Asymmetric Total Synthesis of Brasilicardins

Article abstract of DOI:10.1002/anie.201811403

Brasilicardins, bacterial diterpenoid natural products that display highly potent immunosuppressive activity, are promising immunosuppressant drug candidates. Structurally, they can be described as hybrids of terpenoids, amino acids, and saccharides, and share a characteristic highly strained anti-syn-anti-fused perhydrophenanthrene terpenoid scaffold (ABC-ring system) with two quaternary asymmetric carbon atoms. A unified and stereoselective total synthesis of all four brasilicardins has been designed based on the strategic use of an intramolecular conjugate addition. The ABC-ring system was initially constructed with high stereocontrol by novel intramolecular conjugate additions of Weinreb amides and in situ generated (Z)-vinyl copper species. The late-stage common intermediate was subjected to stereoselective installation of the amino acid component, followed by introduction of the saccharide unit via glycosylation to accomplish the total synthesis of brasilicardins A–D. Our synthesis offers opportunities to synthesize various brasilicardin analogues for biological and pharmacological investigations.

Full text of DOI:10.1002/anie.201811403

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Asymmetric Total Synthesis of Brasilicardins
Authors: Fumihiko Yoshimura, Ryusei Itoh, Makoto Torizuka, Genki
Mori, and Keiji Tanino
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.201811403
Angew. Chem. 10.1002/ange.201811403
Link to VoR: http://dx.doi.org/10.1002/anie.201811403
http://dx.doi.org/10.1002/ange.201811403

10.1002/anie.201811403
Angewandte Chemie International Edition
COMMUNICATION
Asymmetric Total Synthesis of Brasilicardins
Fumihiko Yoshimura,*[a,c] Ryusei Itoh,[b] Makoto Torizuka,[b] Genki Mori,[a] and Keiji Tanino*[a]
Abstract: Brasilicardins are bacterial diterpenoid natural products
that display highly potent immunosuppressive activity; therefore,
they are considered to be promising immunosuppressant drug
candidates. Structurally, they can be described as hybrids of
terpenoids, amino acids, and saccharides, and share a characteristic
highly strained anti-syn-anti-fused perhydrophenanthrene terpenoid
scaffold (ABC-ring system) with two quaternary asymmetric carbon
atoms. Herein, we describe our design of the unified and
stereoselective total synthesis of all four brasilicardins based on the
strategic use of an intramolecular conjugate addition. The ABC-ring
system was initially constructed with high stereocontrol by novel
intramolecular conjugate additions of Weinreb amides and in situ-
generated (Z)-vinyl copper species. The late-stage common
intermediate was subjected to stereoselective installation of the
amino acid component, followed by introduction of the saccharide
unit via glycosylation to accomplish the total synthesis of
brasilicardins A–D. Our synthesis offers opportunities to synthesize
various brasilicardin analogues for biological and pharmacological
investigations.
sources; therefore, efficient chemical syntheses of 1 and its
derivatives and simplified analogues are required to aid further
studies.
OH
O
HO
HO
O
OH
NHAc
O
Me
O
Me
O
R
HO
Me
BH
C
O
2
CO2H
NH2
16
A
Me
HO
Me
H
Me
Brasilicardin A (1): R = OMe
Brasilicardin B (2): R = H
OH
HO
HO
Me
Me
R
O
Me
H
O
CO2H
H
Me
NH2
HO
Me
Brasilicardin C (3): R = OMe
Brasilicardin D (4): R = H
Me
Me
Me
H
O
Sugars
HO
Me
C
A
H
B
Brasilicardins A–D (1–4) are bacterial diterpenoid natural
products isolated from the cultured broth of the actinomycete
Nocardia brasiliensis IFM 0406, which exhibit diverse biological
activities (Figure 1).[1–3] Among these congeners, brasilicardin A
(1) displays potent immunosuppressive activity (IC50, 0.05 nM).
Although the mechanism of the immunosuppressive action of 1
has not been clarified in detail, it has been suggested that it is
induced by amino acid deprivation via the inhibition of the amino
acid transporter system L.[4] Currently used immunosuppressive
clinical agents such as tacrolimus and cyclosporin A often cause
side effects such as nephrotoxicity and arterial hypertension;
therefore, an alternative to them is desired. In this context, 1 is
thus considered to be a promising drug candidate. Thus, 1 has
been studied with keen interest, particularly in to the context of
the development of new immunosuppressive drugs.[5,6] However,
further biological and pharmacological studies of 1 have not been
explored because of the limited availability of 1 from natural
Me
R
Me
H
anti-syn-anti-fused perhydrophenanthrene
skeleton of brasilicardins
Figure 1. Structures of brasilicardins.
Structurally, brasilicardins A–D (1–4) share a unique and
highly strained anti-syn-anti-fused perhydrophenanthrene
diterpenoid carbon scaffold (i.e., the ABC-ring) containing two
angular methyl groups with the B-ring boat conformer (Figure 1).
To this scaffold, different amino acid and saccharide units are
attached.
