A Total Synthesis of Paeoveitol

Article abstract of DOI:10.1021/acs.orglett.6b01736

A four-step total synthesis of paeoveitol (1), a recently disclosed norditerpene natural product from Paeonia vetchii, is reported. This highly concise synthetic route was guided by biosynthetic considerations and enabled by an unusual intermolecular orth

Full text of DOI:10.1021/acs.orglett.6b01736

Letter
pubs.acs.org/OrgLett
A Total Synthesis of Paeoveitol
Lun Xu,† Fengyi Liu,† Li-Wen Xu, Ziwei Gao, and Yu-Ming Zhao*
Key Laboratory of Applied Surface and Colloid Chemistry of MOE & School of Chemistry and Chemical Engineering Shaanxi
Normal University, 620 West Chang’an Ave, Xi’an, 710119, China
S
* Supporting Information
ABSTRACT: A four-step total synthesis of paeoveitol (1), a recently disclosed norditerpene natural product from Paeonia
vetchii, is reported. This highly concise synthetic route was guided by biosynthetic considerations and enabled by an unusual
intermolecular ortho-quinone methide [4 + 2]-cycloaddition reaction, which proceeded with excellent regio- and
diastereoselectivity. Density functional theory (DFT) calculations point to a crucial intermolecular hydrogen bond and π−π
stacking interaction that govern selectivity in this process.
he plant Paeonia veitchii has been used in many traditional
systemic biological screening. Further biological evaluation will
become possible if greater quantities of 1 can be procured
through synthesis.
T
Asian medicines for hundreds of years.1 The root of this
plant has been shown to have intriguing biological activity,
including immunomodulation, anti-inflammation properties,
and hypoglycemic action, among others.2 In 2014, Chen and
co-workers isolated a unique norditerpene natural product
paeoveitol (1) from the root of Paeonia veitchii in racemic
form.3 In addition, five new natural products, paeoveitols A−E,
were also isolated from Paeonia veitchii a year later (Figure 1).4
Structurally, paeoveitol (1) features a 6/5/6/6-fused
tetracyclic framework with three contiguous stereogenic centers
and two free phenol subunits, rendering it a nontrivial synthetic
target. Herein, we report a four-step, protecting group-free total
synthesis of 1 that can be conducted on multigram scale. An
unusual, yet highly selective, ortho-quinone methide (o-QM) [4
+ 2]-cycloaddition reaction employing benzofuran alcohol as a
coupling partner serves as the hallmark bond-forming step in
this synthetic pathway.
Given the fact that many complex yet racemic natural
products are often associated with nonenzymatic pericyclic
reactions,5 as well as the structural similarity between 1 and 2,
we speculated that 1 might be formed in nature from 2 via the
concerted o-QM [4 + 2]-cycloaddition reaction shown
(Scheme 1).6 Executing this potentially biomimetic cyclo-
addition, however, requires addressing several challenges in
chemo- and regioselectivity. As a reactive intermediate, o-QM
can also react with the nucleophilic hydroxyl group of 2 to form
an ether linkage.7 In addition, the regioselectivity of an
intermolecular concerted [4 + 2]-cycloaddition reaction
between an o-QM and 2 has two chemically feasible
possibilities given the nucleophilicity of both the 2- and 3-
positions of a benzofuran. To the best of our knowledge,
currently disclosed intermolecular [4 + 2]-cycloaddition
reactions between o-QMs and electron-rich enol ethers only
Figure 1. Paeoveitol and related natural products from Paeonia veitchii.
The structure of 1 was determined using X-ray crystallographic
analysis, and a biosynthetic route involving two molecules of
the chiral monoterpene paeoniflorin was also postulated by
Chen and co-workers. Preliminary biological evaluation only
studied the antidepressant abilities of 1 and showed no agitating
activity.3 However, the poor accessibility of this material (1.6
mg per 5 kg of dried plant roots) severely limited further and
Received: June 15, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b01736
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 1. Nature-Inspired Synthetic Design
Table 1. Desired [4 + 2]-Cycloaddition: Selected
Optimization
a
b
entry
conditions
yield (%)
1
2
3
4
5
6
7
8
Et2AlCl (0.2 equiv), CH2Cl2, −78 °C → rt
TiCl4 (0.2 equiv), CH2Cl2, −78 °C → rt
SnCl4 (0.2 equiv), CH2Cl2, −78 °C → rt
ZnBr2 (0.2 equiv), CH2Cl2, rt
43
32
<5
51
86
ZnCl2 (0.2 equiv), CH2Cl2, rt
c
p-TsOH (0.2 equiv), CH2Cl2, rt
ZnCl2 (0.2 equiv), CH2Cl2, rt
42
afforded α-oxygen-substituent benzannulated tetrahydropyran
analogues, which if extrapolated to the system at hand,
corresponds to the undesired regioisomer.8
Seeking to evaluate the aforementioned strategy, we initially
conducted a model study for the proposed intermolecular [4 +
2]-cycloaddition reaction (Scheme 2). Using the method
d
87
ZnCl2 (0.1 equiv), CH2Cl2, rt
84
a
Conditions: 5 (0.20 mmol), 6 (0.24 mmol), acid catalyst, solvent (0.1
b
c
d
M). Isolated yield. p-TsOH·H2O was used. Conducted with 4 Å
molecular sieves (30 mg) as an additive.
62% yield.11 Following a similar route as the model study,
paeoveitol D 2 was synthesized in three-steps and on gram-
scale (37% overall yield from inexpensive methylhydroqui-
none). With building block 2 in hand, we then focused on
coupling partner 3. Ketone 13 was easily prepared according to
the reported procedure in two steps.12 Reduction of 13 with
BH3 in THF afforded desired alcohol 3 in 86% yield. In our
hands, it was found that 3 was somewhat unstable in air and
difficult to purify on a larger scale. Thus, the crude reduction
product from 13 was used directly in the key intermolecular [4
+ 2]-cycloaddition reaction without further purification. To our
delight, the desired cycloaddition product paeoveitol 1 was
obtained in 66% yield as a single diastereomer. After a number
of attempts, single crystals of 1 could be obtained from an
acetone/aqueous ammonia solvent mixture. Utilizing this four-
step synthetic sequence, over 2 g of paeoveitol 1 has been
prepared to date.
For probing the high selectivity of the intermolecular
reaction between o-QM and benzofuran alcohols, with an
emphasis on understanding why only one isomer was afforded
in this reaction, density functional theory (DFT) calculations
with M06-2X functional and polarized triple-ζ 6-311+G(d,p)
basis sets were performed (Figure 2, more stable trans-o-QM
(I) was used here as a starting point to discuss the reaction; for
the cis-o-QM case, see the Supporting Information). Intrinsic
reaction coordinate (IRC) calculations were used to confirm
the connection between the reactant, product, and transition
state. The polarizable continuum model (PCM) was employed
to consider the solvent (CH2Cl2) contribution to both energy
and geometry.
Scheme 2. Model Study for the Desired [4 + 2]-
Cycloaddition Reaction
reported by Hossain and co-workers,9 benzofuran ester 4
could be prepared easily from salicylaldehyde and ethyl
diazoacetate in 90% yield. Reduction of both 4 and 2′-
hydroxyacetophenone (not shown) afforded the desired model
alcohols 5 and 6 and set the stage for the key coupling event.
After significant experimentation, it was discovered that
reacting 5 and 6 with zinc chloride in CH2Cl2 produced the
desired cycloaddition product (7) in 86% yield and as a single
isomer. To our delight, X-ray crystallographic analysis of tosyl-
protected 7 confirmed its relative configuration, which was
identical to that found in paeoveitol 1. High yields of 7 only
became possible after exploring a number of conditions and
Lewis acid promoters (Table 1). Although initial evaluation
showed that Lewis acids based on Al, Ti, and Sn were
competent (Table 1, entries 1−3), ZnCl2 exhibited the best
performance (Table 1, entry 5). Further optimization indicated
that the use of 4 Å molecular sieves and a lower catalyst loading
did not affect the efficiency (Table 1, entries 7 and 8).10 It is
noteworthy that the oxa-Michael addition product 8 (Scheme
2) and any other diastereomers were not observed under any of
these conditions.
Starting with the reactant benzofuran alcohol 5 and trans-o-
QM (I), the reaction processes through either an oxa-Michael
addition or a concerted [4 + 2]-cycloaddition pathway. In the
former case, the benzylidene carbon in I and hydroxyl oxygen
in 5 approach each other in two different orientations to reach
two transition states TS0 and TS0′ with activation free energies
of 20.1 and 21.6 kcal/mol. Once the barrier was conquered, the
oxa-Michael addition products MP0 and MP0′ are finally
generated with 12.7 and 12.4 kcal/mol of exergonicity,
respectively. Compared to the oxa-Michael addition pathway,
the concerted [4 + 2]-cycloaddition pathway is more complex
due to the regioselectivity as well as the endo/exo selectivity. In
With the success of our model study, we returned to the total
synthesis of 1 (Scheme 3). Direct Vilsmeier−Haack for-
mylation of commercially available methylhydroquinone, which
was demonstrated to be difficult previously, produced 11 in
B
DOI: 10.1021/acs.orglett.6b01736
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Four Step Total Synthesis of Paeoveitol (1)
the validation of computational results. Therefore, the
concerted [4 + 2]-cycloaddition reaction (more specifically,
via endo-TS4) is the most favorable pathway both kinetically
and thermodynamically.
