Stereoselective Allylstannane Addition for a Convergent Synthesis of a Complex Molecule

Article abstract of DOI:10.1021/acs.orglett.5b03252

A convergent methodology with 13 lineal steps for the synthesis of phormidolides B and C macrocyclic core is described. Stereoselective formation of the tetrahydrofuran (THF) core was achieved using a stereocontrolled allylation reaction. The key step of the synthesis is a (Z)-1,5-anti stereoselective allylstannane addition where a new stereocenter and a trisubstituted double bond are formed simultaneously. Finally, Shiina macrolactonization conditions improved the yield of the final cyclization.

Full text of DOI:10.1021/acs.orglett.5b03252

Letter
pubs.acs.org/OrgLett
Stereoselective Allylstannane Addition for a Convergent Synthesis of
a Complex Molecule
,†,‡,∥
Alejandro Gil,†,‡ Adriana Lorente,†,‡,⊥ Fernando Albericio,†,‡,§ and Mercedes Alvarez*
́
†Institute for Research in Biomedicine, Barcelona Science Park-University of Barcelona, Baldiri Reixac 10, E-08028 Barcelona, Spain
§
‡CIBER-BBN, Networking Centre on Bioengineering, Biomaterials and Nanomedicine, Department of Organic Chemistry,
University of Barcelona, 08028 Barcelona, Spain
∥Laboratory of Organic Chemistry, Faculty of Pharmacy, University of Barcelona, 08028 Barcelona, Spain
S
* Supporting Information
ABSTRACT: A convergent methodology with 13 lineal steps
for the synthesis of phormidolides B and C macrocyclic core is
described. Stereoselective formation of the tetrahydrofuran
(THF) core was achieved using a stereocontrolled allylation
reaction. The key step of the synthesis is a (Z)-1,5-anti
stereoselective allylstannane addition where a new stereocenter
and a trisubstituted double bond are formed simultaneously.
Finally, Shiina macrolactonization conditions improved the yield
of the final cyclization.
arine natural products have become an important source
The retrosynthetical analysis of 1 is based on a more
convergent approach than the previously described by our
group.4 After only two disconnections molecular fragments 2
and 3 were obtained which could be easily synthesized in a few
synthetic steps. The formation of the C6−C7 bond was
envisioned through a stereoselective allylstannane addition to
the aldehyde 2. The end game of the synthesis will be based on
the macrolactonization under Shiina conditions and further
removal of the protecting groups (Scheme 1).
The synthesis of aldehyde 2 (Scheme 2) starts with
commercially available 2-D-deoxyribose. Oxidation at the
anomeric position with bromine followed by chemoselective
sequential protection afforded lactone 5. Then lactone 5 was
transformed into acetyl acetal 6 as a 6:4 mixture of epimers.
The addition of allyltrimethylsilane promoted by BF3·OEt2 to
M
of drugs for the treatment of different illnesses.1 Several
natural products and their simplified analogs have been
approved as drugs for the treatment of various diseases.2 In
particular, compounds with a macrolide motif in their structure
are ideal drug candidates because of their interesting biological
activities.3 In this field, phormidolides B and C were isolated
from Petrosiidae sponge in the coasts of Tanzania.4 They are
cytotoxic in three tumor cell lines in the micromolar range
(HT-29, A-549, and MDA-MB-231) with an unknown
mechanism of action.
Phormidolides are interesting synthetic targets because of
their structural complexity (Figure 1). Their structure can be
Scheme 1. Retrosynthesis of Phormidolides Macrocyclic
Core
Figure 1. Structures of phormidolides B and C.
divided into three smaller fragments to confront their total
synthesis: fatty acids, polyhydroxylated chain, and macrolide
ring. A suitable strategy to synthesize these fragments separately
is vital to complete their total synthesis and confirm all the
stereochemical information. Herein, an improved synthetic
approach to the synthesis of the macrocyclic core 1 is
described.
Received: November 11, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.5b03252
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 2. Synthesis of Aldehyde 2
acetylated compound 6, using the Tran et al. method,5 gave
only the desired 5R diastereomer 7a and its deprotected version
7b. Alcohol 7b can be protected in the crude mixture and then
purified to obtain 7a with an 81% yield and excellent
diastereoselectivity.6 One of the advantages of this new
synthesis is the diastereoselective formation of the C−C
bond in position 5 to furnish the pure enantiomer 7. This
methodology addresses a significant weakness in the previous
synthetic route: the unselective installation of the alkyl
substituent onto the tetrahydrofuran portion of the core
structure.4
In order to elongate the chain, allyl derivative 7a was
subjected to an ozonolysis to reach aldehyde 8. Chiral
phosphonate 97,8 was needed at this point to introduce the
methyl group by means of a diastereoselective 1,4 addition to
afford compound 10 in good yield. Removal of the chiral
auxiliary and oxidation with Dess-Martin periodinane (DMP)9
gave aldehyde 2 in 10 steps with easily scalable procedures.
The synthesis of compound 3 started from previously
reported aldehyde 4.4 Wittig olefination with (1-methoxy-
carbonylethyl)triphenylphosphonium bromide10 and further
methyl ester reduction gave alcohol 11 in good yield.
Formation of the corresponding xanthate and 3,3-sigmatropic
thermal rearrangement furnished compound 12 as an epimeric
mixture. Reaction with Bu3SnH under free radical conditions
produced allylstannane 3 as a Z/E mixture of isomers (7:3
ratio) (Scheme 3).11 This isomeric mixture was used without
separation in the synthetic step described below.
new stereocenter and a trisubstituted double bond were created
with complete selectivity for the desired (Z)-1,5-anti product
13.13−15
The explanation for this high stereoselectivity is consistent
with the mechanism of the reaction depicted in Scheme 4. As
reported previously,11 the stereocenter present in 3 controls the
facial selectivity of the transmetalation to give the allyltin
trichloride A where the prop-1-en-2-yl and OTIPS groups were
in a trans relationship on the six-membered oxastannic ring. It
is worthy to mention that these kinds of six-membered
oxastannic rings have been widely reported11,16,17 using 6-
hydroxystannanes and 6-alkoxystannanes but never using the
oxygen of a more oxidated function such as an ester. When
aldehyde 2 was added to the reacting mixture it approached this
chelated structure to form a new chairlike six-membered
transition structure B where the group next to tin adopts the
preferred axial position to avoid steric hindrance with the apical
chloride on the tin. This fact and the preference of the R group
of the aldehyde to adopt the equatorial position explain the
remote stereocontrol of this reaction to obtain the desired
(3R,7S,Z) diastereomer 13. To the best of our knowledge this
has been the first (Z)-1,5-anti allylstannane stereoselective
addition to create a methylated trisubstituted double bond.
Furthermore, this addition with multifunctionalized big
building blocks such as 2 and 3 shows the utility and
robustness of this methodology for the synthesis of natural
products.
In the final stage of the synthesis, compound 13 had to be
converted to the corresponding seco-acid 14 to perform
cyclization. Standard basic conditions were used to protect the
homoallylic hydroxyl with TBS. After quantitative conversion,
TMSOTf was added to deprotect the tert-butyl ester. Then,
aqueous workup was necessary and the reaction crude was
treated with pyridinium p-toluenesulfonate (PPTS) in MeOH
obtaining hydroxy-acid 14 in 72% yield for this 2 stage-3
chemical transformation procedure with only one purification
(Scheme 5). Following Shiina’s methodology,18 slow addition
of 14 to a solution containing 2-methyl-6-nitrobenzoic
anhydride (MNBA) and 4-dimethylaminopyridine (DMAP)
cleanly afforded macrocycle 15 in 67% yield without formation
of dimeric or trimeric species. In our previously described
synthesis, Yamaguchi’s lactonization conditions afforded the
protected macrocycle in 39% yield due to the formation of
polymeric species.4
With fragments 2 and 3 in hand, the most important reaction
in the synthesis was performed successfully following the
procedure described by Thomas et al.12 The allylstannane
addition of 3 to aldehyde 2 resulted not only with a high yield
but also with the desired stereoselectivity. In one reaction, a
Scheme 3. Synthesis of Allylstannane 3
Macrocycle 15 was transformed into aldehyde 16 by selective
deprotection, followed by oxidation. This aldehyde is an
essential fragment for the total synthesis of phormidolides B
and C using previously studied methodology.19 In addition,
B
DOI: 10.1021/acs.orglett.5b03252
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 4. (Z)-1,5-anti Allylstannane Addition: Mechanistic Explanation of Its Excellent Stereoselectivity
the beginning of the synthesis at a multigram scale.
Furthermore, this procedure does not produce the nondesired
diastereomer that has to be eliminated. Second, the stereo-
selective link of the two big molecular fragments by the
Thomas et al. methodology12 gave excellent synthetic results in
terms of yield and desired stereoselectivity. (Z)-1,5-anti remote
stereocontrol has been achieved through a six-membered
oxastannic ring where the coordinating oxygen atom belongs
to an ester functionality. This fact broadens the scope for this
type of remote stereocontrolled addition with 6-alkyloxy-
carbonylstannanes. The mild conditions used to perform this
reaction and the synthesis of a complex allylstannane opens the
possibility to use this methodology in the late stages of a
synthesis to link large polyfunctionalized molecular entities
with complete stereocontrol.
Scheme 5. Cyclization and Deprotecting Procedures To
Obtain 16 or the Desired Macrolide 1
Finally, the Shiina methodology for macrolactonization
instead of Yamaguchi’s conditions led to an increase of the
yield from 39% to 67% for this important transformation. All of
these changes have allowed us to isolate and characterize
aldehyde 16 and triol 1. NMR chemical shifts of macrocycle 1
are in accordance with the described phormidolides B and C.