The combination of their distinctive biological and
pharmacological properties and the novel chemical structures
has made brasilicardins extremely attractive targets for synthetic
organic chemists.[6–9] However, their complex and unique three-
dimensional structure has hampered their chemical synthesis.
Despite considerable synthetic efforts having been undertaken
for over twenty years, the sole successful total synthesis was the
de novo construction of 1 and 3 by Anada, Hashimoto, and co-
workers in 2017[10] which involved sequential functionalization of
the Wieland–Miescher ketone derivative and a Diels–Alder
reaction/reductive angular methylation sequence[7] (Scheme 1a).
Herein, we report a novel and unified approach that enables the
total synthesis of all brasilicardins (1–4), including the first total
synthesis of 2 and 4.
[a]
Prof. Dr. F. Yoshimura, G. Mori, Prof. Dr. K. Tanino
Department of Chemistry, Faculty of Science
Hokkaido University
Sapporo 060-0810, Japan
E-mail: ktanino@sci.hokudai.ac.jp (K.T.)
R. Itoh, M. Torizuka
Graduate School of Chemical Sciences and Engineering
Hokkaido University
[b]
[c]
Sapporo 060-0810, Japan
Present Address: Prof. Dr. F. Yoshimura
School of Pharmaceutical Sciences
University of Shizuoka
From a synthetic perspective, the challenges posed by
brasilicardins (1–4) are as follows: (1) development of a
stereoselective synthetic methodology for the unique carbocyclic
framework (the ABC-ring), which represents the most important
issue in the synthesis, (2) construction of the amino acid
52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan
E-mail: fumi@u-shizuoka-ken.ac.jp (F.Y.)
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/anie.201811403
Angewandte Chemie International Edition
COMMUNICATION
a
Anada–Hashimoto approach
Our key technology
b
Me
different reactivity
Sugars
OMe
Me
CN
O
Me
H
Me
CN
*
Me
M
CN
H3O+
O
CO2H
base
O
O
O
Me
H
R´
R´
R´
*
OR
OR
R
NH2
N
N
N
1 and 3
Diels–Alder/
reductive angular methylation
HO
Me
H
H
Me
Me
Me
Me
5 (E or Z)
6
7
From Wieland–Miescher ketone derivative
c
Me
Me
Me
Sugars
R
R
Me
Me
Me
H
Construction of
amino acid
C
O
CO2H
NH2
HO
HO
CO2t-Bu
PO
CO2Et
Glycosylation
Me
Me
Me
H
H
H
NHFmoc
HO
Me
PO
Me
H
Me
H
Me
Me
Brasilicardins A–D (1–4):
R = OMe or H
Me
8 (R = OMe)
9 (R = H)
10 (P = MOM)
Me
t-BuO
O
N
Br
Br
Ot-Bu
N
Me
O
Intramolecular
conjugate addition
Intramolecular
conjugate addition
Me
H
Me
Me
PO
PO
CO2Et
PO
PO
CN
Me
PO
CN
Me
B
Me
H
H
PO
H
Me
H
Me
Me
Me
Me
Me
11 (P = MOM)
12 (P = MOM)
13 (P = MOM)
Intramolecular
conjugate addition
Me
H
Me
PO
PO
CN
O
PO
CN
O
HO
OH
A
OMe
OMe
Me Me
N
PO
N
Me
16
Me Me
Me
Me
Me
14 (P = MOM)
15 (P = MOM)
Scheme 1. Background and synthetic plan.
component, and (3) stereoselective installation of the sugar
moiety.
amide groups in the product 7 would facilitate the further
transformations.
Aiming to obtain in-depth structure–activity relationships
(SAR) in future for understanding the mode of action of 1 at the
molecular level, we employed a tactic wherein each ring of the
carbocyclic framework as well as the amino acid and saccharide
components would be constructed in a step-by-step manner.
While a biomimetic polyene cyclization affording the above
framework could have been a potential strategy, a previous study
revealed that such an approach was extremely difficult.[11] As one
of the key steps, we designed an intramolecular conjugate
addition (Michael addition) of a,b-unsaturated N-methoxy-N-
methylamides (commonly named Weinreb amides[12]) bearing an
alkanenitrile side chain (Scheme 1b; 5→7). Despite the poor
reactivities of both a,b-unsaturated amides and a-cyano
carbanions generated from simple non-activated alkanenitriles
as Michael acceptors and Michael donors, respectively,[13,14] we
expected that the Michael addition would proceed smoothly
because the putative and stable chelated intermediate 6 formed
upon the addition would prevent unfavored reversible retro-
additions. In this connection, we previously reported another
approach for preventing similar retro-additions by trapping the
intermediate as the corresponding ketene silyl acetal in an
intramolecular Michael addition of a,b-unsaturated lactones.[15]
In addition, if such additions were to occur with facial
discrimination, the sterically much less demanding cyano group
would induce stereoselective cyclization, resulting in
simultaneous construction of a contiguous quaternary–tertiary
stereocenter. Moreover, the potential different reactivities
between the sterically congested cyano group and Weinreb
With the above considerations in mind, our retrosynthetic
analysis of brasilicardins (1–4) is shown in Scheme 1c. The
saccharide moiety was to be installed through regioselective
glycosylation to the protected aglycons 8 or 9 at the final stage
of the synthesis. Aglycons 8 or 9 were to be derived from ester
10 via construction of the amino acid component. Thus, we
envisioned that the tricyclic core 10 could serve as an advanced
intermediate for our unified synthesis. The requisite core 10
could be synthesized by a Michael addition-based strategy. To
be more specific, 10 could be obtained from the a,b-unsaturated
ester 11 through our novel intramolecular Michael addition
promoted by Me2CuLi.[16] The dibromide 11 could be synthesized
from bicyclic compound 12 via homologation of the side chains.
The B-ring could be constructed through an intramolecular nitrile
Michael addition of (Z)-a,b-unsaturated Weinreb amide 13 as
mentioned above (see, Scheme 1b). Amide 13 would be derived
from nitrile 14. The A-ring could be constructed by a similar
Michael addition of (E)-a,b-unsaturated Weinreb amide 15. This
compound, in turn, could be accessed from commercially
available 2,2-dimethylpropane-1,3-diol (16).
Our first objective focused on the stereoselective synthesis
of the crucial precursor 15 for the first intramolecular nitrile
Michael addition (Scheme 2). Monosilylation of 2,2-
dimethylpropane-1,3-diol (16) followed by Swern oxidation of the
remaining alcohol afforded the corresponding aldehyde, which
underwent one-pot Horner–Wadsworth–Emmons (HWE)
olefination[17] to furnish (E)-a,b-unsaturated ester 17 as the sole
product. Asymmetric dihydroxylation of 17 by using the
This article is protected by copyright. All rights reserved.

10.1002/anie.201811403
Angewandte Chemie International Edition
COMMUNICATION
1) NaH, TBSCl
(DHQ)PHN
K2OsO2(OH)4
K3Fe(CN)6, K2CO3
MeSO2NH2
1) MOMCl, i-Pr2NEt
50 °C
2) LiAlH4, 0 °C
EtO2C
HO
HO
CO2Et
MOMO
MOMO
I
HO
OH
OTBS
OTBS
OTBS
2) (COCl)2, DMSO
Et3N, –78 °C to rt
t-BuOH–H2O, 0 °C
3) I2, PPh3, imidazole
96% (3 steps)
Me Me
Me Me
Me Me
Me Me
>99%
16
then (EtO)2P(O)CH2CO2Et
DBU, LiCl, MeCN, 40 °C
17
18 (95% ee)
19
96% (2 steps)
Me
Me
Me
1) EtCN, LDA
HMPA, –78 °C
MOMO
MOMO
CN
OH
MOMO
MOMO
CN
MOMO
MOMO
CN
O
O
1) TPAP, NMO
NaHMDS
OMe
OMe
O
O
N
N
2) TBAF, 50 °C
98% (2 steps)
2)
THF, –78 °C
93% (dr > 99:1)
H
Me
Me Me
20 (dr = 55:45)
Me Me
(EtO)2P
OMe
Me
14
Me
15 (dr = 60:40)
Me
N
NaHMDS
Me
76% (2 steps)
OH
Me
Me
Me
1) TsOH
acetone–H2O
1) DIBAL, CH2Cl2
–78 °C
1) DIBAL, THF
–78 °C
MOMO
MOMO
CN
MOMO
CN
MOMO
MOMO
OMe
MOMO
2) Ph3P(CH2OMe)Cl
t-BuOK
2) Ph3PCH3Br
t-BuOK
2) DIBAL
H
H
H
Me
Me
Me
Me
21
Me
22
Me
23
93% (2 steps)
84% (2 steps)
EtO2C
OH
Me
Me
Co catalyst 24, TsCN
PhSiH3, 2,6-di-t-Bu-pyridine
MOMO
MOMO
CN
Me
MOMO
MOMO
CN
Me
1) TPAP, NMO
2)
1) LiOH (aq.)
O
O
EtOH
2) t-BuONHMe·HCl
EDCI, HOAt, Et3N
H
H
Me
Me
99% (3 steps)
Me
25 (dr = 50:50)
(o-t-BuC6H4O)2P
Me
OEt
>99% (2 steps)
26
NaHMDS
77% (2 steps)
t-BuO
Me
N
OTIPS
Me
O
Ot-Bu
N
Ot-Bu
N
Me
O
N
O
N
Me
Me
H
Me
H
NaHMDS
TIPSCl
then
TBAF
MOMO
CN
Me
CN
Me
MOMO
MOMO
CN
Me
Co
t-Bu
O
t-Bu
H
H
Et2O, –78 °C
MOMO
88% (dr = 93:7)
H
Me
Me
t-Bu
t-Bu
Me
13 (dr = 50:50)
Me
27
24
12
Scheme 2. Synthesis of the bicyclic intermediate 12. TBS = tert-butyldimethylsilyl, DBU = 1,8-diazabicyclo[5.4.0]undec-7-ene, MOM = methoxymethyl, LDA = lithium
diisopropylamide, HMPA = hexamethylphosphoramide, TPAP = tetrapropylammonium perruthenate, NMO = N-methylmorpholine oxide, NaHMDS = sodium
bis(trimethylsilyl)amide, DIBAL = diisobutylaluminum hydride, EDCI = 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride, HOAt = 1-hydroxy-7-
azabenzotriazole, TIPS = triisopropylsilyl, TBAF = tetrabutylammonium fluoride.
dihydroquinine-based
9’-phenanthryl
ether
ligand
oriented equatorially in an antiparallel dipolar arrangement in the
transition state model (Supporting Information, Scheme S1).
(DHQ)PHN[18,19] quantitatively gave the optically active diol 18
with 95% ee. Protection of the diol in 18 with MOM groups,
followed by reduction of the ethyl ester and iodination of the
resulting alcohol, yielded alkyl iodide 19. Alkylation of 19 with
deprotonated propionitrile and the subsequent desilylation
afforded alcohol 20. This compound underwent oxidation with
TPAP and HWE olefination to give (E)-a,b-unsaturated Weinreb
amide 15 as the substrate for an intramolecular Michael addition.
The successful realization of an intramolecular nitrile Michael
addition led us to the next phase of the synthesis: construction
of the B-ring (Scheme 2). Thus, chemoselective reduction of the
Weinreb amide over the cyano group in 14 with DIBAL followed
by one-carbon homologation of the resulting aldehyde by a Wittig
reagent produced enol methyl ether 21. The product 21 was
further converted to terminal alkene 23 in four steps. Introduction
of the cyano group into 23 was achieved by cobalt-catalyzed
hydrocyanation[20] to give secondary nitrile 25 regioselectively.
Oxidation of alcohol 25 with TPAP afforded the corresponding
aldehyde, which underwent Z-selective HWE reaction[21,22] to
afford (Z)-a,b-unsaturated ester 26 (Z/E > 99:1). Saponification
of 26 and the subsequent amidation produced the modified
Weinreb amide,[23] namely, N-tert-butoxy-N-methylamide 13.
We then focused on the crucial intramolecular Michael
addition of 15 for the formation of the A-ring (Scheme 2). In an
extensive screening of reaction conditions, it was found that
when a THF solution of 15 was treated with NaHMDS at –78 °C,
intramolecular Michael addition proceeded stereoselectively to
furnish the desired product 14 in 93% yield as a single isomer.
The stereochemistry of 14 was unambiguously determined by
nuclear Overhauser effect (NOE) experiments (Supporting
Information, Figure S1). The observed complete stereocontrol for
14 was suggested to arise as a consequence of chelation control,
wherein the keteniminate and a,b-unsaturated amide were both
The second intramolecular nitrile Michael addition of 13 was
the crucial factor in the present synthesis; therefore, we explored
the optimal reaction conditions for this step, focusing on the
solvent, additives, and reaction temperature. We found that this
This article is protected by copyright. All rights reserved.

10.1002/anie.201811403
Angewandte Chemie International Edition
COMMUNICATION
Br
Br
Br
Br
Me
H
Me
H
1) DIBAL, THF
MOMO
1) DIBAL, CH2Cl2
–78 °C
CN
MOMO
MOMO
CO2Et
Me2CuLi
–78 °C
12
H
Me
H
Me
2) CBr4, PPh3, Et3N
MOMO
2) (EtO)2P(O)CH2CO2Et
NaHMDS
Et2O, –78 °C
Me
Me
76% (2 steps)
Me
Me
96% (2 steps)
28
11
Me
Me
Me
14
Cu
Me
H
MOMO
MOMO
CO2Et
CO2Et
H
Me
H
Me
–40 °C
83% (dr > 99:1)
H
Me
Me
29
10
ORTEP of 10
Scheme 3. Stereoselective synthesis of the tricyclic compound 10.
addition occurred upon the treatment of a mixture of 13 and
On obtaining the tricyclic compound 10, we directed our
TIPSCl[15] with NaHMDS in Et2O at –78 °C, which produced O-
silyl N,O-ketene acetal 27 in situ. Upon addition of TBAF to 27,
the desired cyclization product 12 was obtained as a 93:7
inseparable mixture of diastereomeric products (88% yield).
Importantly, in this cyclization, the use of the bulky O-tert-butyl
Weinreb amide, (Z)-configuration of the olefin, and addition of
TIPSCl were all indispensable to achieve a high level of
stereocontrol at the newly formed stereocenters of the product
(for comparison with the cyclization of other precursors, see
Scheme S2 in the Supporting Information). The steric repulsions
between the large substituents are likely to account for the
stereochemical outcome (Supporting Information, Scheme S3).
These key nitrile Michael additions (i.e., 15→14 and 13→12)
were reliably performed on a gram scale, which demonstrated
the significant synthetic utility of this methodology.
attention to installation of the appropriate amino acid portion to
this skeleton (Scheme 4). Initially, we investigated installation of
a b-methoxy amino acid for brasilicardins A and C, which was
succeeded by application of Rama Rao’s procedure[24] (Scheme
4a). Half-reduction of 10 followed by HWE reaction of the
resulting aldehyde yielded a,b-unsaturated tert-butyl ester 30.
Asymmetric dihydroxylation of 30 using (DHQ)PHN-ligand[18]
gave the desired diol 31 as a single diastereomer. Regioselective
nosylation at the C17 alcohol of 31 produced mono-nosylate 32,
and subsequent treatment of 32 with NaN3 afforded a-azide 33
with complete inversion of the configuration. O-Methylation of 33
followed by reduction of the azide and protection of the resulting
amine with a Fmoc group resulted in the formation of the
protected amino acid of brasilicardins A and C. Finally, removal
of both MOM groups with HCl in methanol produced diol 8.
Stereochemistry of the amino acid component was verified by X-
ray crystallographic analysis after the conversion to the
corresponding p-bromobenzamide derivative of 8 (Supporting
Information, Figure S2).
Having developed a method for the construction of the AB-
ring including two quaternary asymmetric stereocenters based
on the strategic nitrile Michael additions, we next focused on the
third intramolecular Michael addition, which leads to the ABC-
ring of brasilicardins (Scheme 3). To this end, Weinreb amide 12
was converted to 1,1-dibromoalkene 11, the requisite cyclization
precursor, in a four-step sequence: (1) chemoselective DIBAL
reduction of the Weinreb amide moiety in 12 to an aldehyde, (2)
Corey–Fuchs olefination, wherein the minor diastereomer of 12
was separated, (3) reduction of the cyano group to an aldehyde,
and (4) HWE olefination of the resulting aldehyde.
On the other hand, installation of the amino acid component
of brasilicardins B and D was performed by asymmetric
alkylation of the chiral Schiff base[25] as the key step (Scheme
4b). Thus, ethyl ester 10 was converted to iodide 34 through
LiAlH4 reduction and iodination of the resulting alcohol. Alkylation
of the readily available chiral Schiff base derived from glycine 35
with 34 proceeded smoothly when KHMDS was used as the
base, providing the alkylation product 36 as the sole isomer. This
compound was converted to diol 9 in three steps, i.e., removal of
the chiral auxiliary group, Fmoc-protection of the resulting free-
amine, and deprotection of the MOM groups. Stereochemistry of
the amino acid portion was confirmed by the modified Mosher’s
method[26] (Supporting Information, Scheme S4).
The third intramolecular Michael addition for constructing the
C-ring proceeded smoothly under the reaction conditions used[16]
(Scheme 3): treating 11 with Me2CuLi (5 equiv) at –78 °C in Et2O
for 30 min generated the (Z)-vinylcopper species 29 in situ. After
the reaction mixture was warmed up to –40 °C and then stirred
at this temperature for 1 h, the subsequent conjugate addition of
29 to the a,b-unsaturated ester moiety proceeded smoothly in
the one-pot system to furnish the tricyclic compound 10 with
controlled stereochemistry at the C14 position (83% yield). The
structure of 10 was verified by X-ray crystallographic analysis
(Scheme 3). In this way, we established a novel method for the
stereoselective construction of the anti-syn-anti-fused
perhydrophenanthrene skeleton (the ABC-ring) by triple
intramolecular Michael additions as the key steps.
The final task in the synthesis was the challenging
glycosylation of aglycons 8 and 9 (Scheme 5). First, the
glycosylation of 8 toward brasilicardin A (1) was explored
(Scheme 5a). Although Anada and Hashimoto’s group
succeeded in glycosylation of the C3-protected aglycon of 1 with
the trichloroacetimidate donor of the disaccharide unit,[8,10] this
procedure could not be applicable to diol 8 because the
formation of complex mixtures including the inseparable C3
This article is protected by copyright. All rights reserved.

10.1002/anie.201811403
Angewandte Chemie International Edition
COMMUNICATION
a
Me
Me
OH
(DHQ)PHN, K2OsO2(OH)4
K3Fe(CN)6, K2CO3, MeSO2NH2
1) DIBAL, CH2Cl2
–90 °C
p-NsCl, Et3N
17
CO2t-Bu
CO2t-Bu
10
H
Me
30
H
Me
31
t-BuOH–H2O, 0 °C
CH2Cl2, –10 °C
2) (EtO)2P(O)CH2CO2t-Bu
NaHMDS
OH
97%
96% (2 steps)
Me
1) Me3OBF4
Me
Me
OH
OMe
OH
2) SnCl2
Me
H
NaN3
HO
HO
CO2t-Bu
then Fmoc-OSu
CO2t-Bu
CO2t-Bu
Me
H
H
Me
32
H
Me
33
DMF, 50 °C
3) HCl, MeOH
91% (3 steps)
NHFmoc
Op-Ns
N3
92% (2 steps)
Me
Me
8 (for Brasilicardins A and C)
HO Me
b
N
CO2t-Bu
Me
Me
35
1) LiAlH4
citric acid (aq.)
KHMDS
CO2t-Bu
I
10
H
Me
36
H
Me
2) I2, PPh3, imidazole
98% (2 steps)
THF, –65 °C
N
34
Me
HO
Me
Me
Me
Me
H
MOMO
CO2t-Bu
HO
HO
CO2t-Bu
NHFmoc
9 (for Brasilicardins B and D)
1) FmocCl, NaHCO3
Me
Me
H
H
NH2
2) HCl, MeOH
88% (4 steps)
MOMO
H
Me
Me
Me
Me
37
Scheme 4. Stereoselective synthesis of the protected brasilicardin aglycons 8 and 9. p-Ns = 4-nitrobenzenesulfonyl, Su = succinimidyl, Fmoc = 9-
fluorenylmethyloxycarbonyl, KHMDS = potassium bis(trimethylsilyl)amide.
glycosylation product and degradation of 8 were observed. To
our delight, after extensive screening of glycosyl donors and their
activation conditions, the metallocene-based activator
Cp2HfCl2/AgOTf[27] allowed the regio- and stereoselective
coupling of peracetyl glycosyl fluoride 38 (2 equiv) with 8 to give
the desired a-glycoside 39 (with 43% of 8 recovered). Because
the prolonged reaction time caused an undesired reaction at the
C3 alcohol, this coupling was carefully stopped before full
consumption of 8, with recovery of unreacted 8. Finally, removal
of the tert-butyl ester using TFA followed by global deprotection
of the acetyl and Fmoc groups furnished brasilicardin A (1) (43%
yield, three steps). Confirming the identity of our synthesized
Consequently, the C3 alcohol was temporary protected as an
easily removable methoxyacetate,[28] and the requisite mono-
alcohols 41 and 42 were synthesized from 8 and 9, respectively,
through a three-step conversion. Toward the synthesis of total
synthesis of brasilicardin C, we found that gold-catalyzed
coupling[29] of glycosyl o-cyclopropylethynylbenzoate 43 with 41
proved to be the most effective method for stereoselective
installation of a rhamnose, providing the glycosylation product 44.
This compound was successfully converted to brasilicardin C (3)
via TFA-mediated removal of tert-butyl ester and the subsequent
global deprotection of the remaining acetyl, methoxyacetyl, and
Fmoc groups with aqueous lithium hydroxide solution. In addition,
brasilicardin D (4) was obtained from 42 via a similar sequence
as that described for 3. Thus, we accomplished the first total
synthesis of brasilicardin B (2) and brasilicardin D (4) in addition
to the total synthesis of brasilicardin A (1) and brasilicardin C (3).
1
sample was challenging because the H-NMR spectrum of the
compound proved highly dependent on the concentration and pH.
However, its identification was achieved using 1H-NMR spectrum
of a mixture of synthetic and natural brasilicardin A in a 1:1 ratio
after both materials were purified by reverse phase semi-
preparative HPLC (Capcell Pak C18 SG120). Further, the fact
that the 13C-NMR, IR, and HRMS spectra and optical rotation
results for the synthesized product matched those of the natural
sample supported successful synthesis of our desired product,
as did the narrow range obtained for the melting temperature of
the mixture. Brasilicardin B (2) was also synthesized from 9 using
the same conditions as those used for 1.
In conclusion, we have developed a stereoselective synthetic
route
to
synthesize
brasilicardins
with
potent
immunosuppressive activity, utilizing which we accomplished the
asymmetric total synthesis of brasilicardin A (39 linear steps,
6.8% yield), brasilicardin B (37 linear steps, 6.5% yield),
brasilicardin C (42 linear steps, 12% yield), and brasilicardin D
(40 linear steps, 14% yield) from readily available commercial
materials. The synthesis features (1) the development of a novel
intramolecular nitrile Michael addition, i.e., stereoselective
intramolecular conjugate addition of an a-cyano carbanion to an
a,b-unsaturated Weinreb amide, which allows simultaneous ring
formation and construction of contiguous quaternary–tertiary
asymmetric stereocenters, (2) a Michael addition-based strategy
for the stereoselective formation of the highly strained
Achievement of total synthesis of 1 and 2 prompted us to
pursue the total synthesis of brasilicardins C (3) and D (4)
(Scheme 5b). Unlike the glycosylation with disaccharide 38,
glycosylation of the glycosyl fluoride donor derived from
rhamnose with diols 8 or 9 did not show any regioselectivity at
the C2 alcohol. Attempted use of other donors also failed.
This article is protected by copyright. All rights reserved.

10.1002/anie.201811403
Angewandte Chemie International Edition
COMMUNICATION
OAc
O
a
OAc
O
AcO
AcO
O
Me
OAc
AcO
AcO
NHAc
R
O
Me
Cp2HfCl2, AgOTf
DTBMP, MS4A
OAc
O
Me
NHAc
HO
HO
CO2t-Bu
O
Me
O
R
2
3
AcO
O
+
Me
Me
H
O
Me
O
O
CO2t-Bu
NHFmoc
CH2Cl2
AcO
F
H
Me
H
Me
Me
NHFmoc
8 (R = OMe)
9 (R = H)
HO
Me
H
Me
38 (α-anomer exclusively)
OH
39 (R = OMe) + recovery of 8 (43%)
+
40 (R = H)
recovery of 9 (61%)
O
HO
HO
O
OH
NHAc
O
Me
R
1) TFA
O
Me
O
HO
Me
H
O
CO2H
NH2
2) ethylenediamine, EtOH
H
Me
HO
Brasilicardin A (1, R = OMe): 43% (3 steps from 8)
Brasilicardin B (2, R = H): 37% (3 steps from 9)
Me
Me
b
Me
1) TBSOTf, 2,6-lutidine
–90 °C
2) MeOCH2COCl, pyridine
R
OAc
Me
Ph3PAuNTf2
MS4A
AcO
HO
CO2t-Bu
AcO
O
O
2
3
8 or 9
Me
+
O
H
Me
MeO
NHFmoc
3) HF·pyridine
CH2Cl2, 0 °C
O
O
Me
H
Me
41 (R = OMe): 86% from 8
42 (R = H): 87% from 9
43 (α/β = 40:60)
OAc
AcO
OH
HO
AcO
HO
Me
Me
Me
O
O
O
Me
R
R
Me
H
Me
1) TFA
O
O
CO2t-Bu
NHFmoc
O
CO2H
NH2
H
Me
H
Me
2) LiOH (aq.), 45 °C
MeO
HO
H
Me
Me
Me
44 (R = OMe)
45 (R = H)
Brasilicardin C (3, R = OMe): 94% (3 steps from 41)
Brasilicardin D (4, R = H): 92% (3 steps from 42)
Me
Scheme 5. Total synthesis of brasilicardins A–D. DTBMP = 2,6-di-tert-butyl-4-methylpyridine, MS4A = molecular sieves 4A, TFA = trifluoroacetic acid. Tf =
trifluoromethanesulfonyl.
anti-syn-anti-fused perhydrophenanthrene skeleton (the ABC-
ring), (3) stereoselective installation of the amino acid
component to the terpenoid core, and (4) regio- and
stereoselective glycosylation using glycosyl fluoride or o-
alkynylbenzoate as the glycosyl donor. Our strategy allowed the
unified synthesis of all brasilicardins from the common late-stage
intermediate utilizing the appropriate installation methods for an
amino acid and glycosylation. The novel synthetic route
developed here would accelerate the synthesis and biological
studies of brasilicardins and a wide variety of their analogues that
were previously inaccessible by syntheses or from natural
products as well as aid in obtaining in-depth SAR toward the
development of new immunosuppressive drugs.
Fukushi and Mr. Yusuke Takata (GC-MS & NMR Laboratory,
Faculty of Agriculture, Hokkaido University) in performing mass
spectral measurements, Dr. Yasuhiro Kumaki (High-Resolution
NMR Laboratory, Graduate School of Science, Hokkaido
University) for the NMR experiments, and Professors Tamotsu
Inabe and Takahiro Suzuki (Hokkaido University) for the X-ray
structural analysis. This work was supported by JSPS KAKENHI
Grant Numbers JP22710206, JP24510290, JP15K01795,
JP18K05339, JP18H01970, and JP15H05842 in Middle
Molecular Strategy.
Keywords: Total synthesis • Brasilicardins • Natural products •
Michael addition • Quaternary stereocenters
[1]
H. Shigemori, H. Komaki, K. Yazawa, Y. Mikami, A. Nemoto, Y. Tanaka,
T. Sasaki, Y. In, T. Ishida, J. Kobayashi, J. Org. Chem. 1998, 63, 6900–
6904.
Acknowledgements
[2]
[3]
K. Komatsu, M. Tsuda, M. Shiro, Y. Tanaka, Y. Mikami, J. Kobayashi,
Bioorg. Med. Chem. 2004, 12, 5545–5551.
We acknowledge gratitude toward Professors Jun’ichi
Kobayashi (Hokkaido University) and Takaaki Kubota (Showa
Pharmaceutical University) for providing an authentic sample of
brasilicardin A. The authors thank Professor Hideyuki Shigemori
(University of Tsukuba) and Dr. Hisayuki Komaki (NBRC) for
helpful discussions regarding the purification protocols of
brasilicardins. Additionally, we acknowledge help from Dr. Eri
H. Komaki, A. Nemoto, Y. Tanaka, H. Takagi, K. Yazawa, Y. Mikami, H.
Shigemori, J. Kobayashi, A. Ando, Y. Nagata, J. Antibiot. 1999, 52, 13–
19.
[4]
T. Usui, Y. Nagumo, A. Watanabe, T. Kubota, K. Komatsu, J. Kobayashi,
H. Osada, Chem. Biol. 2006, 13, 1153–1160.
This article is protected by copyright. All rights reserved.

10.1002/anie.201811403
Angewandte Chemie International Edition
COMMUNICATION
[5]
K. Komatsu, M. Tsuda, Y. Tanaka, Y. Mikami, J. Kobayashi, Bioorg.
Med. Chem. 2005, 13, 1507–1513.
[6]
M. E. Jung, B. T. Chamberlain, P. Koch, K. R. Niazi, Org. Lett. 2015,
17, 3608–3611.
[7]
[8]
[9]
D. M. Coltart, S. J. Danishefsky, Org. Lett. 2003, 5, 1289–1292.
M. E. Jung, P. Koch, Org. Lett. 2011, 13, 3710–3713.
M. E. Jung, F. Perez, C. F. Regan, S. W. Yi, Q. Perron, Angew. Chem.
Int. Ed. 2013, 52, 2060–2062; Angew. Chem. 2013, 125, 2114–2116.
[10] M. Anada, T. Hanari, K. Kakita, Y. Kurosaki, K. Katsuse, Y. Sunadoi, Y.
Jinushi, K. Takeda, S. Matsunaga, S. Hashimoto, Org. Lett. 2017, 19,
5581–5584.
[11] M. Nishizawa, H. Takenaka, K. Hirotsu, T. Higuchi, Y. Hayashi, J. Am.
Chem. Soc. 1984, 106, 4290–4291.
[12] S. Balasubramaniam, I. S. Aidhen, Synthesis 2008, 3707–3738.
[13] K. M. Byrd, Beilstein J. Org. Chem. 2015, 11, 530–562.
[14] S. Arseniyadis, K. S. Kyler, D. S. Watt, Org. React. 1984, 31, 1–364.
[15] F. Yoshimura, M. Torizuka, G. Mori, K. Tanino, Synlett 2012, 23, 251–
254.
[16] K. Tanino, K. Arakawa, M. Satoh, Y. Iwata, M. Miyashita, Tetrahedron
Lett. 2006, 47, 861–864.
[17] M. A. Blanchette, W. Choy, J. T. Davis, A. P. Essenfeld, S. Masamune,
W. R. Roush, T. Sakai, Tetrahedron Lett. 1984, 25, 2183–2186.
[18] K. B. Sharpless, W. Amberg, M. Beller, H. Chen, J. Hartung, Y.
Kawanami, D. Lübben, E. Manoury, Y. Ogino, T. Shibata, T. Ukita, J.
Org. Chem. 1991, 56, 4585–4588.
[19] K. Ohmori, S. Nishiyama, S. Yamamura, Tetrahedron Lett. 1995, 36,
6519–6522.
[20] B. Gasper, E. M. Carreira, Angew. Chem. Int. Ed. 2007, 46, 4519–4522;
Angew. Chem. 2007, 119, 4603–4606.
[21] K. Ando, J. Org. Chem. 1997, 62, 1934–1939.
[22] F. P. Touchard, N. Capelle, M. Mercier, Adv. Synth. Catal. 2005, 347,
707–711.
[23] O. Labeeuw, P. Phansavath, J.-P. Genêt, Tetrahedron Lett. 2004, 45,
7107–7110.
[24] A. V. Rama Rao, T. K. Chakraborty, K. Laxma Reddy, A. Srinivasa Rao,
Tetrahedron Lett. 1994, 35, 5043–5046.
[25] S. Yamada, T. Oguri, T. Shioiri, J. Chem. Soc., Chem. Commun. 1976,
136–137.
[26] I. Ohtani, T. Kusumi, Y. Kashman, H. Kakisawa, J. Am. Chem. Soc.
1991, 113, 4092–4096.
[27] K. Suzuki, H. Maeta, T. Matsumoto, Tetrahedron Lett. 1989, 30, 4853–
4856.
[28] C. B. Reese, J. C. M. Stewart, J. H. van Boom, H. P. M. de Leeuw, J.
Nagel, J. F. M. de Rooy, J. Chem. Soc., Perkin Trans. 1 1975, 934–942.
[29] B. Yu, Acc. Chem. Res. 2018, 51, 507–516.
[30] CCDC 1550053, 1551801, 1566258 (for compounds 10, 28, and p-
bromobenzamide derivative of 8 (i.e., compound S2 in the Supporting
Information)) contain the supplementary crystallographic data for this
paper. These can be obtained free of charge from The Cambridge
Crystallographic Data Centre.
This article is protected by copyright. All rights reserved.

10.1002/anie.201811403
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
COMMUNICATION
OH
Fumihiko Yoshimura,* Ryusei Itoh,
Makoto Torizuka, Genki Mori, Keiji
Tanino*
O
HO
HO
O
OH
OH
HO
NHAc
HO
Me
O
Me
Me
O
O
Me
O
R
R
HO
Me
H
Me
BH
C
O
CO2H
NH2
O
CO2H
NH2
Me
H
O
Sugars
H
Me
A
Me
Me
Page No. – Page No.
HO
Me
C
A
H
HO
HO
B
H
Me
Me
Me
Brasilicardin A: R = OMe
Brasilicardin B: R = H
Me Brasilicardin C: R = OMe
Brasilicardin D: R = H
Me
R
Me
H
Asymmetric Total Synthesis of
Brasilicardins
The asymmetric total synthesis of brasilicardin A, a hybrid natural product with
highly potent immunosuppressive activity, has been developed. The synthesis
features a Michael addition-based strategy for the stereoselective formation of the
highly strained anti-syn-anti-fused perhydrophenanthrene skeleton, and
brasilicardins B–D are also synthesized based on the unified approach.
This article is protected by copyright. All rights reserved.