Biomimetic synthesis of ent -(-)-azonazine and stereochemical reassignment of natural product

Article abstract of DOI:10.1021/ol4015706

The first total synthesis of ent-(-)-azonazine has been accomplished using a hypervalent iodine-mediated biomimetic oxidative cyclization to construct the highly strained core. Based on the results from the completed synthesis, both the relative and absolute configurations of natural (+)-azonazine were revised.

Full text of DOI:10.1021/ol4015706

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Biomimetic Synthesis of ent-(ꢀ)-
Azonazine and Stereochemical
Reassignment of Natural Product
Ji-Chen Zhao, Shun-Ming Yu, Yun Liu, and Zhu-Jun Yao*
State Key Laboratory of Bioorganic and Natural Products Chemistry,
Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences,
345 Lingling Road, Shanghai 200032, China
yaoz@sioc.ac.cn
Received June 4, 2013
ABSTRACT
The first total synthesis of ent-(ꢀ)-azonazine has been accomplished using a hypervalent iodine-mediated biomimetic oxidative cyclization to
construct the highly strained core. Based on the results from the completed synthesis, both the relative and absolute configurations of natural
(þ)-azonazine were revised.
The study of fungi within the genus Aspergillus provides
an abundant source of bioactive small molecules posses-
sing unparalleled structural diversity.1 In 2010, Crews
et al. reported a novel hexacyclic dipeptide, azonazine
(1a, Figure 1), from a Hawaiian marine sediment-derived
fungus, Aspergillus insulicola.2 It was found to show anti-
inflammatory activity by inhibiting NF-κB luciferase and
nitrite production. Structurally, azonazine (1a) contains
a benzofuranoindoline ring system bearing a quaternary
center at the C10 position, which presents similarity to the
core of diazonamide A3 (2, Figure 1). However, a unique
transannular 10-membered ring (F) connecting the tetra-
cyclic core (A/B/C/D) and the diketopiperazine unit (E)
makes the skeleton of azonazine much more rigid than that
of diazonamide A. For its challenging structure and inter-
esting biological activity, azonazine is now becoming a
good platform for new methodology examination and also
an attractive target for total synthesis.4ꢀ6 Herein, we want
Figure 1. Retrosynthetic analysis of proposed azonazine (1a).
(1) Cole, R. J.; Jarvis, B. B.; Schweikert, M. A. Handbook of Secondary
Fungal Metabolites; Academic Press: San Diego, 2003; Vols. IꢀIII.
(2) Wu, Q. X.; Crews, M. S.; Draskovic, M.; Sohn, J.; Johnson, T. A.;
Tenney, K.; Valeriote, F. A.; Yao, X. J.; Bjeldanes, L. F.; Crews, P. Org.
Lett. 2010, 12, 4458.
(3) Lindquist, N.; Fenical, W.; Van Duyne, G. D.; Clardy, J. J. Am.
to report the first total synthesis of (ꢀ)-azonazine, together
with our structural revision and reassignment for the
originally proposed structure of natural (þ)-azonazine.
r
Chem. Soc. 1991, 113, 2303.
10.1021/ol4015706
XXXX American Chemical Society

