised skin by single injection; (4) different molecular
sizes of test compounds (10 m l of either FD-4 (molecular
weight (MW) 4300) solution, CAL (MW 623) solution or
D2O (MW 20.03)) were injected into excised skin by single
In the injection process, the skin was stretched onto a cork
support board with four tissue-mounting pins to simulate
skin tension in vivo. The skin containing injected solution
was then mounted in a vertical diffusion cell (effective diffusion area: 1.77 cm2) with the SC side facing the donor compartment (without drug solution), which was covered with
Paraﬁlm to establish an occlusive condition. The receiver solution was approximately 6.0 ml of pH 7.4 PBS, which was
stirred with a magnetic stirrer bar driven by a constant-speed
synchronous motor and maintained at 32 °C using a thermoregulated water bath throughout the experiment. For the release study of CAL, 1 mM EDTA · 2Na in pH 7.4 PBS was
used instead of pH 7.4 PBS. The receiver solution (0.5 ml)
was withdrawn at predetermined time intervals up to 8 h after
drug administration, and the same volume of PBS was added
to the receiver compartment to keep the volume constant.
After the release study was ended over a period of 8 h, the
amount of drug remaining in the skin was measured by isolating the skin from the diffusion cell. The skin was cut into
small pieces with scissors and ﬁne forceps, and homogenized
(at 12000 rpm for 5 min) with 2 ml PBS to extract the drug
under an ice bath. Acetonitrile (2 ml) was then added and
mixed with skin homogenized solution using a vortex shaker
to precipitate protein. After centrifugation at 15000 rpm for
5 min, the clear supernatant was taken for analysis.
Quantitative Assay The concentrations of FD-4 and
CAL samples were analyzed using a spectroﬂuorophotometer
(RF 5300PC; Shimadzu, Kyoto, Japan) at excitation wavelengths of 495 and 490 nm, respectively, and the identical
ﬂuorescent emission wavelength of 515 nm. D2O was determined by the intensity of O-D stretching vibrational band
at 2512 cm 1 infrared spectroscopic spectra.9)
Diffusivity of FD-4 in Skin. Analysis of Diffusivity from
Drug Release Proﬁles The hollow microneedle system is
deﬁned as a way to directly load drugs into skin and provides
a drug depot in skin, as in topical injection. Thus, the DPK
after the application of this device may be different from that
of conventional topical application.10) We injected the drug
solution into skin by hollow microneedle and observed the
drug release behaviors from skin. The obtained release data
were analyzed using the empirical mathematical model pro- 1990 Vol. 33, No. 12 posed by Peppas (Eq. 1),15,16) which is generally used to explain the drug release by coupling Fickian and non-Fickian
Mt kt n M∞ (1) where Mt is the amount of drug released at time t, and M is
the amount of drug released at inﬁnite time. Mt/M is the
fractional drug release, t is the release time, k is a kinetic
constant incorporating structural and geometric characteristics of the controlled release device, and n is an exponent
which characterizes the mechanism of diffusional release.
Based on the exponent, n, drug transport is classiﬁed as Fickian diffusion for n 0.5, anomalous (non-Fickian) transport
for n 0.5— 1.0, case II transport or zero order (time-independent release) for n 1.0, and super-case-II transport for
n 1.0. This equation is applicable only for data up to 30% of
dose release (Mt/M 0.30). According to Eq. 1, the logarithm of the cumulative amount of drug released each time
was plotted against the logarithm of time to calculate n. Release experiments were carried out under perfect sink conditions and release proﬁles were classiﬁed by considering the
skin to be a homogeneous single membrane.
