The discovery of potent antitumor agent C11-deoxypsymberin/irciniastatin a: Total synthesis and biology of advanced psymberin analogs

Article abstract of DOI:10.1021/ol802772s

Structure-activity relationship (SAR) studies by modification of the unsaturated side chain of potent anticancer marine natural product psymberin/irciniastatin A (1) suggest that substitution at C4 and C5 is important for the cytotoxicity of psymberin, bu

Full text of DOI:10.1021/ol802772s

ORGANIC
LETTERS
2009
Vol. 11, No. 4
867-870
The Discovery of Potent Antitumor
Agent C11-Deoxypsymberin/irciniastatin
A: Total Synthesis and Biology of
Advanced Psymberin Analogs
Xianhai Huang,*,† Ning Shao,† Robert Huryk,‡ Anandan Palani,†
Robert Aslanian,† and Cynthia Seidel-Dugan*,‡
Department of Chemical Research and Department of Oncology Research,
Schering-Plough Research Institute, Kenilworth, New Jersey 07033
xianhai.huang@spcorp.com; cynthia.seidel-dugan@spcorp.com
Received December 10, 2008
ABSTRACT
Structure-activity relationship (SAR) studies by modification of the unsaturated side chain of potent anticancer marine natural product psymberin/
irciniastatin A (1) suggest that substitution at C4 and C5 is important for the cytotoxicity of psymberin, but the terminal double bond is not
essential for activity. An aryl group is a good replacement for the olefin. The total synthesis of structurally simplified C11-deoxypsymberin
(29) was completed, and its activity is consistently more potent than the natural product which provides a unique opportunity for further SAR
studies in the psymberin and pederin family. Preliminary mechanism studies suggest the mode of action of psymberin is through cell apoptosis.
Cancer is a leading cause of death in the United States, and
its incidence continues to rise. Although new anticancer
agents continue to emerge, the pharmaceutical industry
continues to seek more potent and selective drugs with fewer
side effect liabilities. A large number of anticancer agents
derived from marine natural products and their synthetic
analogues are either already in clinical trials or advancing
in preclinical development.1 The discovery of potential
anticancer drugs based on marine natural products will be
very beneficial. Recently we completed the total synthesis2
of the potent anticancer marine natural product psymberin/
irciniastatin A (1)3,4 which is a new member of the pederin
family of natural products and an extremely potent and
selective cytotoxin compared to other pederin natural
products.3a Our convergent approach to psymberin provides
us with a good opportunity and unique template to carry out
rational structure-activity relationship (SAR) studies to
discover potential lead anticancer agents. Because of its
reported high potency and unprecedented selectivity, psym-
berin may have a different mode-of-action from that of
(3) Isolation, see: (a) Cichewicz, R. H.; Valeriote, F. A.; Crews, P. Org.
Lett. 2004, 6, 1951, and references cited therein. (b) Pettit, G. R.; Xu, J. P.;
Chapuis, J. C.; Pettit, R. K.; Tackett, L. P.; Doubek, D. L.; Hooper, J. N. A.;
† Department of Chemical Research.
Schmidt, J. M. J. Med. Chem. 2004, 47, 1149.
‡ Department of Oncology Research.
(4) Other total synthesis, see: (a) Jiang, X.; Garcia-Fortanet, J.; De
Brabander, J. K. J. Am. Chem. Soc. 2005, 127, 11254, and references cited
therein. (b) Smith, A. B.; Jurica, J. A.; Walsh, S. P Org. Lett. 2008, 10,
5625. Formal total synthesis, see: (c) Ning, S.; Kiren, S.; Williams, L. J
(1) Simmons, T. L.; Andrianasolo, E.; McPhail, K.; Flatt, P.; Gerwick,
W. H. Mol. Cancer Ther. 2005, 4, 333.
(2) Huang, X.; Shao, N.; Palani, A.; Aslanian, R.; Buevich, A. Org.
Lett. 2007, 9, 2597.
Org. Lett. 2007, 9, 1093
.
10.1021/ol802772s CCC: $40.75
Published on Web 01/26/2009
2009 American Chemical Society

pederin/mycalamide.5 Herein we report the synthesis of
advanced psymberin analogs and the determination of their
antitumor activity.
Before embarking on analog synthesis, we first tested the
activity of synthetic psymberin 1 and its C(8)-C(9) epimer
2 in different human cancer cell lines to confirm the screening
conditions for psymberin analogs as well as to potentially
understand its mode of action. As expected, synthetic
psymberin (1, C(8), C(9) ) S,S, Scheme 1) displayed very
could be important for the cytotoxicity of natural pysmberin.
In addition, we performed some biological studies to
understand the mode of action of the natural product.
Preliminary results supported the idea that psymberin’s mode
of action is through cell apoptosis (see Supporting Informa-
tion). With the assay conditions established, we decided to
use the readily available HOP62 cell line for testing of
analogs.
On the basis of a recent report from the De Brabander
group,6 the dihydroisocoumarin fragment is vitally important
for pysmberin cytotoxicity. Therefore, we decided to focus
our initial analog synthesis on the modification of the
unsaturated “psymberate” side chain (3 and 4, Scheme 1)
since this is a unique feature of psymberin in its class.3a
These compounds can be synthesized using our recently
developed methodology7 to quickly assemble the tetrahy-
dropyran ring. The synthesis of analogs is described in
Scheme 2 (14a-17a and their epimers 14b-17b). Com-
Scheme 1. Rational Design of Psymberin Analogs by
Modification of the Unsaturated “Psymberate” Side Chain
Scheme 2
.
Synthesis of “Psymberate” Side Chain Modified
Psymberins
potent cytotoxic activity against all tumor cell lines studied
(Table 1) with IC50 values in the subnanomolar range.
Table 1. Antitumor Activity of Synthetic Psymberin and Its
C(8)-C(9) Epimera
1 (IC50 nM)
2 (IC50 nM)
cell line
human tissue type
0.76 ( 0.07
0.30 ( 0.03
0.18 ( 0.02
0.81 ( 0.14
0.42 ( 0.02
0.27 ( 0.01
0.28 ( 0.03
0.28 ( 0.02
0.19 ( 0.02
0.82 ( 0.04
0.84 ( 0.08
6800 ( 244
3800 ( 301
2400 ( 431
4900 ( 187
4600 ( 68
4200 ( 174
3600 ( 155
5200 ( 195
3100 ( 341
4800 ( 177
n.d.
ACHN
DU145
H226
HCT116
HOP62
MB231
MB435
MKN45
PC3
kidney
prostate
lung
colon
lung
breast
melanoma
gastric
prostate
colon
pound 52 was coupled with amides using CuI8 to give the
protected N-acyl enamine, in which the C13, C15 acetate,
and O21 TIPS groups were removed with NaOMe/MeOH
followed by selective acetylation of O21 to give compounds
6-9. Enamides 6-9 cyclized smoothly using the PhI(OAc)2
mediated cyclization reaction7 to give the corresponding
cyclization products 10a-13a and their epimers 10b-13b.
The major two diastereomers (C(8)-C(9) ) S,S and
C(8)-C(9) ) R,R)9 were separately treated with TBAF at
50 °C, and a global deprotection was realized to give the
final products 14a-17a and their epimers 14b-17b.
With compounds 14a-17a and their epimers 14b-17b
in hand, we tested them in the human lung cancer cell line
SW620
NHDF
normal
a The CellTiter-Glo Luminescent Cell Viability Assay (Promega,
Technical bulletin 288) was employed in this study. IC50 data are the mean
value of six experiments with statistical significance calculated.
However, the C(8)-C(9) epimer (2, C(8), C(9) ) R,R)
consistently showed dramatically decreased cytotoxicity. This
data strongly suggested that the C(8)-C(9) stereochemistry
(6) Jiang, X.; Williams, N.; De Brabander, J. K. Org. Lett. 2007, 9,
227.
(7) Huang, X.; Shao, N.; Palani, A.; Aslanian, R. Tetrahedron Lett. 2007,
48, 1967.
(5) Narquizian, R.; Kocienski, P. J. The Pederin Family of Antitumor
Agents: Structures, Synthesis and Biological Activity. In The Role of Natural
products in Drug DiscoVery; Mulzer, J., Bohlmann, R., Eds.; Ernst Schering
Research Foundation Workshop 32; Springer: New York, 2000; pp
25-56.
(8) Jiang, L.; Job, G. E.; Klapars, A.; Buchwald, S. L. Org. Lett. 2003,
5, 3667.
(9) The C(8)-R and C(9)-R stereochemistry was assigned by comparison
1
with H spectra from the psymberin synthesis which were determined by
2D NOE NMR and proton-proton coupling between N7, C8, and C9.
868
Org. Lett., Vol. 11, No. 4, 2009

(HOP62), and the results are summarized in Table 2. We
first chose to examine the role of the side chain. When the
more importantly the synthesis of those natural products can
be simplified by eliminating a heteroatom (O) and a
stereogenic center. The same synthetic sequence to natural
psymberin2 was employed for the synthesis of C11-deox-
ypsymberin.
Table 2. Antitumor Activity of “Psymberate” Side Chain
Modified Psymberin Analogs Against Human Lung Cancer Cell
Line (HOP62)a
Our synthesis started with the preparation of the central
linker 22 (Scheme 3) which was quickly synthesized in 66%
Scheme 3. Total Synthesis of C11-Deoxypsymberin
a The CellTiter-Glo Luminescent Cell Viability Assay (Promega,
Technical bulletin 288) was employed in this study. IC50 data are the mean
value of three experiments with statistical significance calculated. The value
for psymberin in this assay is 0.42 nM.
side chain was truncated to a methyl group, both 14a and
its epimer 14b showed much reduced cytotoxic activity. We
then looked into the importance of the terminal olefin. When
the double bond was converted to a terminal hydroxy group,
compound 15a retained a good level of cytotoxicity (260
nM) but was 650-fold weaker compared to psymberin, while
its epimer at C(8)-C(9) (15b) was not as active (>10 000
nM). This suggests that the double bond plays an important
role in psymberin’s activity. We then used a phenyl ring to
mimic the electronic effect of the double bond with no
substituents at C4 and C5; however, compound 16a and its
epimer 16b showed dramatically decreased activity (>10 000
nM). On the other hand, when we installed substituents on
the phenyl-substituted side chain, compound 17a regained
its cytotoxicity (32 nM), and interestingly, its epimer 17b
also displayed moderate activity (615 nM). This result further
confirms that some π-electron character is favorable for the
cytotoxicity of psymberin and suggests the substitution on
the side chain (C4 and C5) plays an important role in
psymberin activity.
We next turned our attention to the psymberin core
tetrahydropyran ring modification. We noticed3b that Ircini-
astatin B (C11 substituted with dO) consistently showed 10
times stronger cell growth inhibition than Irciniastatin A
(psymberin, C11 substituted with -OH) in most of the cancer
cell lines tested. On the basis of this observation, we decided
to synthesize the C11-deoxypsymberin. By doing this, we
could evaluate the role of the oxygen atom in the activity of
psymberin and its congeners since the C11 oxygen substitu-
tion is present in each of the pederin family members, and
overall yield in three steps from the commercially available
bromide 18 (Scheme 3). A high yield alkylation reaction
between 18 and 19 gave ester 20. Treatment of 20 with
TMSCH2Li in pentane10 gave ketone 21 in a single operation
and was converted to enol ether 22 by treatment with
TMSOTf/Et3N. A substrate-controlled aldol reaction11 be-
tween 22 and 232 gave ketone 24 in good yield (75%, dr >
20:1). Chelation-controlled reduction12 of ketone 24 provided
a secondary alcohol at C13 with excellent diastereoselectivity
(dr ) 15:1)13 which was transferred into 25 (E/Z ) 5/1) via
bisacetylation at C13 and C15, debenzylation, Dess-Martin
oxidation, and Takai vinyl iodide formation.14 Enamide 27
Org. Lett., Vol. 11, No. 4, 2009
869

(E/Z ) 5/1) was synthesized from 25 (E/Z ) 5/1) in three
operations: (1) coupling with 262 with CuI to give protected
N-acyl enamine, (2) removal of C13, C15 acetate and O21
TIPS groups with NaOMe/MeOH, (3) acetylation of O21.
Enamide 27 cyclized smoothly under the PhI(OAc)2-medi-
ated cyclization reaction to give the cyclized product in a
total of 89% yield as a mixture of diastereomers. The mixture
of four pairs of diastereomers (C(8)-C(9) ) S,S; C(8)-C(9)
) R,R; C(8)-C(9) ) R,S; C(8)-C(9) ) S,R)15 was acety-
lated at C15 and debenzylated to give alcohols 28 and
epimers 28a-c which were carefully isolated through
extensive silica gel column chromatography and preparative
thin-layer chromatography (with a dr of 5:1 at C(8) and 1:1
at C(9)). The C1 terminal double bond was revealed by
converting 28 and 28a-c to the o-nitrophenyl selenides
followed by treatment with H2O2 at 50 °C.16 Upon treatment
with TBAF at 50 °C, a global deprotection was realized to
give the final products 29 and epimers 29a-c.
When we subjected compounds 29 and 29a-c to cell
proliferation studies in the HOP62 human lung cancer line,
compound 29 displayed extremely potent activity, and
gratifyingly, its epimers (29a-29c) showed improved activity
compared to its corresponding psymberin epimers. Com-
pound 29 was consistently 3-10-fold more potent than
psymberin across all cancer cell lines tested with IC50 values
in the subnanomolar ranges with epimer 29c showing
excellent activity against all cell lines tested, with IC50 values
in single digit nanomolar ranges (Table 3). These findings
potency. This finding may apply to all members of the
pederin family of natural products.
In conclusion, we have prepared a series of advanced
analogs of the antitumor natural product psymberin by
modification of the “psymberate” unsaturated side chain
using our novel PhI(OAc)2-mediated oxidative cyclization
method7 as the key step and tested them in various human
cancer cell lines. Our results suggest that substitution at C4
and C5 is important for the cytotoxicity of psymberin, but
the terminal double bond is not essential for activity. An
aryl group is a good replacement for the olefin. Diastereomers
at C(8) and C(9) constantly showed decreased activity
compared to their natural isomer. We finished the total
synthesis of C11-deoxypsymberin and discovered that this
compound is consistently more potent than the natural
product psymberin which may be true for the other members
of the psymberin/pederin family; this actually provides an
opportunity for further SAR studies in the psymberin and
pederin family. The activity of our synthetic psymberin is
consistent with that reported in the literature. This natural
product is extremely potent against all human cancer cell
lines tested but showed no selectivity between cell lines, and
its diastereomer (2) at C(8) and C(9) showed much decreased
activity. Preliminary mechanism studies suggest the mode
of action of psymberin is through cell apoptosis which is
therapeutically important. Further SAR studies of psymberin,
C11-deoxypsymberin, and the deoxy-pederin family of
natural products are in progress, and the results will be
reported in due course.
Acknowledgment. We thank Drs. Ismail Kola and John
Piwinski at SPRI for their strong support of the SPRI
postdoctoral program.
Table 3. Anticancer Activity of C11-Deoxypsymberin and Its
Epimersa
Supporting Information Available: Biological assay
protocol, apoptosis studies, experimental details, and spectral
data for all new compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
29
29a
29b
29c
cell
human
(IC50 nM) (IC50 nM) (IC50 nM) (IC50 nM) line tissue type
OL802772S
0.265 ( 0.008 n.d.
0.149 ( 0.005 n.d.
0.034 ( 0.004 n.d.
0.055 ( 0.002 177 ( 6
0.142 ( 0.007 n.d.
0.076 ( 0.004 n.d.
0.073 ( 0.006 n.d.
0.160 ( 0.015 n.d.
0.066 ( 0.004 n.d.
n.d.
n.d.
n.d.
8.7 ( 0.18 ACHN
5.9 ( 0.18 DU145
1.6 ( 0.27 H226
kidney
prostate
lung
(10) Mulzer, J.; Mantoulidis, A.; Ohler, E. J. Org. Chem. 2000, 65, 7456.
(11) Evans, D. A.; Allison, B. D.; Yang, M. G.; Masse, C. E. J. Am.
Chem. Soc. 2001, 123, 10840, and references cited therein.
(12) Evans, D. A.; Hoveyda, A. H. J. Org. Chem. 1990, 55, 5190.
(13) Relative stereohemistry at C(11), C(13), and C(15) was determined
by comparing with intermediate from the total synthesis of psymberin.
(14) Takai, K.; Nitta, K.; Utimoto, K. J. Am. Chem. Soc. 1986, 108,
7408.
46 ( 7 3.0 ( 0.12 HOP62
lung
n.d.
n.d.
n.d.
n.d.
n.d.
5.3 ( 0.15 MB231
breast
3.9 ( 0.48 MKN45 gastric
2.9 ( 0.21 PC3
prostate
colon
normal
6.1 ( 0.22 SW620
3.8 ( 0.10 NHDF
a The CellTiter-Glo Luminescent Cell Viability Assay (Promega,
Technical bulletin 288) was employed in this study. IC50 data are the mean
value of three experiments with statistical significance calculated.
(15) The C(8)- and C(9)-stereochemistry of final products were deter-
mined by 1H spectra comparison with synthetic psymberin and its epimers
which was assigned by 2D NOE NMR and proton-proton coupling between
N7, C8, and C9.
(16) (a) Bernardelli, P.; Moradei, O. M.; Friedrich, D.; Yang, J.; Gallou,
F.; Dyck, B. P.; Doskotch, R. W.; Lange, T.; Paquette, L. A. J. Am. Chem.
Soc. 2001, 123, 9021–9032. (b) Grieco, P. A.; Takigawa, T.; Schillinger,
W. J. J. Org. Chem. 1980, 45, 2247.
should facilitate further SAR studies with psymberin since
it suggests that the C11-oxygen is not essential for good
870
Org. Lett., Vol. 11, No. 4, 2009