Antifungal activity of 2α,3β-functionalized steroids stereoselectively increases with the addition of oligosaccharides

Article abstract of DOI:10.1016/j.bmcl.2011.10.015

Invasive fungal infections pose a significant problem to the immune-compromised. Moreover, increased resistance to common antifungals requires development of novel compounds that can be used to treat invasive fungal infections. Naturally occurring steroid

Full text of DOI:10.1016/j.bmcl.2011.10.015

Bioorganic & Medicinal Chemistry Letters 21 (2011) 7379–7386
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Antifungal activity of 2a,3b-functionalized steroids stereoselectively
increases with the addition of oligosaccharides
Amy Cammarata a, , Sunil Kumar Upadhyay b, , Branko S. Jursic b,c, Donna M. Neumann a,d,e,
⇑
a Department of Pharmacology and Experimental Therapeutics, LSUHSC, 1901 Perdido St. New Orleans, LA 70112, USA
b Department of Chemistry, University of New Orleans, New Orleans, LA 70148, USA
c Stepharm, P O Box 24220, LA 70184, USA
d LSUHSC Eye Center of Excellence, New Orleans, LA, USA
e Department of Genetics, LSUHSC, New Orleans, LA, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Invasive fungal infections pose a significant problem to the immune-compromised. Moreover, increased
resistance to common antifungals requires development of novel compounds that can be used to treat
invasive fungal infections. Naturally occurring steroidal glycosides have been shown to possess a range
of functional antimicrobial properties, but synthetic methodology for their development hinders thor-
ough exploration of this class of molecules and the structural components required for broad spectrum
antifungal activity. In this report, we outline a novel approach to the synthesis of glycoside-linked func-
tionalized 2a,3b-cholestane and spirostane molecules and present data from in vitro screenings of the
antifungal activities against human fungal pathogens and as well as mammalian cell toxicity of these
derivatives.
Received 6 September 2011
Revised 30 September 2011
Accepted 4 October 2011
Available online 12 October 2011
Keywords:
Spirostanes
Cholestanes
Antifungal
Glycosides
Ó 2011 Elsevier Ltd. All rights reserved.
The occurrences of life-threatening fungal infections have con-
tinued to rise over the last two decades, particularly within popu-
lations of immune compromised individuals.1,2 Further, some
clinical therapeutics used in the treatment of invasive fungal
infections can be cytotoxic and a growing number of organisms
are exhibiting increased drug resistance to common antifun-
gals.3–5 Finally, cross-continental travel, combined with both the
increase in immune compromised populations and increased drug
resistance, essentially ensures that opportunistic fungal pathogens
will continue to present significant treatment challenges.3,5–7
Considering these concerns, the design and synthesis of novel anti-
fungal agents remains an area of intense significance.
The exploration of the biological properties of natural products
has successfully elucidated classes of molecules that have potent
antimicrobial activities.8–11 One such class of natural products that
have shown considerable promise as antimicrobials is the sapo-
nin.2 These natural products are typically isolated from plant or
marine species,9 and are generally characterized as steroidal glyco-
sides. Furthermore, there have been reports that indicate that sap-
onins with functionalized 2,3-spirostane moieties linked to an
oligosaccharide exhibit broad spectrum antifungal activity,
through yet to be defined mechanisms.12–14 However, the major
obstacle to the systematic study of 2,3-spirostane glycosides as
potential antifungals is the availability of synthetic preparations
of these target compounds.
In addition to the aforementioned class of saponins, there have
also been reports showing that synthetic modifications to choles-
terol have also yielded a number of promising antimicrobial
compounds. Among these are 6,5-fused cholestane oxazoles, oxy-
genated cholesterol derivatives, cholesterol-bound fatty acid deriv-
atives and cholesterol-hydrazone derivatives.8,15,16 However,
oligosaccharide derivatives of functionalized 2,3-cholestanes have
not yet been explored with respect to their potential for antifungal
activity.
In keeping pace with these findings, we have begun to explore
the possibility that further synthetic modifications to functional-
ized spirostane and cholestane molecules might bring forth insight
into the development of a novel class of structurally diverse com-
pounds that have antifungal activity. Recently, we published our
approaches to the synthesis of functionalized 2,3-cholestane and
spirostane molecules and the in vitro evaluation of their antifungal
activities against pathogenic fungi.17,18 Most of these molecules
exhibited little or no capacity for antifungal activity. We then
hypothesized that a functionalized steroid moiety may be a key
contributor to antifungal activity, but interactions between the ste-
roid and fungal cell wall or membrane may require the presence of
an oligosaccharide, as seen in previously reported active saponins.
In this report, we use our previously synthesized functionalized
⇑
Corresponding author.
E-mail address: dneum1@lsuhsc.edu (D.M. Neumann).
2a,3b-steroids as building blocks to demonstrate that the addition
of sugar residues in the 3b-position of functionalized steroids
Both authors contributed equally to this Letter.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.10.015

