Our exhaustive effort toward the total synthesis of cytotoxic marine nonanolide cytospolide E has been detailed. To achieve this synthesis, we have explored both the ring-closing metathesis and lactonization-based macrocyclization strategies using a variety of precursors. Unfortunately, none of them provided the desired product. The ring-closing metathesis approach provided mainly the macrocycle with Z-olefin, whereas the macrolactonization strategy culminated in 8-epi-9-epi-cytospolide E following the regioselective formation of a 10-membered macrocycle over a 9-membered macrocycle.
Our exhaustive effort toward the total synthesis of cytotoxic marine nonanolide cytospolide E has been detailed. To achieve this synthesis, we have explored both the ring-closing metathesis and lactonization-based macrocyclization strategies using a variety of precursors. Unfortunately, none of them provided the desired product. The ring-closing metathesis approach provided mainly the macrocycle with Z-olefin, whereas the macrolactonization strategy culminated in 8-epi-9-epi-cytospolide E following the regioselective formation of a 10-membered macrocycle over a 9-membered macrocycle.
Nonanolides are a class
of molecules that belong to the family
of secondary metabolites. The broad spectrum of biological activities
(antimalarial, antituberculosis, anticancer, pesticide activity, and
phytotoxic activity) and diverse architectural features of these medium-ring-sized
molecules have created immense interest in the scientific community.[1,2] Several synthetic groups worldwide have contributed to the synthesis
of nonanolides and related molecules.[2] Structurally
nonanolides are 10-membered macrolides. Mostly, the core lactone framework
is appended with a C-9 alkyl group, oxygenated functionalities, and
a skeletal trans-olefin moiety. Zhang and co-workers, in 2011, discovered
first five new cytotoxic nonanolides, cytospolides A–E (1–5) (Figure ), from the crude acetone extract of fungi, Cytospora sp. isolated from Ilex canariensis.[3] These are the first examples of nonanolides
that possess an additional C-2 methyl group compared with the other
existing nonanolides. The structures and absolute configurations of
these natural products were elucidated using detailed NMR, mass spectrometry,
and X-ray crystallographic analysis and by combining time-dependent
density functional theory calculations of
circular dichroism data. Architecturally, cytospolides A–C
(1–3) are the acetyl derivatives
of cytospolide D (4), whereas cytospolide E (5) is the C-2 epimer of cytospolide D. The stereochemical difference
of the C-2 methyl group in cytospolides D and E is manifested in their
cytotoxic activities. Cytospolide E (5) having a C-2
methyl group of (S)-configuration is more toxic (IC50 ∼
10 μM) to human lungs carcinoma cell line A549 compared with
cytospolide D (4), in which the C-2 methyl center is
in the opposite configuration.[3] In continuation
of our program[4] toward the asymmetric synthesis
of bioactive macrocyclic natural products and their analogues, we
embarked on the total synthesis of cytospolide E (5).
A number of synthetic reports on cytospolides D and E exist in the
literature. Yadav,[5] Nanda,[6] and Ramana[7] et al. have shown,
separately, the synthesis of Z-isomer of cytospolide E using the late-stage
ring-closing metathesis (RCM)-based macrocyclization strategy. The
Z-isomer of cytospolide D was also synthesized by Kamal[8] and Nanda[6] et al.
using the RCM approach. However, the total synthesis of naturally
occurring cytospolide D was successfully achieved independently by
Kamal[8] and Nanda[9] et al. using the classical macrolactonization approach. Later, Stark
et al.[10] developed an elegant synthetic
route for the total synthesis of cytospolide D adopting the macrolactonization
strategy. The utility of macrolactonization for the synthesis of cytospolide
E was also tested by Ramana et al.,[7] but
a dimeric 20-membered macrocycle was obtained rather than the actual
natural product. The total synthesis of cytospolide E has not been
successful till date. Thus, exploration of different possible synthetic
routes toward the total synthesis of cytospolide E is essential. In
this article, we report our efforts toward the synthesis of cytospolide
E, in detail, which eventually produced 8-epi-9-epi-cytospolide E along with a previously reported macrocyclic
skeleton with Z-olefin.
Figure 1
Chemical structures of cytospolides A–E
(1–5).
Chemical structures of cytospolides A–E
(1–5).
Results and Discussion
Ring-Closing Metathesis Approach
Scrutiny of literature
revealed that apart from reaction conditions and the nature of catalysts
the skeletal substituent/functionality of the reactant plays a pivotal
role in the stereochemical outcome of the RCM reaction.[11] This prompted us to rely first on the RCM[4e,12]-based macrocyclization strategy even though a number of reports[5−8] toward the synthesis of cytospolides D and E demonstrated its problem
toward E-selective metathesis. Careful analysis of these reports revealed
that in most of the cases both the allylic and bishomoallylic hydroxy
groups of the RCM precursor were protected either as smaller methoxymethyl
(MOM) ethers, as benzyl ethers, or as a mixture of a similar class
of protecting groups like benzyl and p-methoxy benzylether. We anticipated that the changes in the skeletal framework might
help in achieving a favorable conformation toward the E-selective
RCM. This possibility could be explored either by varying the protecting
group on the hydroxy centers or by removing some of the skeletal functionality,
which could be installed in the later stage of synthesis or by modifying
the structure of olefins. Moreover, during the total synthesis of
cytospolide P, we have observed that the presence of bulky allylic tert-butyldimethylsilyl (TBS) ether facilitated the formation
of the E-selective RCM product.[4e] Consolidating
these facts and our own experiences, we planned the retrosynthesis
of cytospolide E (5) as depicted in Scheme , where the role of protecting
groups ought to be investigated with the goal of optimizing the best
possible conditions allowing the total synthesis. Therefore, cytospolide
E was planned to be synthesized from compound 7 via cyclic
intermediate 6 using RCM as one of the key steps. It
was decided that the allylic hydroxy functionality of compound 7 would be protected by a bulky TBS ether as it is adjacent
to the reaction site. Compound 7 could further be constructed
from alcohol 8 and acid 9 using intermolecular
esterification as one of the pivotal steps.
Scheme 1
Initial Retrosynthetic
Analysis of Cytospolide E (5)
Our initial efforts toward the synthesis of cytospolide
E are delineated
in Scheme . We started
our synthesis from known epoxy alcohol 10,[13] which was protected as benzyl ether using BnBr/NaH
to result in compound 11. The epoxy functionality of
compound 11 was then opened using allyl magnesium bromide
in the presence of CuCN to get the corresponding alkenol, which was
further reacted with methoxymethyl chloride (MOMCl)/N,N-diisopropylethylamine (DIPEA) followed by Li/naphthalene
to access another alkenol 8 in good overall yield (61%
in three steps). The known acid 9(14) was then esterified with alkenol 8 using the
Steglich protocol[15] to get ester 7 in 79% yield. The stage was set to carry out the crucial
RCM reaction. G-II and HG-II catalysts were screened under different
reaction conditions. It was observed that in both the cases a linear
homodimer (13) was formed as a major product along with
an inseparable trace amount of 10-membered macrolide (12). The identity of macrolide (12) was confirmed only
by mass spectroscopy, which restricted us to characterize the geometry
of olefin embedded in the cyclic structure. The formation of linear
homodimer 13 as a major product could be reasoned to
be due to the presence of bulky TBS ether. This might have exerted
a steric constraint around the allylic olefin to make it reluctant
toward olefin metathesis. Next, we were keen to see whether the removal
of allylic TBS ether would have any effect on the outcome of the RCM
reaction. Therefore, compound 14 was prepared from compound 7 using tetrabutylammonium fluoride (TBAF). It was then subjected
to the RCM reaction separately using G-II and HG-II catalysts.[12] Unfortunately, in both the cases, compound 15 was formed with Z-selectivity in 46% yield along with 14%
of cyclic dimer 16. This compelled us to look for an
alternative strategy.
Scheme 2
Initial Efforts toward the Total Synthesis
of Cytospolide E
Reagents and conditions: (a)
BnBr, NaH, tetrabutylammonium iodide (TBAI), 0 °C to room temperature
(rt), 4 h, 86%; (b) (i) allyl magnesium chloride, CuCN, tetrahydrofuran
(THF), 0 °C to rt, 2 h, 83%; (ii) MOMCl, DIPEA, CH2Cl2, 0 °C to rt, 3 h, 88%; (iii) Li, naphthalene,
THF, −40 °C, 30 min, 82%; (c) 9, N,N′-diisopropyl carbodiimide (DIC),
4-dimethylaminopyridine (DMAP), CH2Cl2, rt,
4 h, 79%; (d) G-II or HG-II, CH2Cl2, reflux,
13 h, 37% with respect to (wrt) compound 13; (e) TBAF,
THF, 0 °C to rt, 3 h, 91%; (f) G-II, CH2Cl2, reflux, 2.5 h, 46% wrt compound 15, 14% wrt compound 16.
