Synthetic Study on Carthamin: Problem and Solution for Oxidative Dearomatization Approach to Quinol C -Glycoside

Article abstract of DOI:10.1055/s-0035-1562511

Synthetic study on carthamin, a natural red pigment, is described. Attempts at the construction of the key quinol C-glycoside structure by oxidative dearomatization of C-glycosyl phenols were hampered by the competing cleavage of the C1-C2 bond in the sugar moiety. By employing the 2-O-acetyl protection, the projected reaction proceeded smoothly, giving the desired products in good to excellent yields. An interesting photochemical stability/instability of the chalcone geometry in the synthetic compounds was observed.

Full text of DOI:10.1055/s-0035-1562511

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
© Georg Thieme Verlag Stuttgart · New York
2016, 27, A–G
letter
A
T. Hayashi et al.
Letter
Synlett
Synthetic Study on Carthamin: Problem and Solution for Oxida-
tive Dearomatization Approach to Quinol C-Glycoside
Taiki Hayashi
Ken Ohmori*
Keisuke Suzuki*
Department of Chemistry, Tokyo Institute of Technology,
O-okayama, Meguro-ku, Tokyo 152-8551, Japan
kohmori@chem.titech.ac.jp
ksuzuki@chem.titech.ac.jp
Received: 30.04.2016
Accepted after revision: 31.05.2016
Published online: 05.07.2016
HO
HO
OH
HO
OH
OH
DOI: 10.1055/s-0035-1562511; Art ID: st-2016-u0301-l
OH
HO
O
O
HO
OH
HO
HO
OH O
OH
OH
Abstract Synthetic study on carthamin, a natural red pigment, is de-
scribed. Attempts at the construction of the key quinol C-glycoside
structure by oxidative dearomatization of C-glycosyl phenols were ham-
pered by the competing cleavage of the C1–C2 bond in the sugar moi-
ety. By employing the 2-O-acetyl protection, the projected reaction
proceeded smoothly, giving the desired products in good to excellent
yields. An interesting photochemical stability/instability of the chalcone
geometry in the synthetic compounds was observed.
O
O
O
O
carthamin (1)
OR
OR
OR
RO
RO
OR
RO
RO
O
O
this work
Keywords carthamin, quinol C-glycoside, oxidative dearomatization,
chalcone, photoisomerization
OH
HO
OH
R
HO
OH
OR
– C1
Carthamin (1) is a Japanese traditional red pigment, iso-
lated as a constituent of the petals of safflower (Carthamus
tinctorius L.).1 After a long-standing structural investiga-
tions,2 the currently planar accepted structure was pro-
posed in 1979,2c composed of two C-glucosyl quinochal-
cone units that are interconnected through a single carbon
(Scheme 1). Organic synthesis played a significant role in
the elucidation of stereochemistry by providing model
compounds to assign the stereochemistry of 1.2g,h However,
the total synthesis of 1 has remained unachieved due to for-
midable challenges posed by the stereochemical and func-
tional complexity.
In connection with our long-standing interest in the
synthesis of aryl C-glycoside natural products,3 we became
interested in the synthesis of 1. Among several challenges
foreseen, we sought for an effective approach to construct
the C-glycosyl quinochalcone unit A in a form that is hope-
fully ready for further transformation into the final product
1.
oxidative
OH
O
O
x 2
dearomatization
A
B
C-glycosidation
OH
HO
OH
R
RO
RO
RO
O
+
RO
X
OH
C
D
Scheme 1 Retrosynthetic analysis
A reported approach to such unit A exploited the nucle-
ophilic displacement of glycosyl halide C by the anion de-
rived from hydroquinone D,2h which was low yielding by
the harsh reaction conditions and the multiple reaction
sites, including the C- vs. O-glycosylations.2h,4
As a potential alternative to such an approach exploiting
intermolecular dearomatization, we focused on the oxida-
tive dearomatization of C-glycosyl phenol B, with the aryl
C-glycoside linkage preformed. While the latter approach
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–G

B
T. Hayashi et al.
Letter
Synlett
appeared to be promising in view of the extensive use in
natural product synthesis,5 preliminary study uncovered a
hitherto hidden difficulty that may have limited the use of
this reaction in this particular context.
Herein, we report realization of the construction of C-
glycosyl quinochalcone A via oxidative dearomatization of
phloroglucinol derivative B. The trouble was settled by the
choice of the protection pattern for suppressing the side re-
actions. Also described is an interesting photochemical sta-
bility/instability of the chalcone geometry in the synthetic
compounds. Discussion will be made on the origin of pho-
tochemical behavior.
OBn
BnO
OBn
Sug
Sug
OH
MeO
O
MeO
O
O
BnO
PhI(OCOCF3)2
O
+
MeO
OH
MeCN–H2O
(8:1)
OMe
OMe
10%
0 °C, 25 min
(not detected)
6
7
OMe
Sug
4
OH
MeO
+
OPh
I
I
OMe
OCOCF3
23%
I
8
Scheme 2 shows the preparation of model compounds 4
and 5 from D-glucosyl fluoride 26 and phloroglucinol deriv-
ative 3.7 Upon treatment of 2 and 3 with BF3·OEt2 in CH2Cl2
(–78 °C to 0 °C, 1.5 h), the Friedel–Crafts C-glycosylation
proceeded,8 and subsequent removal of the TBDPS group
(n-Bu4NF, THF, room temperature, 1 h) gave a mixture of
isomeric C-glycosides 4 and 5. Separation by silica gel col-
umn chromatography (hexane–EtOAc = 97:3 to 60:40) gave
the ortho-glycosyl phenol 47 in 39% yield and its para coun-
terpart 57 in 48% yield. Compounds 4 and 5 were composed
of the β-anomer,7 respectively.
Scheme 3 Reaction of ortho-C-glycosyl phenol 4
OBn
OBn
BnO
OBn
OBn
BnO
BnO
OBn
BnO
BnO
O
PhI(OCOCF3)2
O
O
+
BnO
OH
MeO
OMe
MeCN–H2O
(3:1)
MeO
OMe
OH
44%
0 °C
10
OH
5
O
1%
9
OBn
OBn
BnO
BnO
OBn
OBn
Scheme 4 Reaction of para-C-glycosyl phenol 5
BnO
BnO
BnO
O
BnO
BnO
O
O
1) BF3·OEt2
MS4A
BnO
F
In order to gain insight, a simpler model compound 1111
was subjected to the same oxidative conditions (0 °C, 40
min, Scheme 5). The desired product 12 was obtained only
in 16% yield, and the arabinose derivative 10 was again ob-
tained in 69% yield. Notably, p-hydroxybenzaldehyde (13)
was obtained as an additional byproduct in 51% yield, sug-
gesting the origin of the problem.
MeO
OH MeO
OMe
CH2Cl2
2
+
2) nBu4NF
MeO
OMe
THF, r.t.
OMe
OH
48%
39%
4
5
OTBDPS
TBDPS = t-C4H9(C6H5)2Si
3
Scheme 2 Preparation of model substrates 4 and 5
OBn
OBn
OBn
OBn
BnO
BnO
H
O
As an initial feasibility study, C-glycosyl phenols 4 and 5
were subjected to oxidative dearomatization (Scheme 3).9
Upon treatment of ortho-C-glycosyl phenol 4 with bis(tri-
fluoroacetoxy)iodobenzene in wet acetonitrile (0 °C, 25
min), none of the desired product 6 was obtained. Many
side products formed including para-quinone 7 (10% yield)
and iodinated phenyl ether 8 (23% yield).10 Formation of 8
could be rationalized by the electrophilic substitution of
phenol 4 at the ortho position by the hypervalent iodine to
form I followed by the intramolecular ipso substitution by
the phenolic oxygen.10
The reaction of para-C-glycosyl counterpart 5 also gave
poor results, giving only a trace amount of quinol C-glyco-
side 9 (Scheme 4). Importantly, however, an arabinose de-
rivative 10 was identified as a side product (44% yield), sug-
gesting that the sugar moiety was damaged during the pro-
cess.
BnO
OBn
+
BnO
BnO
O
O
OH
PhI(OCOCF3)2
+
O
BnO
MeCN–H2O
(8:1)
OH
OH
0 °C
16%
69%
51%
OH
11
O
12
10
13
Scheme 5 Reaction of model substrate 11
Scheme 6 shows the rationale. Upon activation of the
phenol by hypervalent iodine reagent as in A, the electron
donation from the 2-O-lone pair cleaves the C1–C2 bond to
generate quinone methide B. Addition of water and proton-
ation give the corresponding oxonium species C. Trapping
by a water produces hemiacetal D, which releases the arabi-
nose derivative 10 and aldehyde 13. In a formal sense, this
process corresponds to the one-carbon delivery from the
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–G

C
T. Hayashi et al.
Letter
Synlett
For suppressing this undesired reaction, we reasoned
that the 2-O-protecting group of the sugar would be better
replaced by an electron-withdrawing group to reduce the
electron-donating ability of the oxygen lone pair. Along
these lines, we prepared three substrates 14, 16, and 18,
possessing a 2-O-acetyl protection. To our delight, the oxi-
dative dearomatization of these compounds proceeded
cleanly, giving the corresponding quinol C-glycosides 15,
17, and 19 in good to excellent yields (Scheme 7). Thus,
choice of an electron-withdrawing group for the 2-O-pro-
tection proved crucial for effecting the desired reaction.
Having this clue, synthesis of the quinochalcone was ex-
amined (Scheme 8).12 The precursor 23 was synthesized by
the Friedel–Crafts reaction of glycosyl fluoride 2 and silyl
ether 207 (BF3·OEt2, MS 4 Å, CH2Cl2, –78 °C to 0 °C), giving
the desired C-glycoside 21 in regioselective manner.7 The
protecting groups for the sugar moiety in 21 were ex-
changed by acetyl groups, and the protection for the phenol
was changed to a benzyl group to give C-glycoside 22,
which was carried out by the following four steps, (1) hy-
drogenolytic removal of the benzyl groups (H2, Pd(OH)2/C,
room temperature, 2 d), (2) acetylation (Ac2O, pyridine,
room temperature, 1 h), (3) removal of the silyl group (n-
Bu4NF, AcOH, room temperature, 20 min), and (4) protec-
tion of the liberated phenol by a benzyl group (BnBr, K2CO3,
room temperature, overnight). The ethyl group in 22 was
oxidized by using cerium ammonium nitrate in wet aceto-
nitrile (0 °C, 15 min), giving the corresponding benzyl alco-
hol, which was subjected to TPAP oxidation (room tempera-
ture, 1 h)13 to afford methyl ketone 23 in 75% yield in two
steps.
OBn OBn
OBn OBn
OBn OBn
O+
BnO
BnO
BnO
BnO
BnO
2
+
1
O
O
H2O
BnO
HO
O
O
OH
OCOCF3
I
H+
Ph
A
B
C
OBn OBn
BnO
BnO
H
O
BnO
OBn
O
H2O
+
HO
OH
O
OH
10
– BnOH
BnO
OH
13
OH
D
Scheme 6 Mechanism of the side reaction
sugar to the aromatic unit, by which the sugar moiety is
truncated by one carbon and the aromatic acquires a formyl
group.
OAc
OAc
OAc
OAc
AcO
AcO
2
2
AcO
AcO
O
O
OH
PhI(OCOCF3)2
MeCN–H2O (8:1)
0 °C, 75 min
71%
1) H2, Pd(OH)2
OBn
BnO
OH
14
O
OBn 2) Ac2O, pyridine
DMAP
BnO
BnO
BnO
O
15
3) nBu4NF, AcOH
quant. (3 steps)
BF3⋅OEt2
MS4A
BnO
F
BnO
OAc
OAc
OAc
OAc
O
AcO
AcO
2
MeO
OMe
2
2
CH2Cl2
4) BnBr, K2CO3
AcO
AcO
MeO
OMe
O
–78 to 0 °C
98%
O
OH
MeO
OMe
MeO
OMe
TBDPSO
PhI(OCOCF3)2
82%
TBDPSO
21
MeCN–H2O (8:1)
0 °C, 30 min
20
OH
O
92%
16
17
OAc
OAc
OAc
AcO
AcO
OAc
1) (NH4)2Ce(NO3)6
OBn
OBn
OBn
OBn
BnO
BnO
MeCN–H2O (4:1)
0 °C, 15 min
AcO
AcO
O
O
2
2
AcO
AcO
MeO
75%
OMe
O
MeO
OMe
O
O
OH
2) TPAP, NMO
MS4A, CH2Cl2
r.t., 1 h
MeO
OMe
MeO
OMe
PhI(OCOCF3)2
BnO
22
BnO
MeCN–H2O (8:1)
0 °C, 30 min
(2 steps)
23
OH
18
O
89%
TPAP = (n-C3H7)4NRuO4, NMO = 4-methylmorpholine N-oxide
19
Scheme 8 Preparation of methyl ketone 23
Scheme 7 Reaction of 2-O-acyl derivatives
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–G

D
T. Hayashi et al.
Letter
Synlett
The chalcone moiety was constructed by the aldol con-
densation of ketone 23 with aldehyde 2414 under basic con-
ditions (NaOEt, EtOH, reflux, 1 h), where the acetyl groups
were simultaneously removed (Scheme 9). Removal of the
THP group with Dowex 50W (H+) in ethanol (70 °C, 1 h) and
re-acetylation (Ac2O, pyridine, room temperature, over-
night) afforded the desired chalcone 26 in 85% yield in
three steps. Removal of the benzyl group using trifluoro-
acetic acid in toluene (60 °C, 15 min) gave 74% yield of phe-
nol 27 as yellow powders,15 ready for the oxidative dearo-
matization.
The stereochemistry of 29 was assigned by single-crys-
tal X-ray diffraction analysis of the colorless needles after
recrystallization from CH2Cl2–Et2O (Scheme 10), determin-
ing the its tert-alcohol center as R as shown.16 This is epi-
meric to the currently accepted stereochemistry assigned
for the natural product 1, which was deduced from the CD
spectra of model compounds.2g,h Vice versa, the minor iso-
mer 30 was assigned to have the S stereochemistry.17
Upon attempts at the deprotection of these quinochal-
cone isomers, TMSI allowed facile, selective monodemeth-
ylation. Treatment of 29 with TMSI (MeCN, –30 °C, 15 min)
gave triketone 31 in 97% yield as yellow oil. Under the simi-
lar conditions, isomer 30 was converted into triketone 32 as
yellow oil in 80% yield. In both cases, the monodemethyla-
tion proceeded regioselectively at one of the enol ether
moieties that is proximal to the cinnamoyl group.
OAc
AcO
OAc
AcO
OH
OH
HO
O
MeO
OMe
Figure 1 shows a rationale on the difference of two
methyl groups in the susceptibility to the SN2 attack. The
methyl group at the right-hand side is more reactive, be-
cause, as shown in A, the potential nucleofuge is doubly ac-
tivated by two carbonyl groups (red) and, hence, a more po-
larizable (soft) leaving group. By contrast, the effect is
much less pronounced in the leaving group attached to the
left methyl group as in B (blue).
HO
O
MeO
OMe
OTHP
BnO
O
NaOEt
23
EtOH
OTHP
reflux, 1 h
BnO
O
H
O
25
THP = 2-tetrahydropyranyl
24
1) Dowex 50W
EtOH, 70 °C
2) Ac2O
OAc
AcO
OAc
I–
I–
OR
OR
MeO
O
O
CH3
H3C
O
OMe
O
pyridine
AcO
O
DMAP, r.t.
VS
MeO
OMe
OAc
O
O
85%
(3 steps)
LA
LA
RO
O
B
A
26: R = Bn
27: R = H
TFA–toluene (1:1)
60 °C (74%)
Figure 1 Rationale for regioselective demethylation
Scheme 9 Construction of chalcone unit
Triketones 31 and 32, thus obtained, are known com-
1
pounds,2h and both of the H NMR spectra were consistent
Phenol 27 was treated with bis(trifluoroacetoxy)iodo-
benzene in wet acetonitrile in the presence of NaHCO3
(room temperature, 3 h), giving the desired quinochalcone
28 in 64–73% yield as an inseparable mixture of stereoiso-
mers (7:3 to 8:2 ratio, Scheme 10). Upon peracetylation of
28 (Ac2O, pyridine, room temperature, 3 h, 92% combined
yield), the stereoisomers of the corresponding acetate be-
came separable on silica gel TLC (29: Rf = 0.30, 30: Rf = 0.39,
Et2O–acetone = 90:10). Preparative separation by column
chromatography (Et2O–acetone = 95:5 to 80:20) gave the
major isomer 29 (more polar) as white powders and the
minor isomer 30 as colorless oil (less polar). Notably, each
isomer contained the respective Z-isomer (ca. 10%). By
working under photo-shielded conditions at the stage of 28
and 29/30, the double-bond isomerization was minimized
(vide infra).
with the reported data.18 Careful analysis of the HMBC data
(CDCl3) allowed to identify the major contributors of the
keto–enol tautomers in 31 and 32 as shown in Scheme 10.19
Finally, described is an interesting note on the different
geometrical stability of the chalcone moiety in various in-
termediates. As described earlier, the double bonds in the
intermediates 29 and 30 were geometrically labile, easily
underwent photoisomerization. As a telling experiment,
chalcone 29 (E/Z = 96:4) was exposed to weak UV light for a
long time (4 W, CDCl3, 69 h), giving an almost 1:1 mixture
of the E- and Z-isomers (Scheme 11). Upon demethylation,
interestingly, this E/Z mixture of 29 converged into a single
stereoisomer, (E)-triketone 31 in 95% yield. Thus, apparent-
ly a Z → E isomerization occurred during this conversion
under Lewis acidic conditions. Of further note, the E-geom-
etry of 31 remained unchanged even after exposure to
room light or to UV light, which stands in contrast to the
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–G

E
T. Hayashi et al.
Letter
Synlett
Scheme 10 Oxidative dearomatization of chalcone derivative
geometrical lability of chalcones 28 and 29/30, easily un-
dergoing E → Z isomerization even by room light (vide su-
pra).
Concerning the origin of such a high persistence of the
olefin geometry in 31, we have two postulates. One is the
olefin isomerization via an ionic mechanism (Figure 2). In
view of the extensively conjugated system as I, one could
assume sizable contribution of the canonical form II, mak-
ing a facile isomerization possible, where the hydrogen
bonding may play a role. By the steric preference, direction
of the isomerization would be Z → E, and, if rapid enough,
the process would serve to repair the photoinduced stereo-
randomization, making the E-isomer the sole entity steadi-
ly observed.
Sug
AcO
MeO
OMe
O
Sug
AcO
MeO
hν (UV)
OMe
OAc
O
CDCl3
57:43
O
O
(E)-29
(Z)-29
OAc
TMSI, MeCN
–30 °C
Postulate 1
Sug
95%
OAc
AcO
MeO
Sug
O
O
Oδ–
OR
O
O
OR
AcO
MeO
hν
O
(Vis or UV)
+
O
Oδ–
O
CDCl3
O
H
H
H
O
O
E only
H
I
II
(E)-31
(Z)-31
OAc
Figure 2 Ionic mechanism for persistent E-geometry in 31
Scheme 11 Distinct photochemical behavior of 29 and 31
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–G

F
T. Hayashi et al.
Letter
Synlett
In addition to such a backup mechanism, we envisaged
another postulate that the olefin in 31 has an ‘E-stabilizing’
mechanism due to the special structural feature. More spe-
cifically, we focused our attention to the one-way pho-
toisomerization reported by Arai,20 that is, 2′-hydroxychal-
cone (33), unlike normal olefins, undergoes solely the Z → E
photoisomerization, but not vice versa (Figure 3).21 This
phenomenon is attributed to the deactivation from the (E)-
enol tautomer (E)-34* that is most stable and abundant at
the triplet excited state formed via an intramolecular hy-
drogen atom transfer. By the local structure resemblance
including a hydrogen bond, we surmised that a similar
mechanism might work in 31.
In summary, a viable synthetic route to quinol C-glyco-
side has been developed via the oxidative dearomatization
of C-glycosyl phenol derivatives. Further studies toward the
total synthesis are in progress, centering attention to the
stereochemical control and construction of the dimeric
structure.
Acknowledgment
This work was supported by Grant-in-Aid for Specially Promoted Re-
search (No. 23000006) from Japan Society for the Promotion of Sci-
ence (JSPS). We gratefully acknowledge Prof. Shigeo Katumura
(Kwansei Gakuin Univ.), Prof. Kohei Kazuma (Toyama Univ.), Prof.
Dieter Seebach (ETH), and Prof. Takenori Kusumi (Tokyo Institute of
Technology) for helpful discussion on the subject.
Postulate 2
one-way photoisomerization
hν
Supporting Information
hν
Supporting information for this article is available online at
O
O
O
O
H
H
http://dx.doi.org/10.1055/s-0035-1562511.
S
u
p
p
ortiInfogrmoaitn
S
u
p
p
o
nrtogI
f
rmoaitn
(E)-33
(Z)-33
*
References and Notes
O
O
(1) Kametaka, T.; Perkin, A. G. J. Chem. Soc. Trans. 1910, 97, 1415.
(2) (a) Kuroda, C. J. Chem. Soc. 1930, 752. (b) Seshadri, T. R.; Thakur,
R. S. Curr. Sci. 1960, 29, 54. (c) Obara, H.; Onodera, J. Chem. Lett.
1979, 8, 201. (d) Obara, H.; Onodera, J.; Shirasaki, F. Chem. Lett.
1980, 9, 1095. (e) Takahashi, Y.; Miyasaka, N.; Tasaka, S.; Miura,
I.; Urano, S.; Ikura, M.; Hikichi, K.; Matsumoto, T.; Wada, M. Tet-
rahedron Lett. 1982, 23, 5163. (f) Obara, H.; Namai, S.; Machida,
Y. Chem. Lett. 1986, 15, 495. (g) Sato, S.; Obara, H.; Kumazawa,
T.; Onodera, J.; Furuhata, K. Chem. Lett. 1996, 25, 833. (h) Sato,
S.; Kumazawa, T.; Watanabe, H.; Takayanagi, K.; Matsuba, S.;
Onodera, J.; Obara, H.; Furuhata, K. Chem. Lett. 2001, 30, 1318.
(3) Suzuki, K. Chem. Rec. 2010, 10, 291.
H
(E)-34*
geometrically persistent
Sug
AcO
MeO
Sug
O
OAc
MeO
OMe
OAc
O
O
O
O
E
E
H
H
31
27
geometrically labile
Sug
(4) Obara, H.; Matsui, Y.; Namai, S.; Machida, Y. Chem. Lett. 1984,
13, 1397.
Sug
OH
(5) (a) Tamura, Y.; Yakura, T.; Haruta, J.; Kita, Y. J. Org. Chem. 1987,
52, 3927. For contributions from our laboratory, see: (b) Yasui,
Y.; Koga, Y.; Suzuki, K.; Matsumoto, T. Synlett 2004, 615.
(c) Yasui, Y.; Suzuki, K.; Matsumoto, T. Synlett 2004, 619.
(d) Morita, M.; Ohmori, K.; Suzuki, K. Org. Lett. 2015, 17, 5634.
For reviews, see: (e) Pouysegu, L.; Deffieux, D.; Quideau, S. Tet-
rahedron 2010, 66, 2235. (f) Roche, S. P.; Porco, J. A. Jr. Angew.
Chem. Int. Ed. 2011, 50, 4068.
MeO
OMe
OAc MeO
OMe
OAc
OBn
26
O
O
O
E/Z
E/Z
28
Figure 3 One-way photoisomerization
(6) (a) Rosenbrook, Wm. Jr.; Riley, D. A.; Lartey, P. A. Tetrahedron
Lett. 1985, 26, 3. (b) Posner, G. H.; Haines, S. R. Tetrahedron Lett.
1985, 26, 5.
In relation to the latter mechanism, examples that are
directly relevant to the Arai report were already noted in
the earlier stages of this study. Namely, hydroxychalcone 27
sharing a common structure with 33 showed the persistent
E-geometry of the double bond upon photoirradiation. On
the other hand, chalcones 26 and 28 lacking the hydrogen
bonding are geometrically labile toward photoirradiation.
In any event, these observations are interesting in their own
rights, but add an important guide to our effort toward the
total synthesis of 1.
(7) See Supporting Information.
(8) For early examples of the Friedel–Crafts C-glycosylation of phlo-
roglucinol derivatives, see: (a) Schmidt, R. R.; Hoffmann, M. Tet-
rahedron Lett. 1982, 23, 409. (b) Stewart, A. O.; Williams, R. M. J.
Am. Chem. Soc. 1985, 107, 4289. (c) Schmidt, R. R.; Effenberger,
G. Carbohydr. Res. 1987, 171, 59. (d) Matsumoto, T.; Katsuki, M.;
Suzuki, K. Tetrahedron Lett. 1989, 30, 833.
(9) For related examples, see: (a) Sato, S.; Nojiri, T.; Onodera, J. Car-
bohydr. Res. 2005, 340, 389. (b) Mitra, P.; Behera, B.; Maiti, T. K.;
Mal, D. J. Org. Chem. 2013, 78, 9748.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–G

G
T. Hayashi et al.
Letter
Synlett
(10) For the pertinent examples, see: (a) Fox, A. R.; Pausacker, K. H. J.
Chem. Soc. 1957, 295. (b) Spyroudis, S.; Tarantili, P. Tetrahedron
1994, 50, 11541.
(11) Prepared by the nucleophilic addition of the aryllithium species
to D-gluconolactone followed by the hydride reduction. For
details, see the Supporting Information. See also: Lewis, M. D.;
Cha, J. K.; Kishi, Y. J. Am. Chem. Soc. 1982, 104, 4976.
(12) Oyama, K.; Kondo, T. J. Org. Chem. 2004, 69, 5240.
(13) Griffith, W. P.; Ley, S. V.; Whitcombe, G. P.; White, A. D. J. Chem.
Soc., Chem. Commun. 1987, 1625.
(17) The reason for the stereoselectivity in the dearomatization step
is not clarified. Further study is in progress in our laboratory.
(18) The specific rotations of triketones 31 and 32 did not match the
reported values by Sato:2h 31: [α]D –4.5 (c 0.50, CHCl3), cf. lit.
24
[α]D24 –49.8 (c 1.02, CHCl3); 32: [α]D24 –47 (c 0.34, CHCl3), cf. lit.
[α]D24 –6.23 (c 0.995, CHCl3).
(19) For controversy of keto–enol form of quinochalcones, see: Feng,
Z. M.; He, J.; Jiang, J. S.; Chen, Z.; Yang, Y. N.; Zhang, P. C. J. Nat.
Prod. 2013, 76, 270.
(20) Arai, T.; Tokumaru, K. Chem. Rev. 1993, 93, 23.
(14) Solntsev, K. M.; McGrier, P. L.; Fahrni, C. J.; Tolbert, L. M.; Bunz,
U. H. F. Org. Lett. 2008, 10, 2429.
(15) Fletcher, S.; Gunning, P. T. Tetrahedron Lett. 2008, 49, 4817.
(16) Crystallographic data for compound 29 have been deposited
with the accession number CCDC 1472113 and can be obtained
free of charge via www.ccdc.cam.ac.uk/getstructures.
(21) (a) Arai, T.; Norikane, Y. Chem. Lett. 1997, 26, 339. (b) Norikane,
Y.; Itoh, H.; Arai, T. Chem. Lett. 2000, 29, 1094. For theoretical
treatment, see: (c) Norikane, Y.; Itoh, H.; Arai, T. J. Phys. Chem. A
2002, 106, 2766. (d) Norikane, Y.; Nakayama, N.; Tamaoki, N.;
Arai, T.; Nagashima, U. J. Phys. Chem. A 2003, 107, 8659.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–G