7380
A. Cammarata et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7379–7386
Scheme 1. Condensed reaction pathway for the preparation of compounds 11–14.
Table 1
3b-substituted cholestane derivatives and the MIC50 for three yeast species
a,e
Structure
MIC50
(lg/mL)
Candida albicansb
Candida glabratac
Cryptococcus neoformansd
β
HO
NA
NA
NA
H
11a
OH
O
α
β
O
NA
NA
NA
NA
NA
NA
OH
HO
HO
OH
H
11b
OH
β
O
O
β
OH
H
HO
11c

A. Cammarata et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7379–7386
7381
Table 1 (continued)
a,e
Structure
MIC50
(lg/mL)
Candida albicansb
Candida glabratac
Cryptococcus neoformansd
OH
O
α
β
O
α
OH
O
OH
NA
NA
NA
O
H
OH
11d
OH
OH
OH
β
O
O
β
O
O
OH
NA
NA
4
4
H
HO
OH
OH
11e
HO
OH
β
O
O
β
HO
β
>125
>125
HO
HO
O
HO
OH
H
O
11f
OH
⁄⁄Previously reported in Ref. 17.
NA denotes compounds that had (1) a slight reduction in turbidity to no change and; (2) had less than a 50% reduction in growth compared to controls, as measured
spectroscopically by absorption at 530 nm.
a
MIC values are reported only for compounds displaying (1) a prominent decrease in turbidity by visual comparison to the control wells containing no antifungal and; (2) a
>50% reduction in fungal growth compared to controls containing no antifungal, as measured spectroscopically by absorption at 530 nm.
b
ATCC no. 10231.
ATCC no. 48435.
ATCC no. 36556.
All values were determined after incubation at 35 °C for 48 h.
c
d
e
Table 2
2
a,3b-substituted cholestane derivatives and the MIC50 for three yeast species
a,e
Structure
MIC50
(lg/mL)
Candida albicansb
Candida glabratac
Cryptococcus neoformansd
HO
α
β
HO
NA
NA
32 (25% inhibition)
H
12a
OH
HO
α
O
α
β
O
NA
NA
NA
NA
NA
NA
OH
HO
HO
OH
H
12b
HO
α
OH
β
O
O
β
OH
H
HO
12c
OH
HO
O
α
α
β
O
α
OH
O
OH
NA
NA
NA
O
H
OH
12d
OH
OH
(continued on next page)

7382
A. Cammarata et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7379–7386
Table 2 (continued)
a,e
Structure
MIC50
(lg/mL)
Candida albicansb
Candida glabratac
Cryptococcus neoformansd
HO
α
OH
β
O
O
β
O
HO
OH
O
OH
NA
NA
32
H
12e
OH
HO
HO
α
OH
β
O
O
β
HO
β
NA
8
4
HO
HO
O
HO
OH
H
O
OH
⁄⁄Previously reported in Ref. 17.
12f
NA denotes compounds that had (1) a slight reduction in turbidity to no change and; (2) had less than a 50% reduction in growth compared to controls, as measured
spectroscopically by absorption at 530 nm.
a
MIC values are reported only for compounds displaying (1) a prominent decrease in turbidity by visual comparison to the control wells containing no antifungal and; (2) a
>50% reduction in fungal growth compared to controls containing no antifungal, as measured spectroscopically by absorption at 530 nm.
b
ATCC no. 10231.
ATCC no. 48435.
ATCC no. 36556.
All values were determined after incubation at 35 °C for 48 h.
c
d
e
Table 3
3b-substituted spirostane derivatives and the MIC50 for three yeast species
a,e
Structure
MIC50
(lg/mL)
b
Candida albicans
Candida glabratac
Cryptococcus neoformansd
H
H
β
O
H
HO
NA
>62
NA
O
H
H
13a
OH
O
α
H
H
β
O
H
H
O
NA
NA
NA
OH
HO
OH
O
H
H
13b
OH
H
β
O
H
O
O
O
β
8
8
8
HO
HO
OH
O
H
H
13c
OH
O
α
H
H
β
H
O
α
OH
O
OH
NA
NA
NA
O
O
H
OH
H
13d
OH
OH
H
OH
H
β
O
H
O
O
β
O
HO
OH
O
OH
O
16 (25% inhibition)
32
1
H
H
13e
OH
HO