treatment. O-Demethylation of 10 was achieved by treat-
ment with BBr3 in dichloromethane, providing precursor 4
for the key oxidative cyclization.
Scheme 1. Preparation of Diketopiperazine 4a
Table 1. Optimization of Oxidative Cyclization of 4a
results or yield (%)
a BEP = 2-bromo-1-ethyl pyridinium tetrafluoroborate.
conditions
4
11a
11b
From a strategic point of view, construction of the
benzofuranoindoline ringsystem (Figure 1, A/B/C/D rings
of 1a) containing a highly strained transannular 10-mem-
bered ring (F ring) is the most arduous task in the whole
synthesis. To achieve high efficiency in constructing such a
strained multiring system, proper oxidative free-radical
cyclization mimicking the biogenesis pathway2,5b was ta-
ken into consideration. Azonazine (1a, the proposed
structure) could be prepared from the corresponding
indoline 3 by a selective N-acetylation. Formation of the
benzofuranoindoline core of 3 could be achieved by a
biomimetic oxidative cyclization of the diketopiperazine
precursor 4, which could be conveniently prepared from
the commercially available D-Tyr and D-Trp derivatives.
D-Tyrosine derivative 5 and D-tryptopan derivative 8
were employed as the starting materials in our synthesis
(Scheme 1). Double methylation of D-N-Boc-Tyr-OH (5)
provided compound 6,7 which was transformed to D-N-
Me-Tyr(OMe)-OMe hydrochloride salt (7) by treatment
with thionyl chloride in methanol. BEP-mediated coupling8
of 7 with D-N-Boc-Trp-OH (8) provided the dipeptide 9.
Cyclization of 9 to diketopiperazine 10 was carried out by
deprotection of N-Boc functionality and subsequent base
1 PIDA (1 equiv), LiOAc, TFE, ꢀ15 °C, 30 min
2 PIFA (1 equiv), LiOAc, TFE, ꢀ15 °C, 30 min
3 HTIB (1 equiv), LiOAc, TFE, ꢀ15 °C, 30 min
4 PIDA (1 equiv), LiOAc, HFIP, ꢀ15 °C, 30 min
5 PIFA (1 equiv), LiOAc, HFIP, ꢀ15 °C, 30 min
6 HTIB (1 equiv), LiOAc, HFIP, ꢀ15 °C, 30 min
7 Fe(acac)3 (2 equiv), t-BuOK, THF, ꢀ40 °C to rt
8 PIDA (2 equiv), LiOAc, TFE, ꢀ15 °C, 30 min
9 PIDA (2 equiv), LiOAc, TFE, ꢀ30 °C, 30 min
>30
<5
<5
complex
complex
complex
complex
complex
n.r.
0
0
16
15
12
11
a PIDA: (diacetoxyiodo)benzene. PIFA: [bis(trifluoroacetoxy)-
iodo]benzene. HTIB: [hydroxy(tosyloxy)iodo]benzene. TFE: 2,2,2-tri-
fluoroethanol. HFIP: hexafluoroisopropanol.
With 4 in hand, we then turned our attention to the
construction of the benzofuranoindoline ring systems.
Due to the mild and highly selective oxidizing properties,
a considerable number of hypervalent iodine reagents have
been developed and used for various transformations
of complex organic molecules in recent years.5b,9 Several
hypervalent iodine(III) reagents were examined in this
study for the key oxidative cyclization, as well as Fe(acac)3
(Table 1). When diketopiperazine 4 was added into a cold
trifluoroethanol solution of PhI(OAc)2, two unexpected
diastereomeric prodcuts 11a and 11b bearing the hexa-
cyclic core structure of azonazine were generated in very
low isolated yields (entry 1, Table 1). Their structures were
determined by NMR methods and finally confirmed by
X-ray single crystal analyses. The results indicated that an
overoxidation happened at the C14 position. Though the
detailed mechanism is unclear as of yet, compound 15b
(a diastereomer of 3, see Scheme 3 for its structure) is
(4) For a recent study on the synthesis of azonazine, see: Ghosh, S.;
Kinthada, L. K.; Bhunia, S.; Bisai, A. Chem. Commun. 2012, 48, 10132.
(5) For total syntheses of diazonamide A, see: (a) Nicolaou, K. C.;
Bella, M.; Chen, D. Y. K.; Huang, X.; Ling, T.; Snyder, S. A. Angew.
Chem., Int. Ed. 2002, 41, 3495. (b) Burgett, A. W. G.; Li, Q.; Wei, Q.;
Harran, P. G. Angew. Chem., Int. Ed. 2003, 42, 4961. (c) Nicolaou, K. C.;
Bheema Rao, P.; Hao, J.; Reddy, M. V.; Rassias, G.; Huang, X.; Chen,
D. Y. K.; Snyder, S. A. Angew. Chem., Int. Ed. 2003, 42, 1753. (d)
Cheung, C.-M.; Goldberg, F. W.; Magnus, P.; Russell, C. J.; Turnbull,
R.; Lynch, V. J. Am. Chem. Soc. 2007, 129, 12320. (e) Mai, C.-K.;
Sammons, M. F.; Sammakia, T. Angew. Chem., Int. Ed. 2010, 49, 2397.
(f) Knowles, R. R.; Carpenter, J.; Blakey, S. B.; Kayano, A.; Mangion,
I. K.; Sinz, C. J.; MacMillan, D. W. C. Chem. Sci. 2011, 2, 308.
(6) A recent synthesis of benzofluoroindolines: Beaud, R.; Guillot,
R.; Kouklovsky, C.; Vincent, G. Angew. Chem., Int. Ed. 2012, 51, 12546.
(7) Boger, D. L.; Yohannes, D. J. Org. Chem. 1988, 53, 487.
(9) For recent reviews on hypervalent iodines in organic synthesis,
see: (a) Tohma, H.; Kita, Y. Adv. Synth. Catal. 2004, 346, 111. (b)
Zhdankin, V. V.; Stang, P. J. Chem. Rev. 2008, 108, 5299. (c) Merritt,
E. A.; Olofsson, B. Angew. Chem., Int. Ed. 2009, 48, 9052. (d) Dohi, T.;
Ito, M.; Yamaoka, N.; Morimoto, K.; Fujioka, H.; Kita, Y. Tetrahedron
2009, 65, 10797. (e) Zhdankin, V. V. J. Org. Chem. 2011, 76, 1185. (f)
Silva, L. F.; Olofsson, B. Nat. Prod. Rep. 2011, 28, 1722. (g) Duschek, A.;
Kirsch, S. F. Angew. Chem., Int. Ed. 2011, 50, 1524.
(8) (a) Li, P.; Xu, J. C. Tetrahedron 2000, 56, 8119. (b) Li, P.; Xu, J. C.
Chem. Lett. 2000, 29, 204.
B
Org. Lett., Vol. XX, No. XX, XXXX

speculated to be produced at the first stage of the oxidative
cyclization (C, Figure 2). Unfortunately, the produced highly
strained structure induces further PhI(OAc)2-oxidation to
easily happen on the indoline nitrogen (D). Subsequently,
a TFE-addition takes place at the C14 position. As a final
result, all of these PhI(OAc)2-mediated oxidations enable
ketals 11a and 11b to be formed at the C14 position.
Using the cyclized products 11a and 11b, we continued
their transformations to the final products (Scheme 2).
Deprotection of the unusal bis(trifluoroethoxy)acetal at
the C14 of 11a was found to be surprisingly problematic.
After many experimental trials, it was successfully hydro-
lyzed with HOAc/H2O under microwave-assisted condi-
tions, affording a moderate yield of quinone 12a. This
product was immediately reduced with NaBH4 to afford
stable phenol 13a. Phenol 13a was then converted into to
the corresponding triflate 14a under the optimized condi-
tions (Tf2O/Et3N in EtOAc, ꢀ15 °C). Hydrogenative
removal of the triflate functionality of 14a (with a catalytic
amount of Pd(OH)2 on charcol) followed by N-acetylation
under acidic conditions (Ac2O, AcOH) finally afforded the
originally proposed structure of (þ)-azonazine (1a).
Scheme 2. Synthesis of the Proposed Structure of Azonazine
Figure 2. A proposed oxidative process for generation of 11b.
We attempted to improve the reaction by changing
hypervalent iodine reagents (Table 1, entries 2ꢀ3) and
the solvent (entries 4ꢀ6), but all failed. No reaction was
observed when an Fe(III) oxidant was applied (entry 7).
To avoid the overoxidation, modification of the electron
density of the indole moiety of 4 was also attempted.10
Disappointingly, no improvements were achieved. Finally,
we decided to further optimize the conditions with
PhI(OAc)2. The reaction with 2 equiv of PhI(OAc)2 gave
the highest yields of the products at ꢀ15 °C (entry 8), and
further lowering of the reaction temperature provided no
improvement in the isolated yields (entry 9). Relatively
low yields of the cyclized products 11a and 11b suggest that
the high ring strain existing in the congested hexacyclic
core makes the synthesis of such a ring system exception-
ally difficult and inefficient. Crystallographic data of 11a
and 11b (see Supporting Information for the details)
also confirm that the products bore high strain,11 which
is consistent with previous computational calculation by
Crews et al.2
To our surprise, multiple apparent differences were
found in the 1H NMR spectrum of synthetic 1a from those
reported for the natural product. Careful analysis of the
original data2 found that the assignment of the relative
configurations at C10 and C11 might be ambiguous.12 By
comparison with that of 11a, we found the 1H NMR of the
previous intermediate 11b was more similar to the natural
product. We thus decided to synthesize the other C10/C11
diastereomer 1b from 11b for further analysis.
The C10/C11 diastereomeric target 1b was synthesized
with the same steps under similar conditions (Scheme 3).
Hydrolysis of 11b under microwave conditions followed
by NaBH4 reduction of quinone 12b gave phenol 13b.
Deoxygenation of the C14 phenol group of 13b was
accomplished after O-triflation of 13b and subsequent
(10) When the indole nitrogen was protected either by alkyl or Ts, the
corresponding substrates displayed no reactivity in the following oxida-
tions. When a bromine atom was introduced into the benzene ring of the
indole moiety at different positions, the reaction was found to be very
complicated.
(11) Bond angles and dihedral angles of 11a/11b and 1b were highly
distorted according to the crystallographic data (see Table S1 of the
Supporting Information for details).
(12) The authors (ref 2) favored to assign the absolute configuration
of azonazine in parallel to that of diazonamide A (R10/S11), since the
chemical shifts of dihydrobenzofuran of diazonamide A (ref 3) were
virtually identical to those of (þ)-azonazine.
Org. Lett., Vol. XX, No. XX, XXXX
C

