Revision of the absolute configuration of salicylihalamide A through asymmetric total synthesis

Article abstract of DOI:10.1002/1521-3773(20001201)39:23<4308::AID-ANIE4308>3.0.CO;2-4

A highly E-selective ring-closing metathesis is the key to building the macrocyclic salicylate core of (+)-salicylihalamide A (1). The synthesis results in a reassignment of the absolute configuration of natural (-)-salicylihalamide A (2), a structurally

Full text of DOI:10.1002/1521-3773(20001201)39:23<4308::AID-ANIE4308>3.0.CO;2-4

COMMUNICATIONS
scopic experiments by the group who isolated it,[1] was
misassigned.
From the onset, we deemed it crucial to introduce the
sensitive N-9alkenyl)heptadienamide side chain at a late stage
in the synthesis 9Scheme 1). Considering the options, we felt
signals of the keto form are given): d ˆ 4.31 9m, 1H; CHOH), 3.62 9dd,
J ˆ 11.2, 5.1 Hz, 1H; H-6), 3.579dd, J ˆ 11.2, 5.0 Hz, 1H; H-6), 3.41 9s,
2H; H-2), 3.10 9brs, 1H; OH), 2.90 9dd, J ˆ 17.5, 5.0 Hz, 1H; H-4),
2.83 9dd, J ˆ 17.5, 7.3 Hz, 1H; H-4), 1.47 9s, 9H, C9CH)3); 13C NMR
3
975.5 MHz, CDCl3, 208C, only signals of the keto form are given): d ˆ
28.1 9C9CH3)3), 46.6, 48.4, 51.3 9C-2, C-4, C-6), 67.6 9C-5), 82.7
9C9CH3)3), 166.2 9C-1), 202.9 9C-3).
[15] The diastereomeric excess was determined by GC after formation of
the corresponding acetonides 92,2-dimethoxypropane, cat. camphor-
sulfonic acid).
Revision of the Absolute Configuration of
Salicylihalamide A through Asymmetric Total
Synthesis**
Yusheng Wu, Lothar Esser, and Jef K. De Brabander*
Natural products that elicit a specific and unique biological
response in mammalian cells represent valuable tools to
identify, study, and target possible new gene products. In this
context, the recent isolation of salicylihalamides A and B 91
and 2, Scheme 1) from the marine sponge Haliclona sp. is
noteworthy.[1] Pattern-recognition analysis of their unique
differential 60-cell mean-graph screening profiles 9National
Cancer Institute) suggests that the salicylihalamides belong to
a potentially new mechanistic class of antitumor com-
pounds.[1] Since their discovery in 1997, an emerging class of
novel bioactive metabolites have been isolated that structur-
ally relate to the salicylihalamides by virtue of an unprece-
dented highly unsaturated enamide attached to a macrocyclic
salicylate 9generalized structure 3, Scheme 1). These include
the mechanistically related lobatamides,[2] the potent cyto-
static apicularens,[3] and selective inhibitors of oncogene-
transformed cells 9oximidines),[4] as well as compounds that
induce low density lipoprotein 9LDL) receptor gene expres-
sion.[5] The opportunity to develop chemistry in this areaÐ
none of these compounds have been synthesized previouslyÐ
as well as to access variants for mode of action studies,
prompted us to initiate a synthetic program towards this
intriguing class of natural products.[6] Herein, we disclose the
total synthesis of 1 and demonstrate unequivocally that the
absolute configuration of natural 9À)-salicylihalamide A,
Scheme 1. Synthetic strategy for salicylihalamide A. PMBˆ para-methoxy-
benzyl, MOM ˆ methoxymethyl.
that the addition of 1-lithio-1,3-hexadiene 94) to isocyanate 5
would offer the distinct advantage of mild reaction conditions
and control of stereochemistry.[7] Isocyanate 5 was to be
derived from the corresponding E-a,b-unsaturated carboxylic
acid 9acyl azide formation/Curtius rearrangement), in turn
accessible from a C17aldehyde by Horner± Wadsworth ±
Emmons homologation. A Mitsunobu esterification/olefin
ring-closing metathesis 9RCM) tactic would ultimately un-
ravel the 12-membered benzolactone ring into its primary
components, polyol fragment 6 and benzoic acid derivative 7.