A. Cammarata et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7379–7386
7383
Table 3 (continued)
a,e
Structure
MIC50
(lg/mL)
b
Candida albicans
Candida glabratac
Cryptococcus neoformansd
H
H
OH
β
O
H
O
O
β
HO
HO
β
NA
16
125
O
HO
OH
O
H
HO
O
H
13f
OH
⁄⁄Previously reported in Ref. 18
NA denotes compounds that had (1) a slight reduction in turbidity to no change and; (2) had less than a 50% reduction in growth compared to controls, as measured
spectroscopically by absorption at 530 nm.
a
MIC values are reported only for compounds displaying (1) a prominent decrease in turbidity by visual comparison to the control wells containing no antifungal and; (2) a
>50% reduction in fungal growth compared to controls containing no antifungal, as measured spectroscopically by absorption at 530 nm.
b
ATCC no. 10231.
ATCC no. 48435.
ATCC no. 36556.
All values were determined after incubation at 35 °C for 48 h.
c
d
e
contributes to increased antifungal activity in a stereoselective
manner in three species of pathogenic yeasts, Candida albicans,
Crypotococcus neoformans, and Candida glabrata.
thioethers, phosphates, etc.20 were explored on the example of
b glycosylation of cholesterol. We obtained the best results using
a/
O-2,3,4,6-tetra-O-benzoyl-a-D-glucopyranosyl-trichloroacetimi-
Although there are many procedures for glycosylation of ste-
date (2a, Scheme 1) as a glycosylation reagent.21,22 This approach
utilizes the classic Schmidt’s glycosylation reaction conditions.23
Considering our ability to use this method and have control in
roids,19 it was our imperative to select a simple, reliable, and above
all a/b selective glycosylation method that could then be applied to
the preparation of a wide variety of functionalized steroids. For
the preparation of a/b glucose derivatives of cholesterol, we then
these purposes, several classic glycoside donors, such as halofens,
selected this method for the preparation of other functionalized
Table 4
2
a,3b-substituted spirostane derivatives and the MIC50 for three yeast species
a,e
Structure
MIC50
(lg/mL)
b
Candida albicans
Candida glabratac
Cryptococcus neoformansd
HO
α
H
H
β
O
H
HO
HO
HO
NA
NA
NA
O
H
H
14a
HO
OH
OH
O
α
α
H
H
β
O
H
H
O
NA
NA
NA
NA
NA
NA
OH
O
H
H
14b
HO
α
OH
H
β
O
H
O
O
O
β
OH
O
H
HO
H
14c
OH
HO
α
O
α
H
H
β
H
O
α
OH
O
OH
NA
NA
NA
O
O
H
OH
H
14d
OH
OH
HO
α
H
OH
H
β
O
H
O
O
β
O
HO
OH
O
16
8
0.5
OH
O
H
H
14e
OH
HO
(continued on next page)