Analysis using the Peppas model revealed the primary
drug release mechanism of FD-4-loaded skin, which was
Fickian diffusion (see Results and Discussion). In this case,
the Peppas equation can be expressed the same as the simpliﬁed form of the Higuchi diffusion model as follows:
kt 1 / 2 Q (2) where Q is the cumulative amount of FD-4 released from the
FD-4 loaded skin into the bulk receiver solution per unit surface area (nmol/cm2), and k is the kinetic constant indicative
of the release rate (nmol/cm2 h1/2). Thus, the cumulative
amount of drug released is proportional to the square root of
time. According to Higuchi, the diffusion coefﬁcient of FD-4
in skin by employing a hollow microneedle as a delivery
method, Dskin (cm2/h), can be calculated from Eq. 3.17)
Q 2C0 Dskint (3) π where C0 is the initial concentration of FD-4 in skin after direct injection (nmol/ml). Equations 2 and 3 are conﬁned to
the description of the ﬁrst 30% of the release curve.
Analysis of Diffusivity from Skin Permeation Proﬁles
To examine the diffusion coefﬁcient of FD-4 in viable epidermis and dermis layers, Dved, stripped skin excised from
hairless rats was used as model skin, which was supposed to
be a homogenous single membrane (deﬁned as the one-layered diffusion model). The stripped hairless rat skin was set
in a diffusion cell, and FD-4 solution was applied to the
donor cell on the epidermal side of the skin to determine the
skin permeation proﬁles of FD-4.
The concentration of FD-4, Cved, in the stripped skin at a
position, x, and time, t, can be calculated using Fick’s second
law of diffusion (Eq. 4).
∂t Dved ∂2C ved
∂x 2 (4) where Dved is the diffusion coefﬁcient of FD-4 in the viable
epidermis and dermis layers. Based on the differential equa- tion, Eq. 4 can be changed to Eq. 5, as described in detail by
Hada et al.18) and Sugibayashi et al.19)
C vedi, j rDvedC vedi 1 (1 1, j 2rDved )C vedi, j rDvedC vedi 1, j (5) where Cvedi, j is FD-4 concentration at i-th position and j-th
time in the stripped skin, r is D t/D x2; D x is xi 1 xi, and D t is
tj 1 tj. In addition, the skin permeation rate to the receiver
compartment, J, can be expressed by Eq. 6. The cumulative
amount of FD-4 permeated per unit area, Q, is expressed by
Jj Dved Qj Qj 1 Cn C n, j 1, j (6) Δx Jj D t (7) where n is the number of divisions of skin. Then, Jj was calculated using Microsoft® Excel by setting n 10. Qj was calculated from Jj using Eq. 7. The diffusion coefﬁcient, Dved,
was obtained by ﬁtting the observed data using the least
squares method performed by the solver function of Microsoft® Excel. Other conditions of data analysis, not mentioned here, were the same as in the previous study.18,19)
Full-thickness skin of hairless rats was used instead of
stripped skin to calculate the diffusion coefﬁcient of FD-4 in
the SC layer, DSC. A two-layered diffusion model was established herein to analyze the FD-4 permeation proﬁles. FD-4
concentration in the SC layer can be expressed by Fick’s second law of diffusion as follows:
∂CSC DSC ∂t ∂2CSC (8) ∂x 2 FD-4 concentration, Cved, in the viable epidermis and dermis layers at a position, x and time, t can be expressed by Eq.
4. Equation 8 is changed to Eqs. 9 and 10 by differential calculus.18,19)
C vedi, j 1 1 rDSCCSCi (1 1, j rDvedC vedi 1, j (1 2rDSC )CSCi, j
2rDved )C vedi, j rDSCCSCi (9) 1, j rDvedC vedi 1, j (10) Similar analysis was carried out in the two-layered model
using the least squares method to obtain the diffusion coefﬁcient of FD-4 in SC, DSC . In this calculation, diffusivity in
the viable epidermis and dermis layers, Dved, was ﬁxed to the
already estimated value by the stripped skin permeation experiment.