A fully optimized procedure, delivering gram quantities of
alcohol 6, is presented in Scheme 2. We opted for an
OPMB
OPMB
OPMB
CHO
15
a
b-d
O
HO
TBSO
10
96%
72%
8
9
O2S
10 steps
39% overall
N
e
95%
O
11
Me
Me
1
formulated as 1 through Mosherꢁs ester H NMR spectro-
OPMB
OPMB
OPMB
f, g
75%
h-j
80%
HO
TBSO
TBSO
O
[*] Prof. Dr. J. K. De Brabander, Dr. Y. Wu, Dr. L. Esser[‡]
Department of Biochemistry
OMO
M
OMO
M
OH
OH
NX*
University of Texas Southwestern Medical Center at Dallas
5323 Harry Hines Boulevard, Dallas, TX 75390-9038 9USA)
Fax : 9‡1)214-648-6455
6
13
12
Scheme 2. Reagents and conditions: a) 2-dIcr2B9allyl),[8] Et2O, À788C,
then NaOOH, 96%; b) TBSCl, imidazole, DMAP, DMF, 94%; c) cat.
OsO4, NMO, acetone/H2O; d) Pb9OAc)4, pyridine, PhH, 77% 9steps c, d);
e) 11, TiCl4, iPr2NEt, CH2Cl2, À788C, then 10, À788C, 95%; f) MOMCl,
NaI, iPr2NEt, CH2Cl2, 91%; g) LiEt3BH, THF, À788C !RT, 82 %;
h) TsCl, Et3N, DMAP, CH2Cl2, 91% 95% recovery of 13); i) LiEt3BH,
THF, À788C !RT, 90%; j) TBAF, THF, 98%. Icrˆ isocaranyl, TBS ˆ tert-
butyldimethylsilyl, DMAP ˆ 4-dimethylaminopyridine, DMF ˆ dimethyl-
formamide, NMO ˆ 4-methylmorpholine N-oxide, RTˆ room tempera-
ture, Ts ˆ tosyl ˆ toluene-4-sulfonyl, TBAF ˆ tetrabutylammonium fluo-
ride, X* ˆ boranesultam.
E-mail: jdebra@biochem.swmed.edu
‡
[ ] Author to whom correspondence regarding X-ray crystallographic
analysis should be addressed. E-mail: esser@chop.swmed.edu
[**] Financial support provided by the Robert A. Welch Foundation and
junior faculty awards administered through the Howard Hughes
Medical Institute and the University of Texas Southwestern Medical
Center are gratefully acknowledged. We also thank Mr. Yao Hain Ing
9Schering Plough Research Corporation) for providing high-resolu-
tion mass spectrometry data.
4308
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1433-7851/00/3923-4308 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 23

COMMUNICATIONS
enantioselective allylation of aldehyde 8 to set the absolute
stereochemistry at C15.[8] The corresponding homoallyl
alcohol 9, obtained in 96% yield, was silylated followed by
oxidative double-bond cleavage 972% yield, 3 steps). Treat-
ment of the corresponding aldehyde 10 with the in situ
prepared Z-9O)-titanium enolate[9] derived from 92R)-N-94-
pentenoyl)bornanesultam 911)[10] produced exclusively one
diastereomeric aldol product 12 in 95% yield.[11] Reduction of
the corresponding MOM ether delivered primary alcohol 13
975%, 2 steps), which was further converted into 6 by tosylate
formation, reduction, and fluoride-assisted liberation of the
C15 alcohol 980%, 3 steps).
Hydrolysis 9Ba9OH)2 ´ 8H2O) of the major E-methyl ester
18 was followed by complete silylation with excess TBSCl
9Scheme 4). Upon workup, silylester hydrolysis had occurred,
a, b
18
50-70%
CO2H
NCO
TBSO
O
TBSO
O
c, d
OTBS
OTBS
O
O
70%
20
21
22
e, f
Setting the stage for the RCM,[12] fragment 6 was joined
with carboxylic acid 7[13] through a Mitsunobu inversion
9Scheme 3).[14] Exposure of 14 to a catalytic amount of
Grubbsꢁ ruthenium carbene complex 15[15] preferentially
30-40%
Br
NH
HN
O
O
O
OH
OH O
HO
OH
O
O
OMe
MeO
O
OPMB
OMOM
OPMB
CO2H
a
24 (−)-salicylihalamide A
1 (+)-salicylihalamide A
O
HO
[α]D = −35 (c = 0.7, MeOH)[1]
[α]D = +20.8 (c = 0.12, MeOH)
95%
OMO
M
O
Br
7
6
14
PCy3
O
Ph
Cl
Cl
OH O
b
99%
Ru
OH
O
15
PCy3
m.p. = 144 C
23
PMBO
CHO
Scheme 4. Reagents and conditions: a) Ba9OH)2 ´ 8H2O, MeOH;
b) TBSCl, imidazole, DMF, 50 ± 70% 9from 18); c) 9PhO)2P9O)N3, Et3N,
PhH; d) PhH, 808C, 70% 9from 20); e) 22, tBuLi, Et2O, À788C, then add
21, À788C !08C, 55 ± 65%; f) HF ´ pyridine, pyridine/THF, 40 ± 60%.
