An RCM-Based Total Synthesis of the Antibiotic Disciformycin B

Article abstract of DOI:10.1002/anie.202004589

The total synthesis of the potent new antibiotic disciformycin B (2) is described, which shows significant activity against methicillin- and vancomycin-resistant Staphylococcus aureus (MRSA/VRSA) strains. The synthetic route is based on macrocyclization o

Full text of DOI:10.1002/anie.202004589

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: An RCM-Based Total Synthesis of the Antibiotic Disciformycin B
Authors: Karl-Heinz Altmann and Philipp Waser
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202004589
Link to VoR: https://doi.org/10.1002/anie.202004589

10.1002/anie.202004589
Angewandte Chemie International Edition
COMMUNICATION
An RCM-Based Total Synthesis of the Antibiotic Disciformycin B
Philipp Waser,[a] and Karl-Heinz Altmann*[a]
This paper is dedicated to our late friend Maurizio Botta, who left us much too early.
[a]
MSc. P. Waser, Prof. K.-H. Altmann
ETH Zürich
Department of Chemistry and Applied Biosciences
Institute of Pharmaceutical Sciences
HCl H405
Vladimir-Prolog-Weg 4,
CH-8093 Zürich, Switzerland
Supporting information for this article is given via a link at the end of the document.
Abstract: The total synthesis of the potent new antibiotic
disciformycin B (2) is described, which shows significant activity
against methicillin- and vancomycin-resistant Staphylococcus aureus
(MRSA/VRSA) strains. The synthetic route is based on
macrocycle formation by ring-closing alkyne metathesis (RCAM)
between C8 and C9. While this reaction could be accomplished
with remarkable efficiency, the elaboration of the resulting
macrocyclic alkyne into the disciformycin B aglycone proved to be
difficult and less efficient, owing largely to the limited stability of
several intermediates. Fürstner's work also showed that 2 cleanly
isomerizes to the less active 1 under mildly basic conditions.
macrocyclization of
a tetraene substrate to the 12-membered
macrolactone core by ring-closing olefin metathesis (RCM). Although
macrocyclization was accompanied by concomitant cyclopentene
formation by an alternative RCM pathway, conditions could eventually
be established that gave the macrocycle as the major product. Key
steps in the construction of the RCM-substrate included a highly
efficient Evans syn-aldol reaction, the asymmetric Brown allylation of
angelic aldehyde and the stereoselective Zn(BH4)2-mediated 1,2-
reduction of an enone. The synthesis was completed by late-stage
dehydrative glycosylation to introduce the D-arabinofuranosyl moiety
and final chemoselective allylic alcohol oxidation.
Disciformycins A (1) and B (2) are polyketide-derived macrolide
glycosides that were isolated in 2014 by Müller and co-workers
from the myxobacterium Pyxidicoccus fallax strain AndGT8.[1] The
biological assessment of these compounds revealed significant
antibacterial activity against Gram-positive bacteria, including
methicillin- and vancomycin-resistant Staphylococcus aureus
(MRSA/VRSA) strains, with 2 being more active than 1.
Furthermore, the lack of cross-resistance with other antibiotics
classes pointed to a novel mechanism of action for the
disciformycins.[1][2] Concurrently with the disclosure of
disciformycins A (1) and B (2) by Müller and co-workers, Nett and
co-workers reported the isolation of a structurally related pair of
macrolides, gulmirecins A (3) and B (4), from the predatory
bacterium Pyxidicoccus fallax HKI 727. Like disciformycins A (1)
and B (2), the gulmirecins displayed pronounced antibiotic activity
against several strains of Gram-positive bacteria; 3, but not 4, also
showed promising activity against MRSA.[3]
Subsequent to Fürstner's work, Wolling and Kirschning described
an alternative approach towards the disciformycin B aglycone,
employing
Rengarasu and Maier have described the synthesis of a C1-C12
fragment of gulmirecin B (4),[7] and Ichikawa and co-workers have
reported the synthesis of a fully protected version of C5-O dihydro
gulmirecin A. Ichikawa's approach entailed macrocyclization at
C7-C8 by Ni(0)-mediated reductive cyclization of an ynal
intermediate;[8] unfortunately, deprotection of the advanced
gulmirecin A precursor could not be achieved.
a
macrolactonization strategy.[6] More recently,
In the broader context of the ever growing threat posed by
antibiotic-resistant bacteria and the pressing need for the
development of new antibiotics,[4] we considered disciformycins
and gulmirecins important targets for total synthesis. The
chemistry developed in the course of a successful total synthesis
campaign provides the foundation for systematic structure-activity
relationship (SAR) studies and the eventual discovery of analogs
with improved overall profiles for drug development. Similar
considerations have underlain the work of Fürstner and co-
workers, who have reported the only total synthesis of a
disciformycin/gulmirecin so far (disciformycin B (2)),[5] based on
Based on the superior antibiotic activity of disciformycin B (2) over
1, 3, and 4, our own work, as for Fürstner and co-workers, has
focused on disciformycin B (2) as the primary target for total
synthesis. As depicted in Scheme 1, our synthesis was to be
based on macrocyclization at C8-C9 by means of ring-closing
olefin metathesis (RCM) of tetraene 7 as the crucial step.
Obviously, this strategy entailed some risk, due to the possibility
of 5-membered ring formation by RCM between the disubstituted
terminal double bond and the internal double bond in 7
(corresponding to the C3-C4 double bond in 2), which might in
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202004589
Angewandte Chemie International Edition
COMMUNICATION
fact have been expected to be the preferred reaction path.
However, we were hopeful that the desired macrocycle (with a
trans-configured C8-C9 double bond) might still be accessed
using appropriate protecting groups and careful tuning of the
reaction conditions.[9]
mediated esterification[14] of acid 14 with alcohol ent-9 followed by
selective TBS-ether cleavage with CSA in CH2Cl2/MeOH. (For the
synthesis of 14 and ent-9 see the SI).
Scheme 2. a) ent-9, DCC, DMAP, CH2Cl2, –30 °C, 25 h, 77%; dr 92:8; b) CSA,
CH2Cl2/MeOH (1:1), rt, 8.5 h, 68% (80% based on recovered 15), dr 92:8;
c) 1 mM 16, Grubbs II (60 mol%), benzene, 80 °C, 45 h, 17: ca. 21% (E/Z >
95:5, the material contained minor inseparable impurities), 18: 48%. CSA = (±)-
camphor-10-sulfonic acid, DCC = N,N’-dicyclohexylcarbodiimide, DMAP = 4-
dimethylaminopyridine, PMB
= p-methoxybenzyl, TBDPS = tert-butyl-
diphenylsilyl, TBS = tert-butyldimethylsilyl.
While tetraene 16 did undergo RCM to the desired macrocycle 17,
the latter was obtained in a disappointingly low yield of only 21%
at best (Scheme 2); the major product of the reaction was in fact
cyclopentene 18 that could be isolated in 48% yield. However,
differences in RCM efficiency between diastereoisomers of the
same diene have been described in the literature,[15] which
encouraged us to investigate the effect of inverting the
configuration of the C5 stereocenter in 16 from S to R on RCM
efficiency, hence leading to 7 as an alternative RCM substrate
(Scheme 1). As a second issue in our first generation approach
towards 2, the esterification of 14 with ent-9 was accompanied by
significant epimerization and even under optimized conditions
(1.3 equiv. DCC, 0.05 equiv. DMAP, CH2Cl2, –30 °C, 25 h), the
ester product was obtained in an inseparable mixture with ca. 8%
of its C2-epimer (disciformycin numbering). As a consequence,
the assembly of diene 7 was planned to entail Mitsunobu
esterification of acid 8 and alcohol 9 (Scheme 1).
