Synthesis and biological studies of 35-deoxy amphotericin B methyl ester

Article abstract of DOI:10.1002/anie.200800590

(Chemical Presented) An indispensable OH group: The synthesis of 35-deoxy amphotericin B methyl ester was completed by using a novel method for the coupling of the mycosamine to the aglycone. The investigation of the antifungal activity and efflux-inducing ability of this compound provided data that underscore the relevance of the hydroxy group at C35 and supports the involvement of double-barrel ion channels.

Full text of DOI:10.1002/anie.200800590

Angewandte
Chemie
DOI: 10.1002/anie.200800590
Natural Products (2)
Synthesis and Biological Studies of 35-Deoxy Amphotericin B Methyl
Ester**
Alex M. Szpilman, Jeffrey M. Manthorpe, and Erick M. Carreira*
In the preceding communication we disclosed the synthesis of
35-deoxy amphotericin B aglycone in its protected form.[1]
Herein, we describe the key glycosidation that completes the
assembly of 35-deoxy amphotericin B methyl ester. Addi-
tionally, we present our findings concerning its activity in
potassium-efflux assays and in fungi. We discuss the signifi-
cance of our observations to the mechanism of action of the
clinically important antifungal agent amphotericin B (1).
Figure 1. Side views of the proposed single- (a)and double-barrel (b)
ion channels. H-bonds are indicated with dotted lines.
Amphotericin B incorporates mycosamine and hemiketal
In the most widely accepted mechanism, commonly
referred to as the barrel-stave model, amphotericin B is
suggested to form ion channels in the membrane. These ion
channels are believed to be generated through a self-assembly
process involving 4–12 molecules of the polyene macrolide
stabilized by an equal number of sterols.[1,2] Of particular
importance for the present work is the discrepancy between
the overall length of amphotericin B (about 21 Š)[3] and the
average width of the phospholipid membrane (about 43 Š).
Two models have been postulated to reconcile this discrep-
ancy, involving single- and double-barrel channels
(Figure 1).[1,2] In the first of these, the bilayer narrows to
accommodate the ion channel as it spans the membrane
(Figure 1a). According to the second hypothesis, two barrel
units link up in a tail-to-tail fashion, accommodating to the
width of the membrane (Figure 1b). The hydroxy group on
C35 is suggested to play a pivotal role in the stabilization of
the double-barrel construct.[4]
units that characterize its family of macrolide antibiotics.[5]
Mycosamine is a 3,6-dideoxy-3-aza-mannose that is attached
as the b anomer to the amphotericin B aglycone. The
introduction of b-mannoside residues is widely acknowledged
to be one of the most challenging problems in carbohydrate
chemistry.[6] In the synthesis of mycosamine macrolides, the
problem is augmented by the acute acid-sensitivity of the
aglycone 2 and the obstruction of the hydroxy acceptor by its
position in a groove. To ensure formation of the b anomer, the
glycosidic coupling has been carried out following the method
of Schmidt with the C2’-epi-mycosamine donor bearing ester
groups capable of lending anchimeric assistance. Unfortu-
nately, this approach has been shown to provide the
orthoester 6a as a major byproduct of the reaction and
proceeds to low conversion.[7] In our hands, the reaction of
2a[7b] with 3 equiv of 3a and 30 mol% of PPTS afforded 4%
of the desired glycoside 5a, 12% of orthoester 6a, and 75%
of recovered aglycone 2a (Scheme 1, Table 1, entry 1). A
number of other methods were also examined without
success.[8] Orthoesters can be rearranged under acidic con-
ditions to afford the corresponding glycoside.[9] However, the
resilience of the orthoester and the acid sensitivity of the
unsaturated alcohol combined to impair our efforts in this
direction. For example, the use of a wide range of acids
returned starting orthoester (e.g. PPTS) or resulted in
destruction of the material (e.g. BF3).
[*] Dr. A. M. Szpilman, Dr. J. M. Manthorpe, Prof. Dr. E. M. Carreira
ETH Zurich HCI H335
Wolfgang-Pauli-Strasse 10, 8093 Zurich (Switzerland)
Fax: (+41)44-632-1328
E-mail: carreira@org.chem.ethz.ch
Homepage: http://www.carreira.ethz.ch
In the Schmidt glycosidation reaction, glycoside forma-
tion results from attack of aglycone at C1’ on intermediate 4,
while attack at the acetoxonium carbon leads to orthoester
formation. We hypothesized that modifying the steric and
electronic properties of the participating ester might tune the
[**] This research was supported by a grant from the Swiss National
Science Foundation. Postdoctoral scholarships were provided by
the Carlsberg foundation (A.M.S.)and Fonds NATEQ (J.M.M.).
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2008, 47, 4339 –4342
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4339

Communications
Scheme 1. Glycosidation of amphoteronolide and 35-deoxy amphoteronolide (see Table 1 for combinations and conditions).
Table 1: Introduction of the mycosamine moiety (Scheme 1).
rate ester because of a fine balance of factors: The presence of
the electronegative chloride substituent may lead to greater
charge localization on C1’ in 4; the steric bulk of the 2-
chloroisobutyrate is expected to disfavour attack that leads to
orthoester. Additionally, the chloride also enables facile
subsequent hydrolysis of the ester (Scheme 2).
Acceptor
Donor
Activator
[2]
5
6
2
2a
2a
2a
2b
3a
3b
3b
3b
PPTS[a]
PPTS[a]
CMPT[b]
CMPT[c]
7 mm
4%
12%
11%
23%
30%
75%
57%
–
20 mm
20 mm
20 mm
27%
43%
45%
–
Although the studies outlined above resulted in an
increase in formation of glycoside 5a, conversion remained
low. Full conversion could be achieved by use of additional
PPTS, but the yields were unsatisfactory. NMR experiments
provided clues for the poor conversion: Activation of the
trichloroacetimidate 3b by one equivalent of PPTS led to
rapid formation of the b tosylate 7, which on standing is
converted into the more stable a anomer 8. The mixture
proved unreactive in the glycosidation of amphoteronolide
2a.[11] We surmised that replacement of the tosylate with the
less nucleophilic triflate counterion could be productive.
To our surprise, activation of 3b with one equivalent of
pyridinium triflate led to formation of a mixture of the two
anomeric pyridinium salts (9 and 10). These species were also
unreactive towards the aglycone. After extensive experimen-
[a] 30 mol%. [b] 50 mol% CMPT and 100 mol% 2-chloro-6-methyl-
pyridine. [c] 10 mol% CMPT and 20 mol% 2-chloro-6-methyl-pyridine.
CMPT=2-chloro-6-methyl-pyridinium triflate; PPTS=pyridinium para-
toluenesulfonate.
chemoselectivity in favour of glycoside formation (i.e. of R2 in
3/4, Scheme 1).[10] This proved to be the case. Thus, the use of
the auxiliary 2-chloroisobutyrate furnished a 2.5:1 mixture of
glycoside 5a and orthoester 6a. Importantly, a range of other
esters of mycosamine donor 3 failed to improve the yield of
the glycoside, giving predominantly orthoester instead.[10] The
benzoate afforded a 1:1 mixture of glycoside and orthoester,
but this ester could not be subsequently hydrolyzed without
deleterious effects on the sensitive macrolactone. We spec-
ulate that glycosidation is favoured for the 2-chloroisobuty-
Scheme 2. a)K 2CO3, MeOH; 83% yield; b)(CF 3CO)2O, DMSO, (Me2N)2CO, Et3N, CH2Cl2/Et2O, À788C to RT; c)NaBH 4, MeOH; 82% yield over
two steps (brsm); d) HF/pyridine, MeOH, 40 8C; 72%; e)CSA, MeCN/water; 37% (six cycles); f)Bu 3P, THF/MeOH/water; 67%. brsm=based
on recovered starting material; CSA = camphor sulfonic acid.
4340 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 4339 –4342

