Synthesis of eukaryotic translation elongation inhibitor lactimidomycin via Zn(II)-mediated horner-wadsworth-emmons macrocyclization

Article abstract of DOI:10.1021/ol401186f

An enantioselective synthesis of potent eukaryotic translation elongation inhibitor lactimidomycin has been accomplished in 21 linear steps. This synthesis features a Zn(II)-mediated Horner-Wadsworth-Emmons reaction that could be executed on a large scale to provide the highly strained 12-membered lactimidomycin macrolactone.

Full text of DOI:10.1021/ol401186f

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Synthesis of Eukaryotic Translation
Elongation Inhibitor Lactimidomycin via
Zn(II)-Mediated HornerꢀWadsworthꢀ
Emmons Macrocyclization
Brian J. Larsen, Zhankui Sun, and Pavel Nagorny*
Department of Chemistry, University of Michigan, Ann Arbor, Michigan 48109, United States
nagorny@umich.edu
Received April 26, 2013
ABSTRACT
An enantioselective synthesis of potent eukaryotic translation elongation inhibitor lactimidomycin has been accomplished in 21 linear steps. This
synthesis features a Zn(II)-mediated HornerꢀWadsworthꢀEmmons reaction that could be executed on a large scale to provide the highly strained
12-membered lactimidomycin macrolactone.
The glutarimide-containing natural polyketides lactimi-
domycin (1),1a migrastatin (2),1b and isomigrastatin (4)1c
(Figure 1) have attracted considerable attention due to
their promising anticancer activity. Among this group,
migrastatin (2) is by far the most well-studied member of
this family. Recently, Danishefsky and co-workers synthe-
sized migrastatin (2) and developed diverted total syn-
theses (DTS) of more simple, stable and potent analogues
of 2.2 Subsequent in vivo studies demonstrated that some
of these analogues, such as 3, could serve as selective
and nontoxic inhibitors of various types of breast cancer
metastasis.3
While the initial reports indicated that 12-membered
glutarimide-containing macrolides lactimidomycin (1) and
isomigrastatin (4) could also inhibit cancer cell migration
in vitro,4a a recent study suggests that 1 and 2 are acutely
cytotoxic and do not inhibit migration of cancer cells at
subtoxic doses.4b Nevertheless, being a very potent in-
hibitor of the translocation step in eukaryotic protein
translation initiation, 1 has been identified as a promising
(1) (a) Sugawara, K.; Nishiyama, Y.; Toda, S.; Komiyama, N.; Hatori,
M.; Moriyama, T.; Sawada, Y.; Kamei, H.; Konishi, M.; Toshikazu, O.
J. Antibiot. 1992, 45, 1433. (b) Nakae, K.; Yoshimoto, Y.; Sawa, T.;
Homma, T. Y.; Hamada, Y. M.; Takeuchi, T.; Imoto, M. J. Antibiot.
2000, 53, 1130. (c) Woo, E. J.; Starks, C. M.; Carney, C., Jr.; Arslanian, R.;
Cadapan, L.; Zavala, S.; Licari, P. J. Antibiot. 2002, 55, 141.
(2) (a) Gaul, C.; Njardarson, J. T.; Danishefsky, S. J. J. Am. Chem.
Soc. 2003, 125, 6042. (b) Njardarson, J. T.; Gaul, C.; Shan, D.; Huang,
X.-Y.; Danishefsky, S. J. J. Am. Chem. Soc. 2004, 126, 1038. (c) Gaul, C.;
Njardarson, J. T.; Shan, D.; Dorn, D. C.; Wu, K.-D.; Tong, W. P.;
Huang, X.-Y.; Moore, M. A. S.; Danishefsky, S. J. J. Am. Chem. Soc.
2004, 126, 11326. (d) Nagorny, P.; Krauss, I.; Njardarson, J. T.; Perez,
L.; Gaul, C.; Yang, G.; Ouerfelli, O.; Danishefsky, S. J. Tetrahedron
Lett. 2010, 51, 3873.
(3) (a) Shan, D.; Chen, L.; Njardarson, J. T.; Gaul, C.; Ma, X.;
Danishefsky, S. J.; Huang, X.-Y. Proc. Nat. Acad. Sci. U.S.A. 2005, 102,
3772. (b) Oskarsson, T.; Nagorny, P.; Krauss, I. J.; Perez, L.; Mandal, M.;
Yang, G.; Ouerfelli, O.; Xiao, D.; Moore, M. A. S.; Massague, J.;
Danishefsky, S. J. J. Am. Chem. Soc. 2010, 132, 3224. (c) Lecomte, N.;
Njardarson, J. T.; Nagorny, P.; Yang, G.; Downey, R.; Ouerfelli, O.;Moore,
M. A. S.; Danishefsky, S. J. Proc. Nat. Acad. Sci. U.S.A. 2011, 108, 15074.
(4) (a) Ju, J.; Rajski, S. R.; Lim, S.-K.; Seo, J.-W.; Peters, N. R.;
Hoffman, F. M.; Shen, B. J. Am. Chem. Soc. 2009, 131, 1370. (b)
Micoine, K.; Persich, P.; Llaveria, J.; Lam, M.-H.; Maderna, A.;
Loganzo, F.; Furstner, A. Chem.;Eur. J. 2013, 19, 7370.
r
10.1021/ol401186f
XXXX American Chemical Society

antitumor agent with a strong antiproliferative effect on
tumor cells in vivo.1a,5
Scheme 1. Retrosynthetic Analysis
E-selective cyclizations leading to 12-membered macrolac-
tones have little precedence.8 In addition, the presence of
the (E,Z)-diene moiety decreases the overall flexibility of
the macrocycle, which could affect the relative rates for the
intermolecular and intramolecular pathways. To evaluate
the feasibility of the approach depicted in Scheme 1, we
initiated model studies with the unfunctionalized phos-
phonate 6 (Table 1). Intramolecular cyclization of 6 under
standard conditions employing potassium carbonate
(entry 1)9 or lithium hexamethyldisilazide (entry 2)8a as
bases did not resultin the formation of 7, and only diolide 8
was isolated. Subjecting 6 to the standard Masamuneꢀ
Roush olefination protocol10 provided only minor quan-
tities of E-enoate 7 (13% yield) and mostly resulted in
diolide 8 (60% yield). Encouraged by these results, we
decided to evaluate other soft enolization conditions11
surmising that the Lewis acidity of the counterion corre-
lates with its ability to bring together the termini of 6.
Gratifyingly, the combination of zinc(II) trifluorometha-
nesulfonate, TMEDA, and triethylamine12 (entry 4) pro-
moted the E-selective formation of the macrolactone 7
(78% yield), and only a minor quantity of diolide 8 was
observed (9% yield). Additional studies will be conducted
to establish the origins of the strong templating effect
exhibited by Zn2þ. One of the possible rationals for
this effect is that the coordination to the larger in size
and more polarizable Zn2þ cation, relative to the other
cations (entries 1ꢀ3), is less dependent on the geometrical
constrains posed by the macrocyclic ring in the transition
state, leading to the desired macrolactone.
Figure 1. Glutarimide-containing natural products and their
analogues.
A syntheticapproach to1 and its analogues could help in
identifying new anticancer agents with improved stability
and therapeutic properties; however, high ring strain
energy associated with the unsaturated 12-membered
macrolactone of 1 (vide infra) significantly complicates
the preparation of this macrolide. As a consequence,
€
only Furstner group’s approach could provide access to
synthetic 1.4b,6 As part of our program directed to the
discovery of new natural product-based agents for the
treatment of human diseases, we undertook design-
ing a general synthetic approach to 1 as well as to other
12-membered glutarimide-containing natural products.
In developing a viable general strategy, the number of
synthetic steps after the formation of the macrocyclic ring
should be minimized in order to avoid the decomposi-
tion and isomerization reactions of the intermediates.7 To
accomplish this, we proposed to establish both the macro-
cyclization and installation of the enoate double bond in
a single step via intramolecular HornerꢀWadsworthꢀ
Emmons (HWE) cyclization of precursor 5 (Scheme 1).
This manuscript summarizes our studies on stereoselective
total synthesis of lactimidomycin (1).
It is noteworthy that 7 was found to be unstable and
could easily polymerize upon storage in neat state.13 This
instability could be attributed to the highly strained nature
of the lactimidomycin macrocycle. Thus, our computa-
tional studies suggest that 7 is by 14.7 kcal/mol more
While intramolecular olefination reactions could effec-
tively be used for the construction of larger macrocycles,
(9) Napolitano, C.; McArdle, P.; Murphy, P. V. J. Org. Chem. 2010,
75, 7404.
(5) (a) Schneider-Poetsch, T.; Ju, J.; Eyler, D. E.; Dang, Y.; Bhat, S.;
Merrick, W. C.; Green, R.; Shen, B.; Liu, J. O. Nat. Chem. Biol. 2010, 6,
209. (b) Lee, S.; Liu, B.; Lee, S.; Huang, S.-X.; Shen, B.; Qian, S.-B. Proc.
Nat. Acad. Sci. U.S.A. 2012, 109, 14728.
(6) (a) Micoine, K.; Furstner, A. J. Am. Chem. Soc. 2010, 132, 14064.
(b) Gallenkamp, D.; Furstner, A. J. Am. Chem. Soc. 2011, 133, 9232.
(7) (a) Ju, J.; Lim, S.-K.; Jiang, H.; Shen, B. J. Am. Chem. Soc. 2005,
127, 1622. (b) Ju, J.; Lim, S.-K.; Jiang, H.; Seo, J.-W.; Shen, B. J. Am.
Chem. Soc. 2005, 127, 11930. (c) Ju, J.; Lim, S.-K.; Jiang, H.; Seo, J.-W.;
Her, Y.; Shen, B. Org. Lett. 2006, 8, 5865. (d) Ju, J.; Seo, J.-W.; Her, Y.;
Lim, S.-K.; Shen, B. Org. Lett. 2007, 9, 5183.
(10) Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfeld, A. P.;
Masamune, S.; Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25,
2183.
(11) (a) Rathke, M. W.; Nowak, M. J. Org. Chem. 1985, 50, 2624. (b)
Sano, S.; Yokoyama, K.; Fukushima, M.; Yagi, T.; Nagao, Y. Chem.
Commun. 1997, 559. (c) Katayama, M.; Nagase, R.; Mitarai, K.; Misaki,
T.; Tanabe, Y. Synlett 2006, 1, 129. (d) Claridge, T. D. W.; Davies, S. G.;
Lee, J. A.; Nicholson, R. L.; Roberts, P. M.; Russell, A. J.; Smith, A. D.;
Toms, S. M. Org. Lett. 2008, 10, 5437.
(12) Intermolecular version: (a) Schauer, D. J.; Helquist, P. Synthesis
2006, 3654. Application to the formation of macrolactams: (b) Pirrung,
M. C.; Biswas, G.; Ibarra-Rivera, T. R. Org. Lett. 2010, 12, 2402.
(13) Similar observations have been made for lactimidomycin (1):
Schneider-Poetsch, T. Ph.D. Thesis, John Hopkins University,
2009, p 33.
(8) (a) Ando, K.; Narumiya, K.; Takada, H.; Teruya, T. Org. Lett.
2010, 12, 1460. (b) Giesbrecht, H. E.; Knight, B. J.; Tanguileg, N. R.;
Emerson, C. R.; Blakemore, P. R. Synlett 2010, 374. (c) Le Floc’h, Y.;
Yvergnaux, F.; Gree, R. Bull. Soc. Chim. Fr. 1992, 129, 62–70.
B
Org. Lett., Vol. XX, No. XX, XXXX

strained than its fully saturated 12-membered macrolac-
tone analog (cf. the Supporting Information).
The cross-coupling of the iodide 15 and boronic acid 16,
available in one step from 4-pentyn-1-ol, was evaluated
next. It is known that such vinyl iodides could potentially
undergo an intramolecular Heck cyclization, and that this
pathway could be suppressed by an addition of thallium(I)
ethoxide.18 Indeed, subjecting the mixture of 15 and 16 to
Pd(PPh3)4/TlOEt19 resulted in the rapid formation of the
desiredcross-coupling productcontainingan(E,Z)-diene.
In order to avoid stoichiometric quantities of highly toxic
thalium(I) ethoxide, further optimization of this Suzuki
reaction between 15 and 16 was pursued. Gratifyingly,
extensive evaluation of various catalysts and additives
With these encouraging model study results in hand, the
synthesis of lactimidomycin (1) was pursued next (Scheme 2).
These studies commenced with the known vinylogous
aldol reaction of 9 and acetaldehyde to provide the corre-
sponding aldol adduct (93% yield, >20:1 dr).14 Protection
of this adduct (TBSCl, DMAP, ImH) followed by reduc-
tive removal of the auxiliary (NaBH4, 84% yield, two
steps) and ParikhꢀDoering oxidation of the resultant
alcohol provided aldehyde 10. Aldehyde 10 was then
subjected to the Evans syn-selective aldol reaction15 with
chiral oxazolidinone 11 to provide the corresponding aldol
adduct (>20:1 dr), which was converted to silyl ether 12
(79% yield, three steps). With compound 12 in hand, the
synthesis of the (E,Z)-diene was addressed next.
resulted in identifying Pd(dppf)Cl2 CH2Cl2 in combina-
3
tion with sodium hydroxide as the catalyst and base of
choice.20 Thus, the resultant (E,Z)-diene-containing cross-
coupling product was produced in 84% yield as the only
product. This product was further oxidized under Dessꢀ
Martin conditions to provide macrocyclization precursor
17 (84% yield), which was then subjected to the Zn(II)-
mediated intramolecular HWE reaction conditions deter-
mined previously. Remarkably, this cyclization resulted
in the desired (E)-18 (93% yield, 423 mg of product), with
only minor quantities of the corresponding diolide ob-
served (ca. 3% yield). The formal total synthesis was
completed by TBS-group deprotection that resulted in
macrolactone 19 (90% yield), which is the intermediate
in the Furstner’s group synthesis of lactimidomycin.6a The
1H and 13C NMR spectra as well as the optical rotation of
19([R]D =ꢀ234, c= 1.16, CHCl3) matched the previously
published physical characteristics ([R]D = ꢀ233, c = 1,
CHCl3). This compound was then elaborated to lactimi-
domycin following the previously published route.6a Thus,
19 was oxidized to produce unstable ketone 20, which was
noted to decompose in the neat state upon prolonged
storage and was unstable to multiple purifications by
column chromatography. The methyl ketone functionality
of 20 was converted into the corresponding silyl enol ether
(LiHMDS, Et3N, Me3SiCl). Without purification, this
silyl enol ether was subjected to the Mukaiyama aldol
addition reaction with aldehyde 21 catalyzed by oxazabor-
olidinone 22 (100 mol %, EtCN, ꢀ78 °C).
Table 1. Macrocyclization Optimization Studies
7
8
entry
conditions
(yield, %)a (yield, %)a
1
2
3
4
K2CO3, 18-C-6,1.3 mM in toluene
LiHMDS, 1.3 mM in THF
LiCl, DBU, 1.3 mM in THF
0
0
13
78
65
36
60
9
Zn(OTf)2, TMEDA Et3N, 1.7 mM in THF
a Yields are measured over two steps (oxidation of an alcohol
precursor to 6 followed by the HWE cyclization to 7 and 8); 18-C-6 =
18-crown-6; LiHMDS = lithium hexamethyldisilazide; TMEDA =
tetramethylethylenediamine.
Approaches based on cross-metathesis or Wittig olefina-
tion strategies to convert 12 into 17 were evaluated first;
however, little success was achieved in these studies. Even-
tually, we pursued the assembly of the (E,Z)-diene moiety
through a Suzuki cross-coupling strategy. Thus, the chiral
auxiliary of 12 was removed through the formation of
a thioester (EtSLi, 83% yield), which was reduced by
DIBALH to provide the corresponding aldehyde.16 The
following StorkꢀZhao olefination provided cis-vinyl io-
dide 13 as a 14:1 mixture of Z:E isomers (68% yield, two
steps).17 Compound 13 was next converted into phospho-
nate 15 (44ꢀ85%, two steps) by selective removal of
the trimethylsilyl ether followed by the acylation of the
resultant alcohol with acyl chloride 14 (DMAP, pyridine).
The resultant aldol adduct was treated with buffered
HF Py solution in pyridine/THF to remove the trimethyl-
C
3
silyl ether and produce lactimidomycin (1). The 1H and 13
NMR data of the synthetic lactimidomycin were in good
accordance with those previously reported by the Shen7d
6b
and Furstner groups. However, the optical rotation
€
value for synthetic 1 ([R]D = ꢀ20, c = 0.25, DMSO) was
consistent with the corresponding value reported by the
Sugawara group ([R]D = ꢀ20, c = 0.5, DMSO)1a but
varied from the optical rotation values of Shen ([R]D
=
þ23, c = 0.52, DMSO)21 and Furstner ([R]D = þ6.9,
€
(14) (a) Shirokawa, S.-I.; Kamiyama, M.; Nakamura, T.; Okada, M.;
Nakazaki, A.; Hosokawa, S.; Kobayashi, S. J. Am. Chem. Soc. 2004,
126, 13604. (b) Schmauder, A.; Muller, S.; Maier, M. E. Tetrahedron
2008, 64, 6263.
(15) Evans, D. A.; Bartroli, J.; Shih, T. L. J. Am. Chem. Soc. 1981,
103, 2127.
(16) Evans, D. A.;Nagorny, P.;McRae,K.J.;Reynolds,D.J.;Sonntag,
L.-S.; Vounatsos, F.; Xu, R. Angew. Chem., Int. Ed. 2007, 46, 537.
(17) Stork, G.; Zhao, K. Tetrahedron Lett. 1989, 30, 2173.
(18) Beaver-Goodman, K. Ph.D. Thesis, Harvard University, 2003.
(19) Frank, S. A.; Chen, H.; Kunz, R. K.; Schnaderbeck, M. J.;
Roush, W. R. Org. Lett. 2000, 2, 2691.
(20) Potuzak, J. S.; Tan, D. S. Tetrahedron Lett. 2004, 45, 1797.
(21) The CD spectrum and optical rotation data for natural lactimi-
domycin were provided by Professor B. Shen and included with his
permission (e-mail correspondence, February 6, 2013).
Org. Lett., Vol. XX, No. XX, XXXX
C

Scheme 2. Total Synthesis of Lactimidomycin
c = 0.5, DMSO)6 groups.22 In contrast, the CD spectrum
of synthetic 1 was in good agreement with the correspond-
ing spectrum of the natural sample.22 The discrepancies in
the optical rotation could be attributed to the presence of
optically active impurities that arise from the decomposi-
tion of lactimidomycin. Consistent with our observations
for macrolactone 7 as well as with our previous studies, 1 is
unstable and could decompose upon storage in neat state
(vide supra).13
(2) as well as the identification of more stable and active
analogues of 1 and 2 are ongoing.
Acknowledgment. We are grateful to the University of
Michigan and Robert A. Gregg for financial support. We
also thank Professor B. Shen (Department of Chemistry,
Scripps Research Institute at Florida) for sharing the
unpublished optical rotation and CD data for natural
lactimidomycin.
In conclusion, a scalable enantioselective approach to
eukaryotic translation elongation and cancer cell migra-
tion inhibitor lactimidomycin (1) has been developed. This
synthesis relies on a Zn(II)-mediated E-selective Hornerꢀ
WadsworthꢀEmmons reaction to construct the strained
12-membered macrolactone of 1. Synthesis of isomigrastatin
Supporting Information Available. Experimental pro-
cedures, description of computational studies to esti-
mate the ring strain energy of 7, and 1H and 13C NMR
spectra for compounds 1 and 6ꢀ20. This material is
available free of charge via the Internet at http://pubs.
acs.org.
(22) Refer to the Supporting Information for the detailed compar-
ison of 1H and 13C NMR spectra of synthetic and natural 1.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX