Total Synthesis of the Reported Structure of Stresgenin B Enabled by the Diastereoselective Cyanation of an Oxocarbenium

Article abstract of DOI:10.1021/acs.orglett.8b03219

We report the first total synthesis of the reported structure of the heat shock protein expression inhibitor stresgenin B. The synthesis features (1) diastereoselective cyanation of an oxocarbenium intermediate en route to the synthetically challenging α-

Full text of DOI:10.1021/acs.orglett.8b03219

Letter
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
pubs.acs.org/OrgLett
Total Synthesis of the Reported Structure of Stresgenin B Enabled
by the Diastereoselective Cyanation of an Oxocarbenium
Wei Chuen Chan and Kazunori Koide*
Department of Chemistry, University of Pittsburgh, 219 Parkman Avenue, Pittsburgh, Pennsylvania 15260, United States
*
S Supporting Information
ABSTRACT: We report the first total synthesis of the reported
structure of the heat shock protein expression inhibitor stresgenin B.
The synthesis features (1) diastereoselective cyanation of an
oxocarbenium intermediate en route to the synthetically challenging
α-amido dioxolane, (2) Pd-catalyzed hydration of an unstable nitrile,
and (3) late-stage Au-catalyzed Meyer−Schuster rearrangement or
Ce-mediated Peterson olefination to furnish the exocyclic α,β-
unsaturated ester. Our synthetic endeavors allowed us to conclude
that the structure of stresgenin B requires revision.
3
eat shock proteins (HSPs), such as HSP70 and HSP90,
amide-substituted dioxolane (Scheme 1). The absolute
4
H
are molecular chaperones that perform various biological
processes vital for cell survival. Their overexpression has been
implicated in the development, proliferation, and various
aspects of the survival of cancer cellsmany of their client
proteins are oncogenic. HSPs have garnered a tremendous
amount of attention from the research community as cancer
stereochemistry was not reported. The prospects of stresgenin
Scheme 1. Reported Structure and Retrosynthetic Analysis
of Stresgenin B (1)
1
,2
drug targets (∼50 clinical trials as of 2016). Most of the
clinically studied HSP90 inhibitors bind to the N-terminus
(
ATP-binding site) of the protein. It is known that this mode
of inhibition triggers a heat-shock response (HSR) and induces
the syntheses of other HSPs (e.g., HSP27, 40 and 70) that also
have protective activity toward cancer cells. These limitations
1
,2
reduce the efficacies of the HSP90 inhibitors. There are
some C-terminal inhibitors documented, and they do not
2
trigger a HSR. Nevertheless, targeting the biosynthesis of an
HSP, rather than directly targeting the protein, provides a
different promising therapeutic approach that might bypass the
problem of HSR. In the literature, small molecules that inhibit
3
the biosynthesis of HSPs are very rare (e.g., quercetin). It was
discovered that stresgenin B was capable of inhibiting the heat-
induced gene expression of the human HSP70B promoter
3
driven luciferase in Chinese hamster ovary cells. The IC of
5
0
3
stresgenin B in the HSP70B promoter assay was 7.0 μM.
B being a lead for development of HSP expression inhibitors,
combined with its hitherto unknown mode of action,
prompted us to embark on a synthetic project on this
molecule. In this paper, we report the first total synthesis of the
proposed structure of stresgenin B and its Z-isomer to
determine that the proposed chemical structure requires
revision.
Meanwhile, stresgenin B was also found to be cytotoxic against
3
various cancer cell lines with IC values of 2.6−19.5 μM. The
50
closely related IC values suggested that the cytotoxicity may
5
0
be directly correlated with the inhibition of HSP70B
expression. This natural product also inhibited heat-shock-
3
induced syntheses of HSP72/73, HSP90, and HSP110.
The structure of the natural product was proposed based on
1
13
UV, H and C NMR, HMQC, HMBC, and NOESY
spectroscopic experiments. It was reported to contain a rare
Received: October 9, 2018
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b03219
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
To unambiguously determine the absolute stereoconfigura-
tion of stresgenin B (1), we chose a chiral pool synthesis. From
our preliminary studies, we found that cyclopentenone 5 was
not an accessible intermediate, that the formation of the highly
reactive cyclopentenone should be postponed to a late stage,
and that the α,β-unsaturated ester was inaccessible via
traditional olefination methods. After experimentally evaluating
several approaches, we arrived at the synthetic strategy
depicted in Scheme 1. Retrosynthetically, the exocyclic α,β-
unsaturated ester would be formed by a Meyer−Schuster
rearrangement. The primary amide would be derived from an
unstable cyanoketal 4. The nitrile could be installed stereo-
selectively via a neighboring group directed cyanation of an
oxocarbenium intermediate 6.
same in all cases. Additionally, the stereoselectivity improved
with an electron-poor Ar group but appeared to be
independent of the electronics of the R group (Bn or PMB).