MeO
O
O
MeO
O
O
c, d
93%
OMOM
OMOM
16
E:Z = 10:1
17
allowing the corresponding acid 20 to be converted into
isocyanate 21 as shown. We were now in a position to explore
the crucial installation of the dienamide side chain. Incorpo-
ration proceeded smoothly through the addition of hexadi-
enyllithium, prepared in situ from bromide 22[19] by metal ±
halogen exchange 9tBuLi), to a À788C solution of isocyanate
21 9pure compound according to chromatographic analy-
sis).[7b] The total synthesis was completed by deprotection of
the silylether protecting groups with HF ´ pyridine 91:1) in
THF, to afford synthetic salicylihalamide A 91).[20] This
material was found to be identical to natural salicylihalami-
de A according to NMR 9[D6]benzene and [D4]methanol),
IR, and UV spectroscopy, as well as HPLC, and thin layer
chromatography 93 different solvent systems).[21]
We were completely surprised, however, that the optical
rotation of synthetic salicylihalamide A 91) was of opposite
sign 9[a]2D3 ˆ ‡20.8 9c ˆ 0.12, MeOH)) to the optical rotation
recorded for natural salicylihalamide A 9[a]2D3 ˆ À35 9c ˆ 0.7,
MeOH)).[1] Moreover, synthetic 1 was completely ineffective
9up to 20 mM) in arresting the growth of SK-MEL-28, a
human melanoma cancer cell-line reported to be sensitive to
natural salicylihalamides with a GI50 of 100 nM 9GI50 ˆ the
concentration at which growth of 50% of the cells is
inhibited).[1] At this point we had the fortune that para-
bromobenzoate derivative 23[22] provided crystals suitable for
X-ray diffraction studies, confirming the absolute configura-
tion of our synthetic lactones.[23] Based on all the available
evidence, the absolute configuration of natural 9À)-salicyli-
CO2Me
OH
e, f
75-85%
CO2Me
OH
OH O
OH O
O
O
18
19
E:Z = 4:1
Scheme 3. Reagents and conditions: a) DEAD, PPh3, Et2O, 95%;
b) 10 mol% [9Cy3P)2Cl2Ru CHPh] 915), CH2Cl2, 99%; c) DDQ, CH2Cl2/
H2O, 96%; d) Dess ± Martin periodinane, CH2Cl2, 97%; e) trimethyl
phosphonoacetate, NaH, THF, 08C, 90%; f) BBr3, CH2Cl2, À788C, 90%
918:19 ˆ 4:1). DEAD ˆ diethylazodicarboxylate, Cy ˆ cyclohexyl, DDQ ˆ
2,3-dichloro-4,6-dicyano-1,4-benzoquinone.
ˆ
produced the desired E isomer 16 with an impressive 9and
reproducible!) selectivity of 10 ± 11:1 9E:Z).[16] Moving for-
ward, 16 was oxidatively deprotected 9DDQ) and further
oxidized to aldehyde 17 by exposure to Dess ± Martin period-
inane 993%, 2 steps).[17] At this stage, we had achieved an
extremely efficient synthesis of the lactone core of the
salicylihalamides, delivering gram quantities of 17 in 34%
overall yield from aldehyde 8 914 steps).
All that remained to complete the total synthesis was the
introduction of the side chain followed by final deprotection.
Thus, following homologation of aldehyde 17 with trimethyl
phosphonoacetate 9NaH, THF), treatment with BBr3 led to
the corresponding E- and Z-methyl esters 18 and 19 94:1),
which were separated by flash column chromatography 9silica,
EtOAc/hexanes 935/65)).[18]
Angew. Chem. Int. Ed. 2000, 39, No. 23
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1433-7851/00/3923-4309 $ 17.50+.50/0
4309

COMMUNICATIONS
[18] Note that 19 is a useful intermediate for the synthesis of ent-
salicylihalamide B 92).
[19] Bromide 22 was obtained from 1-trimethylsilyl-3-hexen-1-yne 9G.
halamide A was revised unambiguously to the one represent-
ed by structure 24 9ent-1).
In summary, we have accomplished the first synthesis of
9‡)-salicylihalamide A and revised the absolute configuration
of the natural product to 12S,13R,15S. Our approach features
a highly efficient, trans-selective ring-closing olefin metathesis
for the assembly of the 12-membered salicylate skeleton and
can be readily adapted to obtain the natural enantiomer.
Á
Arsequell, F. Camps, G. Fabrias, A. Guerrero, Tetrahedron Lett. 1990,
31, 2739 ± 2742) by treatment with N-bromosuccinimide and AgNO3
according to: T. Nishikawa, S. Shibuya, S. Hosakawa, M. Isobe, Synlett
1994, 485 ± 486, followed by reduction of the corresponding 1-bromo-
3-hexen-1-yne according to the method in: H. C. Brown, C. D. Blue,
D. J. Nelson, N. G. Bhat, J. Org. Chem. 1989, 54, 6064 ± 6067. The
91Z,3Z)-1-bromo-1,3-hexadiene 922) thus obtained was contaminated
with ꢀ20% of 91Z,3E)-1-bromo-1,3-hexadiene but was used as such.
[20] The 22E isomer 9ꢀ20%) was removed by preparative HPLC 9normal
phase, acetone/hexanes 93/7)).[19]
Received: July 11, 2000 [Z15430]
[21] A sample and the NMR spectra of natural 9À)-salicylihalamide A
were kindly provided by Dr. M. R. Boyd, National Cancer Institute
9Frederick, MD, USA).
[1] K. L. Erickson, J. A. Beutler, J. A. Cardellina II, M. R. Boyd, J. Org.
Chem. 1997, 62, 8188 ± 8192.
[22] This derivative was prepared from 16: 1) DDQ, CH2Cl2/H2O, RT;
2) para-bromobenzoic acid, PPh3, DEAD, Et2O, RT; 3) BBr3
93 equiv), CH2Cl2, À788C.
[2] a) T. C. McKee, D. L. Galinis, L. K. Pannell, J. H. Cardellina II, J.
Laakso, C. M. Ireland, L. Murray, R. J. Capon, M. R. Boyd, J. Org.
Chem. 1998, 63, 7805 ± 7810; b) lobatamide A is identical to the
structure of YM-75518; see: K.-I. Suzumura, I. Takahashi, H.
Matsumoto, K. Nagai, B. Setiawan, R. M. Rantiatmodjo, K.-I. Suzuki,
N. Nagano, Tetrahedron Lett. 1997, 38, 7573 ± 7576.
[3] a) B. Kunze, R. Jansen, F. Sasse, G. Höfle, H. Reichenbach, J. Antibiot.
1998, 51, 1075 ± 1080; b) R. Jansen, B. Kunze, H. Reichenbach, G.
Höfle, Eur. J. Org. Chem. 2000, 913 ± 919.
[23] Crystallographic data 9excluding structure factors) for the structure
reported in this paper 923) have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication no.
CCDC-147019. Copies of the data can be obtained free of charge on
application to CCDC, 12 Union Road, Cambridge CB21EZ, UK 9fax:
9‡44)1223-336-033; e-mail: deposit@ccdc.cam.ac.uk).
[4] J. W. Kim, K. Shin-ya, K. Furihata, Y. Hayakawa, H. Seto, J. Org.
Chem. 1999, 64, 153 ± 155.
[5] K. A. Dekker, R. J. Aiello, H. Hirai, T. Inagaki, T. Sakakibara, Y.
Suzuki, J. F. Thompson, Y. Yamauchi, N. Kojima, J. Antibiot. 1998, 51,
14 ± 20.
[6] We recently reported an enantioselective synthesis of side chain
truncated apicularen A; see A. Bhattacharjee, J. K. De Brabander,
Tetrahedron Lett. 2000, accepted.
[7] a) A similar strategy has been proposed recently; see B. B. Snider, F.
Song, Org. Lett. 2000, 2, 407± 408; b) for a related example, see K.
Kuramochi, H. Watanabe, T. Kitahara, Synlett 2000, 397± 399; c) for
an alternative approach using a copper9i)-catalyzed substitution of
vinyl iodides with amides, see R. Shen, J. A. Porco, Jr., Org. Lett. 2000,
2, 1333 ± 1336.
Clean and Efficient Catalytic Reduction of
Perchlorate**
Mahdi M. Abu-Omar,* Lee D. McPherson,
Â
Joachin Arias, and Virginie M. Bereau
[8] H. C. Brown, R. S. Randad, K. S. Bhat, M. Zaidlewicz, U. S. Racherla,
J. Am. Chem. Soc. 1990, 112, 2389 ± 2392.
[9] D. A. Evans, D. L. Rieger, M. T. Bilodeau, F. Urpí, J. Am. Chem. Soc.
1991, 113, 1047± 1049.
[10] W. Oppolzer, T. Osamu, J. Deerberg, Helv. Chim. Acta 1992, 75,
1965 ± 1978.
Even though perchlorate is a strong oxidizing agent
thermodynamically [Eq. 91)], its reactions are very slow
À
‡
À
ClO4 ‡ 2H ‡ 2eÀ À! ClO3 ‡ H2O
91)
[11] Since no diastereomeric aldol products were detected by 1H NMR
spectroscopic analysis of the crude reaction mixture, including C15
epimers which would have arisen from an enantiomeric C15 aldehyde,
we conclude that aldehyde 10 was essentially enantiomerically pure.
[12] For recent reviews on ring-closing metathesis, see a) R. H. Grubbs, S.
Chang, Tetrahedron 1998, 54, 4413 ± 4450; b) A. Fürstner, Top.
Organomet. Chem. 1998, 1, 37± 72; c) S. K. Armstrong, J. Chem.
Soc. Perkin Trans. 1 1998, 371 ± 388; d) M. Schuster, S. Blechert,
Angew. Chem. 1997, 109, 2124 ± 2144; Angew. Chem. Int. Ed. Engl.
1997, 36, 2037± 2056; e) A. Fürstner, Top. Catal. 1997, 4, 285 ± 299.
[13] Prepared from 2,2-dimethyl-5-9trifluoromethanesulfonyl)benzo[1,3]-
dioxin-4-one 9A. Fürstner, I. Konetzki, Tetrahedron 1996, 52, 1507 1 ±
15078) as follows: a) allylSnBu3 91.2 equiv), Pd29dba)3 92 mol%),
tri92-furyl)phosphine 98 mol%), LiCl 93 equiv), 4-methyl-2-pyrrolidi-
Eo ˆ 1.23 V
because it is nonlabile.[1±3] Consequently, perchlorate salts are
often used to adjust ionic strength in kinetics and electro-
chemical investigations. Perchlorate is also a poor complexing
9ªinnocentº) anion.[4] In 1997, ClO À was found in ground and
4
surface waters in several U.S. western states at concentrations
up to 3700 mgLÀ1.[5±7] For example, 30% of the wells sampled
Â
[*] Prof. M. M. Abu-Omar, L. D. McPherson, J. Arias, Dr. V. M. Bereau
Department of Chemistry and Biochemistry
University of California at Los Angeles
405 Hilgard Avenue
ˆ
none, RT, 48 h; b) CH2 CHCH2OMgBr 94 equiv, prepared from allyl
alcohol and EtMgBr), THF, reflux, 3 h, 90% 9steps a, b); c) MeI
93 equiv), K2CO3 91.1 equiv), acetone, RT, 40 h, 98%; d) Pd9Ph3)4
95 mol%), morpholine 910 equiv), THF, RT, 1 h, 96%. dba ˆ
trans,trans-dibenzylideneacetone.
Los Angeles, CA 90095 ± 1569 9USA)
Fax : 9‡1)310-206-9130
E-mail: mao@chem.ucla.edu
[14] O. Mitsunobu, Synthesis 1981, 1 ± 28.
[**] This work was supported by the National Science Foundation
9CAREER Award to M.M.A.-O., CHE-9874857), the University of
California Toxic Substance Research and Teaching Plan
9UCTSR&TP), and the Arnold and Mabel Beckman Foundation
9BYI Award to M.M.A.-O.).
[15] R. H. Grubbs, S. J. Miller, G. C. Fu, Acc. Chem. Res. 1995, 28, 446 ±
452.
[16] Fürstner and co-workers observed a modest selectivity for the E
isomer 9E:Z ˆ 2.3:1) of a related 12-membered resorcylic benzolac-
tone; see A. Fürstner, G. Seidel, N. Kindler, Tetrahedron 1999, 55,
8215 ± 8230.
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/angewandte/ or from the author.
[17] D. B. Dess, J. C. Martin, J. Am. Chem. Soc. 1991, 113, 7277 ± 7287.
4310
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1433-7851/00/3923-4310 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 23