Studies towards the stereoselective α-hydroxylation of flavanones. Biosynthetic significance

Article abstract of DOI:10.1071/CH07325

The enolates of various propiophenones, chromanones, and also analogues of naturally occurring flavanones were stereoselectively hydroxylated at the ?-position, by employing commercially available enantiopure oxaziridines, to afford the desired ?-hydroxylated target molecules in good to exceptional stereoselectivities and in moderate to good chemical yields. A mechanistic rationale is presented to account for the stereoselectivities achieved. These in vitro results were tentatively related to the stereoselective biosynthesis of enantio-enriched dihydroflavonols while questions were raised about the authenticity of certain natural compounds. CSIRO 2008.

Full text of DOI:10.1071/CH07325

Full Paper
CSIRO PUBLISHING
Aust. J. Chem. 2008, 61, 122–130
www.publish.csiro.au/journals/ajc
Studies Towards the Stereoselective α-Hydroxylation
of Flavanones. Biosynthetic Significance
Zola-Michéle Border,A,B Charlene Marais,A Barend C. B. Bezuidenhoudt,A,C
and Jacobus A. SteenkampA
ADepartment of Chemistry, University of the Free State, PO Box 339, Bloemfontein 9300, South Africa.
BDedicated to the husband and close family of Z.-M. Border who passed away tragically
at the age of 31 before she could fully complete this project and her PhD.
CCorresponding author. Email: bezuidbc.sci@ufs.ac.za
The enolates of various propiophenones, chromanones, and also analogues of naturally occurring flavanones were stereo-
selectively hydroxylated at the α-position, by employing commercially available enantiopure oxaziridines, to afford the
desired α-hydroxylated target molecules in good to exceptional stereoselectivities and in moderate to good chemical
yields. A mechanistic rationale is presented to account for the stereoselectivities achieved. These in vitro results were
tentatively related to the stereoselective biosynthesis of enantio-enriched dihydroflavonols while questions were raised
about the authenticity of certain natural compounds.
Manuscript received: 12 September 2007.
Final version: 11 January 2008.
Introduction
Although the central role of the chalcone–flavanone pair has
gained general acceptance, the biogenetic origin of the 3-
hydroxyflavonoids (Scheme 1) may conceivably be attributed
to more than one route, the possibilities considered being via
chalcone epoxides, α-hydroxychalcones, or 3-hydroxylation of
flavanones.
It was suggested decades ago that chalcone epoxides may
play an important role in the stereoselective biosynthesis of
dihydroflavonols[5] and the success that has been achieved by
the in vitro mimicking of the chalcone epoxide route[6–10] has
understandably and justifiably created expectations that this is
indeed the reactive and biosynthetic intermediate towards the
synthesis of the abundant natural dihydroflavonols.
The synthesis of α-hydroxy carbonyl compounds of preselected
stereochemistry is of considerable interest, as this structural
array is featured in many bioactive molecules, such as vast num-
bers of flavonoids. Noteworthy in the structural assembly of
these flavonoids is that the stereochemistry of the majority of
flavonoids resides mainly in C2, C3, and C4 of the heterocyclic
C3-moiety (pyran ring). Emphasis in the present article is, how-
ever, placed on the stereochemistry of the C3 carbon bearing
the hydroxyl functionality, because the absolute configuration
of only this carbon atom is likely to be preserved under mild
reaction conditions.
Although the source of the C15-skeleton of flavonoids has
been beyond dispute for decades,[1–3] the sequence of changes
that result in the formation of a relatively diverse group of com-
pounds based on variation in the oxidation level of the C3-moiety
of the molecule remains a source of uncertainty despite impres-
sive progress in the understanding of flavonoid biosynthesis.[4]
Although evidence to unequivocally corroborate the exis-
tence of the elusive natural chalcone epoxides remains out-
standing, it is likely that revealing and irrefutable evidence
is lacking because of the high reactivity and hence short-
lived existence of such a proposed and unprotected epoxide.
OH
OH
O
O
O
O
OH
OH
2
3
2
OH
O
Chalcone
Flavanone
Dihydroflavonol
1
C2
2R
2S
C2 C3
2
3
4
5
6
7
2R,3R
2R,3S
2S,3S
2S,3R
Scheme 1.
© CSIRO 2008
10.1071/CH07325
0004-9425/08/020122