7384
A. Cammarata et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7379–7386
Table 4 (continued)
a,e
Structure
MIC50
(lg/mL)
b
Candida albicans
Candida glabratac
Cryptococcus neoformansd
HO
α
H
OH
H
β
O
H
O
O
β
HO
HO
β
NT
NT
NT
O
HO
OH
O
H
HO
O
H
14f
OH
⁄⁄Previously reported in Ref. 18
NA denotes compounds that had (1) a slight reduction in turbidity to no change and; (2) had less than a 50% reduction in growth compared to controls, as measured
spectroscopically by absorption at 530 nm.
NT denotes not yet tested.
a
MIC values are reported only for compounds displaying (1) a prominent decrease in turbidity by visual comparison to the control wells containing no antifungal and; (2) a
>50% reduction in fungal growth compared to controls containing no antifungal, as measured spectroscopically by absorption at 530 nm.
b
ATCC no. 10231.
ATCC no. 48435.
ATCC no. 36556.
All values were determined after incubation at 35 °C for 48 h.
c
d
e
steroids presented in this report (11–14, Scheme 1). Briefly, the
corresponding or b trichloroacetamidates were prepared as gly-
In ‘dry’ hydrolysis, trans-esterification followed by ion-exchange
neutralization generates a clean reaction containing the dry sapo-
nin and the easily removable methyl benzoate. In this case, the dry
saponin is a white powder. In contrast, under normal hydrolysis
conditions, the saponin product would be a viscous, semi-solid
material from which it would be difficult to remove water and
other impurtities. A dichloromethane solution of peresterified ste-
roids 7–10 were slowly titrated with freshly prepared 10% CH3O-
Na/CH3OH until the solution reached pH 10. The reaction
mixture was neutralized with dry acidic dowex (50WX8-200),
and the product was then purified by silica gel chromatography.
The antifungal activity of the resulting glycosylated steroids
shown in Scheme 1 were evaluated in vitro using C. albicans,C. neo-
formans, and C. glabrata. All assays were done in accordance with
NCCLS reference documents.29 The results of these screenings are
summarized in Tables 1–4. Of the analogs tested, none of the
a
cosylation reagents (2, Scheme 1) from commercially available
mono and disaccharides in three steps. The first step of the prepa-
ration was total benzoylation of the free saccharide hydroxy
groups with benzoyl chloride in pyridine.24 The second step in-
volved an elegant and selective perbenzoylated saccharide C-1 es-
ter hydrolysis with ethanolamine in dimethylsulfoxide-ethyl
acetate.25 Finally, in the third step of the procedure, the prepared
C-1 deprotected perbenzoylated saccharide was used in a reaction
with trichloroacetamide in dichloromethane with potassium car-
bonate as a base catalyst (Scheme 1). This reaction was highly effi-
cient, and if it was conducted under kinetic conditions, only the
corresponding b acetamidates were prepared. On the other hand,
if the reaction was conducted under thermodynamic conditions,
then the corresponding
a acetamidate was the dominant prod-
uct.26 Purification of the product involved a simple and fast silica
gel filtration followed with the immediate coupling to the corre-
sponding functionalized steroid.
2a,3b functionalized steroids that contained either a mono or
disaccharide that was -linked to the 3b-position of the steroid
a
showed any antifungal activity against the three yeasts tested,
while a number of the b-linked derivatives showed promise. Spe-
cifically, one of the four steroid derivatives having a b-linked glu-
cose showed antifungal activity against all three species at 8 lg/
mL (13c, Table 3). However, the most promising results were ob-
All of the functionalized steroids used in the preparation of our
glycosylated products (11–14, Scheme 1) were prepared in several
steps from commercially available materials using our previously
developed and published methods of synthesis.17,18 Thus, the pre-
viously prepared 2
a,3b substituted steroids 3–6 (Scheme 1) were
served in all but one class of molecules when the addition of the
glycosylated with trichloroacetamidate glycoside donors
2
disaccharide maltose (1,4a-linked disaccharide) was b-linked to
(Scheme 1) under TMSOTf catalyzed conditions (Scheme 1). It
was demonstrated in many instances that sialic acid donors are
good catalysts for glycosylation.27 However, the opposite isomer
of trichloroacetamidate must be used to prepare the steroid deriv-
ative with our desired stereochemistry. It is important to empha-
size that these glycosylation reactions are very sensitive to
moisture; therefore the reaction must be carried out in the
presence of molecular sieves. The final step in the preparation of
our glycosylated steroids involved the hydrolysis of all ester groups
of compounds 7–10 (Scheme 1). This reaction must also be con-
ducted in a manner to control for excess moisture, due to the fact
that upon isolation, the resulting glycosylated steroid will swell in
an excess of water, making the isolation of the targeted compounds
exceptionally difficult.28 To circumvent this problem and simplify
the isolation and purification of glycosidic steroids 11–14, we per-
formed the hydroxyl deprotection in ‘water-free’ conditions. We
developed a simple and highly efficient method of ‘dry’ hydrolysis.
cholestane, spirostane and 2,3-functionalized spirostanes ( 11e,
13e, and 14e Tables 1, 3 and 4). Antifungal activities of these three
derivatives ranged from 32–0.5
species assayed. The exception to this observation was seen in
the 2 ,3b-functionalized cholestane group, where greater anti-
lg/mL, depending on the fungal
a
fungal activity was observed in the disaccharide cellibiose (1,4b-
linked disaccharide) (12f Table 2). Additionally, the presence of a
hydroxyl group in the 2a-position of the spirostane derivatives in-
creased the antifungal activity against all three yeasts (compounds
13e vs 14e). These findings indicate that the stereochemistry of the
glycosidic bond to the steroid may be a key factor in the develop-
ment of novel antifungals. Current work is underway to determine
the antifungal activities of 2a,3a,2b,3b and 2b,3a-functionalized
steroids and glycosidic steroids. Finally, in vitro mammalian cell
toxicity studies were done using two cell lines (Vero cells (ATCC
no. CRL-1651) and rabbit skin cells (ATCC no. CCL-68) on all deriv-
atives that had antifungal activity of less than 32 lg/mL at a range

