Total synthesis of (-)-aspergilazine A

Article abstract of DOI:10.1021/ol5024097

The total synthesis of (-)-aspergilazine A, an alkaloid possessing a rare N1′ to C6 bisindole bond, is described. A palladium-catalyzed N-arylation was used to selectively install the N1′-C6 bond in the presence of three other possible arylation sites. The ligand XPhos displayed a unique capability to efficiently carry out the N-arylation while simultaneously suppressing epimerization of the sensitive C9 stereocenters. This total synthesis has confirmed that aspergilazine A is a dimer of brevianamide F.

Full text of DOI:10.1021/ol5024097

Letter
pubs.acs.org/OrgLett
Total Synthesis of (−)-Aspergilazine A
Emily M. Boyd and Jonathan Sperry*
School of Chemical Sciences, University of Auckland, 23 Symonds Street, Auckland, New Zealand
*
S Supporting Information
ABSTRACT: The total synthesis of (−)-aspergilazine A, an
alkaloid possessing a rare N1′ to C6 bisindole bond, is
described. A palladium-catalyzed N-arylation was used to
selectively install the N1′−C6 bond in the presence of three
other possible arylation sites. The ligand XPhos displayed a
unique capability to efficiently carry out the N-arylation while
simultaneously suppressing epimerization of the sensitive C9
stereocenters. This total synthesis has confirmed that
aspergilazine A is a dimer of brevianamide F.
spergilazine A (1) is a diketopiperazine dimer isolated
Scheme 1. Proposed Synthetic Route to (−)-Aspergilazine
A
A
from the marine-derived fungus Aspergillus taichungensis
1
ZHN-7-707 in 2012 (Figure 1). Aspergilazine A contains a
rare N1′−C6 bisindole bond that, to the best of our
2
knowledge, only exists in one other known natural product.
In continuation of our interest in bisindole alkaloids that
3
possess uncommon dimerization sites, we describe herein a
total synthesis of 1 that employs a palladium-catalyzed N-
arylation to forge the N1′−C6 bond.
unwanted arylation at this site, favoring selective hetero-
coupling between 2 and 3.
The first task was the synthesis of the N-arylation partners
Figure 1. Aspergilazine A (1) and brevianamide F (2).
2
and 3. Brevianamide F (2) is available by thermal
8
deprotection−cyclization of N-Boc-L-Pro-L-Trp-OMe, but
substantial epimerization of the tryptophan stereocenter
Marfey’s analysis of 1 confirmed the presence of L-proline,
and NOESY experiments showed a cis-relationship between
H-9 and H-15 in both diketopiperazines. Aspergilazine A is
thus proposed to be a dimer of cyclo-L-prolyl-L-tryptophyl4,
otherwise known as the natural product brevianamide F (2).
Our planned synthesis of 1 hinged on the selective indole N-
arylation of 6-bromobrevianamide F (3) with brevianamide
F (2), a challenging disconnection due to the presence of four
possible N-arylation sites (2 × indole N−H and 2 ×
diketopiperazine N−H’s, Scheme 1). However, we were
confident of uncovering a catalytic system that would be
selective for the indole and that the presence of the C6
bromide close to the indole N−H in 3 would restrict
(
C9) gave less than satisfactory results. Subjecting N-Fmoc-
9
L-Pro-L-Trp-OMe (4) to piperidine−DMF was the most
reliable method for providing useful quantities of 2 without
epimerization (Scheme 2).
The route in Scheme 2 was used as a guide for the
synthesis of 6-bromobrevianamide F (3). Accordingly, 6-
bromo-L-tryptophan was required and a reported biocatalytic
5
,6
7
10
route seemed attractive (Scheme 3). Enzymatic resolution of
N -acetyl-(±)-tryptophan (5) followed by esterification gave
α
Received: August 14, 2014
Published: September 23, 2014
©
2014 American Chemical Society
5056
dx.doi.org/10.1021/ol5024097 | Org. Lett. 2014, 16, 5056−5059