Histological Imaging To visualize hollow microneedle
penetration into the skin and the pathway of ﬂuid injection, a
small amount of Evans blue was injected into the skin by a
hollow microneedle (Fig. 1B). After removing the microneedle, the skin sample was frozen in isopentane (2-methylbutane) cooled by dry ice with optimal cutting temperature
compound (Tissue-Tek; Sakura Finetek, Torrance, CA,
U.S.A.) in an embedding mold container. A freezing microtome (Leica CM3050S; Finetec, Japan) was used to make the
vertical sections of 10 m m thickness. These sections were examined histologically using a phase-contrast microscope
(IX71; Olympus, Japan).
Statistical Analysis In vitro drug release measurements
were collected from four to six experiments. Values are expressed as the mean standard error (S.E.). Statistical signiﬁcance of differences between groups in the amount of FD-4 December 2010 1991 Fig. 1. Schematic Representation of Hollow Microneedle Insertion into Skin with a 33-Gauge Hypodermic Needle (A) and Histological Section of Back
Skin of Hairless Rat Pierced with a Hollow Microneedle in Vitro (B)
A small amount of Evans blue was injected into skin. The paths of ﬂuid injection are indicated by the presence of blue dye. Dotted lines show where the hollow microneedle was
inserted. A hollow microneedle could be inserted forward or backwards in this histological section. released from skin was examined using one-way analysis of
variance (ANOVA) followed by Student’s t-test. The signiﬁcance level was set at p 0.05.
RESULTS AND DISCUSSION
Characterization of Hollow Microneedle Injection To
investigate piercing of the skin barrier and delivering the
drug through the barrier by a hollow microneedle, FD-4 solution was injected into the excised back skin of hairless rats.
The needle was ﬁxed with a triangular silicone sheet to maintain an angle of insertion (q ) of 40° and constant insertion
depth in the skin barrier, as shown in Fig. 1A. The microneedle tip was inserted easily into the skin using gentle force,
and no bending was observed in the needle tip. Up to 20 m l
(single injection) of FD-4 solution was successfully injected
into skin without any leakage from the skin surface or from
the bottom (dermal) side of the skin. The presence of FD-4
solution in skin was visualized by the appearance of a greenish-yellow region spread around the injection site (data not
shown). Increasing the injection volume resulted in a broader
greenish-yellow region. After injection and removing the
needle, the skin was frozen instantly in liquid nitrogen and
sectioned vertically to observe FD-4 deposition in the skin.
Next, a histological study was performed by examining
skin sections under a light microscope after loading a small
amount of Evans blue into the skin using a hollow microneedle. Figure 1B shows a typical cross section of a skin piece.
The blue dye in the skin corresponded to the needle insertion
across the stratum corneum and upper epidermis as well as
into the superﬁcial dermis.
Effects of Injection Volume and Number of Injections
on the Release Behavior of FD-4 from Skin To evaluate
the effect of the injection volume, 5, 10 and 20 m l FD-4 solution were loaded into the skin using a hollow microneedle.
Approximately 80% FD-4 solution was released from the
skin over 8 h in all cases (Table 1). Figure 2A shows the time
course of the cumulative amounts of FD-4 released per unit
area from skin. The obtained drug release proﬁles were analyzed using the Peppas equation (see ‘Diffusivity of FD-4 in
Skin’) to characterize the release mechanism. Figure 2B
shows the double logarithmic plot of the cumulative amount Table 1. The Percentage of FD-4 Released from Skin over 8 h after Administration by a Hollow Microneedle
Injection volume (m l) % Number of injectionsa) % 5
20 80.2 3.4
84.6 3.4 1 (10 m l)
2 (each 5 m l)
4 (each 2.5 m l) 81.2 1.6
85.2 1.2* a) The total volume of injection was adjusted to 10 m l. ∗ p 0.05 compared with
10 m l single injection. Mean S.E. with 4 to 6 measurements. of FD-4 released against time, where slopes of the lines represent the release exponent (n). As a result, a Fickian release
can be assumed (n is very close to 0.5). As can be seen in
Fig. 2C, linear relationships (r2 0.974—0.998) were obtained when the cumulative amounts of FD-4 released from
skin were plotted against the square root of time. These results indicated that the release of FD-4 followed the Higuchi
diffusion model validated when drug release was less than
30%. The drug release rate (k) can be calculated from the intercept of the Peppas plot. It was shown that increasing the
injection volume from 5 to 20 m l increased the release rate almost proportionally (Table 2). Since diffusion in skin was the
primary mechanism of drug release from skin, FD-4-loaded
skin might be treated as a drug-loaded matrix. In this respect,
the Higuchi model (see ‘Diffusivity of FD-4 in Skin’) was
applicable and further used to evaluate the diffusivity of FD4 in skin (discussed later).
