Asymmetric synthesis of (+)-and (-)-pauciflorol F: Confirmation of absolute stereochemistry

Article abstract of DOI:10.1021/ol401752u

An efficient, formal enantioselective synthesis of (+)-and (-)-pauciflorol F has been achieved using a recently introduced oxazolidinone controlled torquoselective Nazarov reaction. The absolute stereochemistry of pauciflorol F and its biosynthetic precur

Full text of DOI:10.1021/ol401752u

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Asymmetric Synthesis of (þ)- and
(ꢀ)-Pauciflorol F: Confirmation of
Absolute Stereochemistry
Daniel J. Kerr,† Michael Miletic,† Narasimhulu Manchala,† Jonathan M. White,‡ and
Bernard L. Flynn*,†
Medicinal Chemistry, Monash Institute of Pharmaceutical Sciences,
Monash University, 381 Royal Parade, Parkville, Vic 3052, Australia, and Bio21
Institute, School of Chemistry, University of Melbourne, Parkville, Vic 3010, Australia
bernard.flynn@monash.edu
Received June 21, 2013
ABSTRACT
An efficient, formal enantioselective synthesis of (þ)- and (ꢀ)-pauciflorol F has been achieved using a recently introduced oxazolidinone
controlled torquoselective Nazarov reaction. The absolute stereochemistry of pauciflorol F and its biosynthetic precursors has been
unambiguously confirmed using X-ray crystallography.
Resveratrol 1 is an important building block in nature
that is used in the biosynthesis of a diverse array of
bioactive polyphenols.1 Key sources of these natural pro-
ducts are the plant families Vitaceae, Dipterocarpaceae,
Gnetaceae, Cyperaceae, and Leguminosae. To date, sev-
eral hundred such polyphenolic natural products have
been isolated and characterized. Many of these exhibit
valuablebiological activity, suchasantioxidant, antibiotic,
anticancer, anti-inflammatory, anti-HIV, and antifungal
activity.2 Biosynthetically, the loss of an electron or
addition of a proton reveals in 1 the complementary
functionalities of a latent electrophilic quinone methide
and nucleophilic 3,5-dihydroxyphenyl group 1a (Scheme 1).1a
These functionalities play key roles in the oligomerization
of 1 and the further modification of these oligomers.3
Oxidative dimerization of 1 to ε-viniferin (2)3a and quad-
rangularin A (7)3f are key first steps in this scaffold-
divergent synthesis.1a As part of this biosynthetic diver-
gence, 2 is converted into a regioisomer of 7, ampelopsin
D (3).3b There are a number of cyclization, redox, and
oligomerization pathways available to 1, 2, 3, and 7,
which, in combination with glycosylation, form the basis
of much of the structural diversity seen in these natural
products.1 For example, protic activation of the alkene
in regioisomers 3 and 7 results in cyclization onto the
† Monash University.
‡ University of Melbourne.
(1) For reviews on resveratrol and associated oligomers, see: (a)
Snyder, S. A.; ElSohly, A. M.; Kontes, F. Nat. Prod. Rep. 2011, 28, 897–
ꢀ
924. (b) Quideau, S.; Deffieux, D.; Douat-Casassus, C.; Pouysegu, L.
Angew. Chem., Int. Ed. 2011, 50, 586–621. (c) Snyder, S. A. In Biomi-
metic Organic Synthesis; Poupon, E., Nay, B., Eds.; Wiley-VCH: Weinheim,
2011; pp 695ꢀ721. (d) Ito, T.; Tanaka, T.; Iinuma, M.; Iliya, I.; Nakaya,
K.; Ali, Z.; Takahashi, Y.; Sawa, R.; Shirataki, Y.; Murata, J.; Darnaedi,
D. Tetrahedron 2003, 59, 5347–5363.
(2) For reviews on the biological properties of resveratrol oligomers,
see: (c) Pervaiz, S.; Holme, A. L.; Aggarwal, B. B.; Anekonda, T. S.;
Baur, J. A.; Gojkovic-Bukarica, L.; Della Ragione, F.; Kim, A. L.;
Pirola, L.; Saiko, P. Antioxid. Redox Signal. 2009, 11, 2851–2897. (d)
Marques, F. Z.; Markus, M. A.; Morris, B. J. Int. J. Biochem. Cell Biol.
2009, 41, 2125–2128. See also: Wood, J. G.; Rogina, B.; Lavu, S.;
Howitz, K.; Helfand, S. L.; Tatar, M.; Sinclair, D. Nature 2004,
430, 686–689 and references therein.
(3) For original papers describing the isolation of the resveratrol
dimers depicted in Scheme 1, see: (a) 2: Kurihara, K.; Kawabata, J.;
Ichikawa, S.; Mizutani, J. Agric. Biol. Chem. 1990, 54, 1097–1099. (b) 3:
Oshima, Y.; Ueno, Y. Phytochemistry 1993, 33, 179–182. (c) 4: Ito, T.;
Tanaka, T.; Linuma, M.; Nakaya, K.; Takahashi, Y.; Sawa, R.; Murata,
J.; Darnaedi, D. J. Nat. Prod. 2004, 67, 932–937. (d) 5 and 9 Luo, H.-F.;
Zhang, L.-P.; Hu, C.-Q. Tetrahedron 2001, 57, 4849–4854. (e) 6: Oshima,
Y.; Ueno, Y. Tetraherdron 1993, 49, 5801–5804. (f) 7: Adesanya, S. A.;
Nia, R.; Martin, M.-T.; Boukamcha, N.; Montagnac, A.; Pais, M.
J. Nat. Prod. 1999, 62, 1694–1695. (g) 8: Khan, M. A.; Nabi, S. G.;
Prakash, S.; Zaman, A. Phytochemistry 1986, 25, 1945–1948.
r
10.1021/ol401752u
XXXX American Chemical Society

Scheme 1. Some Resveratrol Derived Natural Products3
pendant 3,5-dihydroxyphenyl unit: 3 f ampleopsin
F (6)3e and 7 f pallidol (8).3g On the other hand, oxida-
tion of this alkene in 3 and 7 gives caraphenols B (5) and C
(9),3d,4 respectively, and oxidative cleavage of the alkene
in 3 gives pauciflorol F (4).3c
Cissus quadrangularis L. (Vitaceae) as the (ꢀ)-enan-
tiomer.3f Nonetheless, its presence in other plant families
can be deduced from the isolation biosynthetic products,
such as 8 and (þ)-9 from Caragana sinica (Legumino-
sae).3d,g,7 While 8 has been isolated from both Vitaceae
and Leguminosae, it cannot be used to inform on changes in
absolute stereochemistry of 7, as it has no optical rotation
and 9 has only been isolated from Leguminosae. Since
Caragana sinica also yielded (ꢀ)-6 we can deduce that
(þ)-3 is present in this plant (as expected for Leguminosae)
and that (þ)-5 has the depicted RSR-stereochemistry (not
previously assigned). Since 9 has a very similar structure to 5
(differing only in the location of OH groups), it is pro-
bable that (þ)-9 has the same RSR-stereochemistry as
(þ)-5. On this basis we could conclude that (þ)-7 is present
in Leguminosae and that, like 2, 7 also switches stereo-
chemistry in Vitacae relative to other plant families. How-
ever, notwithstanding their structural similarity, some
uncertainty exists in assigning the same stereochemistry
to 5 and 9 based on the same optical rotation. Moreover,
the current basis for the assignment of absolute stereo-
chemistry of all resveratrol dimers, comparison of the CD
spectra of 2 to other simpler 2-aryl-2,3-dihydrobenzofurans,
The absolute stereochemistry of (ꢀ)-2 was assigned nearly
25 years ago based on its comparable circular dichroism
(CD) spectra to other simpler 2-aryl-2,3-dihydrobenzofur-
ans (not shown).2,5 This assignment has been extended to
the many biosynthetic products derived from 2.6 It has also
been extended to 7 and related biosynthetic products by
comparing the CD spectra of (ꢀ)-3 and (ꢀ)-7 (both from
Vitacae), which are almost identical.6b Interestingly, while
(ꢀ)-2 is present in Dipterocarpaceae, Gnetaceae, Cyper-
aceae, and Leguminosae, (þ)-2 occurs in Vitacae.6 This
stereochemical switch also extends to the many biosyn-
thetic products derived from 2. At this stage it is not yet
known if 7 also switches its absolute stereochemistry in
different plant families, as it has only been isolated from
(4) For a structural revision of the relative stereochemistry of the
caraphenols B and C, see: Snyder, S. A.; Brill, Z. G. Org. Lett. 2011,
13, 5524–5527.
(5) (a) Lins, A. P.; Yoshida, M.; Gottlieb, O. R.; Gottlieb, H. E.;
Kubitzki, K. Bull. Soc. Chim. Belg. 1986, 95, 737–748. (b) Lima, O. A.;
Gottlieb, O. R.; Magalhaes, M. T. Phytochemistry 1972, 11, 2031–2037.
(6) (a) Takaya, Y.; Yan, K.-X.; Terashima, K.; Ito, J.; Niwa, M.
Tetrahedron 2002, 58, 7259–7265. (b) He, S.; Wu, B.; Pan, Y.; Jiang, L.
J. Org. Chem. 2008, 73, 5233–5241. (c) Ito, T.; Abe, N.; Oyama, M.;
Iinuma, M. Tetrahedron Lett. 2009, 50, 2516–2520.
(7) Choi, C. W.; Choi, Y. H.; Cha, M.-R.; Yoo, D. S.; Kim, Y. S.;
Yon, G. h.; Choi, S. U.; Kim, Y. H.; Ryu, S. Y. Bull. Korean Chem. Soc.
2010, 31, 3448–3450.
(8) (a) Polavarapu, P. Chirality 2012, 24, 909–920. (b) Sandstrom, J.
Chirality 2000, 12, 162–171.
B
Org. Lett., Vol. XX, No. XX, XXXX

also has its limitations.5,8 Confirmation of the absolute
stereochemistry of these natural products can come from
total synthesis, so long as the absolute stereochemistry
of relevant substrates and/or synthetic intermediates can
be unambiguously assigned using chiral pool substrates
and/or X-ray crystallography. Such syntheses can also
facilitate investigations into the bioactivity of specific
antipodes. To this end, diarylindanones 10a and 10b are
valuable synthetic targets for asymmetric synthesis.
Snyder and others have synthesized racemic versions of
these and converted them into 3ꢀ5, 7, 8, 10, and 11
(Scheme 2).4,9 Recently, Heo and co-workers reported
an elegant enantioselective synthesis of (þ)-10a and
(þ)-4.10,11 A key step in this synthesis was the baker’s
yeast reduction of indenone 11 to (R)-(þ)-12 (Scheme 3). This
(Scheme 4). Initial conversion of 13 into the gem-dibro-
mostyrene 14 (95%)14 was followed by a one-pot HBr
elimination and coupling (bromoalkyne and oxazolidinone)
to give ynamides 15aꢀc in good yield (78ꢀ91%).15,16 We
elected to use three different, readily available oxazolidi-
none auxiliaries, Aux1ꢀ3, to evaluate which performs
better and to enable the in-parallel generation of opposite
enantiomers of our target product 4 (note: the antipodes
of all three auxiliaries Aux1ꢀ3 are readily available).
Reductive coupling of the ynamides 15aꢀc with acid
chloride 16 gave the arylvinyl ketones 17a (68%), 17b
(73%), and 17c (49%). The yield of this reaction was in
part limited by the regioselectivity of hydrostannylation
(Bu3Sn group R or β to the oxazolidinone), which varied
somewhat for the three auxiliaries used: the R:β ratio for
Aux1 (2.7:1); Aux2 (4.6:1); Aux3 (3.5:1).17 The yield of 17c
(49%) was further compromised by the difficulty in
separating the regioisomers of the coupled product upon
chromatography.
Scheme 2. Use of Indanones in Resveratrol Dimer Synthesis4,9
Nazarov cyclization of 17aꢀc under our standard con-
ditions (10 equiv of MeSO3H, CH2Cl2, 18 °C) gave the
trans-indanones 18aꢀc in good yields (79ꢀ94%). The
stereochemical induction was highest in the case of Aux1
and Aux3, which both gave diastereomeric ratios (dr)
of >40:1 and lower for Aux2 (dr = 20:1), favoring the
Scheme 3. Heo Approach to 410
was followed by R-arylation to give (þ)-10a and demethyla-
tion to (þ)-4. While elegant, two limitations attend this
approach to (þ)-4: the inability to unequivocally verify the
absolute stereochemistry of the product and the inability
to prepare the correct enantiomer, since there is no reliable
alternative to baker’s yeast for this purpose. Herein, we
describe the use of Evans’ oxazolidinones as chiral auxiliaries
in the asymmetric synthesis of both antipodes of 10a and 4.
This approach provides rigorous assignment of the absolute
stereochemistry of pauciflorol F through X-ray crystal struc-
ture analysis and formal, enantiodivergent, entry into a range
of other related resveratrol dimers.
Our enantioselective approach to 4 utilizes our recently
introduced oxazolidinone controlled Nazarov cyclization
process.12,13 Thissynthesis commences with theconversion
of the aldehyde 13 into ynamides 15aꢀc in two steps
(9) (a) Snyder, S. A.; Zografos, A. L.; Lin, Y. Angew. Chem., Int. Ed.
2007, 46, 8186–8191. (b) Snyder, S. A.; Breazzano, A. G.; Ross, A. G.;
Lin, Y.; Zografos, A. L. J. Am. Chem. Soc. 2009, 131, 1753–1765. (c)
Zhong, C.; Zhu, J.; Chang, J.; Sun, X. Tetrahedron Lett. 2011, 52, 2815–
2817.
depicted diastereomers.18 The isolated yield for 18a (79%)
and 18b (87%) is for the major diastereomer, after chro-
matographic separation of the diastereomic mixture.
The isolated yield for 18c (94%) is for the diastereomeric
(10) Lee, B. H.; Choi, Y. L.; Shin, S.; Heo, J.-Y. J. Org. Chem. 2011,
76, 6611–6618.
(11) For a racemic synthesis of 4 involving Nazarov cyclization, see:
(a) Yang, Y.; Philips, D.; Pan, S. J. Org. Chem. 2011, 76, 1902–1905. For
other racemic syntheses of 4, see ref 9a and: (b) Jeffrey, J. L.; Sarpong, R.
Org. Lett. 2009, 11, 5450–5453. (c) Bo, C.; Lu, J.-P.; Xie, X.-G.; She,
X.-G.; Pan, X.-F. Chin. J. Org. Chem. 2006, 26, 1300–1302.
(12) (a) Kerr, D. J.; Miletic, M.; Chaplin, J. H.; White, J. M.; Flynn,
B. L. Org. Lett. 2012, 14, 1732–1735. (b) Flynn, B. L.; Manchala, N.;
Krenske, E. H. J. Am. Chem. Soc. 2013, 135, 9156–9163. (c) Kerr, D. J.;
Flynn, B. L. J. Org. Chem. 2010, 75, 7073–7084.
(13) For reviews on enantioselective Nazarov cyclizations, see: (a)
Vaidya, T.; Eisenberg, R.; Frontier, A. J. ChemCatChem 2011, 3, 1531–
1548. (b) Shimada, N.; Stewart, C.; Tius, M. A. Tetrahedron 2011,
67, 5851–5870.
(14) Xia, Y.; Jin, Y.; Kaur, N.; Choi, Y.; Lee, K. Eur. J. Med. Chem.
2011, 46, 2386–2396.
(15) Zhang, X.; Zhang, Y.; Huang, J.; Hsung, R. P.; Kurtz, K. C. M.;
Oppenheimer, J.; Petersen, M. E.; Sagamanova, I. K.; Shen, L.; Tracey,
M. R. J. Org. Chem. 2006, 71, 4170–4177.
(16) For an alternative method of converting gem-dibromoalkenes to
ynamides, see: Coste, A.; Karthikeyan, G.; Couty, F.; Evano, G. Angew.