Initial Efforts toward the Total Synthesis
of Cytospolide E
Reagents and conditions: (a)
BnBr, NaH, tetrabutylammonium iodide (TBAI), 0 °C to room temperature
(rt), 4 h, 86%; (b) (i) allyl magnesium chloride, CuCN, tetrahydrofuran
(THF), 0 °C to rt, 2 h, 83%; (ii) MOMCl, DIPEA, CH2Cl2, 0 °C to rt, 3 h, 88%; (iii) Li, naphthalene,
THF, −40 °C, 30 min, 82%; (c) 9, N,N′-diisopropyl carbodiimide (DIC),
4-dimethylaminopyridine (DMAP), CH2Cl2, rt,
4 h, 79%; (d) G-II or HG-II, CH2Cl2, reflux,
13 h, 37% with respect to (wrt) compound 13; (e) TBAF,
THF, 0 °C to rt, 3 h, 91%; (f) G-II, CH2Cl2, reflux, 2.5 h, 46% wrt compound 15, 14% wrt compound 16.Then, we planned for possible
structural changes in the RCM precursor
by converting one of the hydroxy groups into keto functionality, from
which the required center could be reinstalled at an advanced stage
of synthesis. During this exercise, we targeted bishomoallylicalcohol
rather than allylic alcohol because the conversion of allylic alcohol
to keto functionality might have sensitized the C-2 methyl group toward
racemization as it would be flanked between two carbonyl functionalities.
Thus, we planned for compound 17, which could be synthesized
from the known acid 19 and alcohol 20 through
RCM precursor 18 (Scheme ). Our synthetic endeavor started from the known compound 22(5) prepared from compound 10(13) via intermediate 21 following the literature protocol, which was then oxidized to the
corresponding keto by Dess−Martin periodinane (DMP) and subsequently
treated with TBAF to get compound 20. This was then esterified
with the known acid 19(5,6) using the Yamaguchi protocol[4e,16] to get ester 18, which was
subjected to RCM separately using G-II and HG-II. Unfortunately, in
both the cases, macrolide 23 with Z-olefin was formed
along with cyclic dimer 24 (Scheme ). This forced us to look for other possibility.
Scheme 3
Efforts toward the Synthesis of Compound 17
Reagents and conditions: (a) tert-butyldimethylsilyl chloride, imidazole, CH2Cl2, 0 °C to rt, 2.5 h, 98%; (b) allyl magnesium
chloride, CuCN, THF, 0 °C to rt, 3 h, 85%; (c) (i) DMP, NaHCO3, CH2Cl2, 0 °C to rt, 3 h, 95%,
(ii) TBAF, THF, 0 °C to rt, 5 h, 92%; (d) 19, 2,4,6-trichlorobenzoyl
chloride, Et3N, DMAP, CH2Cl2, rt,
3.5 h, 78%; (e) G-II, CH2Cl2, reflux, 3.5 h,
51% wrt compound 23, 13% wrt compounds 24.
Efforts toward the Synthesis of Compound 17
Reagents and conditions: (a) tert-butyldimethylsilyl chloride, imidazole, CH2Cl2, 0 °C to rt, 2.5 h, 98%; (b) allyl magnesium
chloride, CuCN, THF, 0 °C to rt, 3 h, 85%; (c) (i) DMP, NaHCO3, CH2Cl2, 0 °C to rt, 3 h, 95%,
(ii) TBAF, THF, 0 °C to rt, 5 h, 92%; (d) 19, 2,4,6-trichlorobenzoyl
chloride, Et3N, DMAP, CH2Cl2, rt,
3.5 h, 78%; (e) G-II, CH2Cl2, reflux, 3.5 h,
51% wrt compound 23, 13% wrt compounds 24.A number of reports[2d−2i] exist in the literature for the synthesis of a wide range of nonanolides
having a skeletal trans-olefin moiety using the RCM strategy. Noticeably,
in most of those cases, the C-2 position is either unsubstituted or
substituted with the oxygenated center. This prompted us to envisage
whether a macrocycle without a C-2 methyl group could be prepared
from which the construction of a complete cytospolide E framework
would be achieved by C-methylation in the advanced stage of synthesis.
Thus, compound 25 was planned as the next target that
could be synthesized from acid 27 and alcohol 8 via intermediate 26 (Scheme ). During the choice of protection groups
of compound 27, we conceived the same logic as that followed
for compound 7 even though the allylic TBS ether made
it reluctant toward RCM. This approach might help evaluating the role
of allylic TBS in making compound 7 inefficient toward
RCM. We started our synthesis from known compound 28(17) prepared from d-aspartic acid following
a literature protocol.[18] Compound 28 was subjected to (2,2,6,6-tetramethylpiperidin-1-yl)oxyl
(TEMPO)/bis(acetoxy)iodobenzene (BAIB)-mediated oxidation[4f] to get acid 27 in 84% yield. Acid 27 was then esterified with alcohol 8 to obtain
ester 26 following the Steglich conditions.[15] Both the G-II and HG-II catalysts were screened
separately for RCM. Unfortunately, in both the cases, compound 29 was formed in 39% yield along with some unidentified side
products. This result demonstrated that apart from the allylic TBS
group the remote C-2 methyl group also played a major role to make
compound 7 (Scheme ) inefficient toward RCM although the reason is not
clearly understood.
Reagents and conditions: (a)
TEMPO, BAIB, CH2Cl2/H2O (1:1), rt,
3 h, 84%; (b) 8, DIC, DMAP, CH2Cl2, rt, 4 h, 81%; (c) G-II, CH2Cl2, reflux, 8
h, 39%.It was observed that changes of protection
groups or functionalities
did not result in any E-selective RCM in our cases. It is evident
in the literature that structures of olefins also play an important
role in the RCM reaction as they participated directly in the formation
of the metallocycle intermediate.[19a] Thus,
we planned to modify the structure of one of the olefins. For this
exercise, we chose initially the C4–C5 olefin. We planned to protect the hydroxy groups with small protecting
groups as it was observed that bulky allylic groups made the RCM reaction
more sluggish without offering the desire selectivity. Therefore,
we planned to synthesize compound 30 from compounds 32 and 8 through the key intermediate 31 utilizing the RCM approach (Scheme ). We started our synthesis from the known intermediate 33(20) prepared from cinammaldehyde
following a literature protocol, which was treated with MOMCl/DIPEA
followed by LiOH·H2O/30% H2O2 to get acid 32. The previously synthesized alcohol 8 was then esterified with acid 32 following
the Steglich protocol[15] to get compound 31 in good yield (83%). Both G-II and HG-II catalysts were
tested. Unfortunately, the phenyl-substituted olefin became inert
toward metathesis in both the cases. The less hindered olefin participated
in cross metathesis to culminate a liner homo dimer 34 almost exclusively, which discouraged us to attempt RCM further
with phenyl-substituted C5′–C6′ olefin. We were thus prompted to consider a different substituted
olefin.
Scheme 5
Efforts toward the Synthesis of Compound 30 Using
a
Phenyl-Substituted Olefin
Reagents and conditions:
(a)
(i) MOMCl, DIPEA, CH2Cl2, 0 °C to rt, 3
h, 89%; (ii) LiOH·H2O, 30% H2O2, THF/H2O (3:1), 0 °C to rt, 8 h, 81%; (b) 8, DIC, DMAP, CH2Cl2, rt, 4 h, 83%;
(c) G-II, CH2Cl2, reflux, 12 h, 38%.
Efforts toward the Synthesis of Compound 30 Using
a
Phenyl-Substituted Olefin
Reagents and conditions:
(a)
(i) MOMCl, DIPEA, CH2Cl2, 0 °C to rt, 3
h, 89%; (ii) LiOH·H2O, 30% H2O2, THF/H2O (3:1), 0 °C to rt, 8 h, 81%; (b) 8, DIC, DMAP, CH2Cl2, rt, 4 h, 83%;
(c) G-II, CH2Cl2, reflux, 12 h, 38%.The use of pinacol boron-substituted olefin in E-selective
cross-olefin
metathesis is well-documented.[19] We were
keen to see whether this approach could be utilized in RCM to achieve
compound 30 (Scheme ). Thus, ester 35 was conceived as the
next RCM precursor, which could be constructed further from acid 19 and alcohol 36. The less hindered C5′–C6′ olefin was selected to be
substituted with pinacol boron as the other olefin with the allylic
substituent itself was sluggish toward RCM in our cases. The previously
prepared olefin 8 was subjected to cross metathesis[21] with vinyl-B(pin) (37) using G-II
to get alcohol 36 in very good yield and selectivity.
Next, alcohol 36 was esterified with known acid 19 using Steglich conditions.[15] Both G-II and HG-II were screened for RCM under different conditions.
The G-II catalyst participated faster in metathesis compared with
the HG-II catalyst like all of the earlier cases. However, it was
quite unfortunate that compound 38 was formed as an exclusive
product. No macrocycle with E-olefin was detected that prompted us to move for another
macrocyclization approach.
Scheme 6
Efforts toward the Synthesis of Compound 30 Using a
B(pin)-Substituted Olefin
At this
point, we decided
to explore the classical macrolactonization strategy for the target
molecule. Ramana et al.[7] also have adopted
this approach during their synthetic study of cytospolide E, where
the C-3 and C-8 hydroxy groups of the seco acid were in a protected
form. This eventually resulted in a dimeric macrocyclic lactone rather
than the required 10-membered macrocycle. We wondered whether multiple
protected alcohol functionalities on the seco acid would have created
an unfavorable situation toward the formation of a 10-membered macrocycle.
