Total Synthesis of Aurantoside G, an N-β-Glycosylated 3-Oligoenoyltetramic Acid from Theonella swinhoei

Article abstract of DOI:10.1002/anie.201604912

The first synthesis of a natural N-glycosylated 3-acyltetramic acid is reported. Aurantoside G (1 g), a deep-red metabolite of the marine sponge Theonella swinhoei, is highly delicate in the pure state. It features a chlorinated dodecapentaenoyl side chain at an l-asparagine-derived tetramic acid, the ring nitrogen atom of which is linked to a β-configured d-xylose. The side chain was built through consecutive Wittig and HWE reactions and used to N-acylate the amino group of an asparaginate that had already been N-xylosylated through a Fukuyama–Mitsunobu reaction. This N-acylation step fixes the β-configuration of the xylose, which is essential for the antifungal activity, but only if the sugar carries bulky, electron-rich protecting groups such as PMB. In the final step, the heterocycle was closed quantitatively through a basic Lacey–Dieckmann condensation of an entirely unprotected precursor.

Full text of DOI:10.1002/anie.201604912

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201604912
German Edition: DOI: 10.1002/ange.201604912
Natural Products
Total Synthesis of Aurantoside G, an N-b-Glycosylated 3-
Oligoenoyltetramic Acid from Theonella swinhoei
Markus Petermichl and Rainer Schobert*
Abstract: The first synthesis of a natural N-glycosylated 3-
acyltetramic acid is reported. Aurantoside G (1g), a deep-red
metabolite of the marine sponge Theonella swinhoei, is highly
delicate in the pure state. It features a chlorinated dodeca-
pentaenoyl side chain at an l-asparagine-derived tetramic acid,
the ring nitrogen atom of which is linked to a b-configured d-
xylose. The side chain was built through consecutive Wittig and
HWE reactions and used to N-acylate the amino group of an
asparaginate that had already been N-xylosylated through
a Fukuyama–Mitsunobu reaction. This N-acylation step fixes
the b-configuration of the xylose, which is essential for the
antifungal activity, but only if the sugar carries bulky, electron-
rich protecting groups such as PMB. In the final step, the
heterocycle was closed quantitatively through a basic Lacey–
Dieckmann condensation of an entirely unprotected precursor.
between 2 and 10 mgmLꢀ1. The aurantosides F, H, and J are
virtually inactive and aurantoside C has not been tested yet.
The mode of antifungal action might be related to that of the
polyene macrolides of the nystatin type. The few tests for
antimicrobial activity have mostly been disappointing.
Natural 3-acyltetramic acids (3-acylpyrrolidine-2,4-diones)
are hybrid polyketide/amino acid metabolites that are pro-
duced by bacteria, molds, fungi, and sponges, and which show
a high incidence and broad spectrum of biological activities.[1]
Owing to the flexibility of the polyketide biosynthesis
machinery, they come in many structural variants and
complexities. Some of them include sugar residues. While
the majority of these compounds are O-glycosylated at the
end of the 3-acyl side-chain, such as the epicoccamides[2] and
ancorinosides,[3] others feature C-glycosylation at this posi-
tion, like the aflastatins,[4] or N-glycosylation at the lactam
nitrogen atom, as in the aurantosides[5] and rubrosides.[6] Only
a few total syntheses of glycotetramates, and none of N-
glycosylated tetramic acids, have been reported, so far.[7] All
known aurantosides A–K (1a–k) have been isolated as
orange-red pigments from lithistid marine sponges of the
genera Theonella (1a,[5a,b] 1b,[5a] 1g–i,[5d] 1j),[5e] Homophymia
(1c),[5c] Siliquariaspongia (1d–f),[5f] or Melophlus (1k).[5g]
They share an asparagine-derived tetramic acid that carries
a mono- or dichlorinated polyenoyl residue at C-3, which is N-
glycosylated with either d-xylose or di- and trisaccharides
comprised of d-xylose, d-arabinose, and 5-deoxyarabinofur-
anose. Some aurantosides show a distinct antifungal effect
