The efficacy of antifungal agents is primarily evaluated
by antifungal susceptibility testing. However, the mini-
mum inhibitory concentrations (MICs) are influenced by
pH, inoculum size, incubation time, incubation temper-
ature, and composition of the medium (2, 3, 5, 13).
Therefore, the clinical effects cannot be predicted from
only the results of the antifungal susceptibility testing
against dermatophytes. The pharmacokinetics of anti-
fungal agents in the stratum corneum are an especially
important influence on the antifungal activity (1, 10, 14).
To further evaluate the therapeutic efficacy of topical
antifungal agents, experimental animal models infected
with Trichophyton mentagrophytes on the back and foot
of guinea pigs have been used widely (9, 15, 17, 18). In
contrast to the in vitro susceptibility testing, these mod-
els are thought to be useful to assess the clinical effect
and are assumed to be suitable for experiments related to
treatment with antifungal agents (16).
It has been reported that the discrepancy between in
vitro and in vivo experiments in terms of assessing the
clinical effects of some antifungal agents is considered to
be attributable to the dependency on the interaction of
antifungal agents with keratinized tissues (10, 12).
Indeed, some of the topical antifungal agents with high
antifungal activity in vitro show rather poor effects in
Therefore, evaluation of the interaction between the
stratum corneum and antifungal agents is important to
ascertain useful clinical medicines. The purpose of this
study was to evaluate the anti-T. mentagrophytes activi-
ty of antifungal agents by a new in vitro evaluation sys-
Microbiol. Immunol., 46(8), 521―525, 2002
Abbreviations: BFZ, bifonazole; ESI, electrospray ionization;
LCZ, lanoconazole; MIC, the minimum inhibitory concentra-
tion; SCE, stratum corneum epidermis; SDA, Sabouraud’s dex-
trose agar; SDB, Sabouraud’s dextrose broth; TBF, terbinafine.
*Address correspondence to Dr. Takuji Nakashima, Pharma-
ceutical R&D Laboratories, POLA Chemical Industries, Inc.,
560 Kashio-cho, Totsuka-ku, Yokohama, Kanagawa 244―0812,
Japan. Fax: ＋81―45―826―7259. E-mail: takuji.nakashima
A Novel Method Using Micropig Stratum Corneum
In Vitro for the Evaluation of Anti-Trichophyton
Takuji Nakashima*, Akira Nozawa, and Toshiro Majima
Pharmaceutical R&D Laboratories, POLA Chemical Industries, Inc., Yokohama, Kanagawa 244―0812, Japan
Received January 17, 2002; in revised form, Aplil 30, 2002. Accepted May 8, 2002
Abstract: Antifungal susceptibility testing under conditions close to clinical status is expected to provide more
helpful information than that obtained by a conventional microdilution method. For this purpose, we devel-
oped a novel method to evaluate anti-Trichophyton mentagrophytes activity of antifungal agents in vitro by
using disks of micropig stratum corneum epidermis (SCE). Basal agar medium containing K2HPO4,
MgSO4, CaCl2and three kinds of antibiotics. Bifonazole (BFZ), lanoconazole (LCZ) or terbinafine (TBF)
was added to the basal agar medium to give serially doubling dilutions ranging from 0.0006 to 10 ?g/ml.
Five-hundred-?l portions of the agar media thus prepared were solidified in wells of flat-bottomed plates.
SCE disks (6 mm in diameter) were placed on surfaces of the agar medium and 104conidia of T. menta-
grophytes were inoculated on each SCE disk. There was very good correlation between the initial concen-
tration of the antifungal agents added to the basal agar medium (?g/ml) and the concentration of the agents
impregnated into the SCE disks (?g/g) (r2＞0.99). The minimum inhibitory concentration (MIC) values of
BFZ, LCZ and TBF were respectively 26-, 10- and 78-times higher than those measured by the standard
microdilution method. From the correlation between the concentration of the agents in the basal medium
and that in the SCE disks, the above MIC values corresponded to the concentrations in SCE disks (?g/g),
832.95 for BFZ, 1.42 for LCZ and 8.87 for TBF. This novel method of antidermatophytic susceptibility test-
ing using SCE would be useful as an in vitro screening of proper antimycotics for topical treatment of der-
Key words: Stratum corneum, Antimycotic, Tinea unguium, In vitro
tem with the stratum corneum.
Materials and Methods
Preparations of antifungal drugs for analysis and
biological studies. Bifonazole (BFZ) was purchased
from Sigma-Aldrich Co., St. Louis, Mo., U.S.A., and
lanoconazole (LCZ) and terbinafine (TBF) were syn-
thesized in our laboratory. All solvents were HPLC
grade and other chemicals were of analytical grade.
Stock solutions of these antifungal agents for analysis
were prepared in methanol at a concentration of 0.1
mg/ml. Stock solutions of these antifungal agents for
biological studies were dissolved in dimethyl sulfoxide
at a concentration of 100 mg/ml, and subsequent dilutions
were made in appropriate volumes of dimethyl sulfoxide.
Preparation of stratum corneum epidermal (SCE)
disk. Heat-separation was introduced as a simple, reliable
method for obtaining the stratum corneum (4). Heating
normal skin of Yukatan micropig (Japan Charles River,
Co., Ltd., Tokyo) at 60 C for 2 min resulted in a distinct
separation of the stratum corneum from the skin tissue.
The pig stratum corneum epidermal (SCE) disks were
prepared by biopsy (φ6 mm, Kai Industries Co., Ltd.,
Gifu, Japan). The average dry weight of the collected
SCE disks was 1.7 mg.
Standard and calibrations solutions for analysis.
Standard spiking solutions were prepared by the appro-
priate dilution of the stock solutions. The spiking solu-
tions were used for the preparation of the calibration
samples. Calibrators were prepared by adding 50 µl
each of the appropriate spiking solutions to antimy-
cotics-free SCE disks. LCZ standard solution was used
as an internal standard for TBF and BFZ, and TBF stan-
dard solution was used for LCZ.
All the calibrators and test samples were dispensed
into glass tubes, and then spiked with 50 µl of internal
standard solution. One milliliter of methanol was added
to the sample and sonicated for 30 min. Subsequently,
0.25 ml of 5 mmol/liter formic acid was added to the tube
and mixed for 10 sec. The mixture was filtrated, and 1 µl
of the solution was injected onto the HPLC column for
Liquid chromatography-mass-mass spectrometry elec-
trospray ionization (LC/MS/MS-ESI) instrumentation.
Tandem mass spectrometry was performed using a Quat-
tro Ultima triple-stage quadripolar MS system equipped
with an electrospray ionization (ESI) interface (Micro-
mass, Altrincham, U.K.). Masslynx NT software (ver-
sion 3.4) was used for data acquisition. The source and
desolvation temperatures were maintained at 130 and 350
C, respectively. The desolvation and cone gas flows
were set at 650 liter/hr and 0 liter/hr, respectively. The
sample cone voltage was optimized for each of the test
compounds: BFZ at 20 V, TRB at 45 V, and LCZ at 70 V.
The electrospray capillary voltage was set at ＋3.0 kV.
MS/MS experiments were carried out with an argon
pressure of approximately 1.3× 10－4mbar in the colli-
sion cell (Q2). The collision energy was optimized for
each analyte: BFZ at 15 V, TRB at 20 V, and LCZ at 22
V. For MRM experiments, both Q1 and Q3 were set at
fixed m/z values for transition (dwell time: 0.5 sec, inter-
channel delay 0.03 sec). Following HPLC separation, the
peak areas of the m/z 311→242.8 product ion of BFZ
and the m/z 292→140.8 product ion of TBF were mea-
sured against the m/z 319.6→149.8 product ion of LCZ.
The peak area of the m/z 319.6→149.8 product ion of
LCZ was measured against the m/z 292→140.8 product
ion of TBF.
Liquid chromatography was performed with an
Alliance system (Waters Corp., Milford, Mass., U.S.A.).
Chromatographic separations were carried out using an
Inertsil ODS-2, 5 µm column (250 mm long, 4.6 mm i.d.,
GL Sciences Inc.). An isocratic mobile phase containing
5 mmol/liter formic acid-methanol (20:80, v/v) was uti-
lized at a flow rate of 1 ml/min. The flow was split
and 20% was applied to MS/MS system. The column
temperature was kept at 45 C.
Calibration curves (y＝mx± b), represented by the
plots of the peak area ratios (y) of antimycotics to inter-
nal standard versus the concentrations (x) of the cali-
bration samples, were generated using the weighted
(1/x2) linear least-squares regression model. Concen-
trations of antimycotics in SCE were calculated from the
resulting peak area ratios and the regression equation of
the calibration curve.
Organisms and cultural conditions. Trichophyton
mentagrophytes TIMM2789 was obtained from Teikyo
University and used in this study. A conidial suspension
of T. mentagrophytes was prepared in sterile physiolog-
ical saline containing 0.05% (w/v) polysorbate 80 (poly-
oxyethylene sorbitan monooleate, Wako Pure Chemical
Industries, Ltd., Tokyo) from cultures grown on
Sabouraud’s dextrose agar (SDA) slant at 28 C for 2
weeks. Following filtration through a sterilized gauze to
remove hyphal fragments and agar blocks, the final
conidial suspension was adjusted to 2× 104conidia/ml
with Sabouraud’s dextrose broth (SDB) (Becton Dick-
inson and Co.) for standard MIC determination and 2×
106conidia/ml with sterile physiological saline for sus-
ceptibility testing of SCE disks.
Susceptibility testing. The MIC for T. mentagro-
phytes TIMM2789 was determined by a microdilution
technique with SDB 96-well flat-bottomed microdilution
plates (Sumitomo Bakelite Co., Ltd., Tokyo). Two-fold
concentrations were dispensed into each well of rows 1
T. NAKASHIMA ET AL
to 15 in 100 µl volumes. That is, 5 µl of the antimycotic
solutions of 0.012―200 µg/ml were added to 95 µl of
SDB. Row 1 contained the highest drug concentration,
and row 15 the lowest drug concentration. Each well was
then inoculated with 100 µl of the conidial suspension in
SDB. All plates were incubated at 28 C for 7 days.
The MIC was defined as the lowest drug concentration at
which visual fungal growth was prevented.
The susceptibility testing for T. mentagrophytes
TIMM2789 with SCE disk was performed in 48-well
flat-bottomed microdilution plates (Sumitomo Bake-
lite). The medium was composed of 1.5% Difco agar
(Becton Dickinson), 0.2%-K2HPO4, 0.005%-MgSO4,
0.005%-CaCl2 (Wako Pure Chemical), 500 µg cyclo-
heximide (Wako Pure Chemical) per ml, 50 µg of
sisomycin (Sigma-Aldrich Co.) per ml, and 100 µg of
chloramphenicol (Wako Pure Chemical) per ml. Antimy-
cotic solutions were added to the medium to provide
concentrations in the range of 0.0006―10 µg/ml. The
medium containing each antimycotic solution was dis-
pensed with a multi-channel pipette into each well of
rows 1 to 15 in 500 µl volumes. Rows 1 and 15 con-
tained the highest drug concentration and the lowest
drug concentration, respectively, and row 16 was drug
free. A SCE disk was then put on each agar plate, and
inoculated with 5 µl of the conidial inoculum suspension
before dilution with sterile physiological saline (104
conidia/disk). All plates were incubated for 7 days at 28
C for visual observation.
Experiments were carried out to clearly assess whether
each of the three antifungal agents, BFZ, LCZ and TBF,
would more preferentially interact with the stratum
corneum in terms of antifungal activity against T. men-
tagrophytes. In this study, the stratum corneum of
Yukatan micropig was used to examine the effect on
the anti-T. mentagrophytes activity of the three antifun-
The antifungal activity evaluation system using stra-
tum corneum epidermis is shown in Fig. 1. The SCE
disks were placed on the medium in which the source of
nutrition for the growth of T. mentagrophytes was not
contained, followed by the inoculation of the conidia of
T. mentagrophytes onto the SCE disk. T. mentagrophytes
did not grow in the medium alone (data not shown).
Therefore, T. mentagrophytes grew only in the presence
of the stratum corneum of Yukatan micropig as a source
Antimycotic Concentration in SCE Disk
The SCE disks were recovered from the medium con-
taining BFZ, LCZ and TBF. Then the concentration of
each antimycotic which had transferred from the medi-
um to the SCE disk was measured by LC/MS/MS. The
correlations between the concentration in the medium and
in the SCE disks are shown in Fig. 2. Increases in the
concentrations of BFZ, LCZ and TBF in the medium cor-
related well with those in the SCE disks and showed
good linearity (r2was ＞0.9927, ＞0.9948, and ＞0.9977,
respectively). Among the three antimycotics, the high-
est concentration was recovered from the SCE disk on the
BFZ-containing medium, and almost equivalent con-
centrations of LCZ and TBF were recovered from the
corresponding SCE disks.
NEW EVALUATION OF ANTIMYCOTICS, IN VITRO
Fig. 1. Schematic presentation of the culture system for evaluating antimycotic activity using the stratum corneum epi-
dermis (SCE) of Yukatan micropig (upper) and an example of the susceptibility testing with SCE (lower). A SCE disk
was placed on the basal agar medium containing BFZ, LCZ or TBF of serial two-fold diluted concentrations, and coni-
dia of T. mentagrophytes were inoculated on the SCE disk.
Anti-Fungal Susceptibility Test
These experiments were carried out in triplicate and
the means calculated. The drug concentration in the
SCE disk was based on the calibration curve by com-
parison with the mean drug concentration in medium
which prevented the growth of T. mentagrophytes. The
susceptibilities of T. mentagrophytes in SCE were eval-
uated using the medium in which the stratum corneum
was the only source of nutrition. In vitro susceptibilities
of T. mentagrophytes TIMM2789 to BFZ, LCZ, and
TBF were examined and are summarized in Table 1.
LCZ and TBF showed an equal MIC, and the MIC of
BFZ was approximately 200-fold higher than those of
LCZ and TBF. The antifungal activities of all the
antimycotics were decreased when measured using the
medium with SCE; that is, the MICs on SDB alone
were decreased by 26-fold in BFZ, by 10-fold in LCZ,
and by 78-fold in TBF. LCZ prevented the growth of T.
mentagrophytes at the lowest drug concentration in SCE
among three antifungal agents.
The habitation locus of the causal fungi in trichophy-
tosis is the stratum corneum. Therefore, in addition to the
in vitro anti-fungal activity of a medicine, the efficacy of
a topical antimycotic chemotherapy is in some part con-
trolled by the pharmacokinetics of keratinized tissue,
especially the stratum corneum. One of the desirable
pharmacokinetic characteristics of an antimycotic pen-
etrating from the skin surface is that the active type of the
agent remains in the stratum corneum until it inhibits the
growth of the dermatophytes. Some studies of percuta-
neous absorption of the antimycotic applied to human or
animal skin have been reported previously (6, 8, 11).
However, the concentration of antimycotics in the
stratum corneum required to prevent the growth of der-
matophytes is unknown. In the relation between keratin
and an antimycotic, the affinity to keratin and the anti-
fungal activity of antimycotics have been reported (1, 10,
16). The antifungal activity is rather reduced in the
presence of keratin. It is considered that the effect of an
antimycotic is provided by the affinity and the concen-
tration of the antimycotic in the stratum corneum.
Therefore, we attempted to examine the affinity of
antimycotics to keratin and the antifungal activity of
the stratum corneum with three antimycotics. As a
practical method, agar containing an antimycotic but
not a source of nutrition was dispensed into each well of
a 48-well plate. The stratum corneums were placed on
the agar, and then the antimycotic in one stratum
corneum was extracted by methanol after incubation.