In conclusion, we have achieved the first total synthesis of
unique norditerpene natural product paeoveitol (1) with a high
level of efficiency and scalability. The synthesis featured a Lewis
acid-catalyzed biomimetic intermolecular [4 + 2]-cycloaddition
reaction. The selectivities in the key step of the synthesis have
also been studied by DFT calculations. Further application of
the strategy reported herein are underway and will be reported
in due course.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications Web site. (PDF) and 1 (CIF) The
Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.6b01736.
Experimental procedures, spectroscopic data, and images
1
of H and 13C NMR spectra for all new compounds
(PDF)
Single crystal X-ray data for compound 9 (CIF)
Single crystal X-ray data for compound 1 (CIF)
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: ymzhao@snnu.edu.cn
Author Contributions
Figure 2. DFT computed energy surface and transition state
structures.
†L.X. and F.L. contributed equally to this work.
Notes
total, four reach paths are found, including two endo- and two
exo-type additions. Among them, the energetically most
favorable channel corresponds to transition state endo-TS4,
which is stabilized by both a hydrogen bond and π−π stacking
of the benzyl moieties. The free energy barrier height with
respect to reactant complex RC is 15.7 kcal/mol, which is lower
than that of the most feasible oxa-Michael addition channel
(TS0, ΔG⧧ = 20.1 kcal/mol). Furthermore, the product of
endo-TS4 channel, endo-P4, is also lower in free energy than
oxa-Michael addition product by ∼10 kcal/mol and is actually
thermodynamically the most favorable product. The M06-2X
predicted structure for endo-P4 is in good agreement with the
experimentally obtained crystal structure, further confirming
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the National Science Foundation
of China (21202074, 21473107), Shaanxi Provincial Natural
Science Foundation (2016JQ2024), Fundamental Research
Funds for the Central Universities (GK201603046), and
generous start-up funds from Shaanxi Normal University. We
thank Dr. Thomas J. Maimone at UC Berkeley for helpful
discussions during the preparation of this manuscript. Dr.
Dong-Xu Xue at Shaanxi Normal University is acknowledged
for X-ray crystallographic analysis.
C
DOI: 10.1021/acs.orglett.6b01736
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
REFERENCES
■
(1) (a) Kaneda, M.; Iitaka, Y.; Shibata, S. Tetrahedron 1972, 28, 4309.
(b) Ho, T. N. Flora of China, 56; Science Press: Beijing, 1979. (c) Lin,
H. C.; Ding, H. Y.; Wu, T. S.; Wu, P. L. Phytochemistry 1996, 41, 237.
(d) Kostova, I. N.; Simeonov, M. F.; Todorova, D. I. Phytochemistry
1998, 47, 1303. (e) Okasaka, M.; Kashiwada, Y.; Kodzhimatov, O. K.;
Ashurmetov, O.; Takaishi, Y. Phytochemistry 2008, 69, 1767.
(2) (a) Zhang, A. P.; Chen, M. Z.; Xu, S. Y. Chin. Pharm. Bull. 1993,
9, 454. (b) Hsu, F. L.; Lai, C. W.; Cheng, J. T. Planta Med. 1997, 63,
323. (c) Ruan, J. L.; Zhao, Z. X.; Zeng, Q. Z.; Qian, Z. M. Chin. Pharm.
Bull. 2003, 19, 965. (d) Wu, S. H.; Yang, S. M.; Wu, D. G.; Cheng, Y.
W.; Peng, Q. Helv. Chim. Acta 2005, 88, 259. (e) Mao, Q. Q.; Ip, S. P.;
Tsai, S. H.; Che, C. T. J. Ethnopharmacol. 2008, 119, 272. (f) Cui, G.
Z. World Phytomed. 2009, 24, 231. (g) Duan, W. J.; Yang, J. Y.; Chen,
L. X.; Zhang, L. J.; Jiang, Z. H.; Cai, X. D.; Zhang, X.; Qiu, F. J. Nat.
Prod. 2009, 72, 1579. (h) Ji, L. X.; Huang, H.; Li, C. Z.; Jiang, M.; Luo,
G. A. Drug Eval. Res. 2010, 33, 233. (i) He, D. Y.; Dai, S. M. Front.
Pharmacol. 2011, 2, 10.