that is not clearly related to their structures. Aurantosides A,
B, E and I are efficacious against Candida albicans (wildtype)
at minimal inhibitory concentrations (MIC) of 1 mgmLꢀ1 or
below, while aurantosides D, G and K show MIC values of
Any synthetic approach to the aurantosides has to address
the issues of their inherent instability,[5e] and the necessity to
control the b-configuration of the N-glycosidic linkage. The
latter seems to play a role in the antifungal effect, which is
apparent from the activity difference between the epimers 1g
and 1j. We now developed a route to aurantoside G (1g), the
key steps of which should also be applicable to the syntheses
of its more complex congeners. Our retrosynthetic approach
is outlined in Scheme 1. To avoid problems associated with
the polarity and metal-chelating propensity of 3-acyltetramic
acids, this functionality was to be generated only in a final
base-induced Lacey–Dieckmann[8] cyclization of the fully
functionalized N-(d-xylosyl)-N-(b-ketotetradecapentaenoy-
l)asparaginate. The latter should be prepared by aminolysis
of thioester 2, as the source of the entire 3-acyl side-chain,
with methyl N-d-xylosylasparaginate (3) according to
a method reported by Ley et al.[9] We intended to build
thioester 2 through HWE olefination of the known b-
ketophosphonate 4 with the chlorinated tetraenal 5, which
should be accessible through a Wittig olefination of carboe-
thoxymethylenetriphenylphosphorane with the aldehyde 3-
[*] M. Sc. M. Petermichl, Prof. Dr. R. Schobert
Organic Chemistry Laboratory, University Bayreuth
Universitaetsstr. 30, 95447 Bayreuth (Germany)
E-mail: Rainer.Schobert@uni-bayreuth.de
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/anie.201604912.
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Scheme 2. Synthesis of 3’. Reagents and conditions: a) K2CO3, MeI,
DMF, 08C!RT, 1 h; 99%. b) piperidine, DMF, RT, 1 h; 99%. c) oNsCl,
NEt3, CH2Cl2, 08C!RT, 16 h; 92%. d) F3CCO2H, CH2Cl2, 08C!RT,
1 h; 89%. e) allylic alcohol, BF3 ꢀOEt2, reflux, 16 h; 77%. f) PMBCl,
NaH, DMF, 08C!RT, 72 h; 86%. g) KOtBu, DMF, 708C, 2 h. h) 0.1m
HCl, acetone, reflux, 1 h; 87% over 2 steps. i) DIAD, PPh3, THF,
ꢀ788C!RT, 16 h; 69%. j) DIPEA, PhSH, DMF, RT, 4 h; 77%.
DMF=N,N-dimethylformamide, DIAD=diisopropyl azodicarboxylate,
DIPEA=N,N-diisopropylamine, Fmoc=fluorenylmethyloxycarbonyl,
Trt =triphenylmethyl.
the corresponding enol ether 15, which was hydrolyzed right
away with hydrochloric acid to leave building block 8’
(Scheme 2, middle row).
Scheme 1. Retrosynthetic approach to aurantoside G (1g). oNs=o-
The Fukuyama–Mitsunobu reaction[14] of 8’ with sulfona-
mide 7 proceeded fast and with complete consumption of the
starting materials. The N-nosylated product amine 16 was
isolated in 69% yield as the pure a-anomer, as apparent from
the NMR spectra. The chemical shifts of the C-1 of the xylose
differed by ca. Dd = 20 ppm for a- and b-anomers, and the
coupling constants 3JHH for 1-H were 3–4 Hz for the a-anomer
and 7–8 Hz for the b-anomer. De-nosylation of 16 with
thiophenol and Hꢀnigꢁs base afforded the key amine 3’ (P =
PMB) in 77% yield, albeit as a 1.5:1.0 mixture of a- and b-
anomers (Scheme 2, bottom).
(O2N)C6H4SO2.
chlorobut-(2Z)-enal, ideally prepared in situ through oxida-
tion of alcohol 6. The N-xylosylasparaginate 3 was to be
prepared by a Fukuyama–Mitsunobu reaction of a tris-
protected d-xylose 8 with N-nosylated methyl asparaginate
7. It was unclear at this point whether the bulky o-nosyl group
would give rise to a- or b-glycosylation and whether such
a preferred configuration might be preserved throughout the
synthesis.
We started out with the synthesis of the unknown methyl
N-(o-nosyl)asparaginate 7 from commercially available bis-
protected asparagine 9 in four steps and 80% overall yield.[10]
The known intermediates, that is, the bisprotected methyl
ester 10 and the a-aminoester 11, were obtained for the first
time in a pure crystalline form rather than as oils[11]
(Scheme 2, top row).
It took some experimentation to identify p-methoxyben-
zyl (PMB) as the optimum protecting group for xylose. PMB
as an electron-releasing group enables the Fukuyama–Mitsu-
nobu reaction of d-xylose with the electron-poor, weakly
nucleophilic amine 7. Moreover, PMB is bulky enough to
direct the subsequent N-acylation of 3 with 2 in favor of the b-
configuration. The new PMB-protected xylose 8’ was pre-
pared from d-xylose in four steps and 58% yield as a 1.7:1.0
mixture of a- and b-anomers. Selective 1-O-allylation[12] of d-
xylose and subsequent benzylation of the resulting glycoside
13 with PMBCl furnished the fully protected xylose 14. The
allyl ether was cleaved[13] by first isomerizing it with base to
Next, thioester 2 was synthesized starting from 3-chlor-
obut-(2Z)-en-1-ol (6), which was prepared through the
reaction of but-2-yn-1-ol (17) with Red-Al and NCS, accord-
ing to an improved adaptation of the method of Poulter
et al.[15] (Scheme 3). We began the stepwise chain elongation
of alcohol 6 by employing a domino oxidation/Wittig
olefination, as described by Taylor et al.,[16] using MnO2 and
carboethoxymethylenetriphenylphosphorane (18), since the
corresponding aldehyde of 6 is rather delicate. The product
ethyl dienoate 19 was reduced to the alcohol 20, and the chain
was lengthened in the same manner by domino oxidation/
Wittig reaction with MnO2 and ylide 18 to give the ethyl
trienoate 21. After a third sequence of reduction and domino
oxidation/Wittig reaction, the ethyl tetraenoate 23 was
obtained. It was reduced to the corresponding alcohol 24,
which was finally oxidized with MnO2 to the desired aldehyde
5. Because of its instability, even when kept under an inert gas
atmosphere in a dark freezer at ꢀ208C, this aldehyde was
submitted right away to a HWE reaction with Leyꢁs S-tert-
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
necessary to completely convert 3’ into b-ketoamide 25. Since
all aurantosides except 1j feature a b-xylose bearing a bulky
arabinose at C-2, the above route to aurantoside G (1g)
should also be applicable to the b-selective synthesis of these
congeners. Cleavage of the PMB groups of 25 with trifluoro-
acetic acid afforded the unprotected N-glycosylated b-ketoa-
mide 26 as a separable 1:1 mixture of keto and enol tautomers.
Only the keto tautomer 26a underwent Lacey–Dieckmann
cyclization which, when stopped after 10 min reaction time,
furnished the chemically and enantiomerically pure auranto-
side G (1g) in virtually quantitative yield. The enol tautomer
26b needed to be re-equilibrated with acid to give the initial
1:1 mixture of both epimers. Like other polyenoyltetramic
acids such as b-lipomycin,[5e,7f] pure synthetic aurantoside G
proved unstable and prone to decomposition, which thwarted
further biological tests.
In summary, the sponge metabolite aurantoside G (1g)
was synthesized as the first example of an N-glycosylated 3-
acyltetramic acid in 3.7% overall yield. The b-configuration
of the d-xylose residue, which is essential for its bioactivity,
was fixed only in the course of the penultimate Ley-type N-
acylation step. The b-selectivity is governed by the bulky
PMB substituents on the xylose. The electron-releasing effect
of the PMB residues was also exploited to promote the
preceding Fukuyama–Mitsunobu glycosylation step. We
expect this synthetic approach to be applicable also to the
other aurantosides, which all bear bulky mono- or oligosac-
charides at C-2 of a b-xylose. We also demonstrated that
Lacey–Dieckmann cyclization may serve as the final step in
the synthesis of even delicate tetramic acids, since it does not
lead to racemization or interfere with extended conjugated
polyene fragments or unprotected sugars.
Scheme 3. Synthesis of 2. Reagents and conditions: a) (i) Red-Al, THF,
08C!RT, 16 h; (ii) NCS, THF, ꢀ788C!08C, 3 h; 82%. b) MnO2,
Ph3P=CHCO2Et (18), CH2Cl2, reflux, 16 h; 63%. c) DIBAL-H, CH2Cl2,
ꢀ788C, 1 h; 75%. d) MnO2, 18, CH2Cl2, reflux, 16 h; 77%. e) DIBAL-
H, CH2Cl2, ꢀ788C, 1 h; 94%. f) MnO2, 18, CH2Cl2, reflux, 16 h; 60%.
g) DIBAL-H, CH2Cl2, ꢀ788C, 1 h; 77%. h) MnO2, CH2Cl2, RT, 3 h.
i) NaH, THF, 08C, 3 h; 84% over 2 steps. Red-Al=sodium bis(2-
methoxyethoxy)aluminumhydride, THF=tetrahydrofuran, DIBAL-
H=diisobutylaluminumhydride.
butyl 4-(diethylphosphono)-3-oxobutanethioate (4)[17] to
afford the thioester 2 in 84% yield (11% over nine steps).
The aminolysis of thioester 2 with a 1.5:1.0 mixture of a-
and b-anomers of methyl N-d-xylosylasparaginate 3’ in the
presence of an excess of silver trifluoroacetate according to
Leyꢁs general method[7a,9] afforded the b-ketoamide 25 as
a pure b-isomer in 49% yield with respect to recovered
unreacted 3’ (Scheme 4). We assume that only the b-anomer
Acknowledgment
We thank Dr. Sebastian Loscher (University Bayreuth) for
helpful discussions on the retrosynthesis.
Keywords: aurantosides · glycoconjugates · natural products ·
tetramic acids · total synthesis
[1] Reviews on tetramic acids: a) H.-G. Henning, A. Gelbin, Adv.
Heterocycl. Chem. 1993, 57, 139 – 185; b) B. J. L. Royles, Chem.
Rev. 1995, 95, 1981 – 2001; c) E. L. Ghisalberti in Studies in
Natural Products Chemistry, Vol. 28/1 (Ed.: Atta-ur-Rahman),
Elsevier, Dordrecht, 2003, pp. 109 – 163; d) “Monopyrrolic nat-
ural compounds including tetramic acid derivatives”: A. Gossa-
uer in Progress in the Chemistry of Organic Natural Products,
Vol. 86 (Eds.: W. Herz, H. Falk, G. W. Kirby), Springer, Wien,
2003, pp. 1 – 188; e) R. Schobert, Naturwissenschaften 2007, 94,
1 – 11; f) R. Schobert, A. Schlenk, Bioorg. Med. Chem. 2008, 16,
4203 – 4221.
[2] a) A. D. Wright, C. Osterhage, G. M. Kçnig, Org. Biomol. Chem.
2003, 1, 507 – 510; b) H. V. K. Wangun, H.-M. Dahse, C.
Hertweck, J. Nat. Prod. 2007, 70, 1800 – 1803.
[3] a) S. Ohta, E. Ohta, S. Ikegami, J. Org. Chem. 1997, 62, 6452 –
6453; b) M. Fujita, Y. Nakao, S. Matsunaga, M. Seiki, Y. Itoh,
Scheme 4. Final steps to aurantoside G (1g): b-selective Ley N-(b-
keto)acylation to give 25, followed by deprotection and Lacey–Die-
ckmann cyclization.
of 3’ enters into a reaction with 2 owing to steric shielding of
the a-anomer by the PMB groups at the xylose, and that
epimerization of the residual a-anomer is not fast enough
under the reaction conditions. This epimerization definitely
took place upon the recovery (extraction, concentration) of
unreacted 3’, which again yielded a mixture of both anomers,
ready to be reacted once more. Several reaction cycles are
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
R. W. M. van Soest, N. Fusetani, Tetrahedron 2001, 57, 1229 –
1234; c) E. Ohta, S. Ohta, S. Ikegami, Tetrahedron 2001, 57,
4699 – 4703.
A. Mori, R. Katsuta, T. Nukada, Tetrahedron Lett. 2014, 55,
4350 – 4354; e) M. L. Hofferberth, R. Brꢀckner, Angew. Chem.
Int. Ed. 2014, 53, 7328 – 7334; Angew. Chem. 2014, 126, 7456 –
7462; f) O. Hartmann, M. Kalesse, Angew. Chem. Int. Ed. 2014,
53, 7335 – 7338; Angew. Chem. 2014, 126, 7463 – 7466.
[8] a) R. N. Lacey, J. Chem. Soc. 1954, 832 – 839; b) R. N. Lacey, J.
Chem. Soc. 1954, 850 – 854.
[4] a) M. Ono, S. Sakuda, A. Suzuki, A. Isogai, J. Antibiot. 1997, 50,
111 – 118; b) M. Ono, S. Sakuda, H. Ikeda, K. Furihata, J.
Nakayama, A. Suzuki, A. Isogai, J. Antibiot. 1998, 51, 1019 –
1028.
[5] a) S. Matsunaga, N. Fusetani, Y. Kato, J. Am. Chem. Soc. 1991,
113, 9690 – 9692; b) E. W. Schmidt, M. K. Harper, D. J. Faulkner,
J. Nat. Prod. 1997, 60, 779 – 782; c) D. Wolf, F. J. Schmitz, F. Qiu,
M. Kelly-Borges, J. Nat. Prod. 1999, 62, 170 – 172; d) A. S.
Ratnayake, R. A. Davis, M. K. Harper, C. A. Veltri, C. D.
Andjelic, L. R. Barrows, C. M. Ireland, J. Nat. Prod. 2005, 68,
104 – 107; e) R. F. Angawi, G. Bavestrello, B. Calcinai, H. A.
Dien, G. Donnarumma, M. A. Tufano, I. Paoletti, E. Grimaldi,
G. Chianese, E. Fattorusso, O. Taglialatela-Scafati, Mar. Drugs
2011, 9, 2809 – 2817; f) N. U. Sata, S. Matsunaga, N. Fusetani,
R. W. M. van Soest, J. Nat. Prod. 1999, 62, 969 – 971; g) R.
Kumar, R. Subramani, K.-D. Feussner, W. Aalbersberg, Mar.
Drugs 2012, 10, 200 – 208.
[9] L. T. Burke, D. J. Dixon, S. V. Ley, F. Rodrꢂguez, Org. Biomol.
Chem. 2005, 3, 274 – 280.
[10] E. Biron, H. Kessler, J. Org. Chem. 2005, 70, 5183 – 5189.
[11] Z. Huang, M. Zhang, S. D. Burton, L. N. Katsakhyan, H. Ji, ACS
Chem. Biol. 2014, 9, 193 – 201.
[12] a) B. K. Gorityala, J. Ma, K. K. Pasunooti, S. Cai, X.-W. Liu,
Green Chem. 2011, 13, 573 – 577; b) T. K. M. Shing, H. M.
Cheng, W. F. Wong, C. S. K. Kwong, J. Li, C. B. S. Lau, P. S.
Leung, C. H. K. Cheng, Org. Lett. 2008, 10, 3145 – 3148.
[13] F. Zhang, A. Vasella, Carbohydr. Res. 2007, 342, 2546 – 2556.
[14] a) T. Fukuyama, C.-K. Jow, M. Cheung, Tetrahedron Lett. 1995,
36, 6373 – 6374; b) C. Huo, C. Wang, M. Zhao, S. Peng, Chem.
Res. Toxicol. 2004, 17, 1112 – 1120.
[6] N. U. Sata, S. Wada, S. Matsunaga, S. Watabe, R. W. M.
van Soest, N. Fusetani, J. Org. Chem. 1999, 64, 2331 – 2339.
[7] a) S. V. Pronin, S. A. Kozmin, J. Am. Chem. Soc. 2010, 132,
14394 – 14396; b) J. J. Beiger, 2013. Total Synthesis of Aflastatin
A. Doctoral dissertation, Harvard University, http://nrs.harvard.
edu/urn-3:HUL.InstRepos:10974701; c) S. Loscher, R. Schobert,
Chem. Eur. J. 2013, 19, 10619 – 10624; d) A. Yajima, A. Kawajiri,
[15] N. A. Heaps, C. D. Poulter, J. Org. Chem. 2011, 76, 1838 – 1843.
[16] X. Wei, R. J. K. Taylor, Tetrahedron Lett. 1998, 39, 3815 – 3818.
[17] S. V. Ley, P. R. Woodward, Tetrahedron Lett. 1987, 28, 345 – 346.
Received: May 19, 2016
Published online: && &&, &&&&
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Natural Products
M. Petermichl,
R. Schobert*
&&&& — &&&&
Total Synthesis of Aurantoside G, an N-b-
Glycosylated 3-Oligoenoyltetramic Acid
from Theonella swinhoei
N-Glycotetramates: The first synthesis of
a naturally occurring N-glycosylated 3-
acyltetramic acid afforded aurantoside G,
a metabolite of the marine sponge Theo-
nella swinhoei, in 3.7% yield over
xylose, which is decisive for the biological
activity, was established through the
directing effect of bulky PMB protecting
groups on the xylose during the Ley-type
N-(b-keto)acylation step.
12 steps. The b-configuration of the d-
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!