The antimycotic in the extraction liquid was quantitated
by LC/MS/MS. Moreover, the conidia of T. mentagro-
phytes were inoculated on another stratum corneum on
the agar. The concentrations of each antimycotic in the
medium and stratum corneum at which T. mentagro-
phytes growth was inhibited after incubation were com-
According to the results, the antimycotic transfered
dose-dependently to the stratum corneum from the medi-
um. The antimycotic concentration in the medium show-
ing antidermatophytic activity increased more than that
in the medium in the standard method which used SDB,
and moreover, the concentrations of BFZ, LCZ, and
TBF in stratum corneum were 832.95 µg/g, 1.42 µg/g,
and 8.87 µg/g, respectively. We reported the therapeutic
T. NAKASHIMA ET AL
Table 1. Comparison of MICs of BFZ, LCZ and TBF against
Trichophyton mentagrophytes measured by the standard micro-
dilution method and the novel SCE disk method
MIC (µg/ml) measured by:
a)Standard microdilution method using Sabouraud dextrose
b)Initial concentration of the antifungals added to the basal agar
c)Concentration of the antifungals in SCE disks impregnated
from the agar medium (µg/g disk). These values were calculated
from the initial concentrations of the antifungals added to the agar
Fig. 2. A linear relationship between the initial concentrations of
antifungal agents added to the basal agar medium (µg/ml) and the
concentrations of the agents in SCE disks impregnated from the
basal medium measured 7 days after incubation at 28 C (µg/g
disk). The correlation coefficient (r2) was over 0.99 for the tested
efficacy of these three antimycotics using a guinea pig
model of tinea pedis with a new medium useful for the
recovery of dermatophytes (7). This medium is capable
of inactivating the antimycotic carried over from recov-
ery culture. The average fungal burden scores of BFZ,
LCZ, and TBF when infected animals received 7-day
treatment with such active creams were ＋6.9, ＋1.4, and
＋2.1, respectively. The therapeutic efficacy correlated
well with the antimycotic concentration in the stratum
In this study, BFZ was the most distributed to the
stratum corneum at a high concentration among the
three antimycotics. TBF and LCZ were distributed to the
same degree. The adsorption power of keratin is very
strong and it adsorbs various substances. A large num-
ber of antimycotics are inactivated once they have com-
bined with keratin protein (17). The results obtained in
this study were consistent with the results of the report-
ed studies that the MICs with keratin particles were
higher than MICs with nutrient medium (1, 10).
Although LCZ and TBF are almost equal in their distri-
bution and MICs on SDB, LCZ was stronger than TBF in
the antifungal activity in the presence of the stratum
corneum. When the concentration of LCZ in the stratum
corneum is set at 1, approximately 587- and 6-fold con-
centrations of BFZ and TBF are required, respectively.
Therefore, the therapeutic efficacy in the topical treatment
of dermatophytosis would depend on the potent anti-
fungal activity of the agent and the interaction with ker-
In conclusion, this susceptibility testing method using
the stratum corneum of micropig would be useful as an
in vitro alternative to in vivo methods for evaluating
antifungal agents. Moreover, it is also suggested that this
method may be an efficient indicator of the antifungal
activity in clinical cases because the therapeutic efficacy
in a guinea pig model of tinea pedis and the antifungal
activity in this method correlated well.
We thank Prof. Shigeji Aoki (The Nippon Dental University,
Niigata, Japan) for his encouragement and critical advice.
1) Ariga, T., Hase, T., and Yokoo, M. 1990. Activity of topical
antifungals on infected sites―Skin permeability and adsorp-
tion to horny materials―. Nishinihon J. Dermatol. 52:
2) Espinel-Ingroff, A. 2001. Comparison of the E-test with
the NCCLS M38-P method for antifungal susceptibility
testing of common and emerging pathogenic filamentous
fungi. J. Clin. Microbiol. 39: 1360―1367.
3) Espinel-Ingroff, A., Dawson, K., Pfaller, M., Anaissie, E.,
Breslin, B., Dixon, D., Fothergill, A., Paetznick, V., Peter, J.,
Rinaldi, M., and Walsh, T. 1995. Comparative and collabo-
rative evaluation of standardization of antifungal suscepti-
bility testing for filamentous fungi. Antimicrob. Agents
Chemother. 39: 314―319.
4) Kassis, V., and Sondergaard, J. 1982. Heat-separation of
normal human skin for epidermal and dermal prostaglandin
analysis. Arch. Dermatol. Res. 273: 301―306.
5) Llop, C., Pujol, I., Aguilar, C., Sala, J., Riba, D., and Guar-
ro, J. 2000. Comparison of three methods of determining
MICs for filamentous fungi using different end point criteria
and incubation periods. Antimicrob. Agents Chemother. 44:
6) Lucker, P.W., Beubler, E., Kukovetz, W.R., and Ritter, W.
1984. Retention time and concentration in human skin of
bifonazole and clotrimazole. Dermatologica 169 (Suppl 1):
7) Nakashima, T., Nozawa, A., Ito, T., Majima, T., and Yama-
guchi, H. 2002. Development of a new medium useful for the
recovery of dermatophytes from clinical specimens by min-
imizing the carryover effect of antifungal agents. Microbiol.
Immunol. 46: 83―88.
8) Niwano, Y., Matsui, M., Tabuchi, T., Kanai, K., Hamaguchi,
H., Miyazaki, T., Uchida, K., and Yamaguchi, H. 1997.
Studies on the antifungal activity of the new imidazole
antimycotic lanoconazole in infected sites. Distribution in the
skin and in vitro activity in the presence of stratum corneum.
Arzneimittelforschung/Drug Res. 47: 1056―1060.
9) Niwano, Y., Kuzuhara, N., Kodama, H., Yoshida, M.,
Miyazaki, T., and Yamaguchi, H. 1998. In vitro and in vivo
antidermatophyte activities of NND-502, a novel optically
active imidazole antimycotic agent. Antimicrob. Agents
Chemother. 42: 967―970.
10) Okeke, C.N., Tsuboi, R., Kawai, M., and Ogawa, H. 2000.
Fluorometric assessment of in vitro antidermatophytic activ-
ities of antimycotics based on their keratin-penetrating
power. J. Clin. Microbiol. 38: 489―491.
11) Patzschke, K., Ritter, W., Siefert, H.M., Weber, H., and
Wegner, L.A. 1983. Pharmacokinetic studies following sys-
temic and topical administration of [14C] bifonazole in man.
Arzneimittelforschung/Drug Res. 33: 745―750.
12) Pierard, G.E., Arrese, J.E., and De Doncker, P. 1995. Anti-
fungal activity of itraconazole and terbinafine in human
stratum corneum: a comparative study. J. Am. Acad. Der-
matol. 32: 429―435.
13) Pujol, I., Guarro, J., Llop, C., Soler, L., and Fernandez-Bal-
lart, J. 1996. Comparison study of broth macrodilution and
microdilution antifungal susceptibility tests for the filamen-
tous fungi. Antimicrob. Agents Chemother. 40: 2106―2110.
14) Takahashi, H. 1994. Problems with the topical antimycotics.
Jpn. J. Med. Mycol. 35: 331―334.
15) Uchida, K., and Yamaguchi, H. 1991. Retention of
amorolfine in the skin tissue as measured by prevention of the
development of experimental dermatophytosis in guinea
pigs. Jpn. J. Antibiot. 44: 1032―1041.
16) Uchida, K., and Yamaguchi, H. 1996. Preclinical therapeu-
tic evaluation of agents for treating dermatophytosis. Jpn. J.
Med. Mycol. 37: 199―205.
17) Uchida, K., and Yamaguchi, H. 1994. Prophylactic effect of
oral terbinafine on the development of infection in guinea
pigs inoculated with Trichophyton mentagrophytes. Jpn. J.
Antibiot. 47: 50―56.
18) Wakabayashi, H., Uchida, K., Yamauchi, K., Teraguchi, S.,
Hayasawa, H., and Yamaguchi, H. 2000. Lactoferrin given in
food facilitates dermatophytosis cure in guinea pig mod-
els. J. Antimicrob. Chemother. 46: 595―602.
NEW EVALUATION OF ANTIMYCOTICS, IN VITRO