Você está na página 1de 10

Medical Mycology 2013, Early Online: 110

Acyclic sesquiterpenes released by Candida albicans


inhibit growth of dermatophytes

Med Mycol Downloaded from informahealthcare.com by RMIT University on 09/08/13


For personal use only.

JOCHEN BRASCH*, FELIX HORTER*, DANIEL FRITSCH*,VERA BECK-JENDROSCHEK*,


ARMIN TRGER & WITTKO FRANCKE
*Department of Dermatology, University Hospitals of Schleswig-Holstein, Kiel, and Institute of Organic Chemistry, Department of
Chemistry, University of Hamburg, Hamburg, Germany

It is unresolved as to whether fungi that share a common skin habitat might in principal
interact. In in vitro screening tests with Candida albicans, Trichophytum rubrum and
other common dermatophytes, we found C. albicans releases volatile compounds that
inhibit growth of the dermatophytes. By applying (enantioselective) gas chromatography combined with mass spectrometry we identified 8 compounds among which stereochemically pure (3R,6E)-2,3-dihydrofarnesol (R-DHF) and (2E,6E)-farnesol (F-ol) were
the main components. Synthetic R-DHF and its enantiomer, (3S,6E)-2,3-dihydrofarnesol
(S-DHF), as well as F-ol were tested for their capacity to inhibit growth of dermatophytes in microtiter-plate assays over 62 h. All three compounds showed significant
and concentration-dependent, to a certain extent even species-specific, inhibitory effects
on T. rubrum, T. mentagrophytes, Microsporum canis and Epidermophyton floccosum.
In general, S-DHF and F-ol had a pronounced effect on the dermatophytes, similar to
or even stronger than that of fluconazole. E. floccosum was completely suppressed by
12.5 g/ml dihydrofarnesol, as was the inhibition caused by 50 g/ml fluconazole. Similarly, S-DHF- was more active against T. rubrum than fluconazole. To the best of our
knowledge, 2,3-dihydrofarnesol has not yet been described as a volatile generated by
microorganisms, and its inhibitory effect on dermatophytes is new to science. However,
the relevance of this compound in interfungal interference in situ is unknown. In contrast,
farnesol is a well-known semiochemical of C. albicans with intraspecific effects and a
clear impact on other microorganisms. Mutual intermicrobial communication based on
fungal volatiles therefore appears to be an exciting field for future investigations.
Keywords Fungi, dihydrofarnesol, farnesol, enantiomers, semiochemicals, antifungals

Introduction
Dermatophytes are highly specialized hyphomycetes and
the causal agents of superficial human skin infections
(tinea), which account for a large share of all infections
observed worldwide. In Germany, dermatophytes isolated
from tinea lesions primarily belong to Trichophyton rubrum,
T. mentagrophytes and Microsporum canis. However, tinea

Received 26 January 2013; Received in final revised form 18 April 2013;


Accepted 9 June 2013.
Correspondence: Jochen Brasch, Department of Dermatology, University Hospital of Schleswig-Holstein, Campus Kiel, Schittenhelmstra e 7,
D-24105 Kiel, Germany. Tel: 49 431 597 1507; Fax: 49 431 597 1611;
E-mail: jbrasch@dermatology.uni-kiel.de

2013 ISHAM

lesions not only harbor dermatophytes but are usually


concomitantly colonized by other microorganisms, which
may cause interspecific interactions. Dermatophytes can
produce antibiotics, toxins and substances triggering cooperative (CAMP-like) haemolytic reactions [1], but the relevance of these properties is still unexplored. On the other
hand, nearly nothing is known about substances released
by associated or accompanying microorganisms that may
interfere with dermatophyte development in tinea. In this
context it should be noted that in intertriginous tinea the
concomitant presence of a dermatophyte and C. albicans
is not unusual.
In the present paper we investigated the effect of
C. albicans on dermatophytes, tackling the potential interactions at three different levels. First, we examined whether
DOI: 10.3109/13693786.2013.814174

2J

Brasch et al.

C. albicans can emit volatile compounds capable of


affecting the growth of dermatophytes. After establishing
that such substances were in fact released by C. albicans
under appropriate conditions, the chemical structures of
the main volatiles were then elucidated. Subsequently,
these candidate compounds were synthesized and tested
for their impact on the growth of dermatophytes in vitro
with fluconazole serving as a reference.

Material and methods

Med Mycol Downloaded from informahealthcare.com by RMIT University on 09/08/13


For personal use only.

Screening for volatile compounds released by C. albicans


that interfere with the growth of dermatophytes
Screening assays were performed with C. albicans and a
dermatophyte in shared Petri dishes, provided with airtight
seals. All fungal strains used in our study were stored in a
Microbank freezer storagebox (Alere, Neston, UK) at
80C for future investigations. The bottom pan of a Petri
dish (Sarstedt, Nmbrecht, Germany) was filled with a
nutrient agar (22 ml, various compositions were tested, see
below) that was inoculated with C. albicans yeast cells. The
lid of the same Petri dish was filled with 13 ml Sabourauds
glucose agar (SGA), inoculated with a suspension of dermatophyte conidia. To allow continuous microscopic observation of dermatophyte growth without opening the Petri
dish, some free space (15 mm diameter) was left in the
center of the agar in the bottom pan. This window allowed
direct viewing of the surface of the lid culture through the
transparent plastic bottom of the dish. After inoculation of
the agars in the bottom pan and lid, the Petri dish was
closed and provided with an air tight seal of parafilm.
Particular care was taken to prevent any direct contact
between the dermatophyte culture in the lid of the Petri
dish and the C. albicans colonies growing in its bottom
pan. Volatiles, however, could freely diffuse in the air space
of the sealed Petri dish. Compounds released from the
organisms in the top and bottom could reach each other
without hindrance.
Six typical strains of C. albicans isolated from different
patients in our hospital were used for the screening experiments. Their identification was based on their morphology
on rice agar, a positive germ tube test, and an unambiguous
pattern in the ID 32 C assimilation test (bioMerieux, Lyon,
France). All strains were grown for 4 days at 37C on SGA
(SAB2-D, bioMerieux, Lyon, France) supplemented with
penicillin 40,000 IE/L (Grnenthal GmbH, Stolberg, Germany), chloramphenicol 40 mg/l (Calbiochem, La Jolla,
CA, USA), and cycloheximid 400 mg/l (AppliChem
GmbH, Darmstadt, Germany). These cultures served as the
source to prepare yeast suspensions in 0.85% NaCl with
McFarland scale no. 5 turbidity. To inoculate the agar in
the bottom of the Petri plates, the agar surfaces were evenly

and entirely moistened with this yeast suspension. Subsequently, the bottom pans with the Candida-inoculated agar
were incubated for 10 h at 25C before they were closed
with the lid containing SGA inoculated with dermatophyte
conidia.
In addition to SGA, various agar media were tested to
support the development of C. albicans in this system,
including peptone agar containing 10 g peptone (Merck,
Darmstadt, Germany), 20 g glucose and 20 agar agar in
1 l water; Wort agar (BD Biosciences, Sparks MD, USA);
and potato-dextrose-agar (Merck, Darmstadt, Germany),
all prepared according to the suppliers instructions.
The dermatophytes used for the screening experiments
were T. rubrum, T. mentagrophytes, T. terrestre and
M. gypseum. For each species we selected a representative
strain that produced abundant conidia, had typical morphologies and characteristic physiological features [2].
The strains were grown for 14 days at 25C on SGA
(SAB2-D, bioMerieux, Lyon, France) supplemented with
penicillin, chloramphenicol and cycloheximide as described
for the yeast agar. Subsequently, the plates were flooded
with a sterile solution of 0.85% NaCl, and conidia were
detached from the mycelium by gently stirring the culture
surface with a sterile glass rod. The resulting suspension
of conidia and mycelial fragments was harvested with a
pipette, and the conidia were separated by filtration of the
suspension through sterile gauze compresses. The filtrate
was centrifuged, the supernatant discarded, and the conidia
resuspended in 0.85% NaCl and filtered again. This procedure was repeated until only conidia were noted by microscopic examination. The suspension was finally adjusted
to a concentration of 103 conidia/l calculated from counts
in a hemocytometer (Neubauer improved counting chamber). For inoculation, 100 l of this suspension were evenly
spread out on the agar in the lid of the Petri dish.
The sealed Petri dishes with a culture of C. albicans in
the bottom plate and a dermatophyte culture in the lid were
incubated at 25C, and the growth of the dermatophyte was
microscopically evaluated after 24 and 48 h. This was done
by inspection through the agar-free window in the bottom
plate without opening the Petri dish. The percentages of
conidia were determined that showed no, initial and full
germination. Control assays were run with Petri dishes in
which the agar in the bottom pan was inoculated with yeastfree 0.85% NaCl only and with Petri dishes in which a flat
bowl with activated carbon was placed in addition to the
fungal cultures to allow adsorption of volatile compounds.
Chemical analysis of volatile compounds released
by C. albicans
After 48 h incubation at room temperature, volatiles
generated by the C. albicans and dermatophyte in the dual
2013 ISHAM, Medical Mycology, Early Online: 110

Med Mycol Downloaded from informahealthcare.com by RMIT University on 09/08/13


For personal use only.

Acyclic sesquiterpenes inhibit growth of dermatophytes

cultures were trapped by solid-phase-micro-extraction


(SPME) [3]. The SPME fibers (polyacrylate, white,
Supelco/Sigma-Aldrich, Buchs, Germany) were inserted
into the plenum of the dishes through a small hole that had
been drilled at the side of the dishes and the fibers left in
place for 2 h to allow adsorption of the volatiles. Analysis
of volatiles was carried out by using a gas chromatograph
7890A (Agilent, Santa Clara, CA, USA) linked to an
Agilent 5975C inert XL MSD (Agilent, Santa Clara, CA,
USA). For conventional separation, a 60 m HP-5 fused
silica capillary was used with helium as the carrier gas at
a constant velocity of 35 cm/sec. Separation conditions
were 3 min at 50C, then programmed to 80C at a rate
of 3C/min, then to 150C at 5C/min, and finally to
300C at 7.5C/min. Tentative assignments of individual
compounds were based on their mass spectra reported in
the libraries Wiley 9th and NIST 2008, as well as on
published plotted spectra [4,5]. Unambiguous structure
elucidation was carried out by comparison of mass spectra and gas chromatographic retention indices of natural
products and authentic standards. Enantioselective GC/
MS was carried out by using a 25 m fused silica capillary
FS-Hydrodex--TBDM (Macherey & Nagel, Dren,
Germany), operated at 140C isothermal and a constant
flow of 1.5 ml/min.
Synthesis of 2,3-dihydrofarnesol enantiomers
The synthesis of optically active 6E-2,3-dihydrofarnesol
was carried out by enantioselective hydrogenation of
commercially available (E,E)-farnesol (Sigma-Aldrich,
Steinheim, Germany) using a Noyori-catalyst of the
bis-(diphenylphosphino)binaphthyl-type [6,7]. Solutions
of about 400 mg (E,E)-farnesol in 5 ml methanol were
hydrogenated at 170 bar during 2 h at room temperature.
Catalysts were chloro[(S)-(-)-2,2-bis(diphenylphosphino)1,1-binaphthyl](p-cymene)ruthenium(II) chloride (10% of
the educt) or the corresponding (R)-enantiomer.
Inhibition assays
Tests were carried out with (3R,6E)-2,3-dihydrofarnesol
(R-DHF) and its enantiomer, (3S,6E)-2,3-dihydrofarnesol
(S-DHF), synthesized as described above and with commercially available farnesol (F-ol; Sigma-Aldrich, Steinheim, Germany). Our F-ol sample consisted of ca. 99.2%
(2E, 6E)-farnesol, 0.7% (2E, 6Z)-plus (2Z, 6E)-farnesol
and 0.1% (2Z, 6Z)-farnesol. Test concentrations of the
applied substances ranged from 12.550 g/ml.
The dermatophyte species tested in these inhibition
assays were T. rubrum, T. mentagrophytes, M. canis and
E. floccosum. Three strains of each of these species that
were recovered from human tinea and maintained in our
2013 ISHAM, Medical Mycology, Early Online: 110

