Biomimetic Synthesis and Structural Revision of Chaxine B and Its Analogues

Article abstract of DOI:10.1021/acs.orglett.6b03724

Chaxine B and its analogues were synthesized from ergosterol in eight steps on the basis of our proposed biosynthetic pathway, which includes a highly site-selective and regioselective Baeyer-Villiger oxidation as the key step. This synthesis enabled the revision of the structures of chaxine B and its analogues.

Full text of DOI:10.1021/acs.orglett.6b03724

Letter
pubs.acs.org/OrgLett
Biomimetic Synthesis and Structural Revision of Chaxine B and Its
Analogues
Yushi Hirata,† Atsuo Nakazaki,† Hirokazu Kawagishi,‡ and Toshio Nishikawa*,†
†Graduate School of Bioagricultural Sciences, Nagoya University, Chikusa-ku, Nagoya 464-8601, Japan
‡Research Institute of Green Science and Technology, Shizuoka University, 836, Ohya, Suruga-ku, Shizuoka, 422-8529, Japan
S
* Supporting Information
ABSTRACT: Chaxine B and its analogues were synthesized from ergosterol in eight steps on the basis of our proposed
biosynthetic pathway, which includes a highly site-selective and regioselective Baeyer−Villiger oxidation as the key step. This
synthesis enabled the revision of the structures of chaxine B and its analogues.
haxine B and its analogues, chaxine C, D, E, and A, as well
as demethylincisterol A3, were isolated from Agrocybe
are available from natural sources.5 The structures of these
compounds were elucidated by an extensive analysis of 2D
NMR and MS spectra, which revealed unprecedented and
highly degraded steroid structures, in which the A and C/D
rings are connected by an ester linkage. Although the relative
stereochemistry between C2′ and C5′ in the A ring of chaxine
B and D was elucidated, its absolute configuration remains to
be determined. In the case of BB, even the relative
stereochemistry of the A ring has not yet been clarified. The
syntheses of chaxine A and demethylincisterol A3 from
ergocalciferol (vitamin D2) have already been reported,6 while
no synthetic studies of an ester type such as BB or chaxines B−
E have been reported, probably due to their stereochemical
ambiguity. Herein, we propose that these chaxines might be
biosynthesized from ergosterol (1 in Scheme 1), which is an
abundant steroid in mushrooms, based on the structural
similarity of their side chains. This assumption enabled us to
propose the absolute stereochemistry of the A ring as shown in
Scheme 1. In order to confirm this stereostructure, we planned
to synthesize chaxine B (denoted 8) from ergosterol (1) on the
basis of a route inspired by our proposed biosynthetic route
outlined below.
C
chaxingu, an edible mushroom that grows only at high altitude
in southern China (Figure 1).1 Chaxine B was also isolated
Figure 1. Originally proposed structures of chaxines A−E,
demethylincisterol A3, and BB (the assignment of the absolute
stereochemistry of the A rings is tentative).1
Scheme 1 shows our proposed biosynthetic route to chaxine
B from ergosterol. Photoinduced electrocyclization of ergoster-
ol (1), a well-known reaction in the biosynthesis of vitamin D2,
provides a mixture of isomers, in which previtamin D2 (2) or
tachysterol (3) may be considered as a biosynthetic
intermediate of chaxines. One of these two compounds may
be oxygenated to enedione 5 through e.g. endoperoxide 4,
which can be obtained by a [4 + 2] cycloaddition with singlet
from Penicillium sp. together with related compound BB.2
Interestingly, demethylincisterol A3, a carboxylic acid moiety of
chaxine B and C, was also isolated from other natural sources,
such as mushrooms, fungi,3 and marine animals.4 Chaxines
show potent suppressive activity toward the formation of
osteroclast, while BB and chaxine B accelerate neural stem cell
propagation. Despite such medicinally important properties,
details of the biological investigations have so far been
hampered by the limited quantities of these compounds that
Received: December 14, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b03724
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 1. Proposed Biosynthetic Route to Chaxine B and Related Compounds from Ergosterol (1)
Scheme 2. Synthesis of BB, Chaxine B and C, and Their Structural Revision
oxygen (1O2). A subsequent Baeyer−Villiger oxidation of 5 at
the C-6 ketone may generate enol ester 6. Epoxidation of the
electron-rich enol ester moiety, followed by acyloxy migration
under acidic conditions, should then afford chaxine B (8) or its
diastereomer at the C-2′ position, which should depend on the
stereochemistry of the epoxide of 7. Chaxine C and
demethylincisterol A3 should then be obtained from 8 upon
dehydration or hydrolysis, respectively.7
In order to emulate the biosynthetic route chemically in the
laboratory, three synthetic challenges have to be resolved,
although those may be managed by enzymes of the producers;
(i) the site-selective and regioselective Baeyer−Villiger
oxidation of 5, (ii) the diastereoselective epoxidation of 6,
and (iii) the stereochemistry of the rearrangement of the enol
ester epoxide 7. To the best of our knowledge, the Baeyer−
Villiger oxidation of a complex system such as 5 has not yet
been reported.8,9 The site-selectivity and regioselectivity of the
Baeyer−Villiger oxidation of 5 are difficult to predict, as they
depend on many factors, e.g. the electron density and steric
bulk of the substituents and stereoelectronic effects. In this
specific case, however, the two ketones connected via the
alkene group might behave independently, considering that the
planar structure required for conjugation should be difficult to
attain on account of the steric and electronic repulsion between
the two ketone moieties. As the migration step is the rate-
determining step of the Baeyer−Villiger oxidation, we assumed
that the migratory aptitude of the substituents might determine
the site-selectivity and regioselectivity of the reaction. As the
B
DOI: 10.1021/acs.orglett.6b03724
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
tetrasubstituted alkene of the A ring possesses the highest
migratory ability among the four possible substituents, we
anticipated the Baeyer−Villiger oxidation of 5 to occur
preferentially at the C-6 ketone to provide the desired product
6. The stereochemistry of the epoxidation of 6 could be
controlled by a metal-assisted catalysis of homoallylic alcohol or
by the steric hindrance arising from the protection of the
hydroxy group.10 Since the silica-gel-promoted rearrangement
of enol ester epoxides has been reported to be highly
stereoselective,11 one of the two diastereomeric epoxides 7
should afford chaxine B, a natural product, which should enable
the unambiguous determination of the stereostructure of the A
ring. With these considerations in mind, we embarked on the
synthesis of chaxine B and its analogues.
(see Figure 1). This result clearly suggests that the structure of
chaxine B needs to be revised.
As the silica-gel-promoted rearrangement of 7β proceeded
highly stereoselectively to afford 8,19 we expected that the
corresponding α-epoxide 7α would provide an epimer at the C-
2′ position of 8, i.e. the revised structure of chaxine B. We
therefore explored the optimum epoxidation conditions giving
7α as the major product. Epoxidation with MCPBA in Et2O
furnished 7β as a single diastereomer.20 After many experi-
ments, we found that oxidation of 6 with dimethyldioxirane
(DMDO) in acetone afforded a diastereomeric mixture of
1
epoxides 7β and 7α (ratio = ca. 3:5 by H NMR analysis),
which was directly exposed to silica gel to afford 8 and 11 in
21% and 40% yield, respectively. The NMR spectra of 11 are
identical to the reported data for chaxine B. Unfortunately, 11
(chaxine B) could not be crystallized, and the stereochemistry
of the A ring of chaxine B had thus to be confirmed by
degradation experiments (Scheme 3). Reduction of the
Tachysterol (3), rather than previtamin D2 (2), was chosen
as the starting material due to its thermal stability and suitability
as a substrate for the oxygenation with 1O2 (Scheme 2).
Photochemical isomerization of ergosterol (1) affords a mixture
of previtamin D2 (2), tachysterol (3), and lumisterol, whereby
the composition of the mixture depends on the wavelength of
the UV light:12 irradiation with short-wavelength UV light (λ =
254 nm) by a low-pressure mercury lamp furnished 3 in 55%
yield as the major product. However, the subsequent reaction
Scheme 3. Chemical Degradation of Chaxine B (11) and the
Stereostructure of the A Ring
1
of 3 with O2 proved to be problematic: endoperoxide 4 was
obtained in poor yield (<35%) under standard conditions using
light and tetraphenylporphyrin (TPP) as a sensitizer, probably
due to overoxidation of the product. In order to suppress the
side reactions, we attempted the reaction with dimethyl
naphthalene-endoperoxide (NEP),13 which has previously
been reported to generate 1O2 quantitatively at room
temperature. Treatment of 3 with NEP (5.5 equiv) in the
presence of di-tert-butylhydroxytoluene (BHT) at room
temperature improved the yield of 4 to ca. 50%.14 In order
to transform 4 into enedione 5, we initially attempted a two-
step procedure comprising the cleavage of the endoperoxide
moiety followed by oxidation. Endoperoxide 4 was treated with
DBU to give 9 as a single product;15 however, the selective
oxidation of the allylic hydroxy group at the C-9 position of 9
was unsuccessful. Therefore, we synthesized enedione 5 by
oxidation of furan 10, which was easily prepared from 9 using
the Paal−Knorr method. Treatment of 9 with TMSOTf in the
presence of molecular sieves (AW-300) afforded 10 in high
yield as an unstable product. Without purification, 10 was
oxidized with MCPBA at −40 °C, and to our delight, we
observed the formation of enedione 5 as well as enol ester 6,
which is the desired product of the Baeyer−Villiger oxidation of
5. The oxidation of 10 with 2.1 equiv of MCPBA furnished enol
ester 6 in 47% yield in three steps from 9. Therefore, the site-
selective and regioselective Baeyer−Villiger oxidation of 5 was
successfully accomplished as anticipated.
synthesized chaxine B with NaBH4, followed by acetylation,
furnished a mixture of 12, 13, and 14.6a Their structures,
including the relative stereochemistry of 12 and 13, which bear
A rings, were determined by extensive analysis of their 2D-
NMR spectra and are shown in the Scheme 3.21 We thus
conclude that the correct structure of chaxine B is 11, an
epimer at the C-2′ position of BB (8). Furthermore, we
determined the structure of chaxine C, by dehydration (MsCl,
Et3N) of 8 and chaxine B (11), which afforded 16 (not shown)
and 15 in 92% and 74% yield, respectively. Comparison of the
1H and 13C NMR spectra of 16 and 15 with those of natural
chaxine C revealed that 15 is identical with chaxine C (Scheme
2), which strongly implies that chaxine C is, as proposed,
biosynthetically derived from chaxine B. Therefore, we propose
that the stereochemistry of the A ring of chaxine D and E is
identical to that of chaxine B and C, respectively.
In summary, we have achieved the eight-step synthesis of
chaxine B and its epimer BB from ergosterol, which was
inspired by our biosynthetic proposal of these natural products.
Unexpectedly, the X-ray analysis of BB resulted in a revision of
its previously reported structure to that of 8, while the structure
of chaxine B was revised to that of 11, which was based on
chemical degradation experiments and analyses of the 2D NMR
spectra (Figure 2). The salient feature of this synthesis is the
oxidative cascade of furan 10 to enol ester 6 with MCPBA,
which includes the formation of enedione and a highly site-
selective and regioselective Baeyer−Villiger oxidation. Unfortu-
nately, an attempted concomitant epoxidation of the enol ester
under the same conditions was unsuccessful due to competitive
Subsequently, we investigated the regio- and stereoselectivity
of the epoxidation of 6. Reaction of 6 with TBHP and a
10,16
catalytic amount of VO(acac)2
provided epoxide 7β as a
single diastereomer,17 which was subsequently treated with
silica gel to give a crystalline product in 67% overall yield from
6. The 1H and 13C NMR spectra of this product were identical
to those of BB. However, an X-ray crystallographic analysis
revealed that its structure is not consistent with a structure that
contains an ester of the secondary alcohol, as reported, but with
the structure of 8, which contains an ester of the tertiary alcohol
of the A ring.18 Confusingly, the relative stereochemistry of the
A ring in 8 is identical to that originally assigned to chaxine B
C
DOI: 10.1021/acs.orglett.6b03724
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(3) For examples, see: (a) Ohta, K.; Yaoita, Y.; Matsuda, N.; Kikuchi,
M. Nat. Med. 1996, 50, 179. (b) Akihisa, T.; Nakamura, Y.; Tagata, M.;
Tokuda, H.; Yasukawa, K.; Uchiyama, E.; Suzuki, T.; Kimura, Y. Chem.
Biodiversity 2007, 4, 224. (c) Ueguchi, Y.; Matsunami, K.; Otsuka, H.;
Kondo, K. J. Nat. Med. 2011, 65, 307.
(4) Mansoor, T. A.; Hong, J.; Lee, C.-O.; Bae, S.-J.; Im, K. S.; Jung, J.
H. J. Nat. Prod. 2005, 68, 331.
(5) Chaxine B, C, D, and E were isolated in 2.1, 1.6, 0.5, and 0.4 mg
respectively from 1.5 kg of dried fruiting bodies of the mushroom.
(6) (a) De Riccardis, F.; Spinella, A.; Izzo, I.; Giordano, A.; Sodano,
G. Tetrahedron Lett. 1995, 36, 4303. (b) Yajima, A.; Kagohara, Y.;
Shikai, K.; Katsuta, R.; Nukada, T. Tetrahedron 2012, 68, 1729.
(7) An alternative biosynthetic pathway for demethylincisterol A3 has
been proposed. For details, see ref 4.
(8) For reviews, see: (a) Krow, G. R. Org. React. 1993, 43, 251.