Taken together, we propose that electrostatic and π−π
interactions are involved. The oxocarbenium 20b (Figure 1)
Figure 1. Proposed stereochemical model.
We commenced our synthesis from commercially available
D-ribose (Scheme 2), and the known epoxide 7 was prepared
might be more sterically sensitive, keeping the Bn group
further from the oxocarbenium, decreasing the electrostatic
stabilization. An electron-rich R group (i.e., PMB, see the SI)
should have stronger electrostatic attraction that is presumably
Scheme 2. Initial Synthetic Route
6
,7
counterbalanced by weaker π−π interaction.
We then decided to pursue a different strategy that might
enable us to both create the dioxolane and reverse the
stereoselectivity (Table 1 and Scheme 3). We envisioned that a
Table 1. Optimization of Cyanation
b
entry
Lewis acid (equiv)
temp (°C)
25
yield (dr) (%)
1
2
3
4
5
6
7
8
9
n-Bu BOTf (1.5)
dec
2
Sc(OTf) (0.20)
25
25
25
−78
−78
−78
−78
−78
16 (7.1:1)
15 (6.9:1)
dec
76 (12:1)
trace (nd)
3
TiCl (1.5)
4
Yb(OTf) ·nH O (0.20)
3
2
TiCl (1.5)
4
c
AlCl (1.5)
3
Et AlCl (1.5)
30 (14:1)
35 (8:1)
2
SnCl (1.5)
4
BF ·OEt (1.5)
91 (8.2:1)
3
2
d
1
0
BF ·OEt (1.5)
−78 to −20
97 (9:1)
3
2
a
Typical reaction conditions: 8 (0.10 mmol, 1 equiv), MeC(OMe)3
(0.15 mmol, 1.5 equiv), CSA (0.0050 mmol, 0.05 equiv), CH Cl , rt,
4
in three steps in 53% overall yield. The epoxide was
2
2
b
1
4
b,c
then TMSCN (4.0 equiv), Lewis acid (x equiv). Determined by H
NMR analysis using mesitylene as external standard. nd = not
determined. 50 mmol scale, combined isolated yield of diaster-
homologated,
protecting group manipulations to give diols 8 and 9 in 60 and
2% yield over three steps, respectively. The direct
the product of which was then subjected to
c
d
6
eomers.
construction of the α-amidodioxolane functionality proved to
be very challenging. The direct condensation of various α-
5
ketoamides and α-ketoesters was futile, notwithstanding the
lack of any stereoselectivity. After much exploration, we settled
on (phenylthio)acetones 10−12 to form the requisite
dioxolane (first-generation route, Scheme 2). Reaction
conditions were optimized to afford a good dr of up to 8.5:1
benzyloxy substituent at C-3 (Scheme 3) might undergo
neighboring-group participation in a cyclic oxocarbenium
8
−10
ion.
We sought to generate the oxocarbenium via an
orthoester. Diol 8 was treated with MeC(OMe) under acidic
3
conditions, and the resulting orthoester 22 was treated with
TMSCN and BF ·OEt to afford cyanoketal 23 (Table 1).
(
with 12). At this juncture, we were uncertain of the
3
2
stereochemical outcome, but it was ascertained by proceeding
with the synthetic scheme. We opted to elaborate the
thiophenyl handle of 15 as we anticipated facile late-stage
PMB deprotection. This sequence allowed us to isolate 19a as
a crystalline solid with the desired stereoconfiguration,
unfortunately as the minor isomer (Scheme 2). When we
varied the electronics of R and Ar groups systematically (see
the SI for additional experiments), the major isomer was the
Other Lewis acids generally proved to be inferior (entries 1−4
and 6−8). TiCl afforded a higher dr (entry 5), but we
4
reasoned the difficulty in workup renders it unsuitable for
scale-up purposes.