In order to further examine the effect of the number of injections, release studies were performed using different injection conditions, i.e. one of 10 m l, two of 5 m l, and four of
2.5 m l. Four 2.5 m l injections showed that a signiﬁcantly
higher amount of FD-4 was released than with one 10 m l and
two 5 m l injections (p 0.05) (data not shown). The percentage of FD-4 release from four 2.5 m l injections was also signiﬁcantly higher than the others, although the total amount of
percent FD-4 release at the end of the release period (t 8 h)
was not signiﬁcantly different between four 2.5 m l injections
and two 5 m l injections (Table 1). This result might suggest
that multiple injections increased total area of FD-4 released
from the skin. The effect of the number of injections on drug
release was analyzed in a similar manner as above. Thus, the
release proﬁles were also found to follow Higuchi kinetics 1992 Fig. 2. Vol. 33, No. 12 Effect of Injection Volume on the Release Proﬁle of FD-4 from Skin after Administration of FD-4 Solution of 5, 10 and 20 m l (A) Accumulated amount of FD-4 released per unit area from skin, (B) and (C) analysis of the effect of injection volume on the FD-4 release proﬁle using Peppas and the simpliﬁed Higuchi model, respectively. Symbols: , 5 m l; , 10 m l; , 20 m l. Solid line represents calculated value. The values in Figs. 2B and C are represented by the following equa–
tions, respectively: ; log Q 0.56 log t t0 0.16 (r2 0.999) and Q 0.75 √t 0.059 (r2 0.998), ; log Q 0.53 log t t0 0.017 (r2 0.967) and Q 1.2 √t 0.12
(r2 0.979), ; log Q 0.48 log t t0 0.46 (r2 0.965) and Q 3.2 √t 0.15 (r2 0.974). Each point represents the mean S.E. of four to six experiments. Table 2. Release Kinetics of FD-4 after Administration into Skin by a
Mt /M ktn
Exponent (n) Kinetic constant (k) r2 Injection volume (m l)
0.965 Number of injections
1 (10 m l)
2 (each 5 m l)
4 (each 2.5 m l) 0.528
0.983 (n 0.528—0.564). The parameters of n and k derived by the
Peppas plot are summarized in Table 2. From the results, dividing the amount of FD-4 loaded into skin by multiple injections was found likely to increase the release rate and total
amount of drug released from skin.
Effect of Hollow Microneedle-Assisted Delivery on the
Diffusivity of FD-4 within Skin We subsequently examined the diffusion coefﬁcient of FD-4 solution in skin (Dskin)
from the gradients of Higuchi plots (see ‘Diffusivity of FD-4
in Skin’). Dskin was calculated to be 6.7 10 4 cm2/h. A comparison was performed with the diffusion coefﬁcient of FD-4
in the stratum corneum layer (DSC) and in viable epidermis
and dermis layers (Dved), which were 5.1 10 7 and 6.0
10 4 cm2/h, respectively. As a result, Dskin was almost the
same as Dved and much higher than DSC. Because the SC provides a marked skin barrier, its resistance to drug diffusion in
this skin layer is the highest, causing the lowest diffusivity of
FD-4. By employing hollow microneedles, the surface of the
skin barrier (SC) was bypassed; thus, FD-4 solution was easily delivered into the lower epidermis or dermis, where the
permeation resistance of FD-4 must be much lower than that
on the SC; therefore, FD-4 can rapidly diffuse through the
skin. These results exhibited the utility of hollow microneedles for overcoming the barrier function of SC and enhancing the delivery of high molecular weight drugs into skin.