A. Cammarata et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7379–7386
7385
Figure 1. Evaluation of mammalian cell toxicity of active compounds. Cytotoxicity was determined using noncancerous vero cells (African green monkey) and rabbit skin
cells in accordance with Promega CellTiter 96 Non-RadioactivCell Proliferation Assay (cat # G4000) on compounds 11e, 12f, 13c, 13e and 14f. Compounds were diluted in
media to the testing dilution amounts: 200 lg/ml, 100 lg/ml, 50 lg/ml and 25 lg/ml. All assays were done in triplicate and the average values are presented, with SD. Cells
were incubated in the presence of the compounds for 24 h at 37 °C and 5% CO2. Tetrazolium dye solution was added to each well and allowed to incubate for 1–4 h.
Stabilization/Stop solution was added and allowed to sit at room temperature for 1 h. Formazan product was scored spectrophotometrically with an automatic plate reader
set at 570 nm. 1Â saponin was used as a control to measure cell death. Wells containing 1% DMSO and no drug were used as a positive control and the average absorbance at
570 nm were set to 100%. The percent of cells viable were calculated as a ratio between the average absorption at 570 nm of the drug containing wells/average absorption at
570 nm of the control (no drug). Ratios were multiplied by 100 to give the values as a percentage. It is important to note that the MIC value shown at the top of each panel
does not reflect the MIC for each fungal species analyzed. Rather, this value is the highest MIC value of the three species tested.
of at least 5–10 times greater than the minimum inhibitory con-
centration in accordance with Promega CellTiter 96 Non-Radioac-
are summarized in Figure 1. Two of the five compounds assayed
(12f and 13c) showed significant toxicity in both cell lines utilized.
In summary, several promising antifungal agents were identified
through the course of these assays. Further work determining the
tivCell Proliferation Assay (cat
#
G4000). This included
compounds11e, 12f, 13c, 13e and 14f. The results of those studies