Thus, we felt the need to explore this approach using a seco acid
with fewer hydroxy-protected functionalities. This could be possible
if the regioselective formation of one macrocycle could be tuned over
other possible macrocycle by taking the advantage of the skeletal
olefin moiety. Niwa et al.[21] have also
adopted a similar concept during the total synthesis of unnatural
(+)-neodidemnilactone having skeletal Z-olefin. They successfully
constructed the 10-membered macrocycle over a possible 9-membered
macrocycle in moderate yield. Therefore, we sketched the retrosynthesis
of cytospoilde E (5) as depicted in Scheme , which could be synthesized
from seco acid 39 where only the C-3 hydroxy would be
suitably protected. Seco acid 39 could be constructed
from compound 41 via intermediate 40 using
modified Crimmins aldol as one of the key steps.
Scheme 7
Initial Retrosynthetic
Analysis of Cytospolide E (5)
Using the Macrolactonization Strategy
To test the hypotheses of regioselective formation of
the 10-membered
ring over the 9-membered macrocycle in the presence of a trans-olefin
moiety embedded in the reacting seco acid, we followed Scheme . We started our synthesis
from cis-4-decen-1-ol (42), which was
converted to a mixture of enantiomers 43 and 43a by sequential treatment with OsO4/N-methylmorpholine-N-oxide (NMO) followed by 2,2-dimethoxypropane (2,2-DMP)/camphorsulfonic
acid (CSA). Next, the mixture of compounds (43 and 43a) was subjected to Swern oxidation followed by Wittig olefination
in the presence of ethyl(triphenylphosphoranylidene)acetate to yield
the corresponding α,β-unsaturated ester. The ester moiety
was reduced with diisobutylaluminium hydride (DIBAL-H) to get the
racemic mixture of allylic alcohols 41 and 41a. This was then oxidized to the corresponding aldehydes using Swern
conditions and subsequently was subjected to TiCl4-mediated
modified Crimmins aldol reaction[4a,22] in the presence
of oxazolidinone 44(22) to get
an inseparable mixture of diastereomers. Interestingly, the 1H and 13C NMR spectra of these two diastereomers were
so identical that it was difficult to sense the presence of a mixture
of compounds at the first glance. The aldol adducts were protected
as TBS ether with tert-butyldimethylsilyl trifluoromethanesulfonate
(TBSOTf)/2,6-lutidine to get the mixture of compounds 40 and 40a, which was then saponified using LiOH·H2O to obtain the corresponding acid as an inseparable mixture.
Next, we have tried different reagents (CSA, AcOH) in different conditions
for selective deprotection of acetonide in the presence of TBS ether,
but none of them functioned well. It was our great relief that Zn(NO3)2·6H2O was found to be effective.[23] An inseparable mixture of seco acids 39 and 39a was obtained, which was then subjected to crucial
macrolactonization following the Yamaguchi protocol.[4d,16d] A complex mixture was formed, which was subsequently desilylated
with TBAF to get lactone 45 as the only isolable product
with 22% yield. The formation of lactone 45 was confirmed
by high-resolution mass spectrometry (HRMS). However, the 1H and 13C NMR data of this lactone did not match with
the reported data of isolated cytospolide E. Thus, we looked first
for the possibility of the formation of a nine-membered lactone. Detailed
two-dimensional (2D)-NMR studies were carried out. It could be argued
that if the C-8 hydroxy group took part in lactone ring formation
then the corresponding C-8 proton must have a correlation with C-7
proton(s). However, the 1H–1H correlated
spectroscopy experiment showed no such correlation but rather a correlation
with C-10 protons. This observation ruled out the possibility of a
nine-membered ring formation. The chance of epimerization of the C-2
center was also not taken into account because its NMR data was not
in accordance with that for cytospolide D,[3] which is a C-2 epimeric product of cytospolide E. This confirmed
that lactone 45 is the 8-epi-9-epi isomer of cytospolide E. It might be the case that the
required seco acid did not cyclize, which would have decomposed during
the reaction. This result prompted us to synthesize the required seco
acid having stereochemically pure C-8 and C-9 hydroxy centers to understand
its fate during macrolactonization.
Scheme 8
Test of Viability
of Regioselective Formation of the 10-Membered
Ring over the 9-Membered Macrocycle
Reagents
and conditions: (a)
(i) OsO4, NMO, BuOH/THF (1:1),
0 °C to rt, 7 h, (ii) 2,2-DMP, CSA, CH2Cl2, 0 °C to rt, 3 h, 84% over two steps; (b) (i) (COCl)2, dimethyl sulfoxide (DMSO), Et3N, CH2Cl2, −78 to 0 °C, 1.5 h, (ii) PPh3=CHCO2Et, toluene, 80 °C, 6 h, (iii) DIBAL-H, CH2Cl2, −78 °C, 30 min, 71% over three steps;
(c) (i) (COCl)2, DMSO, Et3N, CH2Cl2, −78 to 0 °C, 2 h, (ii) 44, TiCl4, DIPEA, N-methylpyrrolidone (NMP), CH2Cl2, 0 °C, 2 h, (iii) 2,6-lutidine, TBSOTf,
CH2Cl2, 0 °C to rt, 30 min, 72% over three
steps; (d) (i) LiOH·H2O, 30% H2O2, THF/H2O (3:1), 0 °C to rt, 10 h, (ii) Zn(NO3)2·6H2O, CH3CN, 6 h,
49% over two steps; (e) (i) 2,4,6-trichlorobenzoyl chloride, Et3N, THF, rt, 6 h, then DMAP, toluene, 80 °C, 10 h, then
12 h at rt, (ii) TBAF, THF, 0 °C to rt, 3 h, 22% over two steps
based on one isomer.
Test of Viability
of Regioselective Formation of the 10-Membered
Ring over the 9-Membered Macrocycle
Reagents
and conditions: (a)
(i) OsO4, NMO, BuOH/THF (1:1),
0 °C to rt, 7 h, (ii) 2,2-DMP, CSA, CH2Cl2, 0 °C to rt, 3 h, 84% over two steps; (b) (i) (COCl)2, dimethyl sulfoxide (DMSO), Et3N, CH2Cl2, −78 to 0 °C, 1.5 h, (ii) PPh3=CHCO2Et, toluene, 80 °C, 6 h, (iii) DIBAL-H, CH2Cl2, −78 °C, 30 min, 71% over three steps;
(c) (i) (COCl)2, DMSO, Et3N, CH2Cl2, −78 to 0 °C, 2 h, (ii) 44, TiCl4, DIPEA, N-methylpyrrolidone (NMP), CH2Cl2, 0 °C, 2 h, (iii) 2,6-lutidine, TBSOTf,
CH2Cl2, 0 °C to rt, 30 min, 72% over three
steps; (d) (i) LiOH·H2O, 30% H2O2, THF/H2O (3:1), 0 °C to rt, 10 h, (ii) Zn(NO3)2·6H2O, CH3CN, 6 h,
49% over two steps; (e) (i) 2,4,6-trichlorobenzoyl chloride, Et3N, THF, rt, 6 h, then DMAP, toluene, 80 °C, 10 h, then
12 h at rt, (ii) TBAF, THF, 0 °C to rt, 3 h, 22% over two steps
based on one isomer.Attempt toward the synthesis
of cytospolide E with a stereogenically
pure substrate is shown in Scheme . Compound 46 was prepared from l-arabinose following the literature procedure.[13] This was then oxidized using Swern conditions followed
by the Wittig olefination to get the corresponding α,β-unsaturated
ester, which was treated further with LiCl/NaBH4. An inseparable
mixture of alcohols with and without the olefin was obtained, which
was finally hydrogenated to ensure the completely saturated product 43 in 70% overall yield. Alcohol 43 was then
converted to seco acid 39 through intermediates 41 and 47 following the similar chemistry as
mentioned above. Next, seco acid 39 was subjected to
macrolactonization. A number of macrolactonization protocols like
Shiina,[4d,16d,24] Yamaguchi,[4d,16d] and Steglich[15] have been attempted under
different reaction conditions. Unfortunately, none of them were effective.
In every case, either the seco acid decomposed or it produced a complex
mixture, which was quite difficult to isolate or characterize.
Scheme 9
Attempts toward the Synthesis of Cytospolide E Using the C-3 Hydroxy
Group in a Protected Form
Reagents and conditions:
(a)
(i) (COCl)2, DMSO, Et3N, CH2Cl2, −78 to 0 °C, 2 h and then PH3P=CHCO2C2H5, rt, 10 h, (ii) LiCl, NaBH4, THF/EtOH (2:1), 0 °C to rt, 7 h, (iii) 10% Pd/C, H2, 10 h, 70% yield over three steps; (b) (i) (COCl)2, DMSO, Et3N, CH2Cl2, −78
to 0 °C, 2 h, (ii) PPh3=CHCO2Et,
toluene, 80 °C, reflux, 6 h, 79%, (iii) DIBAL-H, CH2Cl2, −78 °C, 30 min, 69% over three steps;
(c) (i) (COCl)2, DMSO, Et3N, CH2Cl2, −78 to 0 °C, 2 h, (ii) 44, TiCl4, DIPEA, NMP, CH2Cl2, 0 °C, 2 h,
80% over two steps; (d) (i) 2,6-lutidine, TBSOTf, CH2Cl2, 0 °C to rt, 30 min, (ii) LiOH·H2O,
30% H2O2, THF/H2O (3:1), 0 °C
to rt, 10 h, (iii) Zn(NO3)2·6H2O, CH3CN, 6 h, 53% over three steps; (e) (i) LiCl, NaBH4, THF/EtOH (2:1), 0 °C to rt, 8 h, (ii) p-methoxybenzyl (PMB)-acetal, CSA, CH2Cl2, 0
°C to rt, 1 h, (iii) DIBAL-H, CH2Cl2, 0
°C, 30 min, (iv) TEMPO, BAIB, CH2Cl2/H2O (1:1), 3 h, (v) Zn(NO3)2·6H2O, CH3CN, 80 °C, 2 h, 34% over five steps;
(f) 2-methyl-6-nitrobenzoic anhydride, DMAP, 4 Å MS, CH2Cl2, rt, 12 h.
Attempts toward the Synthesis of Cytospolide E Using the C-3 Hydroxy
Group in a Protected Form
Reagents and conditions:
(a)
(i) (COCl)2, DMSO, Et3N, CH2Cl2, −78 to 0 °C, 2 h and then PH3P=CHCO2C2H5, rt, 10 h, (ii) LiCl, NaBH4, THF/EtOH (2:1), 0 °C to rt, 7 h, (iii) 10% Pd/C, H2, 10 h, 70% yield over three steps; (b) (i) (COCl)2, DMSO, Et3N, CH2Cl2, −78
to 0 °C, 2 h, (ii) PPh3=CHCO2Et,
toluene, 80 °C, reflux, 6 h, 79%, (iii) DIBAL-H, CH2Cl2, −78 °C, 30 min, 69% over three steps;
(c) (i) (COCl)2, DMSO, Et3N, CH2Cl2, −78 to 0 °C, 2 h, (ii) 44, TiCl4, DIPEA, NMP, CH2Cl2, 0 °C, 2 h,
80% over two steps; (d) (i) 2,6-lutidine, TBSOTf, CH2Cl2, 0 °C to rt, 30 min, (ii) LiOH·H2O,
30% H2O2, THF/H2O (3:1), 0 °C
to rt, 10 h, (iii) Zn(NO3)2·6H2O, CH3CN, 6 h, 53% over three steps; (e) (i) LiCl, NaBH4, THF/EtOH (2:1), 0 °C to rt, 8 h, (ii) p-methoxybenzyl (PMB)-acetal, CSA, CH2Cl2, 0
°C to rt, 1 h, (iii) DIBAL-H, CH2Cl2, 0
°C, 30 min, (iv) TEMPO, BAIB, CH2Cl2/H2O (1:1), 3 h, (v) Zn(NO3)2·6H2O, CH3CN, 80 °C, 2 h, 34% over five steps;
(f) 2-methyl-6-nitrobenzoic anhydride, DMAP, 4 Å MS, CH2Cl2, rt, 12 h.We reasoned that
the steric crowding due to the presence of a bulky
TBS ether at the C-3 position might be a cause of unsuccessful macrolactonization.
Accordingly, we planned for seco acid 48 having a relatively
less sterically constrained PMB ether at the C-3 position. Initially,
the protection of the free hydroxy of compound 47 as
a PMB ether was attempted using PMB-chloromidate in the presence of
TfOH or CSA, but the starting material decomposed completely. Next,
compound 47 was reacted with LiCl/NaBH4 to
obtain the corresponding diol, which was further reacted with p-methoxy benzyl dimethoxy acetal in the presence of a catalytic
amount of CSA to furnish the corresponding acetal. This was then opened
regioselectively with DIBAL-H to get the corresponding primary alcohol,
which directly converted to the corresponding acid using TEMPO/BAIB.[4f] Finally, the acid was subjected to selective
acetonide deprotection in the presence of Zn(NO3)2·6H2O[23] to provide the
requisite seco acid 48 in 34% overall yield (in five
steps). The stage was set for crucial macrolactonization. It was observed
that a complex mixture was formed under Steglich conditions, whereas
the staring material decomposed completely under Yamaguchi conditions.
Interestingly, the Shiina macrolactonization protocol provided a mixture
of macrocyclic dimers, confirmed by HRMS. This result suggested that
the cyclization of a C-3 unprotected seco acid might help us to achieve
the desired result. In this context, it did not appear realistic to
proceed with three unprotected hydroxy groups. Therefore, we planned
to protect the C-8 hydroxy group as a MOM ether and manipulated the
precursors accordingly.Efforts toward the synthesis of cytospolide
E using the C-8 hydroxy
group in a protected form is described in Scheme . Previously synthesized compound 22 was treated with MOMCl/DIPEA to get the corresponding MOM
ether and subsequently was subjected to oxidative cleavage using OsO4/NaIO4 to obtain the corresponding aldehyde. The
resultant aldehyde was then reacted with ethyl(triphenylphosphoranylidene)acetate
following the Wittig olefination protocol followed by treatment with
DIBAL-H to yield allylic alcohol 49 in 58% overall yield.
Alcohol 49 was then oxidized to the corresponding aldehyde
using Swern conditions and subsequently was subjected to the modified
Crimmins aldol reaction[22] in the presence
of auxiliary 44(22) to obtain
compound 50 in 77% yield. Next, compound 50 was saponified in the presence of LiOH·H2O/30%H2O2 to get the corresponding acid. Our initial efforts
to deprotect the TBS ether to get the required seco acid 51 were not successful. This compelled us to change our strategy. We
planned first to deprotect the TBS ether and then install the acid
functionality. Therefore, compound 50 was treated initially
with TBAF, which eventually produced an eliminated product rather
than the required desilylated product. We speculated that a milder
desilylating agent might solve this problem. Thus, HF-py in THF was
tested. We were delighted to see that the corresponding desilylated
product was obtained in very good yield (88%), which was subsequently
saponified in the presence of LiOH·H2O/30% H2O2 to obtain seco acid 51 in 84% yield. Seco
acid 51 was then subjected to macrolactonization. Both
Shiina and Yamaguchi protocols were tested. Unfortunately, no satisfactory
results were obtained. In both the cases, the starting material either
remained unreacted as a mixed anhydride or decomposed completely.
Scheme 10
Attempt toward the Synthesis of Cytospolide E Using the C-8 Hydroxy
Group in a Protected Form
Reagents and conditions:
(a)
(i) MOMCl, DIPEA, CH2Cl2, 0 °C to rt, 3
h, (ii) OsO4, NaIO4, NaHCO3, CH3CN/H2O (1:1), 0 °C to rt, 6 h, (iii) PPh3=CHCO2Et, toluene, 80 °C, reflux, 6
h, (iv) DIBAL-H, CH2Cl2, −78 °C,
30 min, 58% over four steps; (b) (i) (COCl)2, DMSO, Et3N, CH2Cl2, −78 to 0 °C,
2 h, (ii) 44, TiCl4, DIPEA, NMP, CH2Cl2, 0 °C, 2 h, 77% over two steps; (c) (i) HF-py,
THF, 0 °C to rt, 3 h, 88%, (ii) LiOH·H2O, 30%
H2O2, THF/H2O (3:1), 6 h, 84%.
Attempt toward the Synthesis of Cytospolide E Using the C-8 Hydroxy
Group in a Protected Form
Reagents and conditions:
(a)
(i) MOMCl, DIPEA, CH2Cl2, 0 °C to rt, 3
h, (ii) OsO4, NaIO4, NaHCO3, CH3CN/H2O (1:1), 0 °C to rt, 6 h, (iii) PPh3=CHCO2Et, toluene, 80 °C, reflux, 6
h, (iv) DIBAL-H, CH2Cl2, −78 °C,
30 min, 58% over four steps; (b) (i) (COCl)2, DMSO, Et3N, CH2Cl2, −78 to 0 °C,
2 h, (ii) 44, TiCl4, DIPEA, NMP, CH2Cl2, 0 °C, 2 h, 77% over two steps; (c) (i) HF-py,
THF, 0 °C to rt, 3 h, 88%, (ii) LiOH·H2O, 30%
H2O2, THF/H2O (3:1), 6 h, 84%.
Conclusions
In summary, we have
documented various synthetic approaches toward
the total synthesis of cytospolide E. The ring-closing metathesis
approach provided a 10-membered macrocycle embedded with skeletal
Z-olefin as a major product. It was further observed that changes
in reaction conditions and the structure of the RCM precursor did
not result in the required E-olefinic product, clearly indicating
that the above-mentioned factors do not affect the transition state
enough to favor an E-selective RCM. Thus, we believe that the RCM-based
macrocyclization strategy is unsuitable for the synthesis of cytospolide
E. On the other hand, the macrolactonization approach did not produce
the required macrocycle either. However, 8-epi-9-epi-cytospolide E was synthesized by the regioselective
macrolactonization of the 10-membered macrocycle over the possible
9-membered macrocycle from an isomeric mixture of seco acids. This
result demonstrates that the stereochemistry of the C-2 methyl group
plays a crucial role in macrolactonization. This is further supported
by the fact that cytospolide D, with an inverted configuration of
the C-2 methyl center relative to that of cytospolide E, was synthesized
by others, albeit the yield was very low. It is noteworthy that a
seco acid having only a C-3 hydroxy protecting group provided a 10-membered
lactone with E-olefin, although C-8 stereochemistry and C-9 stereochemistry
are in opposite configuration relative to the required one. We are
of the opinion that macrolactonization is a viable approach to achieve
the unique architecture of cytospolide E. However, further explorations
based on functional group conversions, especially around the C-8 and
C-9 centers, are essential before a successful synthesis is achieved.
Efforts are presently in progress in that direction in our laboratory
to accomplish the total synthesis of cytospolide E, which will be
reported in due course.
Experimental Section
General Experimental Procedure
All moisture-sensitive
reactions were performed in oven or flame-dried glassware with a Teflon-coated
magnetic stirring bar under an argon atmosphere using dry, freshly
distilled solvents, unless otherwise noted. Air- and moisture-sensitive
liquids were transferred via a gastight syringe and a stainless steel
needle. Reactions were monitored by thin layer chromatography (TLC,
silica gel 60 F254) plates where ethanolic anisaldehyde (with 1% AcOH
and 3.3% conc. H2SO4)-heat and aqueous KMnO4 (with K2CO3 and 10% aqueous NaOH solution)
were used as developing agents. All workup and purification procedures
were carried out with reagent-grade solvents under an ambient atmosphere
unless otherwise stated. Column chromatography was performed using
silica gel 60–120 mesh, 100–200 mesh, and 230–400
mesh. Yields were mentioned as chromatographically and spectroscopically
homogeneous materials unless otherwise stated. Optical rotations were
measured only for pure compounds and not for mixtures using a sodium
(589, D line) lamp and are reported as follows: [α]D25 (c = mg/100 mL, solvent). IR spectra
were recorded as thin films (for liquids). HRMS was performed using
a quadruple-TOF micro-MS system using the electrospray ionization
(ESI) technique. 1H NMR spectra were recorded on 300, 400,
and 500 MHz spectrometers in appropriate solvents and calibrated using
a residual undeuterated solvent as an internal reference, and the
chemical shifts are shown in parts per million (ppm) scales. Multiplicities of NMR signals are
designated as s (singlet), d (doublet), t (triplet), q (quartet),
br (broad), m (multiplet, for unresolved lines), etc. 13C and 2D-NMR spectra were recorded on 75, 100, and 125 MHz spectrometers.
(S)-2-((R)-1-(Benzyloxy)hexyl)oxirane
(11)
To a stirred suspension of NaH (60%, 1.67
g, 41.60 mmol) in anhydrous THF (100 mL) at 0 °C under argon,
a solution of known epoxy alcohol 10 (5.0 g, 34.67 mmol)
dissolved in anhydrous THF (50 mL) was added dropwise. The reaction
mixture was stirred for 1 h at the same temperature. Benzyl bromide
(4.95 mL, 41.60 mmol) was then added over a period of 10 min. Finally,
TBAI (0.64 g, 1.73 mmol) was added, and the reaction mixture was warmed
slowly to room temperature and stirred further for 3 h. The reaction
was then quenched by slow addition of cold water (10 mL). The aqueous
layer was extracted with EtOAc (3 × 100 mL), washed with water
and brine, dried over Na2SO4, filtered, and
concentrated in vacuo. Purification of the crude residue by flash
column chromatography (SiO2, 230–400 mesh, 2–9%
EtOAc in hexane as eluent) afforded benzyl ether 11 (6.98
g, 86%) as a colorless liquid: Rf = 0.50
(5% EtOAc/hexane); 1H NMR (CDCl3, 300 MHz) δ
7.36–7.24 (m, 5H), 4.65 (d, J = 11.4 Hz, 1H),
4.50 (d, J = 11.4 Hz, 1H), 3.28–3.22 (m, 1H),
2.94–2.90 (m, 1H), 2.77 (dd, J = 5.4, 3.9
Hz, 1H), 2.71 (dd, J = 5.4, 2.7 Hz, 1H), 1.67–1.59
(m, 2H), 1.59–1.47 (m, 1H), 1.44–1.25 (m, 5H), 0.88
(t, J = 6.9 Hz, 3H); 13C NMR (CDCl3, 75 MHz) δ 138.7, 128.4, 127.8, 127.7, 78.2, 72.4,
53.7, 45.7, 32.9, 31.9, 24.9, 22.7, 14.1 ppm; IR (neat) νmax 2929, 2859, 1454, 1091, 733, 696 cm–1; HRMS (ESI) m/z calcd for C15H22O2Na [M + Na]+ 257.1518,
found 257.1517.
(5S,6R)-5-(Methoxymethoxy)undec-1-en-6-ol
(8)
To an ice-cold stirred solution of epoxide 11 (2.5 g, 10.67 mmol) in anhydrous THF (30 mL) under argon,
CuCN (143 mg, 1.60 mmol) and allyl magnesium chloride (8.0 mL, 2 M
in THF, 16.0 mmol) were added sequentially. The reaction mixture was
stirred further for 1 h at the same temperature and then for 2 h at
room temperature. The reaction was then quenched with a saturated
aqueous solution of NH4Cl (5 mL) at 0 °C. The mixture
was extracted with EtOAc (2 × 50 mL), washed with water and brine,
dried over Na2SO4, filtered, and concentrated
in vacuo. Purification of the crude residue by flash column chromatography
(SiO2, 100–200 mesh, 5–12% EtOAc in hexane
as eluent) afforded the corresponding compound (2.44 g, 83%) as a
clear liquid: Rf = 0.40 (10% EtOAc/hexane);
[α]D25 = +9.7 (c 1.6,
CHCl3); 1H NMR (CDCl3, 300 MHz) δ
7.35–7.25 (m, 5H), 5.91–5.77 (m, 1H), 5.08–4.96
(m, 2H), 4.57 (dd, J = 14.7, 11.7 Hz, 2H), 3.84–3.77
(m, 1H), 3.38–3.33 (m, 1H), 2.35–2.23 (m, 1H), 2.17–2.03
(m, 2H), 1.63–1.41 (m, 5H), 1.33–1.28 (m, 5H), 0.88
(t, J = 6.9 Hz, 3H); 13C NMR (CDCl3, 75 MHz) δ 138.6, 128.5, 127.9, 127.8, 115.0, 82.4,
72.1, 71.3, 32.1, 31.2, 30.5, 28.9, 25.6, 22.7, 14.2 ppm; IR (neat)
νmax 3451, 2929, 2858, 1454, 1070 cm–1; HRMS (ESI) m/z calcd for C18H28O2Na [M + Na]+ 299.1987,
found 299.1985.To a stirred solution of the above compound
(1.48 g, 5.35 mmol) in anhydrous CH2Cl2 (15
mL) under argon at 0 °C, DIPEA (1.86 mL, 10.7 mmol) and MOMCl
(0.61 mL, 8.0 mmol) were added sequentially. The reaction mixture
was allowed to attain room temperature and stirred further for 3 h
before quenching with water (5 mL). The organic layer was separated,
and the aqueous part was extracted with EtOAc (2 × 30 mL). The
combined organic extract was washed successively with saturated aqueous
NaHCO3, water, and brine; dried; and concentrated in vacuo.
Flash column chromatographic purification (SiO2, 100–200
mesh, 8% EtOAc in hexane as eluent) provided the corresponding MOM
ether (1.51 g, 88%) as a colorless liquid: Rf = 0.60 (5% EtOAc/hexane); 1H NMR (CDCl3, 300 MHz) δ 7.34–7.24 (m, 5H), 5.90–5.76 (m,
1H), 5.07–4.96 (m, 2H), 4.77 (d, J = 6.9 Hz,
1H), 4.72–4.63 (m, 2H), 4.50 (d, J = 11.7
Hz, 1H), 3.74–3.69 (m, 1H), 3.46–3.37 (m, 4H), 2.31–2.19
(m, 1H), 2.14–2.03 (m, 1H), 1.79–1.67 (m, 1H), 1.59–1.42
(m, 3H), 1.35–1.25 (m, 6H), 0.87 (t, J = 6.9
Hz, 3H); 13C NMR (CDCl3, 75 MHz) δ 138.9,
138.6, 128.4, 127.9, 127.6, 114.9, 96.4, 81.2, 78.3, 72.2, 55.9, 32.0,
30.6, 30.3, 30.0, 25.9, 22.7, 14.2 ppm; IR (neat) νmax 2930, 2859, 1718, 1272, 1096, 1026 cm–1; HRMS
(ESI) m/z calcd for C20H32O3Na [M + Na]+ 343.2249, found
343.2246.To a stirred solution of naphthalene (1.69 g, 13.2
mmol) in anhydrous
THF (25 mL) under argon, small pieces of lithium metal (114 mg, 16.5
mmol) were added. The reaction mixture was then stirred at the ambient
temperature for 2 h to form a dark green solution of lithium napthalenide.
The reaction mixture was then cooled to −40 °C, and the
above MOM-protected compound (1.06 g, 3.30 mmol) dissolved in 5 mL
of anhydrous THF was cannulated into it. After being stirred at −40
°C for 30 min, the reaction mixture was quenched with the saturated
aqueous solution of NH4Cl (5 mL) and extracted with EtOAc
(3 × 30 mL). The combined organic layers were washed with water
and brine and dried over Na2SO4, filtered, and
concentrated under reduce pressure. The residue was purified by flash
column chromatography (SiO2, 100–200 mesh, 8–18%
EtOAc in hexane as eluent) to get compound 8 (620 mg,
82%) as a clear oil: Rf = 0.28 (10% EtOAc/hexane);
[α]D24 = +7.9 (c 1.8,
CHCl3); 1H NMR (CDCl3, 300 MHz) δ
5.88–5.74 (m, 1H), 5.06–4.96 (m, 2H), 4.73 (d, J = 6.6 Hz, 1H), 4.64 (d, J = 6.6 Hz, 1H),
3.63–3.59 (m, 1H), 3.53 (dt, J = 9.0, 3.0
Hz, 1H), 3.42 (s, 3H), 2.74 (d, J = 6.3 Hz, 1H),
2.30–2.20 (m, 1H), 2.18–2.01 (m, 1H), 1.75–1.61
(m, 1H), 1.59–1.47 (m, 2H), 1.46–1.39 (m, 2H), 1.37–1.25
(m, 5H), 0.89 (m, 3H); 13C NMR (CDCl3, 75 MHz)
δ 138.3, 115.0, 97.3, 83.5, 73.1, 55.9, 32.0, 31.7, 30.2, 29.4,
26.0, 22.7, 14.1 ppm; HRMS (ESI) m/z calcd for C13H26O3Na [M + Na]+ 253.1780, found 253.1783.
A solution of ester 7 (60
mg, 0.13 mmol) in anhydrous and degassed CH2Cl2 (80 mL) was taken in a round-bottomed flask under argon fitted with
a reflux condenser. The Grubbs second-generation metathesis catalyst
(G-II, 11 mg, 0.013 mmol) was added, and the solution was refluxed
for 13 h. The solvent was evaporated, and the resultant residue was
directly loaded on a silica gel column. Flash column chromatography
(SiO2, 100–200 mesh, 6% EtOAc in hexane as eluent)
provided linear dimeric mixture 13 (43 mg, 37%) along
with trace amount of lactone 12.
To a solution
of compound 14 (50 mg, 0.15 mmol) in anhydrous and degassed
CH2Cl2 (60 mL), Grubbs second-generation metathesis
catalyst (G-II, 12.7 mg, 0.015 mmol) was added, and the solution was
refluxed for 2.5 h. The solvent was evaporated, and the residue was
directly loaded on a silica gel column. Flash column chromatography
(SiO2, 230–400 mesh, 7–24% EtOAc in hexane
as eluent) afforded lactone 15 (21.1 mg, 46%) and cyclic
dimeric products 16 (12.8 mg, 14%).
HRMS (ESI) m/z calcd for C34H60O10Na [M + Na]+ 651.4084,
found 651.4083.
(R)-6-Hydroxyundec-1-en-5-one
(20)
To an ice-cold solution of compound 22 (1.00
g, 3.33 mmol) in anhydrous CH2Cl2 (10 mL) under
argon, NaHCO3 (11.18 g, 13.3 mmol) and DMP (2.12 g, 5.00
mmol) were added sequentially. The reaction mixture was allowed to
attain ambient temperature with constant stirring. After 3 h, the
reaction mixture was quenched with saturated aqueous solution of Na2S2O3, diluted with Et2O (60
mL), and stirred for another 30 min until the two phases separated.
The mixture was then transferred to a separating funnel, and the organic
extract was washed with water and brine, dried over Na2SO4, and concentrated in vacuo. Purification of the crude
residue by flash column chromatography (SiO2, 100–200
mesh, 5% EtOAc in hexane as eluent) furnished the corresponding ketone
(940 mg, 95%) as a pale yellow liquid: Rf = 0.58 (5% EtOAc/hexane); [α]D27 = +24.2
(c 2.3, CHCl3); 1H NMR (CDCl3, 300 MHz) δ 5.87–5.74 (m, 1H), 5.06–4.94
(m, 2H), 4.00 (dd, J = 6.9, 5.4 Hz, 1H), 2.65–2.60
(m, 2H), 2.33–2.26 (m, 2H), 1.66–1.50 (m, 2H), 1.36–1.23
(m, 6H), 0.92–0.81 (m, 12H), 0.05 (s, 3H), 0.03 (s, 3H); 13C NMR (CDCl3, 75 MHz) δ 213.4, 137.5, 115.1,
78.9, 36.7, 35.0, 31.8, 27.3, 25.9, 24.5, 22.6, 18.2, 14.1, −4.7
ppm; IR (neat) νmax 2955, 2929, 2858, 1717, 1092
cm–1; HRMS (ESI) m/z calcd for C17H34O2NaSi [M + Na]+ 321.2226, found 321.2228.The above ketone (460 mg,
1.54 mmol) was dissolved in anhydrous THF (5 mL) under argon at 0
°C. TBAF (2.31 mL, 1 M solution in THF, 2.31 mmol) was added.
The reaction mixture was warmed to room temperature and stirred further
for 5 h prior to quenching with saturated aqueous NH4Cl
solution (2 mL). The resultant mixture was extracted with EtOAc (3
× 15 mL), washed with water and brine, dried over Na2SO4, and concentrated in vacuo. Flash column chromatographic
purification (SiO2, 100–200 mesh, 5% EtOAc in hexane
as eluent) of the crude residue gave compound 20 (261
mg, 92%) as an oily liquid: Rf = 0.32
(5% EtOAc/hexane); [α]D27 = −38.4
(c 1.5, CHCl3); 1H NMR (CDCl3, 300 MHz) δ 5.85–5.71 (m, 1H), 5.07–4.96
(m, 2H), 4.17–4.12 (m, 1H), 3.45 (d, J = 5.1
Hz, 1H), 2.59–2.46 (m, 2H), 2.39–2.32 (m, 2H), 1.81–1.76
(m, 1H), 1.53–1.42 (m, 2H), 1.35–1.24 (m, 5H), 0.89–0.85
(m, 3H); 13C NMR (CDCl3, 75 MHz) δ 211.6,
136.6, 115.8, 76.6, 37.1, 33.8, 31.7, 27.6, 24.6, 22.5, 14.0 ppm;
IR (neat) νmax 3472, 2928, 2860, 1710, 1060 cm–1; HRMS (ESI) m/z calcd for C11H20O2Na [M + Na]+ 184.1463, found 184.1464.
Following the same experimental procedure as described
for the preparation of compound 15, compound 18 (54 mg, 0.16 mmol) dissolved in anhydrous and degassed CH2Cl2 (80 mL) was treated with the Grubbs second-generation
metathesis catalyst (G-II, 13.6 mg, 0.016 mmol) to yield compound 23 (26.4 mg, 51%) and dimeric products 24 (12.4
mg, 13%, purification 230–400 mesh, 7–18% EtOAc in hexane
as eluent) as a brownish liquid.
To a solution of the above alcohol
(520 g, 2.40 mmol) in CH2Cl2 and H2O (1:1, 10 mL), TEMPO (150 mg, 0.96 mmol) and BAIB (3.09 g, 9.60
mmol) were added. The reaction was continued for 3 h at the room temperature,
and then the mixture was diluted with (10 mL) CH2Cl2. The organic layer was separated, and the aqueous layer was
extracted with CH2Cl2 (2 × 15 mL). The
combined organic layers were washed with saturated aqueous Na2S2O3 and brine, dried over Na2SO4, filtered, and the filtrate was concentrated under
reduced pressure. Purification of the crude residue by column chromatography
(SiO2, 100–200 mesh, 12–25% EtOAc in hexane
as eluent) gave compound 27 (465 mg, 84%) as a liquid: Rf = 0.29 (20% EtOAc/hexane); [α]D27 = +2.9 (c 0.8, CHCl3); 1H NMR (CDCl3, 300 MHz) δ 5.90–5.79
(m, 1H), 5.25 (d, J = 17.1 Hz, 1H), 5.10 (d, J = 10.5 Hz, 1H), 4.58 (q, J = 6.0 Hz,
1H), 2.61–2.47 (m, 2H), 0.88 (s, 9H), 0.07 (s, 3H), 0.05 (s,
3H); 13C NMR (CDCl3, 75 MHz) δ 176.8,
139.8, 115.3, 70.7, 43.4, 25.8, 18.2, −4.2, −5.0 ppm;
IR (neat) νmax 2932, 2878, 1720, 1032 cm–1.
To an ice-cold solution of
compound 42 (5.0 g, 32.0 mmol) in a mixture of THF (15
mL), BuOH (15 mL), and water (3 mL),
OsO4 (2% solution in BuOH,
2.5 mL) and NMO (7.5 g, 64.0 mmol) were added, and the reaction mixture
was stirred for 1 h at the same temperature. Next, the reaction mixture
was allowed to warm to room temperature and stirred for another 6
h. It was then quenched with a solution of Na2SO3. The solvent was removed in a rotary evaporator, and the residue
was extracted with EtOAc (3 × 50 mL). The combined organic extracts
were washed with water and brine, dried (Na2SO4), and concentrated in vacuo to give the corresponding triol as a
white solid, which was taken directly to the next reaction without
further purification or characterization.To an ice-cold solution
of the above triol in anhydrous CH2Cl2 (50 mL)
were added 2,2-DMP (10 mL) and CSA (75 mg, 0.32 mmol). The solid was
insoluble initially. However, it dissolved slowly with the progress
of the reaction. After 3 h, the reaction was quenched with a saturated
solution of NaHCO3. The reaction mixture was extracted
with CH2Cl2 (2 × 50 mL), washed with brine,
dried (Na2SO4), and concentrated in vacuo. The
resultant residue was purified by column chromatography (SiO2, 100–200 mesh, 15–28% EtOAc in hexane as eluent) to
afford a mixture of compounds 43 and 43a (12.4 g, 84%) as a colorless liquid: Rf = 0.25 (20% EtOAc/hexane); 1H NMR (500 MHz, CDCl3) δ 4.06–4.04 (m, 2H), 3.70–3.66 (m, 2H),
2.20 (bs, 1H), 1.75–1.55 (m, 2H), 1.53–1.36 (m, 7H),
1.33–1.24 (m, 9H), 0.88 (m, 3H); 13C NMR (100 MHz,
CDCl3) δ 107.6, 78.3, 78.2, 62.9, 32.0, 29.9, 29.8,
28.6, 26.8, 26.0, 22.7, 14.1 ppm; IR (neat) νmax 3407,
2933, 1218, 1029 cm–1; HRMS (ESI) m/z calcd for C13H26O3Na [M + Na]+ 253.1780, found 253.1782.
To a stirred solution
of (COCl)2 (1.79 mL, 20.93 mmol) in anhydrous CH2Cl2 (20 mL) under argon at −78 °C, anhydrous
DMSO (3.05 mL, 41.85 mmol) was added dropwise. After 15 min, compounds 43 and 43a (3.21 g, 13.95 mmol) dissolved in
anhydrous CH2Cl2 (13 mL) were cannulated into
the reaction mixture and the reaction mixture was stirred for 30 min
at −78 °C. Anhydrous Et3N (9.70 mL, 69.75 mmol)
was then added, and the reaction mixture was stirred for another 30
min at the same temperature. The reaction mixture was then allowed
to warm to 0 °C and then quenched with a saturated aqueous solution
of NH4Cl (5 mL). The reaction mixture was extracted with
CH2Cl2 (3 × 50 mL), washed with water and
brine, dried over Na2SO4, and concentrated in
vacuum. Flash chromatography with a short pad of silica gave the corresponding
aldehyde, which was used for the next step without further purification
and characterizations.The above aldehyde was dissolved in dry
toluene, and ethyl(triphenylphosphoranylidene)acetate (9.7 g, 27.9
mmol) was added. The reaction was continued for 6 h at 80 °C.
The mixture was then cooled, and toluene was evaporated in vacuum.
The resultant residue was purified by flash column chromatography
(SiO2, 100–200 mesh, 5–12% EtOAc in hexane
as eluent) to yield the corresponding α,β-unsaturated
ester (3.39 g, 81%) as a colored liquid: Rf = 0.64 (20% EtOAc/hexane); (mixture of isomers) 1H NMR
(500 MHz, CDCl3) δ 6.97 (m, 1H), 5.83 (d, J = 15.5 Hz, 1H), 4.15 (q, J = 7.0 Hz,
2H), 4.04–3.99 (m, 2H), 2.45–2.40 (m, 1H), 2.25–2.22
(m, 1H), 1.64–1.60 (m, 1H), 1.51–1.45 (m, 3H), 1.44–1.36
(m, 4H), 1.37–1.25 (m, 11H), 0.87 (m, 3H); 13C NMR
(125 MHz, CDCl3) δ 166.7, 148.5, 121.8, 107.6, 78.0,
77.4, 60.3, 32.0, 29.6, 28.9, 28.7, 28.6, 26.1, 26.0, 22.6, 14.3,
14.1 ppm; IR (neat) νmax 2935, 1722, 1654, 1217 cm–1; HRMS (ESI) m/z calcd for C17H30O4Na [M + Na]+ 321.2042, found 321.2044.To a cooled solution (−78
°C) of the above ester (3.89
g, 13.05 mmol) in anhydrous CH2Cl2 (15 mL) under
argon, DIBAL-H (1.0 M in toluene, 27.40 mL, 27.40 mmol) was added
dropwise. The reaction mixture was stirred for another 30 min before
quenching it with MeOH (8 mL). The reaction mixture was diluted with
Et2O (60 mL), a saturated solution of sodium potassium
tartrate (20 mL) was added, and the mixture was stirred further for
1 h until the two layers separated well. The solvent was removed,
and the aqueous part was extracted with Et2O (2 ×
50 mL), washed with water and brine, dried over Na2SO4, and concentrated in vacuo. Flash column chromatography (SiO2, 100–200 mesh, 25% EtOAc in hexane as eluent) of the
crude residue afforded an inseparable mixture of allylic alcohols 41 and 41a (2.95 g, 88%) as a colorless oil: Rf = 0.29 (20% EtOAc/hexane); (mixture of isomers) 1H NMR (500 MHz, CDCl3) δ 5.69–5.60
(m, 2H), 4.05–3.96 (m, 4H), 2.25–2.19 (m, 1H), 2.08–1.95
(m, 2H), 1.59–1.46 (m, 3H), 1.45–1.41 (m, 4H), 1.39–1.21
(m, 9H), 0.85 (t, J = 6.5 Hz, 3H); 13C
NMR (100 MHz, CDCl3) δ 132.4, 129.6, 107.5, 78.1,
63.7, 32.0, 29.7, 29.5, 28.9, 28.7, 26.1, 26.0, 22.8, 22.6, 14.1 ppm;
IR (neat) νmax 3351, 2936, 1455, 1217 cm–1; HRMS (ESI) m/z calcd for C15H28O3Na [M + Na]+ 279.1936,
found 279.1939.
Following a similar procedure
as that of the above Swern oxidation, the mixture of compounds 41and 41a (2.01 g, 7.80 mmol) was converted to
the corresponding aldehyde, which was used for the next step without
further purification and characterizations.To a dry round-bottomed
flask under argon, (S)-4-benzyl-3-propionyloxazolidin-2-one
(1.82 g, 7.80 mmol) and anhydrous 40 mL CH2Cl2 were added. After cooling to 0 °C, TiCl4 (0.94 mL,
8.58 mmol) was added dropwise and the solution was allowed to stir
for 15 min. DIPEA (1.63 mL, 9.36 mmol) was added dropwise to the reaction
mixture, and the solution was stirred for 40 min. 1-Methyl-2-pyrrolidinone
(0.75 mL, 7.80 mmol) was added at 0 °C, and the mixture was stirred
for an additional 15 min. The aldehyde from the previous step was
dissolved in anhydrous CH2Cl2 (10 mL) and cannulated
directly to the in situ generated enolate. The reaction mixture was
stirred further for 1 h at the same temperature and finally quenched
with a saturated aqueous solution of NH4Cl. The organic
layer was separated, and the aqueous layer was extracted with CH2Cl2 (2 × 30 mL). The combined organic layers
were dried over Na2SO4, filtered, and concentrated.
Flash column chromatography of the crude residue (SiO2,
230–400, 18–35% EtOAc in hexane as eluent) furnished
an inseparable mixture of corresponding diastereomers (3.01 g, 78.6%
yield over two steps) as a gummy liquid: Rf = 0.33 (25% EtOAc/hexane); (mixture of isomers) 1HNMR
(300 MHz, CDCl3) δ 7.36–7.19 (m, 5H), 5.83–5.74
(m, 1H), 5.52 (dd, J = 15.3, 6.0 Hz, 1H), 4.74–4.66
(m, 1H), 4.46–4.43 (m, 1H), 4.25–4.16 (m, 2H), 4.03–3.98
(m, 2H), 3.90–3.82 (m, 1H), 3.25 (dd, J =
13.3, 3.5 Hz, 1H), 2.87–2.75 (m, 2H), 2.31–2.22 (m,
1H), 2.16–2.06 (m, 1H), 1.53–1.44 (m, 3H), 1.42–1.37
(m, 4H), 1.32–1.23 (m, 12H), 0.91–0.87 (m, 3H); 13C NMR (75 MHz, CDCl3) δ 176.7, 153.2, 135.1,
132.6, 129.5, 129.0, 127.5, 107.4, 78.1, 77.4, 72.6, 66.3, 55.2, 42.9,
37.9, 31.9, 29.7, 29.4, 28.9, 28.7, 26.1, 26.0, 22.6, 14.1, 11.3 ppm;
IR (neat) νmax 3472, 2934, 2860, 1781, 1698, 1455,
1380, 1217, 771 cm–1; HRMS (ESI) m/z calcd for C28H41NO6Na [M + Na]+ 510.2832, found 510.2834.To
an ice-cold solution of the above aldol adducts (2.73 g, 5.61
mmol) in anhydrous CH2Cl2 (20 mL), 2,6-lutidine
(1.53 mL, 14.02 mmol) and TBSOTf (1.54 mL, 6.73 mmol) were added sequentially,
and the reaction mixture was stirred for 30 min at the same temperature
before quenching with a saturated aqueous solution of NaHCO3. The organic layer was separated, and the aqueous layer was extracted
with CH2Cl2 (2 × 30 mL). The combined organic
extracts were washed with aqueous CuSO4, water, and brine,
dried (Na2SO4), filtered, and concentrated in
vacuo. Purification of the crude residue by column chromatography
(SiO2, 100–200 mesh, 10% EtOAc in hexane as eluent)
furnished compounds 40 and 40a (3.05 g,
91%) as a gummy liquid: Rf = 0.39 (10%
EtOAc/hexane); (mixture of isomers) 1H NMR (300 MHz, CDCl3) δ 7.33–7.20 (m, 5H), 5.66–5.58 (m, 1H),
5.50 (dd, J = 15.9, 6.6 Hz, 1H), 4.62–4.55
(m, 1H), 4.29 (t, J = 6.6 Hz, 1H), 4.18–4.09
(m, 2H), 4.03–3.96 (m, 3H), 3.27 (dd, J =
13.5, 3.0 Hz, 1H), 2.77 (dd, J = 13.5, 9.6 Hz, 1H),
2.26–2.19 (m, 1H), 2.11–1.99 (m, 1H), 1.54–1.38
(m, 7H), 1.36–1.24 (m, 9H), 1.20 (d, J = 6.9
Hz, 3H), 0.92–0.86 (m, 12H), 0.02 (s, 3H), 0.00 (s, 3H); 13C NMR (75 MHz, CDCl3) δ 174.9, 153.3, 135.5,
131.7, 131.7, 131.6, 131.6, 129.5, 129.0, 127.4, 107.4, 78.0, 77.3,
75.1, 66.0, 55.7, 44.4, 37.9, 32.0, 29.7, 29.5,29.4, 28.8, 28.7, 26.1,
26.0, 25.9, 25.7, 22.7, 18.2, 14.1, 12.7, −4.0, −4.8
ppm; IR (neat) νmax 2932, 2858, 1781, 1698, 1379,
1217, 774 cm–1; HRMS (ESI) m/z calcd for C34H55NO6Na
[M + Na]+ 624.3696, found 624.3697.
2,4,6-Trichlorobenzoyl
chloride (0.11 mL, 0.71 mmol) was added to a stirred solution of the
mixture of seco acids 39 and 39a (55 mg,
0.14 mmol) and Et3N (0.19 mL, 1.40 mmol) in anhydrous THF
(7 mL) under argon at room temperature. After stirred the reaction
mixture for 6 h, the resultant Et3N·HCl salt was filtered
off quickly and the filtrate was concentrated under reduced pressure.
The mixed anhydride thus obtained was dissolved in anhydrous toluene
(20 mL) and added slowly via a syringe pump over 8 h to a stirred
solution of DMAP (87 mg, 0.71 mmol) in anhydrous toluene (100 mL)
at 80 °C. The syringe was rinsed with another 20 mL of dry toluene, which was added over
2 h at the same temperature. The reaction mixture was then allowed
to attain room temperature and stirred further for 12 h. The reaction
mixture was quenched with saturated aqueous NaHCO3 (5 mL)
solution. The toluene layer was separated, and the aqueous layer was
extracted with EtOAc (2 × 15 mL). The organic extracts were combined,
washed with water and brine, dried (Na2SO4),
and concentrated in vacuo. The attempt to purify the crude residue
by column chromatography (SiO2, 230–400 mesh, 10–15%
EtOAc in hexane as eluent) resulted in a complex inseparable mixture
of compounds (14 mg), which was taken for the next step without further
characterizations.To an ice-cold solution of the above mixture
(14 mg) in anhydrous THF (2 mL), TBAF (1.0 M solution in THF, 0.15
mL) was added. The reaction mixture was then stirred for 3 h at the
same temperature prior to quenching with saturatedNH4Cl
solution (1 mL). The mixture was extracted with EtOAc (2 × 15
mL), washed with water and brine, dried (Na2SO4), filtered, and concentrated in vacuo. Purification of the crude
residue by column chromatography (SiO2, 100–200
mesh, 30% EtOAc in hexane as eluent) yielded lactone 45 (4 mg, 22%, based on a single isomer) as a colorless oil: Rf = 0.45 (40% EtOAc/hexane); [α]29D = +6.13 (c 0.2, CHCl3); 1H NMR (500 MHz, CDCl3) δ 5.50 (dd, J = 16.0, 8.0 Hz, 1H), 5.47–5.41 (m, 1H), 4.77 (ddd, J = 11.0, 7.5, 4.0 Hz, 1H), 3.93 (t, J =
9.0 Hz, 1H), 3.65 (m, 1H), 2.50 (dq, J = 9.0, 6.5
Hz, 1H), 2.31–2.27 (m, 1H), 2.11–2.06 (m, 1H), 1.98–1.93
(m, 1H), 1.84–1.79 (m, 2H), 1.78–1.75 (m, 1H), 1.71–1.66
(m, 1H), 1.56 (m, 1H, merged with H2O), 1.43–1.22
(m, 9H), 0.87 (m, 3H); 13C NMR (75 MHz, CDCl3) 174.7, 133.6, 131.8, 78.1, 77.3, 74.2, 49.4, 37.8, 32.2, 31.9,
28.4, 24.6, 22.6, 14.1, 13.2 ppm; IR (neat) νmax 3359,
3198, 2987, 2945, 2852, 1710, 1665, 1634, 1463 cm–1; HRMS (ESI) m/z calcd for C15H26O4Na [M + Na]+ 293.1729,
found 293.1727.
To a stirred solution of (COCl)2 (4.15 mL, 48.20 mmol) in anhydrous CH2Cl2 (80
mL) under argon at −78 °C, DMSO (6.85 mL, 96.40 mmol)
was added dropwise. After 15 min, alcohol 46 (6.51 g,
32.15 mmol) dissolved in anhydrous CH2Cl2 (25
mL) was cannulated into the reaction mixture, which was stirred for
30 min at the same temperature. Et3N (22.40 mL, 160.66
mmol) was added, and the mixture was stirred for another 30 min at
the same temperature. The reaction mixture was then warmed slowly
to room temperature. The complete formation of the corresponding aldehyde
was confirmed by TLC. Next, ethyl(triphenylphosphoranylidene)acetate
(11.19 g, 32.13 mmol) was added into the reaction mixture and the
reaction mixture was stirred for 10 h at the same temperature. After
completion of the reaction, the solvent was evaporated under reduced
pressure. Flash column chromatography (SiO2, 100–200
mesh, 8% EtOAc in hexane as eluent) of the residue furnished the corresponding
α,β-unsaturated ester (7.01 g, 80.5%) as a colored liquid: Rf = 0.65 (20% EtOAc/hexane); [α]D27 = −6.13 (c 3.3, CHCl3); 1H NMR (500 MHz, CDCl3) δ 6.81 (dd, J = 16.0, 6.0 Hz, 1H), 6.03 (d, J = 16.0
Hz, 1H), 4.60 (m, 1H), 4.20–4.16 (m, 3H), 1.48–1.43
(m, 5H), 1.37–1.23 (m, 12H), 0.87–0.84 (m, 3H); 13C NMR (125 MHz, CDCl3) δ 166.1, 143.9, 123.0,
108.8, 78.5, 77.5, 60.5, 31.7, 30.5, 28.1, 26.0, 25.6, 22.5, 14.3,
14.0 ppm; IR (neat) νmax 2935, 1722, 1456, 1163 cm–1; HRMS (ESI) m/z calcd for C15H26O4Na [M + Na]+ 293.1729, found 293.1728.To an ice-cold solution of
NaBH4 (4.56 g, 120.58 mmol) in 30 mL of THF/ethanol solvent
mixture (1:1), LiCl (5.10 g, 120.58 mmol) was added. The solution
was stirred for 1 h at the same temperature. The above compound (6.52
g, 24.12 mmol) dissolved in anhydrous THF (15 mL) was cannulated,
and the reaction mixture was stirred for another 6 h at room temperature.
The reaction mixture was cooled to 0 °C and then quenched with
a saturated aqueous solution of NH4Cl (10 mL). The solvent
was removed under reduced pressure. The aqueous part was extracted
with EtOAc (3 × 80 mL). The combined organic layers were washed
with brine, dried over Na2SO4, and concentrated
in vacuo. Flash column chromatography (SiO2, 100–200
mesh, 30% EtOAc in hexane as eluent) of the resultant residue afforded
an inseparable mixture of corresponding unsaturated and saturatedalcohols (4.85 g). The mixture was dissolved in EtOAc (30 mL) and
was hydrogenated for 10 h in the presence of 10% Pd/C (500 mg) using
a hydrogen balloon. The reaction mixture was then filtered through
celite and concentrated to furnish compound 43 (4.81
g, 86.7%, yield over two steps) as a colorless liquid: Rf = 0.25 (20% EtOAc/hexane); [α]D28 = −2.0 (c 4.2, CHCl3); 1H NMR (500 MHz, CDCl3) δ 4.06–4.04
(m, 2H), 3.70–3.66 (m, 2H), 2.20 (bs, 1H), 1.75–1.55
(m, 2H), 1.53–1.36 (m, 7H), 1.33–1.24 (m, 9H), 0.88
(m, 3H); 13C NMR (100 MHz, CDCl3) δ 107.6,
78.3, 78.2, 62.9, 32.0, 29.9, 29.8, 28.6, 26.8, 26.0, 22.7, 14.1 ppm;
IR (neat) νmax 3407, 2933, 1218, 1029 cm–1; HRMS (ESI) m/z calcd for C13H26O3Na [M + Na]+ 253.1780,
found 253.1782.
Figure 1
Scheme 1
Scheme 2
Scheme 3
Scheme 4
Scheme 5
Scheme 6
Scheme 7
Scheme 8
Scheme 9
Scheme 10