The small differences in chemical shift are caused by the
presence of the polyhydroxylated chain in the natural products.
Moreover, the lack of this chain would explain the inactivity of
macrolide 1. Further studies to link the polyhydroxylated chain
to the macrolactone are under development.
total deprotection gave macrocycle 1 to perform spectral
characterization to compare with the natural product and
biological tests (IC50 tests).20 Key aspects to control the
deprotection included solvent, temperature, and reaction time.
An excess of a buffered solution of TBAF/AcOH in THF at 40
°C during 24 h produced selective monodeprotection of 15. At
this temperature di- and trideprotected species were not
detected. The obtained alcohol was oxidized with DMP to give
aldehyde 16 with high yield. On the other hand, 1 can be
obtained by raising the temperature in 1,4-dioxane to 90 °C
with a longer reaction time in moderate yield.
In summary, this paper describes a convergent methodology
to synthesize 15, the protected macrocyclic core of
phormidolides B and C, with an important improvement in
the total yield relative to our previous methodology. The
number of linear steps has been reduced from 17 to 13, and the
overall yield has been increased 10 times. Three factors
contributed to this increase in the overall yield. First, the
diastereoselective formation of 7a by addition of allyltrime-
thylsilane to the THF core avoided diastereomer separation at
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.5b03252.
Experimental procedures and characterization of the
described compounds (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: mercedes.alvarez@irbbarcelona.org.
Present Address
⊥University of Zurich, Winterthurerstrasse, 190, 8057, Zurich,
Switzerland.
̈
Notes
The authors declare no competing financial interest.
C
DOI: 10.1021/acs.orglett.5b03252
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
ACKNOWLEDGMENTS
■
This study was partially funded by the MINECO-FEDER
(CTQ2012-30930), the Generalitat de Catalunya (2014 SGR
137), and the Institute for Research in Biomedicine Barcelona
(IRB Barcelona). A.G. thanks to the Spanish Ministry of
Education for the FPI grant.
REFERENCES
■
(1) Blunt, J. W.; Copp, B. R.; Keyzers, R. A.; Munro, M. H. G.;
Prinsep, M. R. Nat. Prod. Rep. 2015, 32, 116−211.
(2) Molinski, T. F.; Dalisay, D. S.; Lievens, S. L.; Saludes, J. P. Nat.
Rev. Drug Discovery 2009, 8, 69−85.
(3) Lorente, A.; Makowski, K.; Albericio, F. Ann. Mar. Biol. Res. 2014,
1, 1003−1013.
́
(4) Lorente, A.; Gil, A.; Fernandez, R.; Cuevas, C.; Albericio, F.;
́
Alvarez, M. Chem. - Eur. J. 2015, 21, 150−156.
(5) Tran, V. T.; Woerpel, K. A. J. Org. Chem. 2013, 78, 6609−6621.
(6) The stereochemistry of the new stereocenter in C5 was
confirmed by NOE 1D experiments (see Supporting Information).
(7) Yokokawa, F.; Asano, T. Tetrahedron 2004, 60, 6859−6880.
(8) Ishizaki, M.; Hara, Y.; Kojima, S.; Hoshino, O.; Funagawara-
machi, I. Heterocycles 1999, 50, 779−790.
(9) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155−4156.
́
(10) Boisse, T.; Rigo, B.; Millet, R.; Henichart, J.-P. Tetrahedron
2007, 63, 10511−10520.
(11) Carey, J. S.; MacCormick, S.; Stanway, S. J.; Teerawutgulrag, A.;
Thomas, E. J. Org. Biomol. Chem. 2011, 9, 3896−3919.
(12) McNeill, A. H.; Thomas, E. J. Synthesis 1994, 1994, 322.
(13) The stereochemistry of the new stereocenter was determined
using Mosher’s derivatization (see Supporting Information).
(14) The configuration of the double bond was confirmed based on
the 13C chemical shift of the olefinic methyl group. Previous published
work confirms that if this chemical shift is above 20 ppm the
configuration is Z (see ref 15).
́
(15) Lorente, A.; Albericio, F.; Alvarez, M. J. Org. Chem. 2014, 79,
10648−10654.
(16) MacCormick, S.; Thomas, E. J. J. Organomet. Chem. 2006, 691,
1437−1440.
(17) Carey, J. S.; Thomas, E. J. J. Chem. Soc., Chem. Commun. 1994,
283−284.
(18) Shiina, I.; Fukui, H.; Sasaki, A. Nat. Protoc. 2007, 2, 2312−2317.
(19) Lamariano-Merketegi, J.; Lorente, A.; Gil, A.; Albericio, F.;
́
Alvarez, M. Eur. J. Org. Chem. 2015, 2015, 235−241.
(20) Compound 1 was inactive. It was tested in HT-29, A-549, and
MDA-MB-231 cell lines under the same conditions as natural
compounds (see ref 4).
D
DOI: 10.1021/acs.orglett.5b03252
Org. Lett. XXXX, XXX, XXX−XXX