Following the provisions of the retrosynthesis summarized in
Scheme 1, the synthesis of carboxylic acid building block 8
departed from N-acyl oxazolidinone 13.[13] As depicted in Scheme
3, the reaction of methacrolein with the dibutylboron enolate
derived from 13 afforded syn-aldol product 19 in excellent yield
(97%) and with high diastereoselectivity (dr > 95:5). Imide 19 was
transformed into the desired Weinreb amide 11 by reaction with
AlMe3 and MeNH(OMe)·HCl and subsequent TBS-protection of
the secondary hydroxy group in 78% overall yield. Reaction of 11
with the vinyllithium species derived from vinyl iodide 12[12] by
iodine-lithium exchange with tBuLi at –78 °C then furnished
ketone 20 in high yield (85%). The latter could be reduced with
freshly prepared Zn(BH4)2[16] at –78 °C to –10 °C, to deliver allylic
alcohol 21 in 75% yield and with excellent diastereoselectivity (dr
> 95:5). The reaction appeared to be scale-dependent, with yields
of 21 of 54%, 75%, and 65% on a 50 mg, 700 mg, and 2 g scale
of 20, respectively. Protection of the newly formed secondary
Scheme 1. Retrosynthesis of disciformycin B (2). PMB = p-methoxybenzyl,
TBDPS = tert-butyldiphenylsilyl, TBS = tert-butyldimethylsilyl.
The RCM-product 6 was to be elaborated into disciformycin B (2)
by -selective glycosylation, cleavage of the PMB-ether at C6,
esterification, global deprotection, and, in the final step,
establishment of the C3-C5 enone moiety by chemoselective
oxidation of the allylic hydroxy group.[10] Formation of the enone
moiety in the final step would eliminate any potential difficulties
arising from the reactivity of this group and/or double bond
migration.[5] Tetraene 7 was to be prepared by Mitsunobu
esterification of acid 8 and alcohol 9 (vide infra); the latter was
envisioned to be accessible from angelic aldehyde (10)[11] by
means of asymmetric allylation. Carboxylic acid 8 would be
obtained from known vinyl iodide 12[12] and Weinreb amide 11 by
addition-elimination followed by diastereoselective carbonyl
reduction. Weinreb amide 11 was to be prepared from N-acyl
oxazolidinone 13[13] by Evans syn-aldol reaction with methacrolein.
The retrosynthesis depicted in Scheme 1 includes some specific
learnings from a prior approach towards 2 that had included
tetraene 16 as the substrate for RCM-based macrocyclization
(Scheme 2). The latter had been obtained by DCC/DMAP-
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202004589
Angewandte Chemie International Edition
COMMUNICATION
hydroxy group as a TBDPS-ether, to deliver 22 in 97% yield,
enabled selective liberation of the primary hydroxy group by TBS-
ether cleavage with HF•pyridine (81% yield). The ensuing primary
alcohol was then converted into the desired carboxylic acid 8 by
treatment with Dess-Martin periodinane[17] followed by oxidation
of the ensuing (crude) aldehyde under Pinnick-Kraus
conditions[18] in excellent overall yield (89%).
Scheme 4. a) LiAlH4, Et2O, 0 °C to rt, 24 h; b) MnO2, CH2Cl2, rt, 16 h, Z/E >
95:5; c) (-)-Ipc2Ballyl, Et2O, -100 °C, 1 h, 52% (three steps), Z/E > 95:5, > 90%
ee; d) 8, DEAD, PPh3, THF, 0 °C to rt, 4 h, 83%, dr > 95:5; e) CSA,
CH2Cl2/MeOH (1:1), rt, 2 h, 57% (82% based on recovered 24); f) 1 mM 7,
Grubbs II (75 mol%), benzene, 80 °C, 6 h, 6: ca. 37% (E/Z > 95:5, contains <
10% linear dimer 25), 26: 31%. CSA = (±)-camphor-10-sulfonic acid, DEAD =
diethyl azodicarboxylate, Ipc = isopinocampheyl, PMB = p-methoxybenzyl,
TBDPS
= tert-butyldiphenylsilyl, TBS = tert-butyldimethylsilyl, THF =
tetrahydrofuran.
Under optimized conditions (i. e. 1 mM 7, 75 mol% catalyst added
portionwise, benzene, 80 °C, 6 h), the desired macrocycle 6 was
obtained in ca. 37% yield; the isolated material contained < 10%
of linear dimer 25 that could not be removed at this point.[24] In
addition, cyclopentene 26 was isolated in 31% yield.
The protected aglycone 6 was then submitted to dehydrative
glycosylation[25] with TBS-protected D-arabinofuranose 27
(prepared from D-arabinose in 2 steps)[26] (Scheme 5). Thus,
treatment of 6 with Ph2SO, Tf2O, and TTBP in CH2Cl2/toluene
(1:10) furnished the desired α-anomer 28 in 60% isolated yield
(with an α/β selectivity of the reaction of 3:1).[27] DDQ-mediated
cleavage of the PMB-ether, followed by esterification of the
ensuing secondary alcohol with isovaleroyl chloride gave the fully
protected intermediate 5 in 51% overall yield from 28.
Not unexpectedly, the subsequent removal of the four silyl
protecting groups proved to be a significant challenge, due to
facile migration of the isovaleroyl residue to the allylic hydroxy
group. Best results were obtained by treatment of 5 with
HF•pyridine in THF/pyridine (1:1) at room temperature for 24 h;
importantly, the reaction needed to be terminated before
complete conversion was reached (i.e. before complete removal
of the more stable TBDPS-group from 30) and a slightly acidic
quench (SiO2) was crucial, as the addition of base caused further
undesired migration. Within these confines, the desired tetrol 31
could be isolated in 51% yield after flash column chromatography
on silica, together with its regioisomer 32 (26%) and TBDPS-
protected intermediate 30 (16%). The undesired ester 32 could
be equilibrated to a mixture of 31 and 32 in a 1:2 ratio (based on
1H-NMR analysis of the crude product mixture) by stirring in
THF/aqueous saturated KHCO3 (1:1), which allows for partial
recycling.
Scheme 3. a) i. 13, Bu2BOTf, NEt3, toluene, –50 °C, 1.5 h; ii. methacrolein,
–78 °C to 0 °C, 2 h, 97%, dr > 95:5; b) i. MeNH(OMe)·HCl, AlMe3, THF, 0 °C
to rt, 30 min; ii. 19, –20 °C to 0 °C, 3 h; c) TBSOTf, 2,6-lutidine, CH2Cl2, 0 °C,
30 min, 78% (two steps); d) i. 12, tBuLi, Et2O, –78 °C, 1 h; ii. 11, –78 °C to
–50 °C, 2 h, 85%, E/Z > 95:5; e) Zn(BH4)2, Et2O, –78 °C to –10 °C, 40 min,
75% (710 mg 20), dr > 95:5; f) TBDPSCl, imidazole, DMAP, CH2Cl2, rt to 50 °C,
18 h, 97%; g) HF·pyridine, THF, 0 °C to 4 °C, 43 h, 81%; h) Dess-Martin
periodinane, NaHCO3, CH2Cl2, rt, 1 h; i) NaClO2, NaH2PO4, 2-methyl-2-
butene/tBuOH (1:1), H2O,
trifluoromethanesulfonyl, THF
0
°C, 1.5 h, 89% (two steps). Tf
tetrahydrofuran, DMAP
=
4-
=
=
dimethylaminopyridine, Ph = phenyl, PMB = p-methoxybenzyl, TBDPS = tert-
butyldiphenylsilyl, TBS = tert-butyldimethylsilyl.
As depicted in Scheme 4, treatment of angelic aldehyde (10)
(obtained from commercially available methyl angelate (23) by
LiAlH4 reduction followed by oxidation of the ensuing alcohol with
activated MnO2 in CH2Cl2)[11] with (-)-Ipc2Ballyl at –100 °C[19] gave
alcohol 9 in good yield (52% overall from 23)[20] with high
enantioselectivity
(> 90% ee)
and
without
noticeable
isomerization of the double bond (Z/E > 95:5). Esterification of
alcohol 9 with carboxylic acid 8 under Mitsunobu conditions[21]
proceeded smoothly to afford tetraene 24 in 83% yield with
complete inversion of configuration (dr > 95:5). In order to prevent
impairment of the subsequent metathesis reaction by the
presence of a bulky substituent adjacent to one of the double
bonds,[22] 24 was then converted into the free secondary alcohol
7
by selective cleavage of the TBS-ether with CSA in
CH2Cl2/MeOH (1:1) in 57% yield (82% based on recovered 24).
With tetraene 7 in hand, the stage was set to investigate the
crucial RCM step. These experiments were focused on optimizing
the ring-closure with Grubbs II catalyst,[23] which had produced
reasonably promising results in initial experiments (see the SI).
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202004589
Angewandte Chemie International Edition
COMMUNICATION
macrocyclic ring-closure by RCM of the multireactive tetraene 7.
The reaction delivered the desired macrocycle 6 as the major
product in 37% yield, closure of the macrocycle thus being
favored over competing cyclopentene formation. Importantly,
macrocyclization was significantly more efficient with 7 than its
diastereomer 16.[30] Other key steps in the synthesis of 2 included
(a) the stereoselective Brown allylation of isomerization-prone
angelic aldehyde (10); (b) the stereoselective reduction of ketone
20, to establish the more favorable configuration at C5
(disciformycin numbering) for RCM-mediated macrocyclization;
(c) the dehydrative glycosylation of the C7-hydroxy group with tris-
TBS-protected D-arabinofuranose; and, finally, (d) the mild
chemoselective oxidation of allylic alcohol 31 to install the
sensitive enone functionality. Our route allows for late-stage
derivatization of the macrocycle, which has provided access to a
number of analogs of disciformycin B (2) for SAR studies. The
results of these studies will be reported in due course.
Acknowledgements
We are indebted to Dr. B. Pfeiffer and Dr. L. Betschart for NMR
support, to Dr. X. Zhang, L. Bertschi, R. Häfliger, and O. Greter
for HRMS spectra, and to Kurt Hauenstein for general technical
support. Generous funding by ETH Zurich is gratefully
acknowledged. We thank Prof. Rolf Müller, Helmholtz Institute of
Pharmaceutical Research Saarland (HIPS), Saarbrücken,
Germany, for providing a sample of natural disciformycin B.
Keywords: disciformycin • macrocycle • natural products • ring-
closing metathesis • total synthesis
[1]
[2]
F. Surup, K. Viehrig, K. I. Mohr, J. Herrmann, R. Jansen, R. Müller,
Angew. Chem. Int. Ed. 2014, 53, 13588–13591; Angew. Chem. 2014,
126, 13806–13809.
The molecular target of 1 and 2 has been suggested to be the bacterial
RNA polymerase: J. Herrmann, O. Kalinina, F. Surup, R. Müller.
Antibacterial disciformycins from myxobacteria as novel RNA
polymerase inhibitors. Presented at the 2nd European Conference on
Natural Products, Frankfurt/Main, Germany, September 6-9, 2015.
S. Schieferdecker, S. König, C. Weigel, H.-M. Dahse, O. Werz, M. Nett,
Chem. Eur. J. 2014, 20, 15933–15940.
Scheme 5. a) 27 (/ = 2.5:1), Ph2SO, Tf2O, TTBP, 3Å MS, CH2Cl2/toluene
(1:10), –78 °C to rt, overnight, 60% (28, α-anomer), α/β = 3:1; b) DDQ,
CH2Cl2/pH 7 buffer (6:1), rt, 22 h, 79%; c) isovaleroyl chloride, pyridine, DMAP,
CH2Cl2, 50 °C, 14 h, 65% after recrystallization; d) HF·pyridine, THF/pyridine
(1:1), 0 °C to rt, 24 h, 51% (31, 61% based on recovered 30), 26% (30), 16%
[3]
[4]
US Department of Health, Center for Disease Control (CDC). Antibiotic
(32);
e)
4-acetylamino-2,2,6,6-tetramethyl-piperidine-1-oxo-ammonium
Resistance
Threats
in
the
United
States
2019.
DOI:
tetrafluoro-borate, SiO2, CH2Cl2, rt, 1 h, 71%. DDQ = 2,3-dichloro-5,6-cyano-p-
benzoquinone, DMAP = 4-dimethylaminopyridine, MS = molecular sieves, Tf2O
= trifluoromethanesulfonic anhydride, THF = tetrahydrofuran, TTBP = 2,4,6-tri-
tert-butyl-pyridine. TBS = tert-butyldimethylsilyl, TBDPS = tert-butyldiphenylsilyl,
PMB = p-methoxybenzyl
http://dx.doi.org/10.15620/cdc:82532
[5]
Y. Kwon, S. Schulthoff, Q. M. Dao, C. Wirtz, A. Fürstner, Chem. Eur. J.
2018, 24, 109–114.
[6]
[7]
[8]
[9]
M. Wolling, A. Kirschning, Eur. J. Org. Chem. 2018, 648–656.
R. Rengarasu, M. E. Maier, Synlett 2019, 30, 1346–1350.
S. Kitahata, A. Katsuyama, S. Ichikawa, Org. Lett. 2020, 22, 2697–2701.
For RCM-based macrocyclizations in competition with the formation of
small rings, see, e. g.: a) D. J. Dixon, A. C. Foster, S. V. Ley, Org. Lett.
2000, 2, 123–125; b) A. Rivkin, Y. S. Cho, A. E. Gabarda, F. Yoshimura,
S. J. Danishefsky, J. Nat. Prod. 2004, 67, 139–143.
Finally, the allylic hydroxy group in tetrol 31 was chemoselectively
oxidized with 4-acetylamino-2,2,6,6-tetramethyl-piperidine-1-
oxoammonium tetrafluoroborate (Bobbitt's salt)[28] to afford
disciformycin B (2) in 71% yield (Scheme 5).[29] Notably, the enone
functionality was quickly generated (1 h at room temperature) in
the presence of three unprotected sugar hydroxy groups and
purification of the reaction mixture included simple SiO2 column
chromatography. No isomerization to the less potent
disciformycin A (1) was observed under these conditions.
[10] For examples, see a) T. Zöllner, P. Gebhardt, R. Beckert, C. Hertweck,
J. Nat. Prod. 2005, 68, 112–114; b) C. K. Hill, J. F. Hartwig, Nat. Chem.
2017, 9, 1213–1221.
[11] U. Vögeli, W. von Philipsborn, Org. Magn. Reson. 1975, 7, 617–627.
[12] The synthesis of 12 followed the approach described for ent-12 in: N.
Fleary-Roberts, G. Lemière, J. Clayden, Tetrahedron 2015, 71, 7204–
7208.
In conclusion, we have completed a convergent total synthesis of
the potent antibiotic disciformycin B (2) that was built around
[13] For the synthesis of ent-13 see: D. A. Evans, J. R. Gage, J. L. Leighton,
A. S. Kim, J. Org. Chem. 1992, 57, 1961–1963.
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202004589
Angewandte Chemie International Edition
COMMUNICATION
[14] B. Neises, W. Steglich, Angew. Chem. Int. Ed. 1978, 17, 522–524;
Angew. Chem. 1978, 90, 556–557.
[15] a) J. Jin, Y. Chen, Y. Li, J. Wu, W.-M. Dai, Org. Lett. 2007, 9, 2585–2588;
b) L. Caggiano, D. Castoldi, R. Beumer, P. Bayón, J. Telser, C. Gennari,
Tetrahedron Lett. 2003, 44, 7913–7919.
[16] W. J. Gensler, F. A. Johnson, A. D. B. Sloan, J. Am. Chem. Soc. 1960,
82, 6074–6081.
[17] D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155–4156.
[18] a) B. O. Lindgren, T. Nilsson, Acta Chem. Scand. 1973, 27, 888–890; b)
G. A. Kraus, M. J. Taschner, J. Org. Chem. 1980, 45, 1175–1176; c) B.
S. Bal, W. E. Childers, Jr., H. W. Pinnick, Tetrahedron 1981, 37, 2091–
2096.
[19]
P. K. Jadhav, K. S. Bhat, P. T. Perumal, H. C. Brown, J. Org. Chem.
1986, 51, 432–439.
[20] Neither the intermediate alcohol nor aldehyde 10 could be extensively
purified, due to their high volatility. In particular, 10 could only be isolated
as a mixture with CH2Cl2. Thus, an accurate yield for the allylation
reaction could not be calculated. However, the allylation was clean and
gave 9 in estimated yields of around 80% from 10 (calculated based on
the CH2Cl2 content of 10).
[21] O. Mitsunobu, Synthesis 1981, 1–28.
[22] For examples, see a) B. M. Trost, G. Dong, J. A. Vance, J. Am. Chem.
Soc. 2007, 129, 4540–4541; b) T. R. Hoye, H. Zhao, Org. Lett. 1999, 1,
1123–1125; c) T. Kawaguchi, N. Funamori, Y. Matsuya, H. Nemoto, J.
Org. Chem. 2004, 69, 505–509; d) K. C. Nicolaou, Y.-P. Sun, R. Guduru,
B. Banerji, D. Y.-K. Chen, J. Am. Chem. Soc. 2008, 130, 3633–3644.
[23] C. Lecourt, S. Dhambri, L. Allievi, Y. Sanogo, N. Zeghbib, R. Ben Othman,
M.-I. Lannou, G. Sorin, J. Ardisson, Nat. Prod. Rep. 2018, 35, 105–124.
[24] Separation was possible after the glycosylation step. See the SI for
details.
[25] B. A. Garcia, J. L. Poole, D. Y. Gin, J. Am. Chem. Soc. 1997, 119, 7597–
7598.
[26] R. E. Lee, K. Mikušová , P. J. Brennan, G. S. Besra, J. Am. Chem. Soc.
1995, 117, 11829–11832.
[27] For the investigation of other conditions see the SI.
[28] J. M. Bobbitt, J. Org. Chem. 1998, 63, 9367–9374.
[29] NMR-spectroscopic data for 2 were in full agreement with those reported
in the literature.[1][5] The specific rotation of synthetic 2 was lower than
reported for the natural material[1] (i. e. [휶]ퟐ푫ퟎ = +35 (c = 0.2, MeOH) for
[
]
푫 = +64 (c = 0.4, MeOH) for natural 2). Fürstner and
synthetic 2 vs.
휶
co-workers have reported [휶]ퟐ푫ퟎ = +188 (c = 0.04, MeOH)[5]. Synthetic 2
co-eluted with natural 2 from a chiral, normal phase Daicel Chiralpak AD-
H column.
[30] As pointed out by one of the reviewers, the difference in the efficiency of
the RCM between dienes 7 and 16 may well be ascribed to differences
in the degree of conformational pre-organization for ring-closure. This
proposal is in line with observations by Ichikawa et al. in their studies
towards the synthesis of gulmirecin A,[8] where they observed
a
pronounced dependence of the efficiency (or even feasibility) of the
Ni(0)-mediated reductive cyclization to the 12-membered ring on the
structure of the ynal precursor.
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202004589
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents
Large or small (ring), that was the question. The antibiotic disciformycin B has been synthesized in 18 linear steps from simple
starting materials. Key step was the macrocyclization of tetraene 7 by RCM. Under optimized conditions, the desired 12-membered
macrolactone was obtained in preference over the cyclopentene arising from competing RCM with the internal double bond. The
synthesis was completed by an exceptionally mild chemoselective oxidation of an allylic hydroxy group.
Institute and/or researcher Twitter usernames: D-CHAB@ETH_DCHAB
6
This article is protected by copyright. All rights reserved.