α-Hydroxylation of Flavanones
123
O-Glc
HO
H
OCH2OMe
2Ј
OMe
OMe
OMe
O-Glc
OH
HO
H
4
O
O
A
O
B
3
H
H
OMe
H
OH
O
H
OMe
OH
8
9
10
Fig. 1. Isolated chalconol glucosides and a typical chalcone epoxide.
results will be related to an outstanding issue concerning the
biosynthetic route to C3-hydroxylated flavonoids.
X
N
X
X
X
N
Results and Discussion
SO2
SO2
O
O
Various oxidative methods have been developed for the syn-
thesis of α-hydroxy carbonyl compounds from ketones.[31–35]
The versatility of metal enolates, the aprotic nature of N-
sulfonyloxaziridines, and their availability in enantiopure form
(e.g. 11) made this protocol ideally suited to the stereoselective
hydroxylation of flavanone analogues (Scheme 2).
(ϩ)-11
(Ϫ)-11
X ϭ H or Cl or OMe
Fig. 2. Available chiral N-sulfonyloxazindines.
Notable, however, is the recent isolation from Trifolium alexan-
drinum by Mohamed et al.[11] of the analogous chalcanol glu-
cosides, 2-methoxy-4,6-dihydroxy-αꢀ-chalcanol-α,β-epoxide-4-
O-β-d-glucopyranoside 8 and 2-methoxy-3,4,6-trihydroxy-αꢀ-
chalcanol-α,β-epoxide-4-O-β-d-glucopyranoside 9, which is
plausibly formed via the enzymatic reduction of the correspond-
ing chalcone epoxides (e.g. 10) (Fig. 1).[12]
Hence, the propiophenone analogues (Scheme 3, 12–19),
reminiscent of the flavanone structural array, were exposed to an
appropriate base (lithium diisopropyl amide (LDA)), the kinetic
enolates oxidized by an oxaziridine 11 (X = Cl), and the prod-
ucts separated and identified. The absolute configuration and
% enantiomeric excesses (% ees) were determined by derivati-
zation of the non-racemic α-hydroxy carbonyl compounds with
the (R)-(+)-Mosher acid chloride, utilizing a modified Mosher’s
method.∗ These results are summarized in Tables 1 and 2.
It is clear from the results that utilization of the
dichloro(camphorylsulfonyl)oxaziridine, (+)-11 (X = Cl), led
to a good stereochemical outcome, while the chemical yields
of the desired products were satisfactory.
Notable, however, is the erosion of the stereogenic integrity
as the hydroxylation of the aromatic ring is increased. Although
the yields (chemical and ees) of compounds 12 and 13 (Table 1,
entries 1 and 2) were good, a decline in % ee for compounds
14 and 15 (Table 1, entries 3 and 4) was evident. Not only were
the yield and % ee disappointingly low for compound 15, but
the absolute configuration was R compared with the S absolute
configuration of the other oxidation products. The most obvi-
ous argument to account for this inversion of configuration is
to embrace the notion that the opposite geometry (E-isomer)
of the enolate is operational under the prevailing conditions
(Scheme 4). It is, however, clear from an investigation by Davis
etal.[36] onthegeometryofvariousenolatesthatitishighlylikely
that the enolate of 15 is in the Z-configuration. In the absence of
evidence to the contrary, it is conceivable that the Z-isomer is the
kinetically favoured product at the prevailing low temperature,
whereas the E-isomer predominates at higher temperatures.
Davis et al.[36] proposed both planar (AS and CR) and spiro
(BS and DR) transition states for the attack of (+)-11 (X = Cl)
on either the Si- or Re-faces of Z-enolates with the transition
state that is more favourable under each unique circumstance
resulting in the major product.
Roux and Ferreira[13] suggested that α-hydroxychalcones,
of which only three have been isolated to date,[14–17] offer
a viable alternative route to 2,3-trans- as well as 2,3-cis-
dihydroflavonols. More recent evidence, however, indicates that
these are presumably not intermediates representing a first step
in the biological oxygenation of either a flavanone or chalcone,
but rather compounds resulting from the isomerisation and/or
ring contraction of already formed dihydroflavonols.[18]
The stereoselective biosynthetic C3 hydroxylation of fla-
vanones by 2-oxoglutarate-dependent dioxygenases (2-ODDs)
has been conclusively evaluated and reported on.[4,19–23] This
preferred enzymatic conversion of the abundant 2S-flavanones
3 into the 2R,3R-trans isomers 4,[24] and to a lesser extent
the 2R,3S-cis isomers 5,[22] is reflected by the preference for
these configurations in nature and is also corroborated by
the exclusive isolation of the 2R,3S-cis isomers 5 of certain
dihydroflavonols.[25–30]
Althoughwe, likethemajorityofchemists, takecognisanceof
the complexity of biochemical transformations, an investigation
into the chemical 3-hydroxylation of flavanones, utilizing mod-
ern synthetic methodologies such as powerful chiral reagents
or catalysts seems desirable. The general applicability of N-
sulfonyloxaziridines in the hydroxylation of enolates to afford
α-hydroxy compounds[31] offered a promising opportunity to
shed more light on this intriguing biosynthetic question. More-
over, with non-racemic oxaziridines 11 (Fig. 2), these hydroxy-
lations can be carried out with high asymmetric induction and
predictable stereochemistry.[31]
In the present paper, we describe details of a comprehen-
sive study of the asymmetric hydroxylation of propiophenone
analogues, 4-chromanones, and also of flavanones using non-
racemic (camphorylsulfonyl)oxaziridines. In addition, these
The prevailing protocol was extended to include the 4-
chromanones (Scheme 5, 20–24, Table 3). Although no obvious
discernible pattern emerged, it is worthwhile to note that sub-
stitution at the α-carbon (cf. 24) had a positive influence on the
∗ As both alcohol isomers are present in the reaction mixture, both of them are esterified with the (R)-(+)-Mosher acid chloride, which enables comparison
of the chemical shifts of corresponding protons of these isomers and thus allocation of the absolute configuration according to the established model. By the
same token, the % ee is conveniently calculated by the comparison of the integration values of the relevant protons (A. F. Hundt, J. F. W. Burger, J. P. Steynberg,
J. A. Steenkamp, D. Ferreira, Tetrahedron Lett. 1990, 31, 5073).

