Tackling the Spiro Tetracyclic Skeleton of Cyanogramide: Incorporation of a Hydantoin Moiety

Article abstract of DOI:10.1021/acs.orglett.8b03540

Wang's enantioselective thiourea-catalyzed spiro-annulation paved the way to the first tetracyclic analog of the marine natural product cyanogramide from the actinobacterium Actinoalloteichus cyanogriseus. The synthesis comprises seven steps starting from

Full text of DOI:10.1021/acs.orglett.8b03540

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Tackling the Spiro Tetracyclic Skeleton of Cyanogramide:
Incorporation of a Hydantoin Moiety
Marco Monecke and Thomas Lindel*
TU Braunschweig, Institute of Organic Chemistry, Hagenring 30, 38106 Braunschweig, Germany
S
* Supporting Information
ABSTRACT: Wang’s enantioselective thiourea-catalyzed spiro-annulation paved the way
to the first tetracyclic analog of the marine natural product cyanogramide from the
actinobacterium Actinoalloteichus cyanogriseus. The synthesis comprises seven steps starting
from an alkylidene indolinone. Installation of the (E)-styryl side chain faced a
regiochemistry problem, circumvented by prior conversion to the hydantoin and
employing Batey’s conditions. Comparison of the ECD spectra of the spiro indoline
pyrrolo[1,2-c]imidazole and cyanogramide confirmed the absolute configuration of the
natural product.
aminal ester 2. We also included the related tetracyclic
compound 4, where the C2 unit of cyanogramide is replaced by
a carbonyl group. The resulting hydantoin moiety of 4 could
be more stable than ring D of the putative, demethylated
cyanogramide derivative 2. The styryl side chain was to be
introduced by Cu-mediated N−C coupling.
For studies toward 3, we chose racemic 3,3′-thiopyrrolidonyl
spiro-oxindole 5, the ethyl ester analog of which had been
obtained by Jiangtao Sun et al.6 and by Rui Wang et al.7 by
Michael addition/cyclization of α-isothiocyanato imides to 2-
oxoindolin-3-ylidene acetates in the presence of DBU (1,8-
diazabicyclo[5.4.0]undec-7-ene, Scheme 2; see the Supporting
he marine indole alkaloid cyanogramide (1, Scheme 1)
Twas isolated from the marine-derived actinobacterium
Actinoalloteichus cyanogriseus by Weiming Zhu et al. in 2014.1
The molecule features a pyrrolo[1,2-c]imidazole system, which
is spiro-fused to an oxindole. Cyanogramide (1) is the only
natural product exhibiting that tetracyclic partial structure.
Besides, only a few tetracyclic structures are known that share
the tricyclic spiro[indoline-3,3′-pyrrole]dione moiety with
compound 1. In all cases, an oxazolidine ring is annulated
instead of the imidazolinone of cyanogramide (1).2−4 Only
one molecule has been described that contains the double
bond present in ring C of cyanogramide.5 While cyanogramide
(1) showed only weak cytotoxicity against the adriamycin- and
vincristine-resistant cancer cell lines, it reversed their resistance
against adriamycin and vincristine at 5 μM concentration. This
effect was attributed to inhibition of the P-glycoprotein, an
ATP dependent drug efflux pump highly expressed in the
tested cell lines, by cyanogramide (1).1
Scheme 2. Synthesis of Spiro-indolone Pyrrolone 9 from
Spiro-thiolactam 5
Initially, we expected that the pyrrolo[1,2-c]imidazole-1,5-
dione system of cyanogramide (1, rings C and D, Scheme 1)
could be accessed by cyclization of an N-acetylated, pyrrolone
carboxamide-based precursor 3, being in equilibrium with
Scheme 1. Spiro-indolone Alkaloid Cyanogramide (1) from
the Marine-Derived Bacterium Actinoalloteichus
cyanogriseus, Putative Synthetic Precursors
Received: November 5, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b03540
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Information). It was possible to convert spiro-pyrrolidine-2-
thione 5 to the corresponding lactam 6 by treatment with
H2O2/HCO2H (97%), followed by conversion to carboxylic
acid 7 with TFA (trifluoroacetic acid). For the subsequent
Barton decarboxylation we were inspired by Williams’
synthesis of spirotryprostatin B and derivatives.8 Decarbox-
ylation of 7 by formation of the acid chloride, in situ
conversion to the Barton ester, reaction with BrCCl3 under
irradiation conditions,8 and reaction with DBU afforded the
α,β-unsaturated spiro-indolone pyrrolone 9 in modest yield
(27%), which was difficult to purify. All our attempts to subject
9 to ammonolysis led to decomposition, which led us to
abandon that pathway.
Scheme 4. Synthesis of TMSE-Protected Spiro-
pyrrolidinone Carboxamide 19 via Thiourea-Catalyzed
Michael Addition/Cyclization
As long as the ester moiety was in place, ammonolysis of 6
was facile and afforded pyrrolidinone carboxamide 10 (Scheme
3). Buchwald coupling of 10 with (E)-β-bromostyrene (11)
Scheme 3. Study on the Relative Reactivity of the Lactam
and Primary Amide Nitrogen Atoms towards Buchwald
Coupling, Starting from Spiro-indolone 6
20
The optical rotatory power of 17 ([α]D = −33.6, c = 1.0)
corresponds well to values reported by Wang et al. for similar
compounds. Oxidative hydrolysis of 17 afforded lactam 18,
followed by ammonolysis to TMSE ester 19.
As concluded from the behavior of spiro-pyrrolone
carboxamide 10 (Scheme 3), regioselective N-styrylation
would require protection of the lactam nitrogen, ideally by
acetylation. Thus, we attempted the acetylation of lactam 19
(Scheme 5). Acetylated product 20 was obtained regioselec-
tively in quantitative yield. The constitution of 20 was secured
by an HMBC correlation of the α-hydrogen 5′-H and the
acetyl carbonyl carbon. Heating of 19 under reflux with 10
equiv of Ac2O led to double N-acetylation affording 24 (82%).
However, there was no tendency of 20 or 24 to undergo
cyclization to a 2-hydroxy-2-methylimidazolidin-4-one, which
would correspond to the core structure of cyanogramide (1).
Surprisingly, Buchwald coupling of N-acetylated lactam 20
with bromostyrene 11 did not occur at the free primary amide
moiety. Instead, we isolated compound 21 (Scheme 5), where
the acetyl was replaced by the styryl group. Apparently, the
acetyl group was cleaved off by Cs2CO3 that preceded the N−
C coupling. Even under Batey’s conditions11 employing
potassium styryl trifluoroborate (22)/Cu(OAc)2/O2 we
observed undesired styrylation of the lactam nitrogen.
Differing from 21, the acetyl group had migrated to the
neighboring primary amide resulting in compound 23. All
attempts to install a styryl unit at the diacetylated tricycle 24
under either Buchwald’s or Batey’s conditions failed.
under microwave conditions with elevated amounts of CuI
(1.1 equiv)/DMEDA (N,N′-dimethylethylenediamine, 2.2
equiv), as applied in our synthesis of parazoanthine F,9
resulted in the formation of the undesired N-styryllactam 12
(45%). The primary amide moiety had remained in place. To
check the general availability of the primary amide moiety for
Buchwald coupling, we reacted 12 again and obtained bis-
styryl product 13 in moderate yield (41%). Thus, both
functional groups did react, but not in the desired order, and
the lactam nitrogen had to be functionalized before Buchwald
coupling.
For further experiments, we replaced the tert.-butyl ester by a
TMSE ((trimethylsilyl)ethyl) ester function, because the tert-
butyl group had shown some instability. We also switched to
the optically active series. TMSE ester 14 (Scheme 4) was
obtained by Wittig reaction of N-methylisatin and the
corresponding acetic acid based phosphorane (see the
Supporting Information). Reaction of 14 with isothiocyanato
compound 1510 in the presence of chiral organocatalyst 167
afforded spiro-indolinone 17 in high yield (88%) and perfect ee
(Scheme 4, >98%, as judged by analytical HPLC; see the
Supporting Information). The relative configuration of product
17 was derived from an intense NOESY correlation between
5′-H and 4-H of the indolone.
We concluded that the fourth ring of the cyanogramide
skeleton had to be formed prior to styrylation and, given our
observation with the acetyl group, introduced a C1 instead of a
C2 unit. While treatment of lactam-carboxamide 19 with
B
DOI: 10.1021/acs.orglett.8b03540
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 5. Reaction of Mono- or Diacetylated Spiro-indoles
20 and 24 under Buchwald’s and Batey’s Conditions,
Leading to Deacetylation or Acetyl Shift, Respectively, and
Styrylation at the Lactam Nitrogen
Scheme 6. Synthesis of Cyanogramide Analog 4 Containing
a Hydantoin Partial Structure
hydantoin moiety (see the Supporting Information). To our
knowledge, hydantoins have not been subject to copper-
mediated N−C coupling.
Several attempts were made to transform the hydantoin
moiety of analog 4 into ring D of cyanogramide (1) by
treatment with organometallic methylation reagents (MeMgCl,
MeLi) or Cp2TiMe2 (Petasis reagent). Unfortunately, none of
these attempts were successful and only led to decomposition
of tetracycle 4.
The ECD spectrum of 4 (MeOH, c = 0.01 mg/mL) proved
to be almost identical to that of the natural product
cyanogramide (1) [λmax (Δε) 276 (3.0), 258 (−4.9), 237
(6.1), 220 (−9.5) nm for 2, compared to λmax (Δε) 310
(−1.4), 277 (1.1), 258 (−5.3), 238 (2.2), 222 (−2.6) nm for
1] providing independent evidence for the correct assignment
of the absolute configuration of 1. The ECD spectra are clearly
dominated by the rigid tetracyclic cores. The optical rotation
of cyanogramide analog 4 is negative ([α]D20 = −21.0, c = 1.0),
as is the reported value of cyanogramide (1, [α]D23 = −96.0, c
= 0.1).1
Even if the cyclic ester aminal moiety of cyanogramide (1)
has still to be assembled, we consider the enantioselective
synthesis of the close structural analog 4 as an important step
forward. As the ester aminal moiety of cyanogramide (1, ring
D) suggests, facile methanolysis of the alkylidene hydantoin
moieties of 27 and 4 could start at the urea moiety. We
performed orienting quantum chemical calculations (DFT,
ωB97X-D, 6-31G*; see the Supporting Information) that
indicate that cyanogramide (1) indeed prefers the observed
methoxyimidazolidinone ring, whereas the hydroxy analog 2
should prefer (19 kJ/mol in the gas phase, 15 kJ/mol in DMF,
continuum model) an open state with the lactam nitrogen
being acetylated. This may also be true for compound 20
trimethylorthoacetate was unsuccessful, CDI (carbonyl-
diimidazole)/DMAP (N,N-dimethylaminopyridine) afforded
tetracyclic hydantoin 25 in very good yield (93%) without
epimerization (Scheme 6). With thiocarbonyldiimidazole we
obtained a smaller yield (72%) and observed partial
epimerization at C5′. The TMSE protecting group of 25 was
removed by treatment with BBr3/DCM affording carboxylic
acid 26 that could be purified conveniently by precipitation
from MeCN. Fluoride sources such as TBAF (tetra-n-
butylammonium fluoride), HF-pyridine, or NH4F failed.
Carboxylic acid 26 was activated with oxalyl chloride, followed
by addition of sodium pyrithione (8)/DMAP. We were able to
observe the resulting, labile Barton ester on the TLC, as
analyzed by TLC/MS. On prolonged standing, the Barton
ester decomposed. We added BrCCl3 in MeCN, irradiated at
370 nm (Rayonet, 1 h),8 and obtained a rather complex
mixture of reaction products, from which tetracycle 27
containing the desired spiro-pyrrolidone ring was isolated
20
(37% after chromatography, [α]D − 41.6, c = 1.5). Differing
from precursor 26, alkylidene hydantoin 27 proved to be
unstable and very difficult to purify.
On applying Buchwald’s conditions to 27 we did not
observe any product. However, we were pleased to find that
under Batey’s conditions (stoichiometric Cu(OAc)2, oxygen)
target compound 4 could be obtained on reaction with
potassium styryl trifluoroborate (22, 28%).11 As in the case of
27, alkylidene hydantoin 4 proved to be very sensitive under
chromatography conditions and was accompanied by small
amounts (12%) of a methylester derived from a degraded
C
DOI: 10.1021/acs.orglett.8b03540
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(Scheme 5) that did not cyclize to the ester aminal. The
methoxy group of cyanogramide could originate from
methanol, which was used as solvent during extraction and
chromatography.1
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b03540.
All relevant experimental procedures, characterization
data, and NMR spectra (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: th.lindel@tu-bs.de.
ORCID
Thomas Lindel: 0000-0002-7551-5266
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
M.M. thanks the Fonds der Chemischen Industrie for a
doctoral stipend.
■
REFERENCES
■
(1) Fu, P.; Kong, F.; Li, X.; Wang, Y.; Zhu, W. Org. Lett. 2014, 16,
3708−3711.
(2) Sen, S.; Potti, V. R.; Surakanti, R.; Murthy, Y. L. N.; Pallepogu,
R. Org. Biomol. Chem. 2011, 9, 358−360.
(3) Cowley, A. R.; Hill, T. J.; Kocis, P.; Moloney, M. G.; Stevenson,
R. D.; Thompson, A. L. Org. Biomol. Chem. 2011, 9, 7042−7056.
(4) Zhou, Y.; Xi, Y.; Zhao, J.; Sheng, X.; Zhang, S.; Zhang, H. Org.
Lett. 2012, 14, 3116−3119.
(5) Meloche, J. L.; Ashfeld, B. L. Angew. Chem., Int. Ed. 2017, 56,
6604−6608.
(6) Wang, L.; Zhang, D.; Li, J.; Xu, G.; Sun, J. RSC Adv. 2014, 4,
44193−44196.
(7) Cao, Y.; Jiang, X.; Liu, L.; Shen, F.; Zhang, F.; Wang, R. Angew.
Chem., Int. Ed. 2011, 50, 9124−9127.
(8) Sebahar, P. R.; Osada, H.; Usui, T.; Williams, R. M. Tetrahedron
2002, 58, 6311−6322.
(9) Wefer, J.; Lindel, T. Eur. J. Org. Chem. 2015, 2015, 6370−6381.
(10) Li, L.; Klauber, E. G.; Seidel, D. J. Am. Chem. Soc. 2008, 130,
12248−12249.
(11) Bolshan, Y.; Batey, R. A. Angew. Chem., Int. Ed. 2008, 47,
2109−2112.
D
DOI: 10.1021/acs.orglett.8b03540
Org. Lett. XXXX, XXX, XXX−XXX