Organic Letters
Letter
Scheme 2. Synthesis of Brevianamide F (2)
Scheme 4. Failed Copper-Catalyzed Arylations of 2 with 3
and 8
Scheme 3. Synthesis of (−)-6-Bromobrevianamide F (3)
induced epimerization. As a result, the focus shifted to
5a,b
palladium-catalyzed N-arylations.
We chose to employ the more robust N-Boc bromide 8 in
the subsequent palladium-catalyzed N-arylation, as it did not
appear to undergo epimerization or degradation as readily as
5a,b
bromide 3 (Scheme 4). Early attempts using Hartwig’s
conditions only epimerized 2 and 8 (Table 1, entries 1 and
5c
2
), and therefore, Buchwald’s dialkylbiaryl phosphine ligands
were trialled. While the N-arylation with DavePhos was
t
16
unsuccessful (entry 3), Bu-XPhos gave the desired product
in 6% yield (entry 4) that rose to 33% by increasing the
9
catalyst and ligand loadings (entries 5−6). Interestingly,
epimerization was only evident at the higher loading (entry
6
) and subsequent efforts focused on eliminating this process.
In particular, a publication detailing that SPhos suppresses
racemization in the Suzuki coupling of α-amino acids
17
6
-bromo-L-tryptophan methyl ester (6) with a 92% enantio-
appeared promising. Indeed, SPhos completely suppressed
epimerization in our system (even at 100 °C) but with no
benefit to the overall yield (entry 7). Further ligand screening
1
1
meric excess. Subsequent amide coupling with N-Fmoc-L-
proline gave dipeptide 7 as a rotameric mixture that upon
9
t
16
exposure to piperidine−DMF underwent concomitant de-
showed that Me BuXPhos performed poorly on both fronts
4
18
protection−cyclization to give 6-bromobrevianamide F (3) as
(epimerization and yield; entry 8) and RuPhos suppressed
epimerization but gave a poor yield of 9 (entry 9). In a
pivotal experiment, XPhos doubled the previous best yield
(entry 10) with no evidence of epimerization and increasing
the catalyst and ligand loadings did not induce epimerization
12,13
a single diastereomer.
A strong NOE correlation between
7,11
H-9 and H-15 confirmed the desired cis-relationship.
With the coupling partners 2 and 3 in hand, the N-arylation
was considered. Due to the low abundance of N-arylindoles in
nature, intermolecular indole N-arylations are uncommon in
total synthesis and hence the utility of this reaction in the6
assembly of complex molecules is somewhat unexplored1.
t
(as in the case of BuXPhos, entries 4−6), affording 9 in 79%
yield (entry 11). We examined the behavior of the bidentate
ligand XantPhos in the N-arylation, obtaining a yield of 24%
with accompanying epimerization (entry 12).
4
Notable exceptions are the total syntheses of psychotrimine
that used Buchwald’s
5
d,e
copper-diamine conditions to install
Encouraged by the successful N-arylation of 2 with 8, we
were eager to attempt the protecting-group-free N-arylation
(Scheme 5). Upon subjecting 6-bromobrevianamide F (3)
and brevianamide F (2) to the optimized conditions from
Table 1, the N-arylated product 1 was obtained in 9% yield
along with an 80% recovery of 2. The yield of this reaction is
limited by instability of bromide 3 under the reaction
conditions, and various attempts at adding bromide 3
portionwise over the course of the reaction did not increase
the yield of 1. Even so, a selective indole N-arylation in the
presence of four possible arylation sites had been accom-
plished, with no epimerization apparent. The product 1 was
spectroscopically identical to the material obtained by
removing the Boc group from dimer 9 (Scheme 5).
the N-arylindole subunit. This previous work was used as a
starting point for the attempted arylation of brevianamide F
(
2) with 6-bromobrevianamide F (3) but was found
5d,e
unsuccessful. Various copper-diamine
and copper-phenan-
1
5
throline catalytic systems resulted in the epimerization of 2
and 3 (C9) at temperatures above 100 °C (Scheme 4).
Conducting the reactions at 90−95 °C suppressed epimeriza-
tion, but no arylation products were observed. The poor
recovery of bromide 3 from all the reactions suggested it was
degrading, and as a result, the N-Boc bromide 8 was
prepared. However, the attempted N-arylation of 8 with 2
under a similar array of copper-catalyzed conditions led to
comparable results: no reaction at 90−95 °C and epimeriza-
tion of 2 at 100 °C (Scheme 4). We were now faced with the
conundrum that not only did the N-arylation fail under
copper catalysis, substrates 2 and 3 were prone to thermally
1
1
The NMR data of synthetic 1 were in full agreement with
1
,11
the natural product.
However, the optical rotation of
synthetic 1 {[α]2 −70.0 (c = 0.10, CH Cl /MeOH 1:1)} was
2
D
2
2
5
057
dx.doi.org/10.1021/ol5024097 | Org. Lett. 2014, 16, 5056−5059