(3) Liang, W. J.; Geng, C. A.; Zhang, X. M.; Chen, H.; Yang, C. Y.;
Rong, G. Q.; Zhao, Y.; Xu, H. B.; Wang, H.; Zhou, N. J.; Ma, Y. B.;
Huang, X. Y.; Chen, J. J. Org. Lett. 2014, 16, 424.
(4) Liang, W. J.; Ma, Y. B.; Geng, C. A.; Huang, X. Y.; Xu, H. B.;
Zhang, X. M.; Chen, J. J. Fitoterapia 2015, 106, 36.
(5) For select examples utilizing pericyclic reaction as key steps in the
syntheses of racemic natural products, see: (a) Chapman, O. L.; Engel,
M. R.; Springer, J. P.; Clardy, J. C. J. Am. Chem. Soc. 1971, 93, 6696.
(b) Nicolaou, K. C.; Petasis, N. A.; Zipkin, R. E.; Uenishi, J. J. Am.
Chem. Soc. 1982, 104, 5555. (c) Nicolaou, K. C.; Petasis, N. A.;
Uenishi, J.; Zipkin, R. E. J. Am. Chem. Soc. 1982, 104, 5557.
(d) Nicolaou, K. C.; Zipkin, R. E.; Petasis, N. A. J. Am. Chem. Soc.
1982, 104, 5558. (e) Nicolaou, K. C.; Petasis, N. A.; Zipkin, R. E. J.
Am. Chem. Soc. 1982, 104, 5560. (f) Drew, S. L.; Lawrence, A. L.;
Sherburn, M. S. Angew. Chem., Int. Ed. 2013, 52, 4221. (g) Strych, S.;
Journot, G.; Pemberton, R. P.; Wang, S. C.; Tantillo, D. J.; Trauner, D.
Angew. Chem., Int. Ed. 2015, 54, 5079.
(6) For a recent review of o-QM, see: (a) Van De Water, R. W.;
Pettus, T. R. R. Tetrahedron 2002, 58, 5367. (b) Pathak, T. P.; Sigman,
M. S. J. Org. Chem. 2011, 76, 9210. (c) Willis, N. J.; Bray, C. D. Chem. -
Eur. J. 2012, 18, 9160. (d) Singh, M. S.; Nagaraju, A.; Anand, N.;
Chowdhury, S. RSC Adv. 2014, 4, 55924. (e) Bai, W. J.; David, J. G.;
Feng, Z. G.; Weaver, M. G.; Wu, K. L.; Pettus, T. R. Acc. Chem. Res.
2014, 47, 3655. (f) Caruana, L.; Fochi, M.; Bernardi, L. Molecules
2015, 20, 11733. (g) Wang, Z.; Sun, J. Synthesis 2015, 47, 3629.
(7) For an example, see: (a) Lai, Z.; Wang, Z.; Sun, J. Org. Lett. 2015,
17, 6058. (b) Jensen, K. H.; Pathak, T. P.; Zhang, Y.; Sigman, M. S. J.
Am. Chem. Soc. 2009, 131, 17074. (c) Zhang, Y.; Sigman, M. S. J. Am.
Chem. Soc. 2007, 129, 3076. (d) Schultz, M. J.; Sigman, M. S. J. Am.
Chem. Soc. 2006, 128, 1460. (e) Gligorich, K. M.; Schultz, M. J.;
Sigman, M. S. J. Am. Chem. Soc. 2006, 128, 2794.
(8) For a recent application of the intramolecular reaction of this
chemistry in total synthesis, see: Lumb, J. P.; Choong, K. C.; Trauner,
D. J. Am. Chem. Soc. 2008, 130, 9230.
(9) Dudley, M. E.; Morshed, M. M.; Hossain, M. M. Synthesis 2006,
10, 1711.
(10) Preliminary screening with some chiral ligands gives low
enantioselectivity (<10% ee).
(11) Mehta, G.; Khan, T. B.; Sunil Kumar, Y. C. Tetrahedron Lett.
2010, 51, 5116.
(12) Macías, F. A.; Marin, D.; Chinchilla, D.; Molinillo, J. M. G.
Tetrahedron Lett. 2002, 43, 6417.
D
DOI: 10.1021/acs.orglett.6b01736
Org. Lett. XXXX, XXX, XXX−XXX