Angewandte
Chemie
tation we identified 2-chloro-6-methyl-pyridinium triflate as
uniquely efficient for the glycosidation. Thus, reaction of
protected 35-deoxy amphoteronolide 2b with donor 3b in the
presence of 10 mol% 2-chloro-6-methyl-pyridinium triflate
and 20 mol% 2-chloro-6-methyl-pyridine afforded the
desired glycoside in 45% yield accompanied by 30% of the
orthoester (Table 1, entry 4). Importantly, glycosidation of
the natural amphoteronolide 2a gave similar yields (entry 3).
With the successful assembly of glycoside 5b, we could
attend to the inversion of the C2’ stereogenic center and
removal of the protecting groups. The 2-chloroisobutyrate
ester was hydrolyzed under mildly basic conditions
(Scheme 2). The configuration at C2’ was inverted by an
oxidation–reduction sequence as described earlier.[7]
Figure 2. K+ efflux from LUV induced by 13 (c)and amphotericin B
methyl ester (a)measured by potentiometry. The substrate was
added externally (as a DMSO solution)to afford a final concentration
of 1 mm. LUV with a diameter of 100 nm and containing 13%
ergosterol and 87% POPC in their membranes were utilized.[15]
LUV=large unilamellar vesicle, POPC=1-palmitoyl-2-oleoyl-sn-glycero-
phosphocholine.
Efflux all but ceased at 0.1 mm or in pure POPC-LUV while at
10 mm a weak efflux could be detected.
The TBS groups were removed from 12 by the action of
HF/pyridine in methanol over 12 h. In contrast, for the
analogous intermediate in the synthesis of amphotericin B,
desilylation required 48 h.[7b] The acetonide groups were
hydrolyzed under acidic conditions. Azide reduction with 1,3-
propanedithiol/triethylamine was very slow (50% conversion
after 24 h) and led to a complex mixture.[7b,12] Fortunately, the
azide could be reduced under neutral conditions using
tributylphosphine to cleanly afford 35-deoxy amphotericin B
methyl ester.
At this point, the stage was set to determine the
consequence of deleting the 35-hydroxy group on the
biological profile. We measured and compared the activity
of 35-deoxy amphotericin B methyl ester (13) and amphoter-
icin B methyl ester against Saccharomyces cerevisiae and
Candida albicans (Table 2). Strikingly, for both strains, 13 was
more than an order of magnitude less active.
We have presented a strategy for the total synthesis of
amphotericin B analogs, which addresses the necessary gly-
cosidation and subsequent elaboration steps. Specifically, we
disclose its application to the synthesis of 35-deoxy ampho-
tericin B methyl ester (13). We also present data that for the
first time experimentally confirms the importance of the
hydroxy group at C35 of amphotericin B in its role as a
fungicide and its ability to cause electrolyte efflux in lip-
osomes. Collectively, these observations are consistent with
the involvement of double-barrel ion channels in the mem-
brane for the activity of amphotericin B and its derivatives.
The approach we have described herein sets the stage for
additional investigations involving the study of conjugates to
various carbohydrates and other small molecules. Further
work to elucidate details of the mechanism of action using
fully synthetic analogs prepared according to the presented
strategy is underway.
Received: February 5, 2008
Published online: April 29, 2008
Table 2: Antifungal activity of amphotericin B methyl ester (AME)and
35-deoxy amphotericin B methyl ester (13).[a]
Entry Compound Saccharomyces cerevisiae
BY4741 [mm]
Candida albicans
CAF2-1 [mm]
Keywords: amphotericin B · antifungal agents ·
mechanism of action · natural products · total synthesis
.
1
2
AME
13
0.25
4.6
0.1
2.6
[1] A. M. Szpilman, D. M. Cereghetti, N. R. Wurtz, J. M. Man-
thorpe, E. M. Carreira, Angew. Chem. 2008, DOI: 10.1002/
ange.200800589; Angew. Chem. Int. Ed. 2008, DOI: 10.1002/
anie.200800589.
[a] Measured according to the National Committee on Clinical Labo-
ratory Standards protocol. See reference [13] for experimental details.
[2] S. C. Hartsel, C. Hatch, W. Ayenew, J. Liposome Res. 1993, 3,
377 – 408, and references therein.
[3] P. Ganis, G. Avitabile, W. Mechlin, C. P. Schaffner, J. Am. Chem.
Soc. 1971, 93, 4560 – 4564.
[4] M. Baginski, H. Resat, J. A. Mccammon, Mol. Pharmacol. 1997,
52, 560 – 570.
[5] Macrolide Antibiotics, Chemistry, Biology and Practice (Ed.: S.
Omura), Academic Press, New York, 1984.
We proceeded to examine the ability of 35-deoxy ampho-
tericin B methyl ester (13) to cause K-efflux from large
unilamellar vesicles (LUV), as monitored by using potassium-
ion selective electrodes.[14,15] We utilized LUV with a mem-
brane made up of POPC with or without ergosterol as a
component (see Figure 2 and the Supporting Information).[16]
35-Deoxy amphotericin B methyl ester (13) at 1 mm concen-
tration showed severely diminished ability to induce the
leakage of K+ compared to amphotericin B methyl ester.
[6] a) A. V. Demchenko, Synlett 2003, 1225 – 1240; b) E. S. H. El
Ashry, N. Rashed, E. S. I. Ibrahim, Curr. Org. Synth. 2005, 2,
175 – 213.
Angew. Chem. Int. Ed. 2008, 47, 4339 –4342
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4341

Communications
[7] a) K. C. Nicolaou, T. K. Chakraborty, Y. Ogawa, R. A. Daines,
[13] V. Paquet, A. A. Volmer, E. M. Carreira, Chem. Eur. J. 2008, 14,
2465 – 2481.
[14] a) P. van Hoogevest, B. de Kruijff, Biochim. Biophys. Acta
Biomembr. 1978, 511, 397 – 407; b) D. S. Orlov, T. Nguyen, R. I.
Lehrer, J. Microbiol. Methods 2002, 49, 325 – 328.
N. S. Simpkins, G. T. Furst, J. Am. Chem. Soc. 1988, 110, 4660 –
4672; b) K. C. Nicolaou, R. A. Daines, Y. Ogawa, T. K. Chakra-
borty, J. Am. Chem. Soc. 1988, 110, 4696 – 4705.
[8] a) R. L. Halcomb, S. J. Danishefsky, J. Am. Chem. Soc. 1989, 111,
6661 – 6666; b) D. Kahne, S. Walker, Y. Cheng, D. Van Engen, J.
Am. Chem. Soc. 1989, 111, 6881 – 6882; c) G. K. Packard, S. D.
Rychnovsky, Org. Lett. 2001, 3, 3393 – 3396.
[9] a) N. K. Kochetkov, A. Y. Khorlin, A. F. Bochkov, Tetrahedron
1967, 23, 693 – 707; b) for a recent review, see: F. Kong,
Carbohydr. Res. 2007, 342, 345 – 373.
[10] A related approach was recently published: H. Yu, D. L.
Williams, H. E. Ensley, Tetrahedron Lett. 2005, 46, 3417 – 3421.
[11] Glycoside tosylates are capable of glycosidating simple alcohols:
R. Eby, C. Schuerch, Carbohydr. Res. 1974, 34, 79 – 90.
[12] A mixture of epimers is formed during the azide reduction. The
saponification of methyl esters of acetonide-protected deriva-
tives of natural amphotericin B (1) leads to epimerization, in
contrast to what is observed for amphotericin B methyl ester.
[15] a) A. Zumbuehl, P. Stano, M. Sohrmann, T. Vigassy, E. Pretsch,
M. Peter, P. Walde, D. Jeannerat, E. M. Carreira, Angew. Chem.
2004, 116, 5293 – 5297; Angew. Chem. Int. Ed. 2004, 43, 5181 –
5185; Angew. Chem. 2004, 116, 5293 – 5297; b) A. Zumbuehl, P.
Stano, M. Sohrmann, M. Peter, P. Walde, E. M. Carreira, Org.
Biomol. Chem. 2007, 5, 1339 – 1342; c) V. Paquet, A. Zumbuehl,
E. M. Carreira, Bioconjugate Chem. 2006, 17, 1460 – 1463; d) A.
Zumbuehl, P. Stano, D. Heer, P. Walde, E. M. Carreira, Org. Lett.
2004, 6, 3683 – 3686.
[16] The width of the phospholipid bilayer of these LUVs is 37 Š and
thus any formation of single-barrel ion channels should be
precluded. See the Supporting Information for additional
measurements.
4342
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 4339 –4342