Organic Letters
Letter
a
Table 1. Palladium-Catalyzed N-Arylation of 8 with 2
b
c
d
entry
catalyst (mol %)
ligand (mol %)
temp (°C)
time (h)
yield 9 (%)
epimerization
epi-2/epi-8
t
1
2
3
4
5
6
7
8
9
Pd(dba) (10)
P( Bu) ·HBF (8)
90
95
40
64
64
46
17
39
45
72
39
29
39
39
0
0
2
3
4
Pd(OAc) (30)
dppf (40)
epi-2/epi-8
epi-2/epi-8
no
2
Pd (dba) (15)
DavePhos (60)
tBuXPhos (20)
tBuXPhos (40)
tBuXPhos (60)
90
0
2
3
Pd (dba) (5)
90
6
2
3
Pd (dba) (10)
90
25
33
28
8
no
2
3
e
Pd (dba) (15)
90
epi-2/epi-8 /epimeric dimers
2
3
Pd (dba) (15)
SPhos (60)
100
90
no
2
3
t
f
Pd (dba) (10)
Me BuXPhos (40)
epi-2/epi-8/epimeric dimers
2
3
4
Pd (dba) (15)
RuPhos (60)
XPhos (40)
XPhos (60)
XantPhos (5)
90
14
62
79
24
no
2
3
1
1
1
0
1
2
Pd (dba) (10)
100
100
no
2
3
Pd (dba) (15)
no
2
3
h
g
Pd (dba) (1.7)
95
epi-2/epi-8 /epimeric dimers
2
3
a
b
All reactions carried out on a 0.02 mmol scale using toluene as solvent. Entries 1 and 2 used Cs CO (1.5 equiv) as the base. Entries 3−12 used
2
3
c
d
K PO (4 equiv). Pd (dba) was superior to Pd(OAc) in entries 3−12 (ref 5b). Epimerization occurs at C9. Epimeric dimers refer to the
3
4
2
3
2
1
presumed formation (by H NMR) of the three compounds 9-epi-9, 9′-epi-9, and 9,9′-di-epi-9, of which only 9-epi-9 could be completely
e
f
g
h
characterized. ∼2% combined yield of the three epimeric dimers. ∼9% yield. ∼20% yield. Catalyst and ligand loadings were taken from ref 19.
Scheme 5. Total Synthesis of (−)-Aspergilazine A (1)
To conclude, a selective palladium-catalyzed indole N-
arylation has been used to complete the total synthesis of the
natural product (−)-aspergilazine A (1). Although the
arylation between 2 and 3 gave a poor yield of
(
−)-aspergilazine A (1), it represents a selective indole N-
arylation in the presence of four possible arylation sites. Using
the protected bromide 8 in the N-arylation gave an excellent
yield of the product 9, which upon facile deprotection gave
(
−)-aspergilazine A (1). The XPhos ligand demonstrated a
unique capability to bring about N-arylation while simulta-
neously suppressing epimerization during the coupling. As
previous examples of intermolecular indole N-arylations in
14
natural products synthesis have relied on copper catalysis,
the use of palladium catalysis herein represents an important
advance in this area.
ASSOCIATED CONTENT
■
*
S
Supporting Information
Experimental procedures, 1D and 2D NMR spectra along
with tabulated spectroscopic data. This material is available
free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
2
5
of opposite sign to the literature value {[α] +72 (c = 0.10,
D
1
*
E-mail: j.sperry@auckland.ac.nz.
CH Cl /MeOH, 1:1)}. After some initial concern, commu-
2
2
nication with the authors of the isolation paper confirmed the
Notes
optical rotation had been reported incorrectly and should
1
b
The authors declare no competing financial interest.
have been quoted as negative.
Hence, the absolute
configuration of (−)-aspergilazine A (1) is as shown and is
indeed a dimer of brevianamide F. Unfortunately, we were
unable to obtain a sample of natural (−)-1 for direct
comparison to the synthetic material.
ACKNOWLEDGMENTS
■
We thank the Royal Society of New Zealand for the award of
a Rutherford Discovery Fellowship (J.S). Professor Dehai Li
5
058
dx.doi.org/10.1021/ol5024097 | Org. Lett. 2014, 16, 5056−5059

Organic Letters
Ocean University of China) is thanked for helpful
Letter
(
discussions regarding the optical rotation of aspergilazine A.
REFERENCES
■
(
1) (a) Cai, S.; Kong, X.; Wang, W.; Zhou, H.; Zhu, T.; Li, D.; Gu,
Q. Tetrahedron Lett. 2012, 53, 2615−2617. (b) Corrigendum: Cai,
S.; Kong, X.; Wang, W.; Zhou, H.; Zhu, T.; Li, D.; Gu, Q.
Tetrahedron Lett. 2014, 55, 5404.
(
2) Cividiclavine: Vining, L. C.; McInnes, A. G.; Smith, D. G.;
Wright, J. L. C.; Taber, W. A. FEMS Symp. 1982, 13, 243−251.
3) (a) Boyd, E. M.; Sperry, J. Tetrahedron Lett. 2012, 53, 3623−
626. (b) Wang, C.; Sperry, J. Chem. Commun. 2013, 4349−4351.
c) Wang, C.; Sperry, J. Tetrahedron 2014, 70, 3430−3439. (d) Blair,
L. M.; Colby Davie, E. A.; Sperry, J. Org. Biomol. Chem. 2014, 12,
878−6884.
4) (a) Birch, A. J.; Russell, R. A. Tetrahedron 1972, 28, 2999−
008. (b) Steyn, P. S. Tetrahedron 1973, 29, 107−120. (c) Kobayashi,
(
3
(
6
(
3
M.; Aoki, S.; Gato, K.; Matsunami, K.; Kurosu, M.; Kitagawa, I.
Chem. Pharm. Bull. 1994, 42, 2449−2451. (d) Medhi, R. B. A.;
Shaaban, K. A.; Rebai, I. K.; Smaoui, S.; Bejar, S.; Mellouli, L. Nat.
Prod. Res. 2009, 23, 1095−1107.
(
́
5) (a) Mann, G.; Hartwig, J. F.; Driver, M. S.; Fernandez-Rivas, C.
J. Am. Chem. Soc. 1998, 120, 827−828. (b) Hartwig, J. F.; Kawatsura,
M.; Hauck, S. I.; Shaughnessy, K. H.; Alcazar-Roma, L. M. J. Org.
Chem. 1999, 64, 5575−5580. (c) Old, D. W.; Harris, M. C.;
Buchwald, S. L. Org. Lett. 2000, 2, 1403−1406. (d) Klapars, A.;
Antilla, J. C.; Huang, X.; Buchwald, S. L. J. Am. Chem. Soc. 2001,
1
23, 7727−7729. (e) Antilla, J. C.; Klapars, A.; Buchwald, S. L. J.
Am. Chem. Soc. 2002, 124, 11684−11688.
6) For reviews on indole N-arylations, see: (a) Joucla, L.;
(
Djakovitch, L. Adv. Synth. Catal. 2009, 351, 673−714. (b) Xu, H.
Mini-Rev. Org. Chem. 2009, 6, 367−377.
(
7) Aspergilazine A numbering has been used for consistency
purposes.
8) (a) Caballero, E.; Avendan
003, 68, 6944−6951. (b) Lopez-Coben
J. D.; Lopez-Alvarado, P.; Ramos, M. T.; Avendano, C.; Menen
(
2
̃
o, C.; Menen
́
dez, J. C. J. Org. Chem.
́
̃
as, A.; Cledera, P.; San
́
chez,
́
́
dez, J.
C. Synthesis 2005, 3412−3422. (c) Kieffer, M. E.; Chuang, K. V.;
Reisman, S. E. J. Am. Chem. Soc. 2013, 135, 5557−5560.
(
9) Gentilucci, L.; Cerisoli, L.; De Marco, R.; Tolomelli, A.
Tetrahedron Lett. 2010, 51, 2576−2579.
(
10) Blaser, G.; Sanderson, J. M.; Batsanov, A. S.; Howard, J. A. K.
Tetrahedron Lett. 2008, 49, 2795−2798.
(
11) See Supporting Information for full details.
(
12) The minor diastereomer (<5%; originating from small
quantities of the D-tryptophan and/or epimerization during the
cyclization) is advantageously removed by flash column chromatog-
raphy to give diastereomerically pure diketopiperazine 3.
(
13) For the synthesis of an enantiopure 6-bromobrevianamide
derivative using an asymmetric hydrogenation, see: Kim, J.;
Movassaghi, M. J. Am. Chem. Soc. 2011, 133, 14940−14943.
(
14) (a) Newhouse, T.; Baran, P. S. J. Am. Chem. Soc. 2008, 130,
1
0886−10887. (b) Newhouse, T.; Lewis, C. A.; Eastman, K. J.;
Baran, P. S. J. Am. Chem. Soc. 2010, 132, 7119−7137. (c) Matsuda,
Y.; Kitajima, M.; Takayama, H. Org. Lett. 2008, 10, 125−128.
(
d) Takahashi, N.; Ito, T.; Matsuda, Y.; Kogure, N.; Kitajima, M.;
Takayama, H. Chem. Commun. 2010, 46, 2501−2503.
(15) Altman, R. A.; Buchwald, S. L. Org. Lett. 2006, 8, 2779−2782.
(16) Anderson, K. W.; Tundel, R. E.; Ikawa, T.; Altman, R. A.;
Buchwald, S. L. Angew. Chem., Int. Ed. 2006, 45, 6523−6527.
17) Prieto, M.; Mayor, S.; Lloyd-Williams, P.; Giralt, E. J. Org.
Chem. 2009, 74, 9202−9205.
18) Maiti, D.; Fors, B. P.; Henderson, J. L.; Nakamura, Y.;
Buchwald, S. L. Chem. Sci. 2011, 2, 57−68.
19) Chapman, C. J.; Matsuno, A.; Frost, C. G.; Willis, M. C. Chem.
Commun. 2007, 38, 3903−3905.
(
(
(
5
059
dx.doi.org/10.1021/ol5024097 | Org. Lett. 2014, 16, 5056−5059