7386
A. Cammarata et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7379–7386
Fig. 1 (continued)
8. Banday, M. R.; Farshori, N. N.; Ahmad, A.; Khan, A. U.; Rauf, A. Eur. J. Med. Chem.
2010, 45, 1459.
9. Kaskiw, M. J.; Tassotto, M. L.; Th’ng, J.; Jiang, Z. H. Bioorg. Med. Chem. 2008, 16,
3209.
10. Renault, S.; De Lucca, A. J.; Boue, S.; Bland, J. M.; Vigo, C. B.; Selitrennikoff, C. P.
Med. Mycol. 2003, 41, 75.
11. Dixon, J. T.; Van Heerden, F. R.; Holzapfel, C. W. Tetrahedron: Assymetry 2005,
16, 392.
role of stereochemistry of the steroid moiety with respect to anti-
fungal activity is currently underway.
Detailed information regarding the syntheses, spectroscopic
characterization and biological evaluation of the steroid-glycosides
presented in this paper can be found in the Supplementary data
provided.
12. De Lucca, A. J.; Boue, S.; Palmgren, M. S.; Maskos, K.; Cleveland, T. E. Can. J.
Microbiol. 2006, 52, 336.
13. De Lucca, A. J.; Bland, J. M.; Vigo, C. B.; Cushion, M.; Selitrennikoff, C. P.; Peter,
J.; Walsh, T. J. Med. Mycol. 2002, 40, 131.
14. De Lucca, A. J.; Jacks, T. J.; Broekaert, W. J. Mycopathologia 1998, 144, 87.
15. Shamsuzzaman, S.; Khan, M. S.; Alam, M.; Tabassum, Z.; Ahmad, A.; Khan, A. U.
Eur. J. Med. Chem. 2010, 45, 1094.
This work was supported in part by funding from the LSUHSC
School of Medicine, LSUHSC Department of Pharmacology and
Experimental Therapeutics, Louisiana Lions Eye Foundation, and
by an unrestricted grant from Research to Prevent Blindness,
New York, NY (Department of Ophthalmology-LSUHSC).
16. Brunel, J. M.; Loncle, C.; Vidal, N.; Dherbomez, M.; Letourneux, Y. Steroids 2005,
70, 907.
Supplementary data
17. Jursic, B. S.; Upadhyay, S. K.; Creech, C. C.; Neumann, D. M. Bioorg. Med. Chem.
Lett. 2010, 20, 7372.
18. Upadhyay, S. K.; Creech, C. C.; Bowdy, K. L.; Stevens, E. D.; Jursic, B. S.;
Neumann, D. M. Bioorg. Med. Chem. Lett. 2011, 21, 2826.
19. Cheung, A. K. J. Virol. 1991, 65, 5260.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.10.015.
20. Levy, D. E.; Fugedi, P. The organic Chemistry of Sugars; CRC Press: Boca Raton, 2006.
21. Ali, I. A. Handbook of Chemical Glycosylation: Advances in Stereoselectivity and
Therapeutic Relevance; Gottingen: Germany, 2004.
22. Zhu, X.; Schmidt, R. R. Handbook of Chemical Glycosylation: Advances in
Stereoselectivity and Therapeutic Relevance; Wiley-VCH: Weinheim Germany, 2008.
23. Wang, Z. Comprehensive Organic Name Reactions and Reagents; Wiley: New
York, 2009.
24. Wang, Z. D.; Mo, Y.; Chiou, C. L.; Liu, M. Molecules 2010, 15, 374.
25. Grynkiewicz, G.; Fokt, I.; Szeja, W.; Fitak, H. J. Chem. Res. 1989, S, 152.
26. Schmidt, R. R.; Michael, J. Tetrahedron Lett. 1984, 25, 821.
27. Ressa, D. K.; Linhardt, R. J. Curr. Org. Synth. 2004, 1, 31.
28. Guclu-Ustandag, O.; Mazza, G. Crit. Rev. Food Sci. 2007, 47, 231.
29. Clinical Laboratory and Standards Institute. M27-A3 2010, 28.
References and notes
1. Ellis, D.; Marriott, D.; Hajjeh, R. A.; Warnock, D.; Meyer, W.; Barton, R. Med.
Mycol. 2000, 38, 173.
2. Coleman, J. J.; Okoli, I.; Tegos, G. P.; Holson, E. B.; Wagner, F. F.; Hamblin, M. R.;
Mylonakis, E. A. C. S. Chem. Biol. 2010, 5, 321.
3. Bow, E. J. Oncology (Williston Park) 2001, 15, 1035.
4. Bow, E. J. Semin. Hematol. 2009, 46, 259.
5. Spanakis, E. K.; Aperis, G.; Mylonakis, E. Clin. Infect. Dis. 2006, 43,
1060.
6. Chakrabarti, A.; Chatterjee, S. S.; Shivaprakash, M. R. Nippon Ishinkin. Gakkai
Zasshi 2008, 49, 165.
7. Cornely, O. A. Infection 2008, 36, 296.