Chem., Int. Ed. 2009, 48, 4381–4385.
(17) See Supporting Information.
(18) The diastereomeric ratios were determined by cleaving the
auxiliary from a crude diastereomeric mixture, followed by chiral HPLC
of the resultant enantiomeric mixture of 12; see Supporting Information.
Org. Lett., Vol. XX, No. XX, XXXX
C

Clark et al.19 To help clarify the source of this discrepancy
we subjected our sample of (ꢀ)-12 ([R]D = ꢀ5.1) to
the steps of R-arylation and demethylation described in
Scheme 2 and noted that it gave (þ)-10a ([R]D = þ136)
and (þ)-4 ([R]D = þ77),17 whereas Heo had reported their
Scheme 4. Synthesis of (þ)- and (ꢀ)-12
sample of (þ)-12 ([R]D = þ8.9) gave (þ)-10a ([R]D
=
þ137) and (þ)-4 ([R]D = þ86).10,20 Thus, the most likely
explanation for this discrepancy is that a mistake had been
made in noting the optical rotation of 12 obtained from the
baker’s yeast reduction of 11 and that it had been incor-
rectly assigned the dextrorotary (þ)-stereochemistry
(no explanation is given for the variation in magnitude of
rotation between equivalent compounds of similar ee).
In conclusion, the oxazolidinone controlled Nazarov
reaction has been applied to a formal synthesis of both
antipodes of 4 (Aux2: 7 steps, 28%, 99% ee). This is
comparable in overall efficiency to the recent synthesis of
the unnatural enantiomer, (þ)-4, by Heo and co-workers
(6 steps, 33%, 99% ee).10 Of the three oxazolidinones
used in this study, Aux1 and Aux2 appear the most
useful, with Aux3 being less efficient in the reductive
coupling step. This synthesis has confirmed the absolute
stereochemistry of (ꢀ)-4 and is consistent with earlier CD
studies on 2 and on it being the biosynthetic source of 4.
This methodology has the potential to provide enantiodi-
vergent access to many resveratrol oligomers, based on
earlier described approaches using racemic 10a and 10b
(Scheme 1).4,9
mixture obtained directly from the Nazarov reaction (base
wash, no chromatography). Reductive cleavage of the aux-
iliaries (Aux1ꢀ3) from 18aꢀc using lithium naphthalenide
(LiNph) produced indanones (R)-(ꢀ)-12 and (S)-(þ)-12
in good yield (84ꢀ92%). While the enantiomeric purity
of (S)-(þ)-12 obtained from 18c was 95% ee (reflecting
the dr of 42:1), higher enantiomeric purity (g99% ee) was
obtained for (R)-(ꢀ)-12 and (S)-(þ)-12, obtained from
18a and 18b, which had undergone prior chromato-
graphic purification of the major diastereomers.
Figure 1. ORTEP of 18a (arbitrary numbering).
The stereochemical assignment of diastereomers 18aꢀc
was based on the X-ray crystal structure analysis of 18a
(Figure 1). Accordingly, we assigned (þ)-12 the S-stereo-
chemistry. This conflicts with Heo and co-workers who
assigned (þ)-12 the R-stereochemistry based on analogous
baker’s yeast reductions of indanones performed by
Supporting Information Available. X-ray crystal struc-
ture data for 18a, preparative procedures and spectro-
scopic data for all compounds, and chiral HPLC traces
for different samples of (þ)- and (ꢀ)-12. This material is
available free of charge via the Internet at http://pubs.acs.org.
(20) All optical rotations were performed in MeOH (c = 0.4 or 0.5) at
20 °C.
(19) (a) Clark, W. M.; Kassick, A. J.; Plotkin, M. A.; Eldridge, A. M.;
Lantos, I. Org. Lett. 1999, 1, 1839–1842. (b) Clark, W. M.; Tickner-
Eldridge, A. M.; Huang, G. K.; Pridgen, L. N.; Olsen, M. A.; Mills, R. J.;
Lantos, I.; Baine, N. H. J. Am. Chem. Soc. 1998, 120, 4550–4551.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX