The effect of ion pairing on the skin permeation of amlodipine

Article abstract of DOI:10.1691/ph.2008.7302

The purpose of the present study was to evaluate the effect of ion pairing on the skin permeation of amlodipine. Amlodipine base (AM) was first prepared from amlodipine besilate (AM-B), then amlodipine adipate (AM-A), amlodipine oxalate (AM-O) and amlodipine maleate (AM-M) were prepared using AM and the corresponding organic acids. Differential scanning calorimetry (DSC) thermogram studies demonstrated the formation of complexes between AM and the various acids. In vitro percutaneous absorption of AM and its complexes was evaluated through excised rat skin using 2-chamber diffusion cells. The results showed that AM had the greatest steady-state flux and lowest permeability coefficient of the five compounds from the El system (ethanol : isopropyl myristate (IPM) = 2:8), and its four complexes all exhibited a lower flux and higher permeability coefficient than AM.

Full text of DOI:10.1691/ph.2008.7302

ORIGINAL ARTICLES
Department of Pharmaceutics, Shenyang Pharmaceutical University, People’s Republic of China
The effect of ion pairing on the skin permeation of amlodipine
Yuxuan Jiang, Liang Fang, Xicao Niu, Rui Ma, Zhonggui He
Received September 18, 2007, accepted November 2, 2007
Prof. Liang Fang, Department of Pharmaceutics, Shenyang Pharmaceutical University,
103 Wenhua Road, Shenyang 110016, People’s Republic of China
fangliang2003@yahoo.com
Pharmazie 63: 356–360 (2008)
doi: 10.1691/ph.2008.7302
The purpose of the present study was to evaluate the effect of ion pairing on the skin permeation of
amlodipine. Amlodipine base (AM) was first prepared from amlodipine besilate (AM-B), then amlodi-
pine adipate (AM-A), amlodipine oxalate (AM-O) and amlodipine maleate (AM-M) were prepared using
AM and the corresponding organic acids. Differential scanning calorimetry (DSC) thermogram studies
demonstrated the formation of complexes between AM and the various acids. In vitro percutaneous
absorption of AM and its complexes was evaluated through excised rat skin using 2-chamber diffusion
cells. The results showed that AM had the greatest steady-state flux and lowest permeability coeffi-
cient of the five compounds from the EI system (ethanol : isopropyl myristate (IPM) ¼ 2 : 8), and its four
complexes all exhibited a lower flux and higher permeability coefficient than AM.
1. Introduction
2. Investigations, results and discussion
Amlodipine belongs to the dihydropyridine class of cal-
cium channel blockers, and it is used to treat hypertension
and relieve symptoms of angina pectoris (Haria et al.
1995). Patients suffering from these diseases usually re-
quire long-term medication and patient compliance is very
important. However, in the case of amlodipine, the only
available route is oral administration that requires daily
dosing. Transdermal drug delivery systems (TDDS) pro-
vide an attractive approach for long-term treatment and
have many advantages over traditional therapies (Abraham
1995). The daily dose of amlodipine ranges from 2.5 to
10 mg, which is very suitable for TDDS. Some research
has been carried out to verify the feasibility of developing
an effective transdermal system for AM-B (Francoeur and
Potts 1988). In addition, McDaid and Deasy have dis-
cussed the formulation development of a reservoir-type
transdermal system containing AM (McDaid and Deasy
1996). However, the steady-state flux they obtained needs
further improvement in order to attain the clinical levels
required in patients.
Many studies have highlighted the importance of ion pair
formation for TDDS (Fang et al. 2003; Cheong and Choi
2002; Kadono et al. 1998). When an ion pair contains
both a cation and an anion, it may have a residual de-
gree of hydrophobicity. The ion pair diffuses into the
aqueous viable epidermis, then dissociates into its
charged species which partition into the epidermis and
then diffuse forward (Megwa et al. 2000a, b; Valenta
et al. 2000). Accordingly, AM was first prepared from
amlodipine besilate (AM-B) in this investigation, and
then amlodipine maleate (AM-M), amlodipine oxalate
(AM-O) and amlodipine adipate (AM-A) were prepared
from AM. Next, the skin permeability of AM and its
four complexes through excised rat skin was studied
using 2-chamber diffusion cells.
