Biomimetic Total Synthesis of Paeoveitol

Article abstract of DOI:10.1021/acs.orglett.6b02228

A highly stereocontrolled synthesis of paeoveitol has been developed in 26% yield, in 7 steps from commercially available materials. The synthetic strategy was inspired primarily by the biogenetic hypothesis and was enabled by hetero-Diels-Alder cycloaddi

Full text of DOI:10.1021/acs.orglett.6b02228

Letter
pubs.acs.org/OrgLett
Biomimetic Total Synthesis of Paeoveitol
Yuhan Zhang, Yonghong Guo, Zhongle Li, and Zhixiang Xie*
State Key Laboratory of Applied Organic Chemistry & College of Chemistry and Chemical Engineering, Lanzhou University,
Lanzhou 730000, P. R. China
S
* Supporting Information
ABSTRACT: A highly stereocontrolled synthesis of paeovei-
tol has been developed in 26% yield, in 7 steps from
commercially available materials. The synthetic strategy was
inspired primarily by the biogenetic hypothesis and was
enabled by hetero-Diels−Alder cycloaddition to construct the
target molecular framework.
he root of Paeonia veitchii (Chuan-Chi-Shao) has a long
history of use as the composition in many traditional
and the structural novelty prompted us to undertake synthetic
studies of paeoveitol. Herein, we report a biomimetic total
synthesis of paeoveitol, which was enabled by hetero-Diels−
Alder cycloaddition. Very recently, Zhao and co-workers
reported the first total synthesis of paeoveitol employing a
similar hetero-Diels−Alder cycloaddition.4
T
medicine formulas (for example Sini-San and Xiaoyao-Wan) for
the treatment of depression-like disorders in China.1 Its genus,
Paeonia (Paeoniaceae), is gaining interest in the medicinal
chemistry and drug discovery community. As the characteristic
components of the genus Paeonia, Paeoniflorin-related
monoterpene glycosides with a cagelike unit are rich and
generally recognized as the active constituents including
activities in immunomodulation, anti-inflammation, hypoglyce-
mic action, antidepression, and circadian regulation.2 The
monoterpene, diterpene, and triterpenoids are also isolated
from this genus. Paeoveitol, a pair of norditerpene enantiomers,
were isolated from the root of Paeonia veitchii in 2014 by Chen
et al.3 (Figure 1). Extensive NMR spectroscopic analyses and
In the isolation paper, Chen and co-workers postulated that
paeoveitol might biosynthetically be derived from two
molecules of paeoniflorin.3 The intermediates were assumed
to be obtained by hydrolysis of glucosyl, loss of benzoyl groups,
cleavage of the hemiketal−acetal linkage, and the four-
membered ring from paeoniflorin. The structure of paeoveitol
was then established via a series of dehydration, cyclization,
oxidation, and decarboxylation derived from the intermediates.3
However, the structure of paeoniflorin is thought to be more
complex than the intermediates. Generally, most biogenetic
pathways always approached the natural products from simple
molecules.5 Based on this realization, the biogenesis precursor
of paeoveitol may not be the paeoniflorin. Recently, Chen and
co-workers have isolated five monoterpene paeoveitols A−E
from the same genus plant6 (Figure 1). Inspired by the finding
that the benzofuran moiety of paeoveitol is paeoveitol D, which
was isolated from the same species, we deduced that paeoveitol
D would be one of the precursors of paeoveitol. With one
partner (paeoveitol D) in mind, the target molecule
(paeoveitol) should be the mergence of the paeoveitol D and
the other partner by hetero-Diels−Alder reaction (Scheme 1).
The other partner would be easily transformed to ortho-
quinone methides (o-QMs) via acid catalysis, base catalysis,
oxidation, thermolysis, photolysis, or tautomerization.7 Thus,
paeoveitol A was assumed to be the other precursor partner of
paeoveitol. There existed two basic assumptions: one was that
the paeoveitol A directly transformed to the o-QM intermediate
via a series of dehydration, oxidation, and decarboxylation
(Scheme 1, path a); the other was that the paeoveitol A was
transformed to paeoniflorin A at first8 and then to the o-QM
intermediate in the next step (Scheme 1, path b). Technically,
Figure 1. Paeoveitol, paeoniflorin, and paeoveitol A−E.
X-ray crystallographic analysis of this molecule revealed that its
structure was featured by an unprecedented [3, 2-b] fused 2, 3-
dihydrobenzofuran and chromane in a tetracyclic framework
with three contiguous stereogenic centers. After being
separated by chiral HPLC, the absolute configurations of
(+)-paeoveitol and (−)-paeoveitol were determined on the
basis of electronic circular dichroism (ECD).
Unfortunately, the bioactivity of paeoveitol was not reported
in detail due to the limited supply in the original paper that
isolated it. The combination of the uncharacterized bioactivity
Received: July 28, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b02228
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
situ followed by addition of the dienophile to form benzopyrane
in high yields and excellent stereoselectivities.14 This strategy
happened to coincide with the idea of asymmetric synthesis of
paeoveitol. Reaction setups were referred to by Schneider.14a
However, it did not work (Scheme 3, condition A). The reason
Scheme 1. Biogenetic Pathway Proposed for Paeoveitol
Scheme 3. Model Reaction between Benzofuran and
Compound 3
paeoniflorin A has a very similar structure with the o-QM
intermediate and successfully been isolated from Paeonia
suffruticosa,8 belonging to the same genus of paeoveitol.
Thus, the paeoveitol A or the related compounds that might be
transformed to o-QMs, together with paeoveitol D, were used
to build paeoveitol via hetero-Diels−Alder reaction (Scheme
1).
Hetero-Diels−Alder reaction of o-QMs has been proven to
be an efficient method for the preparation of a wide variety of
chromane motifs (benzopyrans) based on the dienophile with
high chemo-, regio-, diastereo-, and enantioselectivity.7,9 This
powerful reaction has also been used in linchpin reactions for
the construction of complex natural products.10 As shown in
Scheme 1, we envisioned that if the dienophile simple natural
product paeoveitol D and o-QM could be merged together by
hetero-Diels−Alder reaction, a short and biomimetic synthesis
of paeoveitol could be realized. However, benzofuran as the
dienophile occurring in a hetero-Diels−Alder reaction is rarely
reported. Pioneering work for benzofuran as the dienophile
with o-naphthoquinone methides was reported by Sajiki and co-
workers.11 There are two challenges in this reaction: first, the
paeoveitol D has a benzofuran motif, which is inactive due to
aromaticity; second, the regioselectivity is unpredictable in the
hetero-Diels−Alder reaction when using the benzofuran motif
as the dienophile to react with o-QM.
might be that the benzofuran is an aromatic ring and inactive as
a dienophile. In order to make the reaction occur, the reaction
temperature was elevated. After careful consideration, dichloro-
methane as solvent is not suitable for increasing the
temperature of reaction. The solvent was changed from
dichloromethane to toluene. Using toluene as solvent with
10% BINOL-based phosphoric acid catalyst, the cycloaddition
product 5 was obtained as a single diastereoisomer in good
yield with excellent regio- and diastereoselectivity at 60 °C.
However, the enantioselectivity is poor. The relative config-
uration of 5 was determined on the nuclear overhauser effect
(NOE) and confirmed by transforming 5 to compound 5′,
which was determined by X-ray crystallography analysis. The
success of the model study made us more confident to pursue
further exploration.
Efficient methods for o-QM intermediate generation from 2-
(hydroxymethyl)phenol or o-methylene-acetoxy-phenols have
been developed and applied.12 Based on these methodologies,
an o-QM precursor was designed and prepared. As illustrated in
Scheme 2, the synthesis of the precursor of o-QM was started
With the success of our model study, our efforts were
focused on synthesis of the paeoveitol D (Scheme 4). The
Scheme 4. Synthesis of Paeoveitol D
Scheme 2. Synthesis the Precursor of the o-Quinone
Methide
from commercially available 2-methylhydroquinone 1. The
compound 2 was prepared in 98% over two steps using the
known method.13 To avoid forming p-quinone, compound 2
was selectively protected by acetate and then reduced the
ketone to the corresponding alcohol by NaBH4 to obtain the
compound 3.
