Impact of absolute stereochemistry on the antiangiogenic and antifungal activities of itraconazole

Article abstract of DOI:10.1021/ml1000068

Itraconazole is used clinically as an antifungal agent and has recently been shown to possess antiangiogenic acitivity. Itraconazole has three chiral centers that give rise to eight stereoisomers. The complete role of stereochemistry in the two activities of itraconazole, however, has not been addressed adequately. For the first time, all eight stereoisomers of itraconazole (1a?h) have been synthesized and evaluated for activity against human endothelial cell proliferation and for antifungal activity against five fungal strains. Distinct antiangiogenic and antifungal activity profiles of the trans stereoisomers, especially 1e and 1f, suggest different molecular mechanisms underlying the antiangiogenic and antifungal activities of itraconazole.

Full text of DOI:10.1021/ml1000068

pubs.acs.org/acsmedchemlett
Impact of Absolute Stereochemistry on the
Antiangiogenic and Antifungal Activities of Itraconazole
Wei Shi,† Benjamin A. Nacev,†,§ Shridhar Bhat,† and Jun O. Liu*,†,‡
†Department of Pharmacology and Molecular Sciences, ‡Department of Oncology, and §Medical Scientist Training Program,
Johns Hopkins School of Medicine, 725 North Wolfe Street, Baltimore, Maryland 21205
ABSTRACT Itraconazole is used clinically as an antifungal agent and has recently
been shown to possess antiangiogenic acitivity. Itraconazole has three chiral
centers that give rise to eight stereoisomers. The complete role of stereochemistry
in the two activities of itraconazole, however, has not been addressed adequately.
For the first time, all eight stereoisomers of itraconazole (1a-h) have been
synthesized and evaluated for activity against human endothelial cell proliferation
and for antifungal activity against five fungal strains. Distinct antiangiogenic and
antifungal activity profiles of the trans stereoisomers, especially 1e and 1f, suggest
different molecular mechanisms underlying the antiangiogenic and antifungal
activities of itraconazole.
KEYWORDS Itraconazole, stereochemistry, diastereomers, angiogenesis, antifungal
activity
e have previously reported that itraconazole
(Figure 1), an antifungal drug, potently inhibits
W
in vitro proliferation of human umbilical vein
endothelial cells (HUVEC) and angiogenesis in vivo.1 The
target responsible for itraconazole's antifungal activity is
lanosterol 14R-demethylase (14DM), a key enzyme involved
in the biosynthesis of ergosterol, which is required for the
integrity of the fungal cell membrane.2 However, the role of
human 14DM in the inhibition of angiogenesis by itracona-
zole remains unclear. The poor correlation between human
14DM inhibition and antiangiogenic activity for several
structurally related potent azole antifungal drugs implies
that an other primary molecular target(s) might be respon-
sible for the antiangiogenic activity of itraconazole with
14DM making only a partial contribution.1,3,4
Figure 1. Structures of the eight itraconazole diastereomers aris-
ing from three stereogenic centers numbered 2, 4, and 20. The cis
designation denotes that the two substituents (in the blue boxes)
are on the same side of the 1,3-dioxolane ring, while the trans
designation denotes the opposite orientation.
Itraconazole contains three chiral centers, giving rise to a
total of eight stereoisomers. The dioxolane ring harbors two
chiral centers, while the third one marked as 20 resides on the
sec-butyl side chain appended to the triazolone ring
(Figure 1). As an antifungal drug, itraconazole is adminis-
tered in clinical formulations as a 1:1:1:1 mixture of four cis-
stereoisomers (1a-d).5 Although the antifungal activity6
and metabolism5 of individual cis stereoisomers of itraco-
nazole have been reported, the activity of the trans stereo-
isomers, 1e-h, have not been disclosed to date. Further-
more, itraconazole has not been examined for antiangiogenic
activity in any of its stereochemically pure forms. Our previ-
ous effort was limited to the synthesis and determination
of the antiangiogenic activity of the epimeric mixtures of
4R- and 4S-cis-itraconazole.1 To systematically explore the
effect of absolute stereochemistry at every chiral center of
itraconazole on both antifungal and antiangiogenic activity,
we synthesized all of the eight stereoisomers and compared
their antiangiogenic and antifungal activities.