laboratory were used in the tests. All had been identified by


their typical morphology and physiological characteristics [2].
To allow rapid evaluation of dermatophyte growth, inhibition assays were started with germinated conidia. This
procedure has proved successful in investigations aimed to
quantify an interference with dermatophyte growth independent of medical applications [8]. Conidia were first
harvested from culture plates as described above for the
screening assays and germination induced by incubating
the purified conidia in Sabourauds glucose broth (SAB
B-D, bioMerieux, Lyon, France) supplemented with 40,000
units/l penicillin G and 40 mg/l chloramphenicol at 26C
until initial germination was detected microscopically. This
required 6 h for E. floccosum and M. canis and 12 h for
T. rubrum and T. mentagrophytes. After germination had
started, conidia were cleaned from the nutrient broth by
three cycles of centrifugation for 5 min at 2500 rpm (Megafuge 1.0, Heraeus, Hanau, Germany) and washing with
10 mM sodium phosphate buffer (Merck, Darmstadt, Germany). Washed germinating conidia were resuspended in
phosphate buffer, and their concentration was calculated
from counts in a hemocytometer (Neubauer improved
counting chamber). The suspension was appropriately
diluted with phosphate buffer until a final concentration of
105 conidia/ml phosphate buffer was confirmed by direct
counting. These suspensions were used as inocula for the
following inhibition assays.
Portions of the suspensions of germinating conidia
(50 l/well) were transferred to the wells of 96-well microtiter-plates (Sarstedt, Nmbrecht, Germany). Then either
50 l of R-DHF, S-DHF, or F-ol in 0.1% aqueous solution
of Tween 20 (Sigma Aldrich GmbH, Steinheim, Germany)
was added to the well. For control assays, 50 l of only
0.1% Tween, or only phosphate buffer or 200 g/ml fluconazole (Fragon Produtos para Indstria de Borracha,
Guarulhos, Spain) in phosphate buffer were used. Once the
wells were filled, the light extinctions (related to turbidity
caused by mycelial density) of the separate wells were
measured photometrically at a wavelength of 620 nm
(E620 nm, photometer Tecan sunrise, Tecan Group Ltd, Mnnedorf, Switzerland), and then checked microscopically to
exclude any contaminations. To complete the well filling,
100 l Sabourauds broth or 100 l phosphate buffer for
controls was added. In these wells three different concentrations of R-DHF, S-DHF and F-ol were tested (12.5,
25 and 50 g/ml), with the final fluconazole concentration
(50 g/ml) in the control wells. The completely filled
microtiter plates were then incubated at 20C, and the
extinctions of the wells were measured photometrically
after 20, 26, 32, 38, 44, 50 and 62 h. At the end of the
experiment the contents of the wells were inoculated onto
agar media to exclude bacterial contamination. All experiments were carried out in duplicate.

Med Mycol Downloaded from informahealthcare.com by RMIT University on 09/08/13


For personal use only.

4J

Brasch et al.

The resulting data were statistically evaluated according


to instructions and with computerized support supplied by
the Institute of Medical Informatics and Statistics of the
University Medical Center Schleswig Holstein, Campus
Kiel, using IBM SPSS statistics software (IBM Corp., New
York, USA). After a descriptive pilot analysis revealed that
within the four distinct species of dermatophytes the results
showed no relevant differences for the separate strains, data
were pooled for each species for further statistics. A bifactorial analysis of variance was done (factors: time and
tested substance (R-DHF, S-DHF, F-ol, n 24 and 26, and
fluconazole and nutrient broth, n 76 and 78). In addition,
the differences between the effects elicited by distinct
agents at certain points of time were assessed by unifactorial analyses of variance and paired comparisons (n 24
and 26, respectively). Due to the explorative approach
of our assays no adjustment of P-values was made. A
P value 0.5 was considered statistically significant.

Results
Screening for volatile compounds released by C. albicans
interfering with dermatophyte growth
In screening experiments with cultures of six distinct
strains of C. albicans in the bottom plate and conidia of
T. terrestre in the lid of a sealed Petri dish, 2030% of the
conidia showed initial or marked germination after 24 h.
In comparison, in control dishes in which the bottom plate
was not inoculated with C. albicans, about 90% of the
conidia developed initial or marked germination after
24 h. The percentages of germinating conidia were slightly
different for the distinct strains of C. albicans tested and
also for the other species of dermatophytes used (1 strain
each of T. rubrum, T. mentagrophytes and M. gypseum) but,
in general, all effects were similar for all fungi. Since these
screening experiments were only carried out to generally
check whether the exposure to C. albicans cultures had any
effect on the germination of dermatophyte conidia, no
attempts were made to quantify these effects or to exactly
compare them for the different fungal strains and species
used in this experimental investigation.
Instead, in order to determine a suitable Candida-agar
for further tests, additional pilot experiments were performed with T. terrestre on SGA in the lid of the Petri dish
and C. albicans was grown on various agars media in the
bottom plates. After 24 h, the percentages of conidia
without germination were uniformly in the range of 70%
when SGA, Wort-agar, peptone-agar or potato-glucose
agar was used for the growth of C. albicans in the bottom
of the Petri plates. This clearly indicates that there were no
major differences among the different media with regard
to their capacity to stimulate the release of inhibitory
factors by C. albicans.

Subsequent tests were carried out with combined cultures of C. albicans and dermatophytes, both on SGA, with
and without placing a flat open bowl containing activated
carbon into the sealed Petri dishes. These experiments
were carried out with T. terrestre to determine whether the
activated carbon could adsorb any growth-inhibitory factors from the air space in the Petri dish. After 24 h, the
percentage of ungerminated conidia was in the range of
70% in Petri dishes without activated carbon but was
reduced to approximately 5% in Petri dishes containing
activated carbon.
As a result of these pilot experiments we found that
C. albicans grown on SGA (and on other nutrients) releases
volatile compounds that inhibit growth of dermatophytes.
No such inhibition was detected in control cultures without
Candida and in culture systems with activated carbon
adsorbing chemicals from the air space of the cultures.
Therefore, we concluded that the inhibiting agents were
released by C. albicans into the atmosphere.
Because the test systems we applied for these screening
assays did not allow quantitative measurements, chemical
analysis of inhibitory factors was performed as a next
step. The objectives of the latter studies was to first, the
identification and synthesis of the inhibiting volatiles and
secondly, bioassays with pure, synthetic compounds.
Analysis of volatile factors released by C. albicans
As shown in Figure 1, eight volatile compounds were identified in the head space of C. albicans the peaks of which
are marked by numbers in Figure 1. Main components
were 6E-2,3-dihydrofarnesol (6) and (2E,6E)-farnesol (8),
whereas (E)--farnesene (3), (E)-nerolidol (5) and (2Z,6E)or (2E,6Z)-farnesol (7) were found to be minor components. In fact, peak 7 may represent two stereoisomers of
farnesol, but unambiguous assignment was impossible
because of overlapping at the backside of peak 6. Enantioselective gas chromatography, using the modified cyclodextrin hydrodex--TBDM as the stationary phase, proved
to be suitable for the separation of the 6E-2,3-dihydrofarneseol enantiomers. Under the applied conditions, the (S)enantiomer eluted after the (R)-enantiomer, the two
compounds being separated by a factor of rt S-DHF:
rt R-DHF 1.01. Comparison of retention times (GC/MS)
of the synthetic products with that of the natural volatile
revealed C. albicans to release stereochemically pure
(3R,6E)-2,3-dihydrofarnesol.
Synthesis of 2,3-dihydrofarnesol enantiomers
After having gained some experience, the hydrogenation
reaction proceeded well, furnishing chemically pure compounds in yields between 96 and 98%. Enantioselective
2013 ISHAM, Medical Mycology, Early Online: 110

Acyclic sesquiterpenes inhibit growth of dermatophytes

100

1
2
3
4
5
6
7
8

-Phenylethanol
n-Decanal
(E)--Farnesene
ar-Curcumene
(E)-Nerolidol
(E)-2,3-Dihydrofarnesol
(Z,E)- or (E,Z)-Farnesol
(E,E)-Farnesol

7
3
2

5
4

Med Mycol Downloaded from informahealthcare.com by RMIT University on 09/08/13


For personal use only.

0
14
Fig. 1

16

18

20

22

24

26

28

30

32

34

36

[min]

Gaschromatogram of volatile compounds released by Candida albicans.

GC/MS showed the products to be 9899% enantiomerically pure.


Inhibition assays with 2,3-dihydrofarnesol enantiomers
In the tested dermatophyte species, results were quite similar for the distinct strains, allowing pooled analyses for
each species. In summary, R-DHF, S-DHF and F-ol showed
significant and concentration-dependent inhibitory effects
on the growth of T. rubrum, T. mentagrophytes, M. canis
and E. floccosum. In general, it can be stated that for most
of the species and points of time S-DHF and F-ol had stronger inhibitory effects than R-DHF. Overall, E. floccosum
was the most susceptible, followed by T. mentagrophytes,
T. rubrum and M. canis in decreasing order. Table 1 shows
the results in detail for the tested dermatophyte species,
agents and concentrations. For example, growth of E. floccosum was at almost all points of time virtually completely
suppressed by 12.5 g/ml of both dihydrofarnesol
enantiomers, similar to the inhibition by 50 g/ml of
fluconazole, whereas M. canis was markedly less susceptible especially to R-DHF. However, even upon M. canis
12.5 g/ml of S-DHF had a stronger effect than 50 g/ml
of fluconazole within the total period of incubation.
Since T. rubrum is the most common dermatophyte in
Germany, it was selected to demonstrate inhibition obtained
with distinct compounds (Fig. 2). The effects of R-DHF,
S-DHF and F-ol, applied in the mid-level concentration (25
g/ml), and of 50 g/ml fluconazole are quantified by the
different extinctions at the distinct points of time. The
value of extinction increases with growing mycelial density and, therefore, allows assessing fungal growth. It is
obvious from Figure 2 that in particular, S-DHF, and F-ol
2013 ISHAM, Medical Mycology, Early Online: 110

had marked inhibitory effects on T. rubrum and that S-DHF


was more active than fluconazole. Similar dose-response
relations were found for the other dermatophyte species the
quantified results of which are presented in Tables 14.
All species were clearly inhibited by all agents as compared to their growth in pure nutrient broth (Tables 14).
T. rubrum was inhibited by S-DHF and F-ol more strongly
than by fluconazole, whereas R-DHF showed a lesser (but
still significant) effect. For T. mentagrophytes S-DHF was
the most active inhibitor next to fluconazole, whereas F-ol
and S-DHF were both more active on M. canis than fluconazole. E. floccosum was strongly inhibited by all agents,
with similar effects of R-DHF and S-DHF as compared to
fluconazole and a slightly lesser effect of F-ol. Because of
the different growth rates of the distinct dermatophyte species in pure nutrient broth, comparisons of their extinctions
at a certain point of time are not helpful to assess interspecies differences of susceptibilities to the various agents.
However, from the data presented above it follows that in
general S-DHF is a more active inhibitor than R-DHF and
that E. floccosum is the most susceptible species.

Discussion
Our data show that C. albicans, grown under appropriate
conditions, releases (2E,6E)-farnesol and (3R,6E)-2,3dihydrofarnesol, which act, in vitro, as strong growth
inhibitors on dermatophytes. To the best of our knowledge,
2,3-dihydrofarnesol has not been previously described as
a volatile produced by microorganisms and its inhibitory
effect on dermatophyte development is new to science.
Both of its enantiomers proved to be biologically active,
and the non-natural (3S,6E)-2,3-dihydrofarnesol (S-DHF)

6J

Brasch et al.

Table 1 Arithmetic means of extinction with confidence intervals measured for Trichophyton rubrum in pure nutrient broth and with addition of R-DHF,
S-DHF and F-ol (each substance in concentrations of 12.5, 25 and 50 g/ml) and fluconazole (50 g/ml) at 062 h in comparison to pure nutrient broth.
n 2426 for S-DHF, R-DHF and F-ol, n 7478 for fluconazole and pure nutrient broth. **indicates P 0.01 for differences between extinction in
broth containing agent and pure broth (control).
Time [h]

Substance
Pure nutrient broth
R-DHF (12.5 g/ml)
R-DHF (25 g/ml)
R-DHF (50 g/ml)**

Med Mycol Downloaded from informahealthcare.com by RMIT University on 09/08/13


For personal use only.

S-DHF (12.5 g/ml)**


S-DHF (25 g/ml)**
S-DHF (50 g/ml)**
F-ol (12.5 g/ml)**
F-ol (25 g/ml)**
F-ol (50 g/ml)**
Fluconazole (50 g/ml)**

20

26

32

38

44

50

62

0.054
0.0530.054
0.054
0.0520.055
0.053
0.0520.054
0.053
0.0520.055
0.054
0.0530.055
0.054
0.0530.056
0.056
0.0550.057
0.055
0.0530.056
0.055
0.0540.056
0.057
0.0550.058
0.055
0.0550.056

0.068
0.0670.070
0.07
0.0670.073
0.066
0.0640.069
0.058
0.0550.061
0.055
0.0520.057
0.054
0.0510.056
0.055
0.0520.057
0.054
0.0520.057
0.054
0.0510.057
0.055
0.0520.058
0.058
0.0560.059

0.084
0.0810.087
0.088
0.0820.093
0.08
0.0750.085
0.062
0.0580.067
0.056
0.0510.060
0.054
0.0500.058
0.055
0.0500.059
0.054
0.0500.058
0.054
0.0490.058
0.055
0.0500.059
0.059
0.0560.062

0.108
0.1030.112
0.114
0.1070.122
0.104
0.0960.111
0.071
0.0640.077
0.057
0.0510.064
0.054
0.0480.061
0.055
0.0480.061
0..054
0.0470.060
0.054
0.0470.060
0.055
0.0490.061
0.061
0.0570.066

0.132
0.1270.137
0.145
0.1360.154
0.13
0.1220.139
0.085
0.0770.093
0.06
0.0520.067
0.055
0.0470.062
0.055
0.0470.062
0.054
0.0460.061
0.054
0.0460.061
0.055
0.0470.062
0.065
0.0600.070

0.151
0.1450.157
0.172
0.1610.182
0.154
0.1440.164
0.103
0.0920.113
0.064
0.0560.073
0.056
0.0470.064
0.055
0.0460.063
0.053
0.0450.062
0.053
0.0450.062
0.055
0.0460.063
0.07
0.0640.076

0.168
0.1610.175
0.194
0.1820.206
0.174
0.1620.186
0.116
0.1040.129
0.071
0.0610.082
0.057
0.0470.067
0.055
0.0440.065
0.053
0.0430.064
0.053
0.0430.064
0.054
0.0440.064
0.077
0.0700.084

0.201
0.1910.211
0.236
0.2190.253
0.211
0.1950.228
0.14
0.1220.158
0.097
0.0810.113
0.063
0.0480.078
0.055
0.0400.070
0.056
0.0410.072
0.053
0.0380.068
0.054
0.0390.069
0.1
0.0900.110

was found particularly inhibiting in some of our bioassays.


However, we did not detect this stereoisomer among the
volatiles of C. albicans.
Farnesol is one of the most common sesquiterpenes in
nature. This alcohol is also produced by C. albicans and
has a diversity of effects on this fungus [915]. It acts as
a quorum-sensing compound playing a role in the intercellular communication of this yeast [9,11,14,15]. Furthermore, farnesol per se has antimicrobial effects [1622] and
can reduce the minimum inhibitory concentrations of some
antifungals [23,24].
2,3-Dihydrofarnesol is relatively widespread in plants,
especially as a volatile generated by flowers [2527]. It has
also been identified in the secretion of the temporal gland
of African elephants [28], in ants [29,30], in stingless social
bees [31], and in several bumble bee species [4,3234].
While 2,3-dihydrofarnesol used as marking pheromone
of male bumble bees constantly shows (3S)-configuration
[5,34] enantiomeric compositions of 2,3-dihydrofarnesol
from other sources have not been determined.
Little is known about effects of 2,3-dihydrofarnesol on
microorganisms. A patent kept by Takasago Intern. Corp.
describes some antimicrobial properties of the (3S,6E)enantiomer, but experimental details and results were not
provided [35]. While we could not detect this stereoisomer

in C. albicans, its inhibiting activities against dermatophytes are not too surprising as there are other examples
showing that the enantiomer of a certain natural product
may show similar biological activities. Cases are known
that among four possible stereoisomers the naturally occurring compound is the least active one [36,37].
In the chemical ecology of C. albicans farnesol acts as
an intraspecific semiochemical in the context of quorum
sensing (see above). Similar to farnesol its 2,3-dihydroderivative may as well play such a role. In addition,
both compounds could be associated with defense and
struggle for space during competition with dermatophytes
in tinea lesions colonized by C. albicans. However, we
have no data yet to substantiate the relevance of such activities in vivo. The question whether the development of
tinea lesions can be influenced by the presence of C. albicans is exciting but cannot be answered based on solid data
currently available. Since dermatophytes and C. albicans
can occur within the same skin areas, interference may
happen. Our findings now give new reason to turn more
attention to this clinically relevant aspect. The search for
mutual interfungal communication based on pure acyclic
sesquiterpenes and on mixtures, possibly accounting for
synergistic effects, will be an exciting field for further
investigations.
2013 ISHAM, Medical Mycology, Early Online: 110

Acyclic sesquiterpenes inhibit growth of dermatophytes

Effect of R-DHF on Trichophyton rubrum

(a)

(b)

.210

.210

.190

R-DHF (25 g/ml), n=26

E620 nm

E620 nm

.150

.150
.130

.130
.110

.110

.090

.090
.070

.070

.050

.050

.030

S-DHF (25 g/ml), n=26

.170

.170

.030

5 10 15 20 25 30 35 40 45 50 55 60

Time [h]

(c)

Effect of F-ol on Trichophyton rubrum

Controls; growth of Trichophyton rubrum in pure


nutrient broth and with fluconazole.

(d)
.230
.210

.210
F-ol (25 g/ml), n=26

.190

.150

.150

E620 nm

.170

.130

.130

.110

.110

.090

.090

.070

.070

.050

.050
0

5 10 15 20 25 30 35 40 45 50 55 60
Time [h]

Pure nutrient broth, n=78


Fluconazole (50 g/ml), n=78

.190

.170

.030

5 10 15 20 25 30 35 40 45 50 55 60
Time [h]

.230

E620 nm

Med Mycol Downloaded from informahealthcare.com by RMIT University on 09/08/13


For personal use only.

Effect of S-DHF on Trichophyton rubrum


.230

.230

.190

.030

5 10 15 20 25 30 35 40 45 50 55 60
Time [h]

Fig. 2 Effects of 25 g/ml of R-DHF, S-DHF, and F-ol in comparison with controls (pure nutrient broth and 50 g/ml fluconazole) on the growth
of Trichophyton rubrum as expressed by increasing extinction E620 nm (mean values with confidence intervals) within 62 h. (a) Effect of R-DHF on
T. rubrum; R-DHF (25 g/ml), n 26. (b) Effect of S-DHF on T. rubrum; S-DHF (25 g/ml), n 26. (c) Effect of F-ol on Trichophyton
rubrum; F-ol (25 g/ml), n 26; (d) Controls; growth of T. rubrum in pure nutrient broth and with fluconazole. Pure nutrient broth, n 78;
Fluconazole (50 g/ml), n 78.

2013 ISHAM, Medical Mycology, Early Online: 110

8J

Brasch et al.

Table 2 Arithmetic means of extinction with confidence intervals measured for Trichophyton mentagrophytes in pure nutrient broth and with addition
of R-DHF, S-DHF and F-ol (each substance in concentrations of 12.5, 25 and 50 g/ml) and fluconazole (50 g/ml) at 062 h in comparison to pure
nutrient broth. n 2426 for S-DHF, R-DHF and F-ol, n 7478 for fluconazole and pure nutrient broth. **indicates P 0.01 for differences between
extinction in broth containing agent and pure broth (control).
Time [h]

Substance
Pure nutrient broth
R-DHF (12.5 g/ml)
R-DHF (25 g/ml)
R-DHF (50 g/ml)

Med Mycol Downloaded from informahealthcare.com by RMIT University on 09/08/13


For personal use only.

S-DHF (12.5 g/ml)**


S-DHF (25 g/ml)**
S-DHF (50 g/ml)
F-ol (12.5 g/ml)**
F-ol (25 g/ml)**
F-ol (50 g/ml)
Fluconazole (50 g/ml)**

20

26

32

38

44

50

62

0.05
0.0490.050
0.049
0.0480.050
0.049
0.0480.050
0.049
0.0480.050
0.05
0.0490.051
0.05
0.0500.051
0.051
0.0500.052
0.05
0.0500.051
0.051
0.0500.052
0.053
0.0520.054
0.052
0.0510.052

0.05
0.0490.051
0.05
0.0480.051
0.049
0.0470.050
0.049
0.0480.050
0.05
0.0490.051
0.05
0.0490.051
0.05
0.0490.052
0.05
0.0490.051
0.05
0.0490.051
0.052
0.0510.053
0.05
0.0490.051

0.052
0.0500.054
0.05
0.0470.054
0.049
0.0450.052
0.051
0.0470.055
0.05
0.0460.053
0.05
0.0460.053
0.05
0.0470.054
0.05
0.0460.054
0.05
0.0460.054
0.052
0.0490.056
0.05
0.0480.052

0.056
0.0520.059
0.054
0.0470.060
0.049
0.0430.056
0.052
0.0450.059
0.05
0.0440.056
0.05
0.0440.056
0.053
0.0460.059
0.05
0.0440.056
0.05
0.0440.056
0.054
0.0480.061
0.05
0.0460.054

0.06
0.0540.066
0.058
0.0480.068
0.051
0.0410.060
0.057
0.0460.067
0.05
0.0410.059
0.05
0.0410.059
0.056
0.0460.066
0.05
0.0410.059
0.05
0.0410.059
0.058
0.0480.068
0.05
0.0440.056

0.068
0.0600.075
0.063
0.0500.076
0.053
0.0410.065
0.06
0.0450.075
0.05
0.0380.063
0.05
0.0380.062
0.057
0.0440.071
0.05
0.0370.062
0.05
0.0370.062
0.061
0.0470.075
0.05
0.0420.058

0.079
0.0700.088
0.073
0.0570.089
0.059
0.0440.074
0.064
0.0460.082
0.051
0.0360.066
0.05
0.0350.065
0.059
0.0420.075
0.05
0.0350.065
0.05
0.0350.065
0.064
0.0460.083
0.051
0.0420.060

0.127
0.1150.138
0.123
0.1030.143
0.099
0.0800.118
0.076
0.0520.100
0.054
0.0350.073
0.05
0.0310.069
0.06
0.0400.081
0.05
0.0310.069
0.05
0.0310.069
0.07
0.0460.094
0.055
0.0440.067

Table 3 Arithmetic means of extinction with confidence intervals measured for Microsporum canis in pure nutrient broth and with addition of R-DHF,
S-DHF and F-ol (each substance in concentrations of 12.5, 25 and 50 g/ml) and fluconazole (50 g/ml) at 062 h in comparison to pure nutrient broth.
n 2426 for S-DHF, R-DHF and F-ol, n 7478 for fluconazole and pure nutrient broth. **indicates P 0.01 for differences between extinction in
broth containing agent and pure broth (control).
Time [h]

Substance
Pure nutrient broth
R-DHF (12.5 g/ml)
R-DHF (25 g/ml)
R-DHF (50 g/ml)**
S-DHF (12.5 g/ml)**
S-DHF (25 g/ml)**
S-DHF (50 g/ml)**
F-ol (12.5 g/ml)**
F-ol (25 g/ml)**
F-ol (50 g/ml)**
Fluconazole (50 g/ml)**

20

26

32

38

44

50

62

0.06
0.0580.061
0.058
0.0560.060
0.058
0.0550.060
0.057
0.0550.060
0.06
0.0580.062
0.061
0.0580.063
0.062
0.0600.064
0.059
0.0570.061
0.06
0.0580.063
0.064
0.0620.067
0.061
0.0600.062

0.085
0.0800.089
0.082
0.0740.090
0.076
0.0680.084
0.064
0.0560.071
0.063
0.0560.070
0.061
0.0540.068
0.061
0.0540.069
0.058
0.0510.066
0.059
0.0510.066
0.062
0.0540.069
0.068
0.0640.073

0.106
0.1000.112
0.106
0.0950.116
0.096
0.0860.106
0.073
0.0630.083
0.069
0.0590.078
0.062
0.0530.071
0.062
0.0530.071
0.058
0.0490.068
0.058
0.0480.067
0.061
0.0520.071
0.076
0.0700.082

0.13
0.1230.137
0.132
0.1200.143
0.12
0.1080.132
0.089
0.0780.101
0.079
0.0680.090
0.065
0.0540.076
0.066
0.0550.076
0.058
0.0470.070
0.058
0.0460.069
0.06
0.0490.072
0.09
0.0840.097

0.151
0.1430.159
0.156
0.1420.170
0.145
0.1300.159
0.11
0.0950.124
0.092
0.0780.105
0.072
0.0590.085
0.07
0.0570.084
0.059
0.0460.072
0.057
0.0440.071
0.06
0.0470.074
0.109
0.1010.117

0.172
0.1630.181
0.185
0.1680.201
0.172
0.1560.189
0.135
0.1180.151
0.109
0.0930.125
0.083
0.0670.098
0.077
0.0610.093
0.061
0.0450.077
0.058
0.0420.074
0.061
0.0450.076
0.132
0.1230.142

0.189
0.1780.199
0.207
0.1890.226
0.196
0.1770.214
0.156
0.1380.175
0.124
0.1060.142
0.093
0.0750.110
0.084
0.0660.102
0.064
0.0460.081
0.058
0.0400.075
0.06
0.0420.077
0.155
0.1440.165

0.224
0.2120.236
0.26
0.2390.281
0.253
0.2310.274
0.203
0.1820.225
0.161
0.1410.182
0.121
0.1000.141
0.102
0.0810.123
0.078
0.0570.099
0.057
0.0370.078
0.059
0.0390.080
0.21
0.1980.222

2013 ISHAM, Medical Mycology, Early Online: 110

Acyclic sesquiterpenes inhibit growth of dermatophytes

Table 4 Arithmetic means of extinction with confidence intervals measured for Epidermophyton floccosum in pure nutrient broth and with addition
of R-DHF, S-DHF and F-ol (each substance in concentrations of 12.5, 25 and 50 g/ml) and fluconazole (50 g/ml) at 062 h in comparison to pure
nutrient broth. n 2426 for S-DHF, R-DHF and F-ol, n 7478 for fluconazole and pure nutrient broth. **indicates P 0.01 for differences between
extinction in broth containing agent and pure broth (control).
Time [h]

Substance
Pure nutrient broth
R-DHF (12.5 g/ml)**
R-DHF (25 g/ml)**
R-DHF (50 g/ml)**

Med Mycol Downloaded from informahealthcare.com by RMIT University on 09/08/13


For personal use only.

S-DHF (12.5 g/ml)**


S-DHF (25 g/ml)**
S-DHF (50 g/ml)**
F-ol (12.5 g/ml)**
F-ol (25 g/ml)**
F-ol (50 g/ml)**
Fluconazole (50 g/ml)**

20

26

32

38

44

50

62

0.055
0.0540.055
0.055
0.0540.056
0.054
0.0530.055
0.055
0.0540.056
0.056
0.0550.057
0.057
0.0560.057
0.06
0.0590.061
0.056
0.0550.056
0.06
0.0590.060
0.069
0.0680.070
0.056
0.0560.057

0.057
0.0560.057
0.055
0.0530.056
0.054
0.0520.055
0.056
0.0540.057
0.055
0.0540.056
0.056
0.0550.057
0.058
0.0570.059
0.055
0.0530.056
0.058
0.0570.059
0.065
0.0630.066
0.056
0.0550.056

0.058
0.0580.059
0.055
0.0530.056
0.053
0.0510.054
0.054
0.0530.055
0.055
0.0530.056
0.055
0.0540.056
0.058
0.0560.059
0.054
0.0530.055
0.057
0.0560.059
0.063
0.0620.065
0.056
0.0550.057

0.062
0.0610.063
0.055
0.0530.057
0.053
0.0510.054
0.054
0.0520.055
0.055
0.0530.056
0.055
0.0530.057
0.058
0.0560.059
0.054
0.0520.056
0.057
0.0550.059
0.062
0.0600.064
0.057
0.0550.058

0.069
0.0670.070
0.056
0.0530.059
0.053
0.0500.055
0.053
0.0510.056
0.054
0.0510.057
0.054
0.0510.057
0.057
0.0540.060
0.053
0.0500.056
0.057
0.0540.059
0.062
0.0590.065
0.057
0.0550.058

0.08
0.0770.082
0.057
0.0530.062
0.052
0.0480.057
0.053
0.0490.057
0.054
0.0500.059
0.054
0.0500.058
0.057
0.0520.061
0.053
0.0490.058
0.056
0.0520.061
0.061
0.0570.066
0.057
0.0540.059

0.097
0.0940.101
0.059
0.0530.066
0.053
0.0460.059
0.053
0.0460.059
0.055
0.0480.061
0.054
0.0470.060
0.057
0.0500.063
0.054
0.0470.060
0.056
0.0500.063
0.061
0.0550.068
0.056
0.0530.060

0.134
0.1280.139
0.069
0.0600.079
0.054
0.0450.063
0.054
0.0450.063
0.055
0.0460.065
0.056
0.0470.065
0.056
0.0470.066
0.055
0.0460.064
0.056
0.0470.066
0.063
0.0540.072
0.056
0.0500.061

Acknowledgements
We like to thank Mrs A. Preschke and Mrs K. Voss for
technical assistance, Dr J. Hedderich (Institute for Medical
Informatics and Statistics, University of Kiel) for his professional advising on statistical matters and for statistical
analyses and Mrs K. Houghton for correcting language
errors.
Declaration of interest: The authors report no conflicts of
interest. The authors alone are responsible for the content
and the writing of the paper.

References
1 Schaufuss P, Brasch J, Steller U. Dermatophytes can trigger cooperative (CAMP-like) haemolytic reactions. Brit J Dermatol 2005; 153:
584590.
2 de Hoog GS, Guarro J, Gen J, Figueras MJ. Atlas of Clinical Fungi,
2nd edn. Utrecht, The Netherlands: Centraalbureau voor Schimmelcultures, and Reus, Spain: Universitat Rovira i Virgili, 2000.
3 Pawliszyn J. Solid Phase Microextraction: Theory and Practice, 1997:
New York: Wiley.
4 Bergstrm G, Kullenberg B, Stllberg-Stenhagen S, et al. Studies on
natural compounds II. Identification of a 2,3-dihydrofarnesol as the
main component of the marking perfume of male bumble bees of the
species Bombus terrestris L. Arkiv fr Kemi 1968; 28: 453469.

2013 ISHAM, Medical Mycology, Early Online: 110

5 Ahlquist L, Stllberg-Stenhagen S. Studies on natural odoriferous


compounds III. Synthesis of ()- and ()-3,7,11-trimethyl-6-trans10-dodecadien-1-ol. Acta Chem Scand 1971; 25: 16851694.
6 Takaya H, Ohta T, Sayo N, et al. Enantioselective hydrogenation of
allylic and homoallylic alcohols. J Amer Chem Soc 1987; 109:
15961597.
7 Mashima K, Kusano K, Ohta T, et al. Synthesis of new cationic
BINAP-ruthenium(II)complexes and their use in asymmetric
hydrogenation[BINAP 2,2 -bis(diphenylphosphino)-1,1 binaphthyl]. J Chem Soc Chem Commun 1989; 15: 12081210.
8 Fritz P, Beck-Jendroschek V, Brasch J. Inhibition of dermatophytes
by the antimicrobial peptides human -defensin-2, ribonuclease 7
and psoriasin. Med Mycol 2012; 50: 579584.
9 Ramage G, Saville SP, Wickes BL, et al. Inhibition of Candida albicans biofilm formation by farnesol, a quorum-sensing molecule. Appl
Environ Microbiol 2002; 68: 54595463.
10 Davis-Hanna A, Piispanen AE, Stateva LI, et al. Farnesol and dodecanol effects on the Candida albicans Ras1-cAMP signalling pathway
and the regulation of morphogenesis. Mol Microbiol 2008; 67: 4762.
11 De Sordi L, Mhlschlegel FA. Quorum sensing and fungal-bacterial
interactions in Candida albicans: a communicative network regulating microbial coexistence and virulence. FEMS Yeast Res 2009; 9:
990999.
12 Cugini C, Morales DK, Hogan DA. Candida albicans-produced
farnesol stimulates Pseudomonas quinolone signal production in
LasR-defective Pseudomonas aeruginosa strains. Microbiology 2010;
156: 30963107.
13 Lindsay AK, Deveau A, Piispanen AE, et al. Farnesol and cyclic AMP
signaling effects on the hypha-to-yeast transition in Candida albicans.
Eukaryot Cell 2012; 11: 12191225.

Med Mycol Downloaded from informahealthcare.com by RMIT University on 09/08/13


For personal use only.

10J

Brasch et al.

14 Albuquerque P, Casadevall A. Quorum sensing in fungi a review.


Med Mycol 2012; 50: 337345.
15 Weber K, Schulz B, Ruhnke M. The quorum-sensing molecule E,
E-farnesol its variable secretion and its impact on the growth and
metabolism of Candida species. Yeast 2010; 27: 727739.
16 Akiyama H, Oono T, Huh WK, et al. Actions of farnesol and xylitol
against Staphylococcus aureus. Chemotherapy 2002; 48: 122128.
17 Inoue Y, Shiraishi A, Hada T, et al. The antibacterial effects of terpene
alcohols on Staphylococcus aureus and their mode of action. FEMS
Microbiol Lett 2004; 237: 325331.
18 Semighini CP, Hornby JM, Dumitru R, et al. Farnesol-induced apoptosis in Aspergillus nidulans reveals a possible mechanism for antagonistic interactions between fungi. Mol Microbiol 2006; 59: 753764.
19 Jabra-Rizk MA, Meiller TF, James CE, et al. Effect of farnesol on
Staphylococcus aureus biofilm formation and antimicrobial susceptibility. Antimicrob Agents Chemother 2006; 50: 14631469.
20 Weber K, Sohr R, Schulz B, Fleischhacker M, et al. Secretion of E,
E-farnesol and biofilm formation in eight different Candida species.
Antimicrob Agents Chemother 2008; 52: 18591861.
21 Derengowski LS, De-Souza-Silva C, Braz SV, et al. Antimicrobial
effect of farnesol, a Candida albicans quorum sensing molecule, on
Paracoccidioides brasiliensis growth and morphogenesis. Ann Clin
Microbiol Antimicrob 2009; 8: 13.
22 Dinamarco TM, Goldman MH, Goldman GH. Farnesol-induced cell
death in the filamentous fungus Aspergillus nidulans. Biochem Soc
Trans 2011; 39:15441548.
23 Sharma M, Prasad R. The quorum-sensing molecule farnesol is a
modulator of drug efflux mediated by ABC multidrug transporters
and synergizes with drugs in Candida albicans. Antimicrob Agents
Chemother 2011; 55: 48344843.
24 Cordeiro RA, Teixeira CE, Brilhante RS, et al. Minimum inhibitory
concentrations of amphotericin B, azoles and caspofungin against
Candida species are reduced by farnesol. Med Mycol 2013; 51:
5359.

25 Reisch J, Bergenthal D, Adesina SK, et al. Constituents of Fagara


macrophylla and Zanthoxylum rigidifolcum pericarps. J Nat Prod
1986; 49: 11691171.
26 Knudsen JT, Eriksson R, Gershenzon J, et al. Diversity and distribution of floral scent. The Bot Rev 2006; 72: 120.
27 El-Sayed AM. 2012. The Pherobase: database of insect pheromones
and semiochemicals. http://www.pherobase.com
28 Goodwin TE, Rasmussen EL, Guinn AC, et al. African elephant sesquiterpenes. J Nat Prod 1999; 62: 15701572.
29 Bagnres AG, Billen J, Morgan ED. Volatile secretion of Dufour gland
of workers of an army ant, Dorylus (annoma) molestus. J Chem Ecol
1991; 17: 16331639.
30 Keegans SJ, Billen J, Morgan ED, et al. Volatile glandular secretions
of three species of new world army ants, Ecitn burchelli, Labidus
coecus, and Labidus praedator. J Chem Ecol 1993; 19: 27052719.
31 Francke W, Lbke G, Schrder W, et al. Identification of oxygen containing volatile cephalic secretions of workers of Brazilian stingless
bees. J Braz Chem Soc 2000; 11: 562571.
32 Bergstrm G, Bergmann P, Schmidt JO. Labial gland chemistry of
three species of bumble bees (Hymenoptera: Apidae) from North
America. Bioorg Med Chem 1996; 4: 515519.
33 Urbanov K, Halk J, Hovorka O, et al. Marking pheromones of
the cuckoo bumble bee males (Hymenoptera, Apoidea, Bombus Latreille): composition of labial gland secretions of six species found in
the Czech Republic. Biochem System Ecol 2004; 32: 10251045.
34 Luxov A, Urbanov K, Valterov I, et al. Absolute configuration of
chiral terpenes in marking pheromones of bumble bees and cuckoo
bumble bees. Chirality 2004; 16: 228233.
35 Harada M, Matsuda H, Yamamoto T. Takasago Inter. Corp. Tokyo
17.10.2002. Document Identification DE 6962 1673T2 EP Nr. 073 1160.
36 Mori K. Significance of chirality in pheromone science. Bioorg Med
Chem 2007; 15: 75057523.
37 Mori K. Bioactive natural products and chirality. Chirality 2011; 23:
449462.

This paper was first published online on Early Online on 23 July 2013.

2013 ISHAM, Medical Mycology, Early Online: 110

Você também pode gostar