124
Z.-M. Border et al.
OMe
OMe
O
O
O
O
MeO
MeO
OH
Scheme 2.
R2
R1
R2
R1
R2
R1
O
OH
OH
LDA
ϩ
(ϩ)-11, X ϭ Cl
R3
O
R3
O
R3
S
R
R1 ϭ R2 ϭ R3 ϭ H
12
13
14
15
16
17
18
19
R1 ϭ R3 ϭ H, R2 ϭ OMe
R1 ϭ H, R2 ϭ R3 ϭ OMe
R1 ϭ R2 ϭ R3 ϭ OMe
Scheme 3.
Table 1. Oxidation of propiophenones
or
ee, enantiomeric excess
O
Ϫ
O
Ϫ
Entry
Substrate
Product
Yield [%]
ee [%]
Absolute
configuration
E
Z
1
2
3
12
13
14
15
16
17
18
19
89
95
91
45
88
91
70
28
S
S
S
R
Scheme 4.
enantioselectivity of the reaction (ee > 95%). Further, little or no
substitution on the aromatic ring seemed to negatively impact
on the enantioselectivity as well as the yield of the reaction
(cf. 20 and 21). In contrast to the propiophenones (see above),
chromanones can only form the enolates with the E-geometry.
The stereochemistry of the stereogenic centres was confirmed
by the Mosher protocol (Table 4) and that, together with the
various yields, can be rationalized in terms of the transition states
proposed by Davis et al.[36] and Davis and Chen.[31]
Noteworthy is the very satisfactory outcome of the hydroxy-
lation of chromanone 24, which may conceivably be attributed
to, among others, the greater stability of the more substituted
enolate compared with the enolate of chromanone 20.
Flavanones (e.g. 2 and 3) represent a small but salient group
of compounds in the flavonoid family. They are considered to
be the biogenetic precursors of the C3 hydroxylated analogues
(dihydroflavonols, e.g. 4–7). Hence, stereoselective hydroxyla-
tion of those flavanones via their respective enolates and the
utilization of a chiral oxidant could possibly shed some light on
this biosynthetic pathway.
Indeed, oxidation of the enolate of racemic 30 (Scheme 6)
with chiral oxaziridine (+)-11 (X = Cl), afforded the expected
dihydroflavonol 32, in 26% yield, with the 2,3-trans-2R,3R con-
figuration dominating (57% ee) as established by the Mosher
protocol (Table 5). Extension of the established protocol to
the racemic substituted flavanone 31 led to the expected di-
hydroflavonol 33 in 13% yield, with the 2,3-trans-2R,3R iso-
mer in excess (63% ee). Conspicuous is the absence of the
diastereomeric 2,3-cis isomers, reflecting a truly amazing 100%
diastereoselectivity in favour of the 2,3-trans isomers. The
absence of the thermodynamically less stable 2,3-cis isomers
is likely explicable in terms of a strong and supporting syner-
gism between a preferred fit of the oxaziridine and the formed
enolate and a dynamic stereo-directing effect of the C2 phenyl
ring. Such a transitional spatial arrangement would govern a
4
biased electrophilic attack of the oxaziridine such that the ‘oxy-
gen’ is delivered to the opposite side of the bulky phenyl ring,
hence the exclusive formation of the 2,3-trans products. This
notable ability of the C2 phenyl group to direct selected reagents
to the ‘opposite’ side of itself (2,3-trans-type intermediate) is a
structural feature in certain flavonoids that plays a pivotal role
in many of their reactions, details of which will be discussed
elsewhere.
A disappointing feature of these reactions was the low
yields. This is, however, explicable in terms of a competitive
base-catalyzed opening of the heterocyclic ring to afford the cor-
responding chalcones, the intermediate shifting progressively
from an enolate (34) to a carbanion (35) (Scheme 7) with
increasing base concentration.[37]
Based on these observations, it was envisaged that the selec-
tion of a suitable substrate could lead to an outstanding degree
of stereoselection. Indeed, oxidation of enantiopure flavanone
37 afforded the expected dihydroflavonol 32 (Scheme 8) in
56% yield, and only the trans-2R,3R stereoisomer in 100%
diastereomeric excess (de), but also in 100% ee. Although the
likelihood of scrambling of the established chirality of the enan-
tiopure products via ring-opening or enolization is possible
under the prevailing conditions, the high % ee attained is an
unmistakable indication of a highly successful stereoselective
α-hydroxylation.
Conclusion
In conclusion, it was established (see above) that the enolate
of flavanones are stable under mild conditions and that the
stereochemistry of the formed enantioenriched dihydroflavonol
remained intact under mild workup and separation conditions.