(b) Arends, G.-J.; ten Brink, I. W. C. E.; Sheldon, R. A. Chem. Rev.
2004, 104, 4105.
(9) For a review of biological Baeyer−Villiger oxidations, see: Leisch,
H.; Morley, K.; Lau, P. C. K. Chem. Rev. 2011, 111, 4165.
(10) (a) Sharpless, K. B.; Michaelson, R. C. J. Am. Chem. Soc. 1973,
95, 6136. (b) Hoveyda, A. M.; Evans, D. A.; Fu, G. C. Chem. Rev.
1993, 93, 1307.
Figure 2. Summary of the structural revision of chaxine B and BB
(*the assignment of the absolute stereochemistry of the A rings is
tentative).
(11) For the stereochemistry of the rearrangement of epoxy enol
esters similar to 7, see: Tranchimand, S.; Faure, B.; Naubron, J.-V.;
Alphand, V.; Archelas, A.; Iacazio, G. Eur. J. Org. Chem. 2012, 2012,
4365.
(12) (a) Havinga, E.; De Kock, R. J.; Rappoldt, M. P. Tetrahedron
1960, 11, 276. (b) Malatesta, V.; Willis, C.; Hackett, P. A. J. Am. Chem.
Soc. 1981, 103, 6781. (c) Dauben, W. G.; Phillips, R. B. J. Am. Chem.
Soc. 1982, 104, 355.
(13) (a) Wasserman, H. H.; Larsen, D. L. J. Chem. Soc., Chem.
Commun. 1972, 253. (b) Wasserman, H. H.; Wiberg, K. B.; Larsen, D.
L.; Parr, J. J. Org. Chem. 2005, 70, 105.
(14) The stereochemistry of the endoperoxide of 4 was determined
by NOESY correlation between the proton at the C-9 and the angular
methyl protons.
epoxidation of the side chain. Since this synthetic route is very
short (operationally seven steps), it should provide a variety of
chaxine analogues for biological investigations. This study thus
strongly supports the notion that ergosterol is the biosynthetic
precursor of chaxines and that the synthetic route presented
herein might indeed be similar to the actual biosynthetic route.
Therefore, these results should contribute to the elucidation of
the biosynthesis of chaxines in the near future.
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.6b03724.
(15) We expected that the Baeyer−Villiger oxidation of 9 might
occur selectively at the C-6 ketone at this stage; however, the
attempted Baeyer−Villiger oxidation with MCPBA resulted in an
epoxidation of the A ring in a good yield.
Experimental procedures, characterization data, copies of
1H and 13C NMR spectra for all new compounds (PDF)
(16) Jeker, O. F.; Carreira, E. M. Angew. Chem., Int. Ed. 2012, 51,
3474.
(17) The diastereoselectivity was determined by 1H NMR analysis of
the crude product.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: nisikawa@agr.nagoya-u.ac.jp.
ORCID
(18) CCDC-1508011 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.
uk/data_request/cif.
(19) Judging from the structures of 7β and BB (8), the acyloxy
migration proceeded with inversion of the configuration at the C-2′
position, which is consistent with the conclusions of ref 11.
(20) The epoxidation with MCPBA in CH2Cl2 as a conventional
solvent gave a substantial amount of byproducts, including epoxides of
the side chain.
Atsuo Nakazaki: 0000-0002-0025-0523
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors acknowledge Dr. H. Onodera (Kyowa Hakko
Kirin Co., Ltd.) for providing NMR spectra of BB. This work
was supported by the Grant-in-Aid for Scientific Research on
Innovative Areas ‘‘Chemical Biology of Natural Products’’22
(MEXT) and by the Yamada Science Foundation.
(21) Details of the structure determination are included in the
Supporting Information.
(22) Ueda, M. Chem. Lett. 2012, 41, 658.
REFERENCES
■
(1) (a) Chaxine, A.; Kawagishi, H.; Akachi, T.; Ogawa, T.; Masuda,
K.; Yamaguchi, K.; Yazawa, K.; Takahashi, M. Heterocycles 2006, 69,
253. (b) Chaxine, B.-E.; Choi, J.-H.; Ogawa, A.; Abe, N.; Masuda, K.;
Koyama, T.; Yazawa, K.; Kawagishi, H. Tetrahedron 2009, 65, 9850.
(2) Onodera, H.; Ichimura, M.; Baba, K.; Agatsuma, T.; Sasho, S.;
Susuki, M.; Iwamoto, S.; Kakita, S. WO 2009/096445 A1. Since BB is
used for the compound name in this patent, we used the same name in
this report.
D
DOI: 10.1021/acs.orglett.6b03724
Org. Lett. XXXX, XXX, XXX−XXX