Effect of the Different Molecular Sizes of Compounds
on the Release Proﬁle from Skin Study of the transport of
drug substances within the deeper skin layers lags behind
that of transport in SC. Only SC has been recognized as a
rate-limiting barrier for skin permeation of drugs, particularly hydrophilic drugs. By employing hollow microneedles,
drugs are directly loaded into the lower epidermis or dermis
layer and bypass the SC, as explained above. Thus, hydrophilic molecules can be expected to readily diffuse
through the lower layers of skin which contain a larger
amount of water than the SC. In this study, we then determined the effect of molecular weight on release from skin
after administration by hollow microneedles. These studies
were conducted with D2O (MW 20.03), CAL (MW 623) and
FD-4 (MW 4300). As a result, more than 90.0% D2O was released from treated skin within 2 h (data not shown), and
90.3% CAL was released over 8 h, which was signiﬁcantly
higher than the 81.2% FD-4 released (p 0.05). These results
clearly showed that the molecular sizes of drug compounds
affect drug release from skin. Alternatively, molecular size
has a marked inﬂuence on drug diffusivity within the deeper
layers of skin, with increasing bulky molecules decreasing in
Effect of Different Formulations on the Release Proﬁle
of CAL from Skin We began our discussion by demonstrating the utility of hollow microneedle for transdermal hydrophilic macromolecular delivery and various parameters
affecting release behavior from FD-4-loaded skin. Thereafter, the feasibility of different formulations for controlling
drug release after loading into skin by hollow microneedles
was addressed. For this purpose, three formulations, solution,
o/w emulsion, and liposome suspension, were investigated.
Although drug delivery by a hollow microneedle system
completely evades the barrier function in SC, low diffusion
of a large molecular drug through the deeper skin layers may
limit drug delivery to the deep epidermis and dermis. To
make a logical comparison, therefore, a low molecular
weight of the candidate drug was chosen to avoid the problem of skin permeability. Calcein (MW 623) was selected as December 2010 1993 mechanism of FD-4-loaded skin was found to be Fickian diffusion. Various parameters based on the hollow microneedle
system were examined to optimize the delivery device. Overall, the higher the amount of FD-4 delivered, the faster the
drug release. Dividing the amount of FD-4 loaded into skin
by multiple injections likely increased the total amount of
drug released and the release rate from skin. Furthermore, a
formulation strategy can be applied with a hollow microneedle system to modify drug release through skin. Further
studies are needed to obtain additional details for designing
a sophisticated hollow microneedle array-patch system to
effectively enhance and precisely control macromolecular
delivery across the skin. Fig. 3. Effect of Formulations on the Release Proﬁle of CAL from Skin Symbols: , solution; , emulsion;
mean S.E. of four to six experiments. , liposome. Each point represents the the penetrant for this comparison. Figure 3 shows different
CAL release proﬁles among three formulations after delivery
using a hollow microneedle system. The total amount of
CAL released over 8 h was ranked in the order of solution
(90.3%) emulsion (70.4%) liposome (3.6%). As the
lipophilic content of the formulation was increased, the drug
release considerably declined. CAL was solubilized in the
buffer vehicle and in the external phase of oil/water (o/w)
emulsion to freely diffuse through the skin, therefore resulting in a greater amount of drug released. In the liposome suspension, CAL release occurred through two steps: i) disrupting vesicle structures in the skin to freely liberate CAL or
partitioning of CAL out of lipid vesicles; and ii) diffusion
through the skin to bulk receiver solution, together with the
rigid vesicle structure of traditional liposomes, which may
explain the lower drug release. These results suggest the possibility of controlling drug release from the skin after hollow
microneedle administration by employing a formulation
The present study demonstrated the utilization of hollow
microneedles for macromolecular drug delivery through
skin. We used a single 33-gauge hypodermic needle instead
of hollow microneedle array. The needle was inserted into
the skin with an angle of 40 degree. The present results may
not extrapolate to those by microneedle array. However, the
size of 33-gauge hypodermic needle used in this experiment
resembles to the hollow microneedle reported by other researchers. Furthermore, both of our device and typical hollow microneedle array could avoid stratum corneum barrier
function by creating pores. The advantage of single needle
device is capable to investigate several factors such as administration volume, formulation, and depth penetration of
The diffusivity of FD-4 after administration by a hollow
microneedle was much larger than that in the SC. Release Acknowledgements This research was supported by the
Commission of Higher Education and the Thailand Research
Funds through the Royal Golden Jubilee Ph.D. Program
(Grant No. PHD/0104/2549) and Grant No. RSA 5280001.
As a visiting academic, N. Wonglertnirant would like to
thank Professor K. Sugibayashi for an invitation to work in
his laboratory, Faculty of Pharmaceutical Sciences, Josai
University which supports this work and Dr. H. Todo for his
help and encouragement.
1) Henry S., McAllister D. V Allen M. G., Prausnitz M. R., J. Pharm.
Sci., 87, 922—925 (1998).
2) McAllister D. V Wang P. M., Davis S. P., Park J. H., Canatella P. J.,
Allen M. G., Prausnitz M. R., Proc. Natl. Acad. Sci. U.S.A., 100,
3) Cormier M., Johnson B., Ameri M., Nyam K., Libiran L., Zhang D.
D., Daddona P., J. Controlled Release, 97, 503—511 (2004).
4) Al-Qallaf B., Das D. B., Mori D., Cui Z., Philos. Trans. A Math. Phys.
Eng. Sci., 365, 2951—2967 (2007).
5) Martanto W., Davis S. P., Holiday N. R., Wang J., Gill H. S., Prausnitz
M. R., Pharm. Res., 21, 947—952 (2004).
6) Wu X. M., Todo H., Sugibayashi K., Int. J. Pharm., 316, 102—108
7) Verbaan F. J., Bal S. M., van den Berg D. J., Groenink W. H., Verpoorten H., Lüttge R., Bouwstra J. A., J. Controlled Release, 117,
8) Kaushik S., Hord A. H., Denson D. D., McAllister D. V Smitra S.,
Allen M. G., Prausnitz M. R., Anesth. Analg., 92, 502—504 (2001).
9) Wu X. M., Todo H., Sugibayashi K., J. Controlled Release, 118, 189—
10) Yoshida D., Todo H., Hasegawa T., Sugibayashi K., Int. J. Pharm., 337,
11) Al-Qallaf B., Das D. B., Mori D., Cui Z., Philos. Trans. A Math. Phys.
Eng. Sci., 365, 2951—2967 (2007).
12) Wang P. M., Cornwell M., Hill J., Prausnitz M. R., J. Invest. Dermatol.,
126, 1080—1087 (2006).
13) Harvey A. J., Kaestner S. A., Sutter D. E., Harvey N. G., Mikszta J. A.,
Pettis R. J., Pharm. Res., 2010, in press.
14) Szoka F. Jr., Papahadjopoulos D., Proc. Natl. Acad. Sci. U.S.A., 75,
15) Peppas N. A., Pharm. Acta Helv., 60, 110—111 (1985).
16) Costa P., Sousa Lobo J. M., Eur. J. Pharm. Sci., 13, 123—133 (2001).
17) Higuchi W. I., J. Pharm. Sci., 51, 802—804 (1962).
18) Hada N., Hasegawa T., Takahashi H., Ishibashi T., Sugibayashi K., J.
Controlled Release, 108, 341—350 (2005).
19) Sugibayashi K., Todo H., Oshizaka T., Owada Y., Pharm. Res., 27,
134—142 (2009). ...
View Full Document
- Spring '08