Scheme 3. Completion of the Total Synthesis (ꢀ)-Azonazine (1b)
Scheme 4. Synthesis of ent-Azonazine Analogue 16
to explore the fluorinated analogues and derivatives of
azonazine, which are usually difficult to achieve with
traditional modification from the natural product. For
instance, we could easily synthesize the C-14 trifluoroethoxy
derivative 16 of ent-azonazine in two steps from ketal 11b
by reduction with zinc/HOAc under mild conditions14 and
subsequent N-acetylation (Scheme 4).
In conclusion, we have successfully accomplished the
first total synthesis of (ꢀ)-azonazine, applying a biomi-
metic oxidative cyclization as the key step to construct
the highly strained hexacyclic core. With the spectro-
scopic evidence and X-ray crystal data of the enantiopure
diastereomeric products, the structure of natural (þ)-
azonazine was revised as (2S,10R,11S,19S). In addition,
the intermediate produced in this synthesis could be con-
verted into the fluorinated azonazine derivative in short
convenient steps. The biological properties of these natural
and unnatural products, as well as the fluorinated deriva-
tives, are currently under investigation in our laboratory
and will be reported in due course.
hydrogenation. Finally, a high-yield N-acetylation of 15b
provided the final product 1b, whose structure was deter-
mined by NMR methods and confirmed by X-ray single
crystal diffraction. With this solid evidence, the absolute
chemistry of 1b, a C10,C11-diastereomer of the originally
proposed azonazine (1a), was unambiguously assigned as
(2R,10S,11R,19R) (Scheme 3). With the exception of the
opposite optical rotation {synthetic: [R]27D ꢀ299.3 (c 0.11,
Acknowledgment. Financial support from the MOST
973 project (2010CB833200) and NSFC grant (21032002)
is greatly appreciated.
MeOH); natural: [R]23 þ295.0 (c 0.1, MeOH)}, all the
D
physical data of synthetic 1b were in good agreement with
those reported for the natural product.2 Therefore, the
absolute configuration of natural (þ)-azonazine should be
revised as (2S,10R,11S,19S)-1.
Selective introduction of fluorine atom(s) to bioactive
natural and unnatural products is of extreme interest in
pharmaceutical research.13 The intermediates accidently
generated in this synthesis provide a particular opportunity
Supporting Information Available. Experimental details
and charaterizations of new compounds, NMR copies of
new compounds (PDF), and X-ray data of compounds 1b,
11a, and 11b (CIF). This material is available free of
charge via the Internet at http://pubs.acs.org.
(14) Ghaffarzadeh, M.; Joghan, S. S.; Faraji, F. Tetrahedron Lett.
2012, 53, 203.
(13) For a review on fluorine in medicinal chemistry, see: Purser, S.;
Moore, P. R.; Swallow, S.; Gouverneur, V. Chem. Soc. Rev. 2008, 37,
320.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX