Total synthesis of cladoniamide G

Article abstract of DOI:10.1021/ol400055v

The total synthesis of cladoniamide G, a cytotoxic compound against MCF-7 breast cancer cells (10 μg/mL), was accomplished. Key steps in the sequence include oxidative dimerization of 3-acetoxy-5-chloroindole and a tandem process incorporating three steps

Full text of DOI:10.1021/ol400055v

ORGANIC
LETTERS
2013
Vol. 15, No. 5
1152–1154
Total Synthesis of Cladoniamide G
Benjamin C. Loosley, Raymond J. Andersen,* and Gregory R. Dake*
Department of Chemistry, 2036 Main Mall, University of British Columbia, Vancouver,
B.C., Canada, V6T 1Z1
gdake@chem.ubc.ca; randersn@mail.ubc.ca
Received January 8, 2013
ABSTRACT
The total synthesis of cladoniamide G, a cytotoxic compound against MCF-7 breast cancer cells (10 μg/mL), was accomplished. Key steps in the
sequence include oxidative dimerization of 3-acetoxy-5-chloroindole and a tandem process incorporating three steps: bimolecular carbonyl
addition, lactam formation, and carbamate removal.
Natural products containing a 2,20-bisindole skeleton
areknowntoderivefroma variety of marineand terrestrial
organisms.1 Recently, the Andersen group reported the
isolation and structural elucidation of a new class of
indolotrypoline alkaloids, the cladoniamides, from ex-
tracts of Streptomyces uncialis harbored within the lichen
Cladonia uncialis found near the Pitt River, British
Columbia (Figure 1).2 Ryan has reported the biogenetic
gene cluster for the cladoniamides that suggests they arise
biosynthetically through a tryptophan dimer known as a
biosynthetic precursor to the well-described indolocarba-
zole class of natural products (e.g., staurosporine and
rebeccamycin).3 A characteristic difference between the
indolotryptoline and the indolocarbazole alkaloid classes
is the relative orientation of the bisindole subunit;one of
the indole fragments is flipped within the indolotryptoline.
Compounds within the cladoniamide family differ by
the number of carbon atoms (21 or 22), the position of
functional groups, oxidation level, and halogen substitu-
tion. The presence of chloride substituents has been sug-
gested to be a critical prerequisite for biological activity, as
cladoniamide G (1) is weakly cytotoxic against MCF-7
breast cancer cells (10 μg/mL in vitro) compared to the
activity of cladoniamide F. Similarly, cladoniamide A (3),
featuring an imide functional group, possesses potent activ-
ity (8.8 ng/mL) against human colon cancer HCT116 cells,
while cladoniamide C (5) lacks this activity.4
We were interested in developing a synthetic approach
to the cladioniamide set of natural products that would
enable simple manipulations in order to generate a set of
structural analogs with more significant biological po-
tency. Cladoniamide G (1) was selected as a target to
provide a context for an initial tactical approach. Clado-
niamide G displayed, at the onset of the work, the most
significant biological activity.5 The lessons learned during
this exercise would be utilized in second generation ap-
proaches to other, perhaps more structurally complicated
natural and artificial compounds.
Our analysis of 1 involved establishing the central ring
connecting the two indole partners at a late stage of the
synthesis (eq 1). Condensation between a biselectrophilic
(1) Sanchez, C.; Mendez, C.; Salas, J. A. Nat. Prod. Rep. 2003, 23,
1007–1045.
(2) Williams, D. E.; Davies, J.; Patrick, B. O.; Bottriell, H.; Tarling,
T.; Roberge, M.; Andersen, R. J. Org. Lett. 2008, 10, 3501–3504.
(3) (a) Ryan, K. PLoSONE 2011, 6, e23964. (b) Nakano, H.; Omura,
S. J. Antibiot. 2009, 62, 17–26. (c) Ryan, K. S.; Drennan, C. L. Chem.
Biol. 2009, 16, 351–364.
(4) Chang, F.-Y.; Brady, S. F. J. Am. Chem. Soc. 2011, 133, 9996–
9999.
(5) The synthesis of 3 and its methylated analog BE-54017 has
recently been achieved: Kimura, T.; Kanagaki, S.; Matsui, Y.; Imoto,
M.; Watanabe, T.; Shibasaki, M. Org. Lett. 2012, 14, 4418–4421.
r
10.1021/ol400055v
Published on Web 02/15/2013
2013 American Chemical Society

synthon as represented by A and the C2ꢀC20 bisindole
6 would generate the carbon skeleton of 1. The key
C2ꢀC20 bisindole6wouldbeconstructedusingestablished
indigo dye chemistry.6 5,50-Dichloroindigo is a relatively
expensive starting material,7 and consequently some initial
experiments to establish the feasibility of this approach
were undertaken using indigo as the starting material.
Acetate 7 is readily available from the chemical reduc-
tion of indigo using tin metal, acetic anhydride, and acetic
acid (eq 2). The acetate function could be converted to a
methyl ether in 51% yield through saponification using
tetrabutylammonium hydroxide in the presence of methyl
iodide. Early experiments had demonstrated the high re-
activity of electrophiles at C-3 of the desoxygenated indole.
Thus, the reaction with 7 or 8 with diethyl 2-oxomalonate
in ethyl acetate led to carbonyl addition product 9 or 10 in
82% and 90% yields respectively.
Figure 1. Cladoniamides and biosynthetic relatives.
hydrazine and sodium hydroxide with a dimethyl sulfate
quench.9 This reaction converted 14 to 6 directly in 34%
yield, a 7-fold improvement. Reaction of 6 with diethyl
oxomalonate in hot ethyl acetate to give 16 was followed
by its treatment with DBU to generate ethyl ester 17.
Interestingly, the reaction of 17 with methylamine
(5 equiv) generated diamide 18 in 88% yield. We surmised
that limiting the amount of methylamine to 1 equiv or
lower would allow for the formation of 1 regardless of the
site selectivity of the nucleophilic attack of methylamine.
In the event, limiting the quantity of methylamine to
1 equiv or lower still produced 18 and unreacted 17. The
use of a more hindered nucleophile such as N-methylben-
zylamine or modifying solvents or temperatures was not
successful. Diamide 18 or its O-tert-butyldimethylsilyl
ether derivative could not be manipulated under a variety
of conditions to afford the desired lactam.
At this point, we were forced to track backward to use
a conjunctive electrophilic reagent with 6 that contained
three distinct carbonyl functions. Diester 19, produced
through a Knoevenagel condensation between dimethyl
malonate and benzaldehyde,10 was carefully saponified to
generate carboxylic acid 20 (Scheme 2).11 Standard manip-
ulations generated carbamate 21. Oxidative cleavage of 21
with ozone in the absence of methanol as a cosolvent
Attempts to construct the lactam ring of the indolotryp-
toline core using boron trifluoride etherate activation of 9
were unsuccessful.8 Treatment of 10 with DBU, however,
generated the desired β-ester lactam in 53% yield. The
sequence of experiments in eqs 2 and 3 had provided us
with (a) the means to convert the acetate within 7 to a
methyl ether prior to lactam formation and (b) the appar-
ent need for basic conditions to generate the lactam ring
within 11. With this information in hand, we were ready to
utilize 5-chloroindole as a starting material.
Commercially available 5-chloroindole (12) was converted
to 3-acetoxy-5-chloroindole (13) by initial iodination fol-
lowed by a iodideꢀacetate exchange process promoted
by silver(I) (Scheme 1). Subjecting 13 to sodium hydroxide
in ethanol led, as precedented, to the formation of 5,50-
dichloroindigo (14).6 The reduction of 14 using tin metal as
precedented for indigo led to a disappointing 22% yield of
bisindole 15. Using the procedure used to process 7, the
saponification and methylation of 15 resulted in methyl
ether 6, but only in 21% yield. The attempts to improve the
yields for each of these processes were uniformly unsuc-
cessful. In response, a one-pot procedure was developed
that utilizes the reduction of indigo compounds using
(6) Tanoue, Y.; Sakata, K.; Hashimoto, M.; Hamada, M.; Kai, N.;
Nagai, T. Dyes Pigm. 2004, 62, 101–105.
(9) (a) Borsche, W.; Meyer, R. Ber. 1921, 54, 2854–2856. (b) Bergman,
J.; Egestad, B.; Eklund, N. Tetrahedron Lett. 1978, 19, 3147–3150.
(10) Smith, A. B.; Liu, Z. Org. Lett. 2008, 10, 4363–4365.
(11) Flipo, M.; Beghyn, T.; Lehoux, V.; Florent, I.; Deprez, B. P.;
Deprez-Poulain, R. F. J. Med. Chem. 2007, 50, 1322–1334.
(7) 5,50-Dichloroindigo: $200 CAD/gram; Indigo: $1.40 CAD/gram.
(8) Spectroscopic experiments on the reaction products suggest
dehydrative cyclizations to form polycycles took place.
Org. Lett., Vol. 15, No. 5, 2013
1153

Scheme 1. Approach to 1
Scheme 2. Successful Construction of 1
conversion of starting materials to 1, isolated in 81% yield.
This synthetically prepared material shared all spectro-
scopic data with those of the natural compound.
Access to synthetic cladoniamide G has been achieved
using a sequence with a low step count (9 steps, 5 in the
longest linear sequence, 15% from 5-chloroindole, 27%
from dimethyl malonate). The information gathered in this
exercise will be useful in second generation approaches to
structurally related natural products and their analogs that
could display interesting medicinal and biological chem-
istry properties. Of specific interest to us for further study
are N-glycosyl derivatives of 1. We are also interested in
examining the chemistry of other members of the clado-
niamide family. These studies are underway, and results
will be reported in due course.
generated, after workup with triphenylphosphine, linchpin
reagent 22. It is recommended that 22 (and similar com-
pounds) be used promptly. During storage, these com-
pounds were noted to convert to carbonyl hydrates.
The reaction of 6 with 22 proceeded efficiently in ethyl
acetate at reflux. The examination of the reaction mixture
after 12 h using NMR spectroscopy resulted in a gratifying
observation;the target compound 1 was a significant
product along with unreacted 6. The carbonyl addition
adduct formed in situ unexpectedly underwent additional
spontaneous lactam formation and tert-butyl carbamate
deprotection. Running the reaction between 6 and 22 for
a longer reaction time, 72 h, led to essentially complete
Acknowledgment. The authors thank the University of
British Columbia, the Natural Science and Engineering
ResearchCouncil(NSERC) ofCanada, theCanadaFoun-
dation for Innovation (CFI), and the Canadian Cancer
Society Research Institute for the financial support of our
programs.
Supporting Information Available. Experimental pro-
cedures and characterization data for previously unre-
ported compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
The authors declare no competing financial interest.
1154
Org. Lett., Vol. 15, No. 5, 2013