DFT calculations were undertaken to further understand the
factors governing the stereoselectivity of cyanation. We located
two transition states, TS1 and TS2 (Figure 2A). The distance
of 3.16 or 3.30 Å (Figure 2A) between the benzyl ether oxygen
B
DOI: 10.1021/acs.orglett.8b03219
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Second-Generation Synthetic Scheme
transformation is robust, scalable (up to 50 mmol), and may
find broader applications for the conversion of nitriles to
11,12
primary amides in complex molecules syntheses.
With 24
in hand, we proceeded to forge the cyclopentenone moiety.
Debenzylation was also problematic; the results were either
degradation of starting materials or difficult workups and
irreproducible yields on larger scales (see the SI for the
conditions tested). We found that the rarely employed
combination of BF ·OEt and EtSH afforded the desired
3
2
1
3
allylic alcohol 25 in 64% yield. Ring-closing metathesis
RCM) using nitro-Grela catalyst was uneventful, and the
14
(
ensuing oxidation could be performed in a one-pot fashion
using MnO to give 3 in 84%.
2
Preliminary experiments had revealed that the predominant,
undesired reactivity of enone 3 was conjugate addition,
presumably due to the ring strain of the cyclopentenone
moiety. Traditional olefination conditions (e.g., Wittig/
Horner−Wadsworth−Emmons, Reformatsky, Julia, Peterson)
resulted in either no conversion, conjugate addition, or
degradation of starting material under more forcing conditions.
We reasoned that a hard and sterically unhindered metal
acetylide might preferentially undergo 1,2-addition with 3
1
5
(
Scheme 4). Thus, we were able to synthesize the
Scheme 4. End-Game of Total Synthesis
propargylic alcohol 2 as a single diastereomer in 59% isolated
yield. The propargylic alcohol underwent a smooth Meyer−
Schuster rearrangement in excess MeOH to give the reported
structure (1a) and its Z isomer (1b), albeit with modest E/Z
16
selectivity (see the SI for a mechanistic rationale). More
conveniently, we also found that Ce-mediated Peterson
olefination of ketone 3 could directly afford 1a and 1b in
Figure 2. Calculated transition states and proposed stereochemical
model. Transition states were computed using M06-2x/6-311+
4
0% combined yield (unoptimized). To our knowledge, this is
+
G(d,p)/SMD(dichloromethane)//B3LYP/6-31G(d).
the first example of a direct olefination using an organocerium
15
enolate. The structure of 1a was unambiguously confirmed
and oxocarbenium suggests that it is a noncovalent interaction,
not a formal neighboring-group participation. We attribute
by X-ray analysis.
The H and C NMR spectra of both 1a and 1b did not
8
1
13
1
3
the difference in energies of the transition states to a
combination of electrostatics and sterics. In TS2, the benzyl
ether is further away from the oxocarbenium to accommodate
the incoming nucleophile, thereby weakening the electrostatic
interaction. The stereochemical model is reminiscent of the
reports by Woerpel et al. on alkoxy-directed/accelerated
match those reported for stresgenin B. In particular, C NMR
spectrum of 1a shows striking dissimilarities for the chemical
shifts of C2−C5, with differences ranging from 4.5 to 26.0 ppm
(see the SI). These differences cannot be accounted by mere
stereochemical mis-assignments (i.e., epimers, diastereomers).
Furthermore, the coupling constant reported for the olefin C−
H’s was 9.7 Hz, while those of 1a and 1b were 6.0 and 5.6 Hz,
respectively. Based on literature surveys and our observations
with disubstituted cyclopentene systems, the endocyclic olefin
has signature coupling constants of 5.0−6.0 Hz. These strongly
suggest the natural product contains a ring structure other than
10
nucleophilic additions to acetal-derived oxocarbeniums. In
our case, we note that this may be the first example of the use
of neighboring-group stabilization as a stereocontrolling factor
in a dioxolane derivative synthesis. The conversion of the
nitrile to primary amide 24 (Scheme 3) was met with
difficulties because of the dioxolane ring opening under widely
used reaction conditions, such as alkaline H O and Parkins’
17,18
a cyclopentene.
We first envisioned two possible constitutional isomers CS1
and CS2 (Table 2). Our calculated chemical shifts for C-2 to
2
2
1
1
19
catalyst, [PtH{(PMe O) H}(PMe OH)].
2
2
2
After some trials, we discovered that an anhydrous hydration
methodology developed by Chang and Lee using Wilkinson’s
catalyst, (Ph P) RhCl, provided trace amounts of the desired
C-5 suggested the plausibility of a cyclohexadiene motif.
Changing the position of the methoxy ester substituent
afforded better results for C-2 to C-5 with CS2 over CS1.
Therefore, we chose to focus on this constitutional isomer
CS2. Diastereomeric structure CS3 gave larger MAD and
3
3
1
2
primary amide 24. Screening of transition-metal catalysts
revealed that Pd(OAc) to be the catalyst of choice. This
2
C
DOI: 10.1021/acs.orglett.8b03219
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 2. Comparison of 13C NMR Data
In conclusion, we have accomplished the first total synthesis
of the reported structure of stresgenin B. The success of our
synthesis hinges upon a benzyloxy neighboring-group-con-
trolled diastereoselective cyanation of a cyclic oxocarbenium
ion. We found that by switching to an acyclic oxocarbenium,
the diastereoselectivity could be reversed. A notable feature of
our work is the presence of a free primary amide through the
late-stage of the synthesis. We also investigated plausible
structures of stresgenin B via computational methods.
ASSOCIATED CONTENT
Supporting Information
■
*
S
position
natural
CS1
CS2
CS3
CS4
CS5
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b03219.
O−CH3
52.2
164.5
130.0
136.2
124.4
129.1
79.2
80.9
22.1
109.8
172.1
51.2
166.1
123.3
138.7
120.4
139.8
74.0
68.4
21.6
102.1
169.3
5.0
51.6
164.2
126.7
137.6
122.8
125.7
69.8
71.2
21.2
101.8
170.1
3.7
51.4
164.5
125.9
140.4
126.1
125.2
70.0
73.6
20.3
104.3
168.2
3.9
51.4
164.1
129.0
141.8
131.0
131.2
78.6
80.4
20.7
110.8
175.9
2.2
51.3
163.8
129.0
141.2
124.4
131.6
78.9
80.1
21.1
110.9
168.3
1.6
C-1
C-2
C-3
C-4
C-5
C-6
C-7
C-3′
C-2′
Experimental procedures, computational details and
characterization data for new compounds (PDF)
Accession Codes
CCDC 1863016−1863019 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
C-1′ a
MAD
RMSD
b
6.2
5.1
4.7
3.0
2.2
a
b
MAD = mean absolute deviation. RMSD = root-mean-square
deviation. Calculated using GIAO/mPW1PW91/6-311+G(2d,p)/
AUTHOR INFORMATION
Corresponding Author
■
19
SMD(chloroform)//B3LYP/6-31+G(d,p).
*
E-mail: koide@pitt.edu.
ORCID
RMSD than CS2. The chemical shifts for C-6, C-7, and C-2′ of
CS2 and CS3 still exhibited large deviations.
Wei Chuen Chan: 0000-0003-0199-8960
Kazunori Koide: 0000-0001-8894-8485
Notes
We thus wondered whether the dioxolane ring could be
trans-fused to the cyclohexadiene (CS4 and CS5). We were
initially skeptical as we were not aware of any literature
precedence for this trans-fused 6/5-ring system. Despite the
lack of literature support, the calculated NMR data of CS5
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
1
■
showed the lowest MAD and RMSD ( H NMR: MAD = 0.10
This research was supported in part by the University of
Pittsburgh Center for Research Computing through the
resources provided and the US National Science Foundation
and RMSD = 0.12; see the SI). Moreover, we reasoned that a
trans-fused bicycle might be less susceptible to aromatization.
Akagawa et al. used the UV−vis spectrum of stresgenin B with
a peak at 275 nm as a basis for the presence of a fully
(
CHE-1506942). We thank Drs. Steven Geib, Godugu
3b
Bhaskar, and Damodaran Krishnan Achary for assistance
with X-ray crystallography, mass spectrometry, and NMR
spectroscopy analyses, respectively. We thank Michael P. Cook
conjugated dienoate. However, Figure 5 of the original
3
a
patent does not clearly show such a peak. The structure of
CS5 does not support their NOESY correlations. Specifically,
(
University of Pittsburgh) for preliminary synthetic studies.
the authors reported NOE correlation between ester CH and
3
We are also grateful to Prof. Paul E. Floreancig (University of
α-olefin C−H; they assigned this olefin proton to the peak at
Pittsburgh) for his insightful suggestions. Pd(OAc) was a
generous gift from Dr. Thomas J. Colacot (Johnson Matthey).
7
.43 ppmwe believe this to be a mis-assignmentthe peak
2
at 7.43 ppm must be due to an olefin C−H conjugated to the
ester. If CS5 is the correct structure, the reported correlation
REFERENCES
would refer to the ester CH and β-vinyl C−H (position 3,
3
■
7
.46 ppm). Nevertheless, the calculated average distance
(1) (a) Calderwood, S. K.; Gong, J. Trends Biochem. Sci. 2016, 41,
311−323. (b) Wu, J.; Liu, T.; Rios, Z.; Mei, Q.; Lin, X.; Cao, S.
Trends Pharmacol. Sci. 2017, 38, 226−256.
between ester CH and β-olefin C−H (conformer 1 of CS5)
3
is ∼4.27 Å, indicating that the NOE enhancement might be
relatively weak. In addition, the authors reported NOE
correlations between the dioxolane methyl group and the
peaks at 7.43 and 6.43 ppm; none of our candidate structures
can account for these correlations. At this point, without access
to the original NOESY experiment data, it is not possible to
determine whether these NOE enhancements are significant.
While our DFT calculations support CS5, other constitutional
isomers and diastereomers cannot be fully excluded.
(
2) (a) Butler, L. M.; Ferraldeschi, R.; Armstrong, H. K.; Centenera,
M. M.; Workman, P. Mol. Cancer Res. 2015, 13, 1445−1451.
(
b) Armstrong, H. K.; Koay, Y. C.; Irani, S.; Das, R.; Nassar, Z. D.;
The Australian Prostate CancerBioresource; Selth, L. A.; Centenera,
M. M.; McAlpine, S. R.; Butler, L. M. Prostate 2016, 76, 1546−1559.
(
3) (a) Akagawa, H.; Mizuno, S. Japanese Patent P2000-197498,
2000. (b) Akagawa, H.; Takano, Y.; Ishii, A.; Mizuno, S.; Izui, R.;
Sameshima, T.; Kawamura, N.; Dobashi, K.; Yoshioka, T. J. Antibiot.
1999, 52, 960−970.
D
DOI: 10.1021/acs.orglett.8b03219
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(
4) (a) Srihari, P.; Kumaraswamy, B.; Yadav, J. S. Tetrahedron 2009,
ppm. We observed similar overestimations when we tested the
method on the reported structure, compared to our experimental
spectra. We think there is a systematic error of overestimating carbons
conjugated (β and δ carbons) to an electron-withdrawing group .
65, 6304−6309. (b) Alcaraz, L.; Cridland, A.; Kinchin, E. Org. Lett.
2001, 3, 4051−4053. (c) Bode, J. W.; Carreira, E. M. J. Org. Chem.
2001, 66, 6410−6424.
(
5) (a) Wardrop, D. J.; Velter, A. I.; Forslund, R. E. Org. Lett. 2001,
3
2
, 2261−2264. (b) Wardrop, D. J.; Forslund, R. E. Tetrahedron Lett.
002, 43, 737−739. (c) Wardrop, D. J.; Forslund, R. E.; Landrie, C.
L.; Velter, A. I.; Wink, D.; Surve, B. Tetrahedron: Asymmetry 2003, 14,
29−940.
6) Liu, L.; Floreancig, P. E. Angew. Chem., Int. Ed. 2010, 49, 5894−
897.
7) (a) Hunter, C. A.; Sanders, J. K. M. J. Am. Chem. Soc. 1990, 112,
9
(
5
(
5
525−5534. (b) Cozzi, F.; Cinquini, M.; Annunziata, R.; Dwyer, T.;
Siegel, J. S. J. Am. Chem. Soc. 1992, 114, 5729−5733. (c) Ringer, A.
L.; Sinnokrot, M. O.; Lively, R. P.; Sherrill, C. D. Chem. - Eur. J. 2006,
1
2
1
2, 3821−3828. (d) Wheeler, S. E.; Houk, K. N. J. Am. Chem. Soc.
008, 130, 10854−10855. (e) Wheeler, S. E. J. Am. Chem. Soc. 2011,
33, 10262−10274. We favor the T-shaped configuration as we
believe it might be more conformationally compatible in our system.
(
(
8) Molander, G. A.; Haar, J. P. J. Am. Chem. Soc. 1993, 115, 40−49.
9) (a) Mukaiyama, T.; Murakami, M. Synthesis 1987, 1043−1054.
(
b) Hayashi, Y.; Wariishi, K.; Mukaiyama, T. Chem. Lett. 1987, 16,
1
243−1246.
10) (a) Lavinda, O.; Tran, V. T.; Woerpel, K. A. Org. Biomol. Chem.
014, 12, 7083−7091. (b) Garcia, A.; Otte, D. A.; Salamant, W. A.;
Sanzone, J. R.; Woerpel, K. A. J. Org. Chem. 2015, 80, 4470−4480.
c) Garcia, A.; Sanzone, J. R.; Woerpel, K. A. Angew. Chem., Int. Ed.
015, 54, 12087−12090.
11) For a review on the applications of Parkins’ catalyst, see:
a) Cadierno, V. Appl. Sci. 2015, 5, 380−401. (b) For a recent
application of Parkins’ catalyst (substoichiometric) in total synthesis,
see: Richter, M. J. R.; Schneider, M.; Brandstatter, M.; Krautwald, S.;
Carreira, E. M. J. Am. Chem. Soc. 2018, 140, 16704−16710.
12) (a) Lee, J.; Kim, M.; Chang, S. B.; Lee, H. Y. Org. Lett. 2009,
1, 5598−5601. (b) For a recent application of this method using
(
2
(
2
(
(
̈
(
1
Wilkinson’s catalyst in total synthesis, see: Kou, K. G. M.; Kulyk, S.;
Marth, C. J.; Lee, J. C.; Doering, N. A.; Li, B. X.; Gallego, G. M.;
Lebold, T. P.; Sarpong, R. J. Am. Chem. Soc. 2017, 139, 13882−13896.
(
13) Fuji, K.; Ichikawa, K.; Node, M.; Fujita, E. J. Org. Chem. 1979,
4
(
4, 1661−1664.
14) (a) Grela, K.; Harutyunyan, S.; Michrowska, A. Angew. Chem.,
Int. Ed. 2002, 41, 4038−4040. (b) Michrowska, A.; Bujok, R.;
Harutyunyan, S.; Sashuk, V.; Dolgonos, G.; Grela, K. J. Am. Chem. Soc.
2
(
1
004, 126, 9318−9325.
15) (a) Imamoto, T.; Sugiura, Y. J. Phys. Org. Chem. 1989, 2, 93−
02. (b) Imamoto, T.; Sugiura, Y. J. Organomet. Chem. 1985, 285,
C21−C23. (c) Imamoto, T.; Kusumoto, T.; Tawarayama, Y.; Sugiura,
Y.; Mita, T.; Hatanaka, Y.; Yokoyama, M. J. Org. Chem. 1984, 49,
3
904−3912. (d) Liu, H. J.; Shia, K. S.; Shang, X.; Zhu, B. Y.
Tetrahedron 1999, 55, 3803−3830.
16) (a) For a recent review, see: Engel, D. A.; Dudley, G. B. Org.
(
Biomol. Chem. 2009, 7, 4149−4158. (b) Engel, D. A.; Lopez, S. S.;
Dudley, G. B. Tetrahedron 2008, 64, 6988−6996. We subjected the
isolated product (1) to standard reaction conditions; we observed no
transesterification. We concluded that the methoxy exchange occurred
prior to the formation of the ester product, in agreement with ref 16b.
See the Supporting Information for a mechanistic rationale of the
alkoxy exchange.
(
17) For a review on the role of total synthesis in structural revision,
see: Nicolaou, K. C.; Snyder, S. A. Angew. Chem., Int. Ed. 2005, 44,
012−1044.
18) For a structurally related natural product, see: Zhu, B.;
1
(
Morioka, M.; Nakamura, H.; Naganawa, H.; Muraoka, Y.; Okami, Y.;
Umezawa, H. J. Antibiot. 1984, 37, 673−674.
(
19) Lodewyk, M. W.; Siebert, M. R.; Tantillo, D. J. Chem. Rev.
2
012, 112, 1839−1862 We note that for constitutional isomer CS2
3
there are 2 stereoisomers, four of which are NMR-distinguishable
diastereomers, corresponding to CS2−CS5. We also note that for
CS5 the chemical shift for C-3 still has an absolute deviation of 5.0
E
DOI: 10.1021/acs.orglett.8b03219
Org. Lett. XXXX, XXX, XXX−XXX