With the precursor of the o-QM 3 in hand, we started our
initial investigation to utilize benzofuran 4 and compound 3 as
a model reaction. According to the literature, BINOL-based
phosphoric acids as a Brønsted acid can catalyze ortho-hydroxy
benzhydryl alcohol transferred to o-QMs which generated in
synthesis commenced from the commercially available p-
toluquinone 6. Treatment of p-toluquinone 6 with enamine
7, which was prepared from propionaldehyde and morpholine,
provided the addition products 8 and 8′ in a red color. The
compound 8 and 8′ were directly used without further
purification. After evaporation of the dichloromethane, the
residue was refluxed in THF in the presence of diluted
hydrochloric acid. Compounds 9 and 10 were obtained in 21%
and 72% yield, respectively.15 Selective allylic oxidation of
B
DOI: 10.1021/acs.orglett.6b02228
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
methyl at the furan of 10 was performed with 1.2 equiv of SeO2
in 1,4-dioxane under reflux conditions. This attempt resulted in
a complex mixture.16 Presumably this result was due to the free
phenol function group, which could be oxidized under the same
conditions.17 Notably, when acetyl derivative 11 was performed
under the same conditions, only two components were
achieved: the corresponding allylic alcohol and allylic aldehyde.
Fortunately when the crude mixture was treated with NaBH4
and K2CO3 via reduction and deacetylization, paeoveitol D was
obtained as a single product.
both depending on the reaction conditions. As shown in
Scheme 5, compound 15 was the sole product at room
temperature (∼25 °C) and compound 16 was the sole product
at 120 °C. By running the reaction at 80 °C, both compound
15 and 16 were obtained in an about 1:1 ratio, respectively.
Notably, the compound 15 may convert to thermodynamically
more stable 16 via a [2,3]-sigmatropic-type rearrangement
under the same conditions, and without BINOL-based
phosphoric acid, the compound 15 is stable even at 120 °C.
As this kind of reaction did not work, we gave up this
methodology and sought other possible solutions. The o-
naphthoquinone methides which are formed from FeCl3-
catalyzed ring opening through a Diels−Alder reaction between
benzynes and furans provided the highly functionalized
naphthalene derivatives.11 However, using FeCl3 as the Lewis
acid catalyst, the hetero-Diels−Alder reaction of paeoveitol D
with compound 3 caused a complex product and no main
product. When paeoveitol D was replaced with compound 10,
the reaction did not work. Unfortunately, we did not try other
Lewis acids; very recently Zhao reported ZnCl2 as catalyst
successfully promoted cycloaddition which also is a Lewis acid.4
Following Zhao’s procedure, we obtained the desired cyclo-
addition products in 65% yield between 3 and paeoveitol D, so
the failure of the approaches is due to the wrong Lewis acid
promoter.
With paeoveitol D in hand, the biomimetic synthesis of
paeoveitol was investigated via hetero-Diels−Alder reaction
(Scheme 5). Using model study conditions, paeoveitol D was
Scheme 5. Attempt to Synthesize Paeoveitol via BINOL-
Based Phosphoric Acid Catalyst
The final [4 + 2]-cycloaddition turned out to be the most
difficult task. FeCl3 as the Lewis acid and BINOL-based
phosphoric acid as the Brønsted acid catalyzed o-QMs, which
generated in situ, would not react with paeoveitol D to obtain
the desired [4 + 2]-cycloaddition product. Those results can be
understood as the BINOL-based phosphoric acid catalyzed to
produce the o-QMs in the zwitterionic aromatic resonance
structures rather than in neutral molecules.9 Thermolysis, the
most common method for o-QMs generation in neutral
molecules, led to the elimination of a stable molecule, such
as water, acetic acid, etc.20 Our group has found a series of
efficient methods of o-QMs generation for the syntheses of
racemic spiroketals thermally.21 Using compound 3 as the
precursor of o-QM, the result is unsatisfactory. We then turned
to prepare o-QM via elimination of acetic acid. Compound 17
was designed and prepared for the precursor of o-QM. Using
acetic anhydride in the presence of BF3·Et2O, the alcoholic
hydroxyl groups of compound 3 were selectively acetylated to
afford compound 17 (Scheme 6).22 Following our procedure,21
the paeoveitol D and compound 17 reacted in analytical
toluene at 130 °C in a sealed tube, giving the [4 + 2]-
cycloaddition product 18 as a single diastereoisomer in 55%
yield. The final step of deprotection was achieved by adding
K2CO3 in MeOH, from which we successfully obtained the
treated with compound 3 in the presence of a BINOL-based
phosphoric acid catalyst. Compound 12 was obtained as a
single product in 50% yield. The formation of 12 suggested that
the conjugated addition of the benzofuran to the o-QM,
followed by rearomatization of the benzofuran, dehydration,
and trapping of the resulting cationic species by the phenol
oxygen (in a SN2-like reaction). In order to resist the
dehydration, compound 10 was replaced with paeoveitol D.
However, it still did not work. Only compound 13 was
produced to undergo conjugate addition to the o-QM and
rearomatization in good yield. Presumably compound 5 derives
from the stepwise electrophilic addition of a resonance-
stabilized benzylic carbocation to nucleophilic benzofuran,
followed by capture of the benzylic carbocation intermediate by
a phenoxide ion.18 In those cases, the C−O bond is not formed
due to the electron-withdrawing group factors of acetate, which
may decrease the nucleophilicity of the phenol (phenoxide
ion).19 Elimination of the electron-withdrawing mesomeric
effect on the acetate fragment of the precursor 3 was expected
to increase the nucleophilicity of the phenol, thus encouraging
the key acid mediated cyclization to form the benzopyran ring.
So we changed the −Ac group to the more electron-donating
triethylsilyl group for further investigation. The compound 14
was prepared from compound 2 via TES protection and
reduction. And our results revealed that the reaction between
paeoveitol D and compound 14 produced compound 15, 16, or
Scheme 6. Total Synthesis of Paeoveitol
C
DOI: 10.1021/acs.orglett.6b02228
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
1
paeoveitol. The H and 13C NMR spectra were in agreement
(11) Sawama, Y.; Kawajiri, T.; Asai, S.; Yasukawa, N.; Shishido, Y.;
Monguchi, Y.; Sajiki, H. J. Org. Chem. 2015, 80, 5556−5565.
(12) For recent references, see: (a) Hsiao, C. C.; Liao, H. H.;
Rueping, M. Angew. Chem., Int. Ed. 2014, 53, 13258−13263. (b) Saha,
S.; Schneider, C. Chem. - Eur. J. 2015, 21, 2348−2352. (c) Alamsetti, S.
K.; Spanka, M.; Schneider, C. Angew. Chem., Int. Ed. 2016, 55, 2392−
2396.
with literature data.3
In summary, we achieved total synthesis of the new
norditerpene paeoveitol in 7 steps in 26% yield. The key step
in the total synthesis is the final hetero-Diels−Alder cyclo-
addition to construct the target molecular framework.
(13) (a) Macías, F. A.; Marín, D.; Chinchilla, D.; Molinillo, J. M. G.
Tetrahedron Lett. 2002, 43, 6417−6420. (b) Macías, F. A.; Chinchilla,
D.; Molinillo, J. M. G.; Marín, D.; Varela, R. M.; Torres, A. Tetrahedron
2003, 59, 1679−1683.
(14) (a) El-Sepelgy, O.; Haseloff, S.; Alamsetti, S. K.; Schneider, C.
Angew. Chem., Int. Ed. 2014, 53, 7923−7927. (b) Saha, S.; Schneider,
C. Org. Lett. 2015, 17, 648−651. (c) Zhao, J. J.; Sun, S. B.; He, S. H.;
Wu, Q.; Shi, F. Angew. Chem., Int. Ed. 2015, 54, 5460−5464.
(15) Bian, J. L.; Deng, B.; Xu, L. L.; Xu, X. L.; Wang, N.; Huc, T. H.;
Yao, Z. Y.; Du, J. Y.; Yang, L.; Lei, Y. H.; Li, X.; Sun, H. P.; Zhang, X.
J.; You, Q. D. Eur. J. Med. Chem. 2014, 82, 56−67.
(16) Paul, N. M.; Taylor, M.; Kumar, R.; Deschamps, J. R.; Luedtke,
R. R.; Newman, A. H. J. Med. Chem. 2008, 51, 6095−6109.
(17) King, T. J.; Read, G. J. Chem. Soc. 1961, 5090−5094.
(18) (a) Marsini, M. A.; Huang, Y. D.; Lindsey, C. C.; Wu, K.-L.;
Pettus, T. R. R. Org. Lett. 2008, 10, 1477−1480. (b) Toteva, M. M.;
Richard, J. P. Adv. Phys. Org. Chem. 2011, 45, 39−91.
(19) Rathwell, D. C. K.; Yang, S. H.; Tsang, K. Y.; Brimble, M. A.
Angew. Chem., Int. Ed. 2009, 48, 7996−8000.
(20) (a) Pettigrew, J. D.; Freeman, R. P.; Wilson, P. D. Can. J. Chem.
2004, 82, 1640−1648. (b) Pettigrew, J. D.; Bexrud, J. A.; Freeman, R.
P.; Wilson, P. D. Heterocycles 2004, 62, 445−452. (c) Lawrence, A. L.;
Adlington, R. M.; Baldwin, J. E.; Lee, V.; Kershaw, J. A.; Thompson, A.
L. Org. Lett. 2010, 12, 1676−1679.
(21) (a) Zhou, G. L.; Zheng, D. P.; Da, S. J.; Xie, Z. X.; Li, Y.
Tetrahedron Lett. 2006, 47, 3349−3352. (b) Zhou, G. L.; Zhu, J. R.;
Xie, Z. X.; Li, Y. Org. Lett. 2008, 10, 721−724. (c) Tian, T.; Li, L. Q.;
Xue, J. J.; Zhang, J.; Li, Y. J. Org. Chem. 2015, 80, 4189−4200.
(22) Nagao, Y.; Fujita, E.; Kohno, T.; Yagi, M. Chem. Pharm. Bull.
1981, 29, 3202−3207.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.6b02228.
Copies of 1H, 13C spectra for all new compounds (PDF)
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: xiezx@lzu.edu.cn.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We gratefully acknowledge financial support from the NSFC
(Grant No. 21272104), the “111” Program of MOE, the
program for Changjiang Scholars and Innovative Research
Team in University (PCSIRT: IRT_15R28), the Program for
New Century Excellent Talents in University (NCET-12-0247
and lzujbky-2012-56), and the Fundamental Research Funds
for the Central Universities (No. lzujbky-2013-ct02). We
sincerely thank Prof. Quanxiang Wu (Lanzhou University) for
helpful discussions.
REFERENCES
■
(1) (a) Xie, Z. L.; Wang, X. H. Chin. J. Inf. Tradit. Chin. Med. 2005,
12, 8−9. (b) Chang, H. S.; Liang, J. C.; Sun, J. N.; Wang, Q. G.; Li, J. J.
Beijing Univ. Tradit. Chin. Med. 2006, 29, 451−453. (c) Bao, L.; Chen,
J.; Huang, L.; Chen, W.; Lin, Q.; Yao, X. S.; Hiroshi, K. J. Chin. Med.
Mater. 2008, 31, 1360−1364. (d) Li, L.; Zhang, X. H.; Yan, T. T.; Li,
Q.; Liu, S. K.; Wu, L. L.; Yan, C.; Tan, H. M. Neural Regener. Res.
2010, 5, 1396−1402. (e) Pharmacopoeia of the People’s Republic of
China 2015, Chinese edition; China Medical Science and Technology
Press: 2015; pp 1151−1156.
(2) (a) Zhang, A. P.; Chen, M. Z.; Xu, S. Y. Chin. Pharmacol. Bull.
1993, 9, 454−457. (b) Hsu, F. L.; Lai, C. W.; Cheng, J. T. Planta Med.
1997, 63, 323−325. (c) Mao, Q. Q.; Ip, S. P.; Tsai, S. H.; Che, C. T. J.
Ethnopharmacol. 2008, 119, 272−275. (d) Cui, G. Z. World Phytomed.
2009, 24, 231−233. (e) 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, 424−427.
(4) Xu, L.; Liu, F. Y.; Xu, L.-W.; Gao, Z. W.; Zhao, Y.-M. Org. Lett.
2016, 18, 3698−3701.
(5) (a) In Modern Natural Product Chemistry; Wang, F. P., Ed.;
Science Press: 2009. (b) In The Chemistry of the Monoterpenoid and
Sesquiterpene; Shi, Y. P., Ed.; Chemical Industry Press: 2008.
(6) Liang, W.-J.; Ma, Y.-B.; Geng, C.-A.; Huang, X.-Y.; Xu, H.-B.;
Zhang, X.-M.; Chen, J.-J. Fitoterapia 2015, 106, 36−40.
(7) Bai, W.-J.; David, J. G.; Feng, Z.-G.; Weaver, M. G.; Wu, K.-L.;
Pettus, T. R. R. Acc. Chem. Res. 2014, 47, 3655−3664.
(8) Ding, L. Q.; Jiang, Z. H.; Liu, Y.; Chen, L. X.; Zhao, Q.; Yao, X.
S.; Zhao, F.; Qiu, F. Fitoterapia 2012, 83, 1598−1603.
(9) For a review, see: Caruana, L.; Fochi, M.; Bernardi, L. Molecules
2015, 20, 11733−11764.
(10) Willis, N. J.; Bray, C. D. Chem. - Eur. J. 2012, 18, 9160−9173.
D
DOI: 10.1021/acs.orglett.6b02228
Org. Lett. XXXX, XXX, XXX−XXX