We began the total synthesis by reducing the nitro group
in N-(4-methoxyphenyl)-N0-(4-nitrophenyl)-piperazine 2
(Scheme 1)1 using a palldium-catalyzed transfer hydro-
genation. Instead of ammonium formate that we had
used in our earlier synthesis, the use of hydrazine as a
hydrogen source7 produced much higher yields of aniline
3. Palladium-catalyzed reduction of the nitro group to the
amino group was achieved in a better yield than our
earlier synthesis using ammonium formate as a hydrogen
Received Date: January 10, 2010
Accepted Date: March 29, 2010
Published on Web Date: April 12, 2010
r
2010 American Chemical Society
155
DOI: 10.1021/ml1000068 ACS Med. Chem. Lett. 2010, 1, 155–159
|

pubs.acs.org/acsmedchemlett
Scheme 1. Synthesis of Eight Itraconazole Stereoisomersa
a Reagents and conditions: (a) 10% Pd/C, NH2NH2 H2O, EtOH, reflux, 99%. (b) (i) Phenyl chloroformate, pyridine, CH3CN; (ii) NH2NH2 H2O,
3
3
1,4-dioxane, reflux; (iii) formamidine acetate, AcOH, DMF, 80 °C, 81% over 3 steps. (c) TsCl, Et3N, DMAP, CH2Cl2, 99%. (d) K2CO3, 18-Crown-6, DMSO,
room temperature, 73%. (e) 48% Aqueous HBr, 110 °C, 91%. (f) TfOH, toluene, room temperature, 60 h, 63% for cis and 19% for trans based on NMR.
(g) NaH, DMSO, 50 °C f 85 °C, 62%.
source.1 Without further purification, the amino group in
3 was transformed into the triazolone via the phenylcar-
bamate and semicarbazide intermediates.8 A small
amount of acetic acid was added to facilitate the cycliza-
tion of the semicarbazide with formamidine acetate at a
lower temperature than usual for this type of ring closure,
thereby minimizing side reactions.9 The stereochemistry
at the 20 position in 1a-h was inherited from the optically
pure starting material, (R)-(-)-2-butanol (5a) or (S)-(þ)-2-
butanol (5b). The chiral tosylate 6a or 6b was obtained in
almost quantitative yield by reacting 5a or 5b with tosyl
chloride in the presence of triethylamine and DMAP.
To achieve a stereospecific N-alkylation of triazolone 4
by tosylate displacement of 6a or 6b, the proton abstrac-
tion of triazolone nitrogen was conducted using potas-
sium carbonate in conjunction with 18-crown-6. Such an
application of crown ethers is known to enhance the
nucleophilicity of the nitrogen anion by forming loose
ion pairs.10 In addition to driving the reaction to proceed
at room temperature, this avoided a potential SN1 reac-
tion as well as minimized the elimination reaction of 6a or
6b. A clean SN2 inversion leading to either 7a or 7b in an
enantiomerically pure form has been documented be-
fore.11 , 1 2 The desired stereochemical outcome at this step
in our synthesis was also confirmed later by chiral high-
performance liquid chromatography (HPLC) analysis of
1a and 1b (Table 1 and the Supporting Information).
Removal of the methyl group by heating 7a or 7b in
concentrated aqueous HBr at 110 °C yielded 8a or 8b
containing a free phenol ready for the final coupling.8
Construction of the 1,3-dioxolane ring in 11a-d was
achieved by acid-assisted ketalization of 2,20,40-trichloroace-
Table 1. Chrial HPLC Analysis Data and Optical Rotation of
Itraconazole Stereoisomers
retention
diastereomeric
purity
[R]D in
CHCl3
compounds
time (min)a
1a (2S,4R,20S)
1d (2R,4S,20R)
1b (2S,4R,20R)
1c (2R,4S,20S)
1e (2S,4S,20S)
1h (2R,4R,20R)
1f (2S,4S,20R)
1g (2R,4R,20S)
43.017
42.520
46.333
42.547
-5.5
þ5.7
cis
-12.3
þ12.6
>98%
38.933
39.573
37.800
39.253
-13.2
þ13.3
-19.1
þ18.6
trans
a HPLC conditions are almost identical to those in ref 5.
tophenone 9 with optically pure glyceryl tosylate 10a or
10b.10 While the stereochemistry at C-4 in 11a-d emanates
from the chiral starting material 10a or 10b, C-2 is the new
chiral center that gets created during ketalization and there-
fore numbered in blue. The ratio of cis versus trans diaster-
eomers 11a/11c or 11b/11d is dictated by the steric effects,
and actually, a preponderance of cis-dioxolane was obser-
ved. The cis diastereomer (11a or 11b)13 was separated from
the trans diastereomer (11c or 11d) and further purified by
the well-established double recrystallization approach.10 To
date, however, among the trans diastereomers, only 11c
could be traced in the literature to a fleeting citation, and
surprisingly, 11d is unknown. After a meticulous thin-layer
chromatographic analysis, it was found that the tosylate salts
of the trans diastereomer predominantly remained in the
ethyl acetate solution during the purification process of the
cis product. By running a gradient column (50:1 f 5:1 CH2Cl2-
acetone), 11c was isolated from 11a and other side products
with purity that was sufficient for NMR characterization.
r
2010 American Chemical Society
156
DOI: 10.1021/ml1000068 ACS Med. Chem. Lett. 2010, 1, 155–159
|

pubs.acs.org/acsmedchemlett
Figure 2. 1H and 13C NMR of 1,3-dioxolane intermediates 11a-d. (A) 1H NMR for cis-11a and 11b. (B) 1H NMR for trans-11c and 11d.
Table 2. Potency of Itraconazole Stereoisomers in Biological Assays
Candida glabrata
Saccharomyces cerevisiae
C. albicans
(10261)
C. neoformans
HUVEC
(BY4741)
(H99)
BG1
B92
Rb
MIC80
Rb
MIC80
Rb
MIC80
Rb
MIC80
Rb
MIC80
Rb
a
c
c
c
c
c
compounds
IC50
racemic itraconazoled
1a (2S,4R,20S)
1b (2S,4R,20R)
1c (2R,4S,20S)
1d (2R,4S,20R)
1e (2S,4S,20S)
1f (2S,4S,20R)
93 (76, 115.3)
74 (62, 115)
1
0.5
2
1
0.0156
0.0312
0.0156
0.0312
0.0312
0.0312
0.0156
0.5
1
2
0.0625
0.125
0.25
0.125
0.25
4
1
2
0.5
0.5
0.5
0.5
0.5
1
1
0.5
0.5
0.5
0.5
0.5
1
1
0.8
1.1
4
2
1
1
1
1
106 (82, 138)
1
1
4
147 (123, 177) 1.6
236 (183, 305) 2.5
2
4
2
2
1
1
1
2
2
4
1
1
289 (212, 393)
361 (181, 719)
3.1
3.9
1
2
2
64
64
16
32
2
2
1
2
1
4
0.5
1
0.5
1
1g (2R,4R,20S)
1h (2R,4R,20R)
301 (173, 525) 3.2
346 (236, 508) 3.7
>4
>4
>8
>8
32
32
1
>4
>4
>8
>8
>4
>4
>8
>8
0.5
2
a nM (95% CI). b Ratios of IC50 or MIC80 of stereoisomer/racemic mixture. c μg/mL. d Mixture of the four cis-diastereomers from Sigma-Aldrich.
Similarly, pure 11d was also obtained. The 1H NMR spectra
(Figure 2A,B) illustrate the obvious differences not only in
the aromatic region (7.0-8.5 ppm) but also in the aliphatic
region (3.4-4.8 ppm), arising from different 1,3-dioxolane
ring conformations in the cis and trans diastereomers.
With the building blocks in hand, the displacement of the
O-tosyl group in 11a-d with the phenolic oxygen in 8a,b
under basic conditions afforded final products 1a-h.8 Be-
cause the chiral centers on the 1,3-dioxolane ring are well
separated from the third chiral center on the sec-butyl side
chain, it is not surprising that the cis diastereomers could be
well distinguished from the trans diastereomers by NMR,
whereas all of the cis stereoisomers 1a-d or the trans
stereoisomers 1e-h exhibited identical 1H and 13C NMR
spectra. To provide further support for the successful synthe-
sis and purification of all eight stereoisomers, the optical
rotation and chiral HPLC profile of each stereoisomer was
measured (Table 1and Supporting Information).
After the quality of all stereoisomers was confirmed, the
potency of each stereoisomer against HUVEC proliferation
and fungal growth was determined (Table 2). HUVEC were
incubated with drug or vehicle alone for 24 h and then pul-
sed for 6 h with [3H]-thymidine, the incorporation of which
was taken as a readout of cell proliferation. Inhibition of
fungal growth was assayed by incubating five yeast strains
with 2-fold serial dilutions of each stereoisomer for 30-
60 h depending on the strain (see the methods in the Sup-
porting Information) and then measuring the OD600 of the
r
2010 American Chemical Society
157
DOI: 10.1021/ml1000068 ACS Med. Chem. Lett. 2010, 1, 155–159
|

pubs.acs.org/acsmedchemlett
culture to quantitate growth. The minimum concentration
capableofinhibiting growth by 80% (MIC80) was determined.
The influence of stereochemistry on the inhibition of
HUVEC proliferation by itraconazole was minor. The differ-
ence in potency between 1a and 1f, the most and least
potent stereoisomers, respectively, was only slightly greater
than 4-fold. The most relevant stereochemical determinant
of potency in HUVEC was the configuration of the dioxolane
ring, with the cis diastereomers exhibiting higher potency
than the trans series by several fold. We note that the cis-4R
diastereomer is slightly more potent than the cis-4S isomer.
This is contrary to our previous report.1 After carefully
examining the individual steps of the previous synthesis, it
appears that the stereochemical centers were misassigned.
Hence, we also wish to redress our earlier account with the
stereochemical assignments and the corresponding assay
data in this letter. In contrast to HUVEC, the potency of
itraconazole against fungal proliferation was highly influ-
enced by stereochemistry (Table 2). We observed a diffe-
rence in potency of up to 32-fold between stereoisomers in
one fungal strain. In four out of five strains tested, the least
potent stereoisomers by a margin of at least 4-32-fold were
two of the trans isomers, 1g and 1h. On the other hand, the
other trans pair 1e and 1f was about as potent as the cis
diastereomers (1a-d). The exception was Cryptococcus
neoformans in which 1e and 1f were 2-fold less potent than
1g and 1h and 32-fold less potent than the best inhibitor.
In the case of dioxolane-containing azole antifungals like
itraconazole, ketoconazole, and terconazole, it has been
noted long ago that the cis diastereomeric pairs exhibit
much higher antifungal potency over their trans counter-
parts, and thus, for efficacy reasons, they have been used
clinically as mixtures of cis diastereomers. Docking studies
performed based on the published fluconazole-MtCYP51
(referred to as 14DM for the human enzyme) crystal struc-
ture have offered an explanation to this effect.14 Rupp et al.
analyzed homology-modeled CaCYP51 complexed with dif-
ferent stereoisomers of ketoconazole.15 Interestingly, they
concluded that the cis pair (2S,4R and 2R,4S) and only one of
the trans pair, namely, 2S,4S-ketoconazole, avidly bind to
CaCYP51, which is in good agreement with the reported IC50
values of the stereoisomers of ketoconazole against Candida
albicans.16 Antifungal activities that we measured for the
eight stereoisomers of itraconazole against the three asco-
mycetes perfectly match the pattern observed with ketoco-
nazole. It is possible that the CYP51 enzymes of ascomycetes
poorly bind the 2R,4R-itraconazole, whereas in the case of
phylogenetically distant C. neoformans, this scenario of bind-
ing among the trans pairs is quite the opposite. This may also
be explained by the expression of a stereoselective efflux
pump or catabolic enzyme in this strain. Taken together,
these data indicate that unlike HUVEC inhibition, the sensi-
tivity of fungal growth to itraconazole is dictated not by
cis-trans configuration of the dioxolane ring but instead by
the absolute stereochemistry at the 2 and 4 carbons. The
only commonality that we observed for the role of stereo-
chemistry in HUVEC and fungal inhibition was that the
stereochemistry at the 20 position had little influence on
potency in either case.
In summary, all of the cis diastereomers that make up the
commercial itraconazole exhibited high potency in both
HUVEC and fungal inhibition. All of the trans diastereoi-
somers were less potent in HUVEC proliferation than were
the cis diastereoisomers. In contrast, one pair of trans
diastereoisomers, 1e and 1f, was roughly as potent as the
cis diastereomers with respect to antifungal activity against
four out of five strains. The lack of correlation between
HUVEC and fungal sensitivity to optically pure itraconazole
stereoisomers suggests that human 14DM is not likely to be
the major target for the antiangiogenic activity of itracona-
zole. Indeed, we have recently found that the inhibitory
effect of itraconazole on endothelial cells results largely from
its inhibition of cholesterol trafficking through the lysosomal
compartment, leading to inhibition of the mTOR pathway.17
This work provides previously unavailable data on the role of
stereochemistry in the potency of itraconazole against an
emerging therapeutic target for this drug, angiogenesis. We
demonstrated that compounds 1a and 1b possess the great-
est antiangiogenic potential and should therefore be used as
lead compounds for further optimization of itraconazole as
an antiangiogenic drug.
SUPPORTING INFORMATION AVAILABLE Experimental
procedure, analytical data, and 1H and 13C NMR spectra for new
compounds 11c,d and 1e-h. HRMS and HPLC data for final
compounds 1a-h. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author: *To whom correspondence should be
addressed. E-mail: joliu@jhu.edu.
Funding Sources: This work was supported by NCI, FAMRI, the
Commonwealth Foundation, and the NIH Medical Scientist Training
Program Grant T32GM07309 (B.A.N.). It was also supported in part
by Grant UL1 RR 025005 from the National Center for Research
Resources (NCRR), a component of the National Institutes of Health
(NIH) and NIH Roadmap for Medical Research, and its contents are
solely the responsibility of the authors and do not necessarily
represent the official view of NCRR or NIH.
ACKNOWLEDGMENT We are grateful to Drs. Peter Espen-
shade and Brendan Cormack and Clara Bien for providing us with
the fungal strains and for helpful advice on conducting the fungal
growth assays. We also thank Professor Gerald Hart for the use of
equipment.
REFERENCES
(1)
(2)
(3)
Chong, C. R.; Xu, J.; Lu, J.; Bhat, S.; Sullivan, D. J., Jr.; Liu, J. O.
Inhibition of Angiogenesis by the Antifungal Drug Itracona-
zole. ACS Chem. Biol. 2007, 2, 263–270.
Vanden Bossche, H.; Marichal, P.; Gorrens, J.; Coene, M. C.
Biochemical basis for the activity and selectivity of oral
antifungal drugs. Br. J. Clin. Pract. Suppl. 1990, 71, 41–46.
David, C. L.; Diane, E. K.; Michael, R. W.; Maria, S.; Damjana,
R.; Steven, L. K. Characteristics of the heterologously expressed
r
2010 American Chemical Society
158
DOI: 10.1021/ml1000068 ACS Med. Chem. Lett. 2010, 1, 155–159
|

pubs.acs.org/acsmedchemlett
human lanosterol 14R-demethylase (other names: P45014DM,
CYP51, P45051) and inhibition of the purified human and
Candida albicans CYP51 with azole antifungal agents. Yeast
1999, 15, 755–763.
(17) Xu, J.; Dang, Y.; Ren, R. Y.; Liu, J. O. Cholesterol trafficking is
required for mTOR activation in endothelial cells. Proc. Natl.
Acad. Sci. U.S.A. 2010, 107, 4764–4769.
€
Trosken, E. R.; Michael Arand, M. A.; Zarn, J. A.; Patten, C.;
(4)
(5)
€
Volkel, W.; Lutz, W. K. Comparison of lanosterol-14R-
demethylase (CYP51) of human and Candida albicans for
inhibition by different antifungal azoles. Toxicology 2006,
228, 24–32.
Kunze, K. L.; Nelson, W. L.; Kharasch, E. D.; Thummel, K. E.;
Isoherranen, N. Stereochemcial aspects of itraconazole me-
tabolism in vitro and in vivo. Drug Metab. Dispos. 2006, 34,
583–590.
(6)
(7)
Koch, P.; Rossi, R. F., Jr.; Senanayake, C. H.; Wald, S. A. 2R,4S-
Hydroxyitraconazole isomers. WO 00/66100, 2000.
€
€
Hoglund, L. P. J.; Silver, S.; Engstrom, M. T.; Salo, H.; Tauber,
€
ꢀ
A.; Kyyronen, H.-K.; Saarenketo, P.; Hoffren, A.-M.; Kokko,
K.; Pohjanoksa, K.; Sallinen, J.; Savola, J.-M.; Wurster, S.;
Kallatsa, O. A Structure-Activity Relationship of Quinoline
Derivatives as Potent and Selective R2C-Adrenoceptor Anta-
gonists. J. Med. Chem. 2006, 49, 6351–6363.
(8)
(9)
Heeres, J.; Backx, L. J. J.; Cutsem, J. V. Antimycotic Azoles. 7.
Synthesis and Antifungal Properties of a Series of Novel
Triazol-3-ones. J. Med. Chem. 1984, 27, 894–900.
Sheng, C.; Zhang, W.; Ji, H.; Zhang, M.; Song, Y.; Xu, H.; Zhu,
€
J.; Miao, Z.; Jiang, Q.; Yao, J.; Zhou, Y.; Zhu, J.; Lu, J. Structure-
Based Optimization of Azole Antifungal Agents by CoMFA,
CoMSIA, and Molecular Docking. J. Med. Chem. 2006, 49,
2512–2515.
(10) Tanoury, G. J.; Hett, R.; Wilkinson, H. S.; Wald, S. A.;
Senanayake, C. H. Total synthesis of (2R,4S,20S,30R)-hydro-
xyitraconazole: Implementations of a recycle protocol and a
mild and safe phase-transfer reagent for preparation of the
key chiral units. Tetrahedron: Asymmetry 2003, 14, 3487–
3493.
(11) Tanoury, G. J.; Senanayake, C. H.; Hett, R.; Kuhn, A. M.;
Kessler, D. W.; Wald, S. A. Pd-Catalyzed Aminations of Aryl
Triazoiones: Effective Synthesis of Hydroxyitraconazole En-
antiomers. Tetrahedron Lett. 1998, 39, 6845–6848.
(12) Pinori, M. L. M.; Modena, D.; Mascagni, P. Azole Derivative
Useful as Antifungal Agents with Reduced Interaction with
Metabolic Cytochromes. WO 2005040156, 2005.
(13) cis- and trans-Designations describe the geometrical relation-
ship of triazolylmethyl (C-2) and oxymethyl (C-4) substitu-
ents, either on the same face (cis) or the opposite face (trans)
of the dioxolane ring. While going from tosyl derivatives
(11a-d) all the way to itraconazole stereoisomers 1a-h, the
cis and trans relationship does not change. However, accord-
ing to Cahn-Ingold-Prelog priority rules, the R and S nota-
tions corresponding to C-4 of the dioxolane ring do change.
For example, cis-2S,4S tosylate derivative (11a) eventually
leads to cis-2S,4R,20S (1a) or cis-2S,4R,20R (1b) itraconazole
stereoisomers.
(14) Podust, L. M.; Poulos, T. L.; Waterman, M. R. Crystal structure
of cytochrome P450 14R-sterol demethylase (CYP51) from
Mycobacterium tuberculosis in complex with azole inhibi-
tors. Proc. Nat. Acad. Sci. 2001, 98, 3068–3073.
(15) Rupp, B.; Raub, S.; Marian, C.; Holtje, H.-D. Molecular design
of two sterol 14R-demethylase homology models and their
interactions with the azole antifungals ketoconazole and
bifonazole. J. Comput.-Aided Mol. Des. 2005, 19, 149–163.
(16) Rotstein, D. M.; Kertesz, D. J.; Walker, K. A.; Swinney, D. C.
Stereoisomers of ketoconazole: Preparation and biological
activity. J. Med. Chem. 1992, 35, 2818–2825.
r
2010 American Chemical Society
159
DOI: 10.1021/ml1000068 ACS Med. Chem. Lett. 2010, 1, 155–159
|