2.1. Comparison studies among AM and its complexes
Figure 1 shows the DSC curves of AM and its complexes at
a heating rate of 5 ꢀC/min over the experimental tempera-
ture range 20–300 ꢀC. The DSC curves of AM and its
complexes exhibit characteristic sharp endothermic peaks
corresponding to their melting points. Table 1 summarizes
the melting points and the enthalpies of fusion for AM
and its complexes. The melting peaks of the complexes
are all shifted to higher temperatures compared with AM.
The rank order of the melting points is as follows:
AM < AM-A < AM-M < AM-O < AM-B. The changes in
the melting point and the enthalpy of fusion confirm the
formation of complexes.
Saturated solutions were used in our experiments, because
suspensions had the same maximum thermodynamic activ-
ity and were able to promote maximum flux in equili-
brium systems (Barry 2001). Figure 2 shows the permea-
tion profiles of AM and its complexes from the EI system
through excised rat skin. The fluxes of AM and its com-
plexes were in the following order: AM > AM-B >
AM-M > AM-A > AM-O. AM has the highest flux,
1.14 ꢁ 0.09 mmol/cm2/h. It is obvious that complex for-
mation has a negative effect on the flux. The solubilities
Table 1: DSC melting characteristics of AM and its com-
plexes
m.p. (ꢀC)
Enthalpy of fusion (kJ/mol)
AM
142.32
200.41
170.63
186.71
174.23
25.79
33.59
44.84
40.92
20.23
AM-B
AM-A
AM-O
AM-M
356
Pharmazie 63 (2008) 5

ORIGINAL ARTICLES
AM
10
8
AM-B
AM-A
AM-O
AM-M
6
4
2
0
0
2
4
6
8
10
Time (h)
Fig. 2: Permeation profiles of AM and its complexes from the EI system
through excised rat skin. Each data point represents the average
value of three permeation experiments
AM, its solubility in the EI system is much higher than
that of its complexes. Consequently the permeability coef-
ficient of AM is the lowest of the five compounds.
The excised rat skin contains the epidermis (divided into
the stratum corneum (SC) and the viable epidermis) and
the dermis. The viable epidermis and dermis (ED) are si-
milar and regarded as an aqueous protein gel (Scheuplein
1976). Drugs penetrate the skin according to the following
proces. Firstly, drugs partition between the EI system and
the SC, then diffuse through the SC. Following this, they
partition between the SC and the ED, and finally diffuse
through the ED. After drugs reach the dermis, they are
absorbed by the blood capillaries within minutes.
The permeability through excised rat skin can be defined
by its resistance (Fang et al. 2003).
R ¼ RSC þ RED
ð1Þ
Fig. 1: DSC curves of AM and its complexes at a heating rate of 5 ꢀC /min
Where R, RSC and RED are the resistance of excised skin
(full thickness), SC and ED, respectively. The resistance
can be expressed as the reciprocal of the permeability
coefficient, P, which is the product of the partition coeffi-
cient and the diffusion coefficient divided by the thickness
of the membrane.
in the EI (ethanol : isopropylmyristate) system and the per-
meation coefficients of AM and its complexes are shown
in Tables 2 and 3, respectively. Despite the highest flux of
Table 2: Physicochemical properties relating to the transder-
mal delivery of AM and its complexes (mean ꢀ S.E.,
n ¼ 3)
1
1
1
hSC
hED
R ¼
¼
þ
¼
þ
ð2Þ
P
PSC PED DSCKSC DEDKED
Where P, PSC and PED are the permeability coefficients
through excised skin, SC and ED; DSC and DED are the
corresponding diffusion coefficients; KSC and KED are the
partition coefficients between the EI system and SC, SC
and ED separately. Also, hSC and hED are the thickness of
SC and ED, and they are assumed to be relatively con-
stant. The cells of the SC are cornified and closely
packed, so the SC is the principal barrier in the pene-
Drug
log KO/W
Solubility in the EI system Solubility in water
(mmol/ml)
(mmol/ml)
AM
1.50 ꢁ 0.01 154.05 ꢁ 6.95
0.39 ꢁ 0.02
3.37 ꢁ 0.01
5.57 ꢁ 0.25
3.56 ꢁ 0.02
3.97 ꢁ 0.01
AM-B
AM-A
AM-O
AM-M
0.65 ꢁ 0.03
ꢂ0.22 ꢁ 0.02
ꢂ0.34 ꢁ 0.04
0.25 ꢁ 0.02
4.88 ꢁ 0.14
0.65 ꢁ 0.001
0.15 ꢁ 0.003
0.59 ꢁ 0.010
Table 3: Permeation parameters, skin contents and partition coefficients between the skin and the EI system (log KS/EI) of AM
and its complexes (mean ꢀ S.E., n ¼ 3)
Drug
log P
Js
Q10
tlag
(h)
Skin content (mmol/g)
log KS/EI
(ml/g)
(mmol/cm2/h)
(mmol/cm2)
AM
ꢂ2.13 ꢁ 0.04
ꢂ0.67 ꢁ 0.01
ꢂ0.12 ꢁ 0.02
0.49 ꢁ 0.02
1.14 ꢁ 0.09
1.05 ꢁ 0.03
0.49 ꢁ 0.03
0.48 ꢁ 0.02
0.58 ꢁ 0.03
8.67 ꢁ 0.73
7.14 ꢁ 0.24
3.73 ꢁ 0.14
3.48 ꢁ 0.007
3.12 ꢁ 0.13
2.20 ꢁ 0.39
3.23 ꢁ 0.09
2.27 ꢁ 0.22
2.84 ꢁ 0.36
4.66 ꢁ 0.10
74.72 ꢁ 4.56
50.01 ꢁ 3.66
15.10 ꢁ 1.18
4.13 ꢁ 0.48
10.84 ꢁ 0.95
0.49 ꢁ 0.03
10.26 ꢁ 0.75
23.40 ꢁ 1.83
26.90 ꢁ 3.12
18.28 ꢁ 1.60
AM-B
AM-A
AM-O
AM-M
ꢂ0.01 ꢁ 0.02
Pharmazie 63 (2008) 5
357

ORIGINAL ARTICLES
acid is a potent H-bonding group. However, the formation
Y = 0.0906X – 1.9261
R = 0.9578
of complexes masks the carboxylic group, which makes
H-bonding with the SC impossible. Thus the H-bonding
ability of AM and its complexes is almost identical, indi-
0.5
0
AM-O
AM-M
cating that DSC-AM ꢄ DSC-complexes
.
AM-A
-0.5
-1
AM-B
H
H
C
N
NH
2
3
O
H
CO
O
CH
3
3
-1.5
-2
O
O
Cl
AM
-2.5
0
5
10
15
20
25
30
Amlodipine base
log KS/EI (ml/g)
Fig. 3: Relationship between the permeability coefficients (log P) and skin/
EI system partition coefficients (log KS/EI) of AM and its com-
plexes. Each data point represents the average value of three per-
meation experiments
For lipophilic drugs, the partition from the SC into the ED
becomes a rate-limiting step as far as skin penetration is
concerned. The SC is essentially a lipidic layer. Lipophilic
drugs pass through the SC, and then they must transfer
directly into the aqueous medium –– ED. Therefore lipo-
philic drugs will remain in the SC (Moss et al. 2002).
There is an inverse relationship between log P and log KO/W
(Fig. 5). This result agrees well with a previous study by
Fang et al. (2003). Their investigation came to the conclu-
sion that in the case of isopropyl myristate (IPM), the ex-
perimentally determined permeability coefficients gener-
ally decreased with the increasing n-octanol/water
partition coefficients of the anti-inflammatory drugs having
a wide range of lipophilicity (log KO/W ¼ 0.5 ꢅ 4.88).
tration (Abraham et al. 1995). In contrast, the ED is an
aqueous environment for the permeants. Therefore,
DSC ꢃ DED. Three parameters, KSC, DSC and KED, will be
discussed below separately to explain the differences in
the fluxes and permeability coefficients of AM and its
complexes.
The rank order of the partition coefficients of AM and its
complexes between the EI system and the SC is AM-O >
AM-A > AM-M > AM-B > AM (Table 3 and Fig. 3),
which is almost identical to the rank order of the perme-
ability coefficients of the five compounds. A linear corre-
lation can be obtained between the steady-state fluxes of
AM and its complexes and the skin contents of the five
compounds (Fig. 4). The flux of AM or each complex
through the skin seems to be related directly to the drug
amount in the skin. This result strongly suggests that the
partition from the EI system into the skin, or the SC, is
the main factor governing the permeation of AM and its
complexes.
AM, with a log KO/W ¼ 1.50, is lipophilic, so the partition
from the SC into the ED becomes a barrier for the trans-
dermal passage of AM. On the other hand, its complexes,
having a log KO/W ¼ ꢂ0.34 ꢅ 0.65, are hydrophilic. These
complexes can partition into the ED much easier than
AM after diffusion through the SC. This means that
K
ED-AM < KED-complexes. This could be another factor re-
sulting in a lower permeability coefficient of AM com-
pared with its complexes.
In conclusion, the DSC thermograms provide an evidence
of the formation of complexes. The method of ion pairing
between AM and organic acids does not increase the flux
of AM through excised rat skin. The highest flux is ob-
tained by AM, but it has the lowest permeability coeffi-
cient. The four complexes studied all exhibited a lower
flux and higher permeability coefficient than AM. The ex-
Many studies have shown that the major negative determi-
nant of diffusion across the SC is the H-bonding of the
penetrants (e.g. Pugh et al. 1996; Cronin et al. 1998).
Based on the theory of functional groups as predictors of
diffusion, the order of retardation coefficients of functional
groups is acid > alcohol > phenol > ketone > ether. The
carboxylic group of adipic acid, oxalic acid and maleic
Y = –1.2941X – 0.0118
1
Y = 0.0103X + 0.4297
R = 0.9585
1.2
R = 0.9689
AM-O
0.5
AM
AM-M
0
–0.5
-1
AM-B
1
AM-A
AM-B
0.8
–1.5
-2
AM-M
0.6
AM
AM-A
–2.5
AM-O
0.4
–0.5
0
0.5
1
1.5
0
20
40
60
80
log KO/W
Skin content (µmol/g)
Fig. 5: Relationship between the permeability coefficients (log P) and
n-octanol/water partition coefficients (log KO/W) of AM and its
complexes. Each data point represents the average value of three
permeation experiments
Fig. 4: Relationship between the steady-state fluxes and skin contents of
AM and its complexes. Each data point represents the average va-
lue of three permeation experiments
358
Pharmazie 63 (2008) 5

ORIGINAL ARTICLES
3.2. Methods
periment involving skin uptake proves that the partition
between the SC and the EI system is the most significant
factor governing the permeation of AM and its complexes.
In addition, the lipophilicity of the drugs, as indicated by
the log KO/W value, influences the partition from the SC
into the ED. The highest lipophilicity of AM, among the
five compounds tested, is also a negative factor leading to
the lowest permeability coefficient.
3.2.1. Preparation of amlodipine and its complexes
AM was prepared from AM-B. AM-B was dissolved in methanol at 50 ꢀC,
treated with an equal molar amount of aqueous sodium hydroxide solution,
stirred and allowed to crystallize out at 4 ꢀC. The precipitated free base
was collected by filtration and then dried in an oven for 3 h (McDaid and
Deasy 1996). Its complexes were prepared from free base and the corre-
sponding acids by dissolving them in methanol separately at 50 ꢀC, mixing
and stirring. The methods of crystallization, filtration and drying were the
same as those used for the preparation of free base.
3.2.2. DSC
2.2. Comparison between previous reports and the present
study
DSC measurement was performed using a Shimadzu DSC-60 thermal ana-
lyzer. Samples of 3–5 mg were placed in a standard aluminum crucible
fitted with a perforated lid for scanning. An empty pan was used as a
reference. The samples were heated at a rate of 5 ꢀC ꢆ minꢂ1 over a tem-
perature range of 20–300 ꢀC. The melting points and the enthalpies of
fusion are listed in Table 1.
Many researchers have reported the enhancing effect of
ion pairing on the transdermal flux of drug (e.g. Fang
et al. 2003; Cheong and Choi 2002). However, an oppo-
site result was obtained in the current study. The increased
drug flux induced by ion pairing in the previous investiga-
tions can be attributed to the improved permeation proper-
ties of drug. Comparing the previous reports with the pre-
sent study, two points of transport properties are markedly
different.
(1) The melting points of complexes are lower than that
of the corresponding parent compound in the previous
research. In contrast, the melting point of AM
(142.32 ꢀC) is the lowest of the five compounds.
(2) The formation of complexes leads to a significant in-
crease in aqueous solubility and minor decrease in
liposolubility in the previous investigations. However,
the liposolubilities of amlodipine complexes are much
lower than that of AM, which is indicated by the solu-
bilities in the EI system and the n-octanol/water parti-
tion coefficients of five compounds. Furthermore, the
aqueous solubilities of amlodipine complexes are not
increased significantly compared with AM (Table 2).
Consequently, the flux of AM is higher than any other ion
pairs evaluated, which is different from the previous stu-
dies.
3.2.3. Solubility
The solubilities of AM and its complexes in the EI system and water were
determined at 32 ꢀC which reflects the body surface temperature. An ex-
cess of AM or each complex was dispersed into approximate 2 ml solution
in a sealed glass vial. Each vial was shaken in a water bath for 48 h, and
then 1 ml was transferred to a polypropylene micro-vial and centrifuged.
The concentration of AM and its complexes in the samples was assayed
by HPLC after appropriate dilution with methanol.
3.2.4. n-Octanol/water partition coefficient
The partition coefficients of AM and its complexes were determined in
n-octanol/water. Initially, n-octanol and water were shaken together, the then
allowed to stand for 24 h to ensure mutual saturation. A precise amount of
AM or each complex was dissolved in n-octanol, and then an equal volume
of water was added. After that, the system was shaken in a water bath at
32 ꢀC for 48 h. The initial concentration in n-octanol was defined as C. The
residual concentration in n-octanol was Cw measured by HPLC and hence
the amount partitioned into aqueous phase was C ꢂ Cw. The partition coeffi-
cient (KO/W) could be obtained from the following formula:
KO=W ¼ Cw=ðC ꢂ CwÞ
ð3Þ
3.2.5. In vitro transdermal experiments from the EI system
The experimental equipment consisted of a donor and a receiver compart-
ment of equal volume. The effective diffusion area was 0.95 cm2. About
5 cm2 of abdominal skin was mounted between the two diffusion cells
which were stirred magnetically and thermoregulated with a water jacket at
32 ꢀC. The donor cell was filled with 2.5 ml drug suspension (an excess of
drug was added to the EI system) and the receiver side contained 2.5 ml
receiver solution (PEG400 : pH7.4 phosphate buffer ¼ 4 : 6). Two ml sam-
ples were withdrawn at regular intervals from the receiver side and an
equal volume of receiver medium was immediately added to keep the vo-
lume constant. The concentration in each sample was determined by
HPLC after centrifugation.
3. Experimental
3.1. Materials
3.1.1. Drugs and vehicles
AM-B was kindly donated by Ningxia Kangya Drug Manufacturing Co.,
Ltd. (Ningxia, China). NaOH and oxalic acid were supplied by Shenyang
Zhengxin High-Technologies Research Institute Reagent Department (She-
nyang, China). Maleic acid and adipic acid were obtained from Tianjin
Standard Chemicals Co., Ltd. (Tianjin, China) and Shenyang Dongxing
Chemical Factory (Shenyang, China), respectively. n-Octanol, polyethylene
glycol 400 (PEG 400) and ethanol absolute were purchased separately
from Beijing Fuxing Chemical Factory (Beijing, China), Tianjin Bodi Che-
micals Co., Ltd. (Tianjin, China) and Tianjin Baishi Chemicals Co., Ltd.
(Tianjin, China). IPM was obtained from China National Medicines Co.,
Ltd. (Beijing, China). HPLC grade methanol was obtained from Yuwang
Chemicals Co., Ltd. (Shandong, China). All the other chemicals and sol-
vents were of analytical reagent grade.
3.2.6. Skin uptake experiments
A procedure similar to that used for the in vitro permeation experiment
was performed to determine the amount of AM and its complexes in the
skin. The excised rat skin was mounted between 2-chamber diffusion cells
that connected to a water bath at 32 ꢀC. The dermis side was covered with
a plastic sheet, and the SC side was filled with 2.5 ml suspension of AM
or each complex. After exposure to the drug solution for 10 h, the excess
drug on the skin surface was gently wiped off with methanol swabs. The
treated skin site (0.95 cm2) was punched out, cut, soaked in methanol for
48 h, and homogenized. After centrifugation at 16000ꢇg for 5 min, the
supernatant was analyzed by HPLC. The drug concentration in the treated
skin (skin content) was determined as the amount of drug in the skin
divided by the weight of skin. The partition coefficient between the skin
and the EI system (KS/EI) was calculated by dividing the skin content by
the solubility of drug in the EI system.
3.1.2. Animals
Male Wistar rats weighing 200 ꢁ 20 g (6–8 weeks old) used in all the
experiments were supplied by the Experimental Animal Center of She-
nyang Pharmaceutical University (Shenyang, China). The experiments
were performed in accordance with the guidelines for animal use published
by the Life Science Research Center of Shenyang Pharmaceutical Univer-
sity. The abdominal skin of the rats was used in the transdermal experi-
ments. The hair of the abdominal skin was carefully clipped under anesthe-
sia with urethane (20%, w/w, i.p.) and about 5 cm2 (circle of 2.5 cm
diameter) of full thickness skin was excised from the shaved site. The fat
and sub-dermal tissues were removed with surgical scissors. The skin was
washed with normal saline and kept frozen at ꢂ20 ꢀC. The skin was
checked to ensure that no obvious defects were present prior to the experi-
ments.
3.2.7. Data analysis
The amount of drug that had penetrated at each sampling interval was
obtained from the measured concentration and volume of the receiver
phase. The cumulative amount of drug was calculated by the following
formula:
n
P
Q ¼
ð2:5Ci ꢂ 0:5Ciꢂ2Þ=A i ¼ 2; 4; 6 . . .
ð4Þ
i ¼ 2
Pharmazie 63 (2008) 5
359

ORIGINAL ARTICLES
where Q is the cumulative amount penetrated; Ci is the concentration in
Cronin MTD, Dearden JC, Gupta R, Moss GP (1998) An investigation of
the mechanism of flux across polydimethylsiloxane membranes by use
of quantitative structure-permeability relationships. J Pharm Pharmacol
50: 143–152.
Fang L, Numajiri S, Kobayashi D, Morimoto Y (2003) The use of com-
plexation with alkanolamines to facilitate skin permeation of mefenamic
acid. Int J Pharm 262: 13–22.
Francoeur ML, Potts RO (1988) EP Application 271983.
Haria M, Wagstaff AJ (1995) Amlodipine-a reappraisal of its pharmacolo-
gical properties and therapeutic use in cardiovascular disease. Drugs 50:
560–586.
the receiver compartment at the time i; and A is the effective diffusion
area (0.95 cm2). Experiments were performed for 10 h. The cumulative
amount penetrated from unit area vs. time was plotted. The steady-state
flux (Js) was calculated from the slope of the linear portion of the plotted
curve. The lag time (tlag) was the intercept obtained by extrapolation of the
linear portion to the time axis. The permeability coefficient (P) was ob-
tained by dividing Js by the drug solubility in the EI system.
3.2.8. Analytical method
The concentrations of AM and its complexes were measured by HPLC.
The HPLC equipment consisted of a HITACHI L7110 pump, a HITACHI
UV–VIS L-7420 detector (both from Hitachi High-Technologies Corpora-
tion, Tokyo, Japan), and a HT-220A column temperature controller (Tian-
jin, China). Analysis was performed on a 5-mm ODS column (200 mm ꢇ
4.6 mm, DIKMA technologies, Beijing, China) operated at 40 ꢀC. The mo-
bile phase was methanol-0.03 mol/L KH2PO4 buffer solution (7 : 3) at a
flow rate of 1 ml/min. The wavelength of the detector was 238 nm and the
internal standard was 8 mg/ml propylparaben. Calibration curves were con-
structed using the peak area ratio between AM-B and propylparaben vs.
the molar concentration of AM-B in solution. Comparison of quantity of
AM and its complexes permeated was carried out using molar concentra-
tions. The retention times of the drug and the internal standard were ap-
proximately 4.3 min and 5.8 min respectively.
Kadono M, Kubo K, Miyazaki H, Tojyo N, Nakagawa S, Miyashita K,
Zmanishi P, Rytting JH, Mayumi T (1998) Enhanced in vitro percuta-
neous penetration of salicylate by ion pair formation with alkylamines.
Biol Pharm Bull 21: 599–603.
McDaid DM, Deasy PB (1996) Formulation development of a transdermal
drug delivery system for amlodipine base. Int J Pharm 133: 71–83.
Megwa SA, Cross SE, Benson HAE, Roberts MS (2000a) Ion-pair forma-
tion as a strategy to enhance topical delivery of salicylic acid. J Pharm
Pharmacol 52: 919–928.
Megwa SA, Cross SE, Whitehouse MW, Benson HAE, Roberts MS
(2000b) Effect of ion pairing with alkylamines on the in vitro dermal
penetration and local tissue disposition of salicylates. J Pharm Pharma-
col 52: 929–940.
Moss GP, Dearden JC, Patel H, Cronin MTD (2002) Quantitative struc-
ture-permeability relationship (QSPRs) for percutaneous absorption.
Toxicol in Vitro 16: 299–317.
Pugh WJ, Roberts MS, Hadgraft J (1996) Epidermal permeability – pene-
tration structure relationship: 3. The effect of hydrogen bonding interac-
tions and molecular size on diffusion across the stratum corneum. Int J
Pharm 138: 149–165.
Scheuplein RJ (1976) Mechanism of percutaneous absorption. II. Transient
diffusion and the relative importance of various routes of skin permea-
tion. J Invest Dermatol 48: 79–88.
Acknowledgements: We would like to thank Mr. Yasunori Morimoto (Fa-
culty of Pharmaceutical Sciences, Josai University, Japan) for his kind gifts
of 2-chamber diffusion cells and a synchronous motor.
Reference
Abraham MH, Chadha HS, Mitchell RC (1995) The factors that influence
skin penetration of solutes. J Pharm Pharmacol 47: 8–16.
Barry BW (2001) Novel mechanisms and devices to enable successful
transdermal drug delivery. Eur J Pharm Sci 14: 101–104.
Cheong H, Choi H (2002) Enhanced percutaneous absorption of piroxicam
via salt formation with ethanolamines. Pharm Res 19: 1375–1380.
Valenta C, Siman U, Kratzel M, Hadgraft J (2000) The dermal delivery of
lignocaine: influence of ion pairing. Int J Pharm 197: 77–85.
360
Pharmazie 63 (2008) 5