α-Hydroxylation of Flavanones
125
Table 2. 1H NMR data for the (S)-(−)-MTPA esters of 16–19 in CDCl3 (J values in Hz)
(R) and (S) notation refers to the configuration of the newly formed chiral centre. MTPA, α-methoxy-α-(trifluoromethyl)phenyl acetic
Proton
16
17
18
19
ArH
7.96–7.36 (10H, m)
H2ꢀ, H6ꢀ (R)
H2ꢀ, H6ꢀ (S)
H3ꢀ, H5ꢀ (R)
H3ꢀ, H5ꢀ (S)
H6ꢀ (R)
7.93 (2H, d, J 9.0)
7.89 (2H, d, J 9.0)
6.95 (2H, d, J 9.0)
6.95 (2H, d, J 9.0)
6.07 (2H, s)
6.01 (2H, s)
7.90 (1H, d, J 8.8)
7.87 (1H, d, J 8.8)
6.56 (1H, dd, J 2.5, 8.8)
6.54 (1H, dd, J 2.5, 8.8)
6.44 (1H, d, J 2.5)
6.43 (1H, d, J 2.5)
7.71–7.59 (2H, m)
7.42–7.37 (3H, m)
H6ꢀ (S)
H5ꢀ (R)
H5ꢀ (S)
H3ꢀ (R)
H3ꢀ (S)
MTPA ArH
7.68–7.56 (2H, m)
7.42–7.38 (3H, m)
6.09 (1H, q, J 7.0)
7.59–7.48 (2H, m)
7.54–7.30 (3H, m)
5.96 (1H, q, J 7.0)
H2
H2 (R)
H2 (S)
H3 (S)
H3 (R)
MTPAOMe (R)
MTPAOMe (S)
OMe (S)
OMe (R)
6.13 (1H, q, J 7.0)
6.13 (1H, q, J 7.0)
1.62 (3H, d, J 7.0)
1.55 (3H, d, J 7.0)
3.64 (3H, m)
6.17 (1H, q, J 7.0)
6.19 (1H, q, J 7.0)
1.53 (3H, d, J 6.8)
1.45 (3H, d, J 6.8)
3.65 (3H, m)
3.58 (3H, m)
3.89 (3H, s), 3.84 (3H, s)
3.89 (3H, s), 3.85 (3H, s)
1.61 (3H, d, J 7.0)
1.53 (3H, d, J 7.0)
3.66 (3H, m)
3.58 (3H, m)
3.85 (3H, s)
1.52 (3H, d, J 7.5)
1.42 (3H, d, J 7.5)
3.53 (3H, m)
3.57 (3H, m)
3.48 (3H, m)
3.81 and 3.80 (9H, 2 × s)
3.86 (3H, s)
3.74 and 3.73 (9H, 2 × s)
R5
O
R5
R1
R5
R5
R5
OH
R4
R1
R2
R1
O
O
O
R5
NHMDS
(ϩ)-11, X ϭ Cl
ϩ
OH
R2
R2
R4
R4
R3
O
R3
R3
O
R
S
R1 ϭ R2 ϭ R3 ϭ R4 ϭ R5 ϭ H
R1 ϭ OMe, R2 ϭ R3 ϭ R4 ϭ R5 ϭ H
R1 ϭ R3 ϭ OMe, R2 ϭ R4 ϭ R5 ϭ H
20
21
22
23
24
25
26
27
28
29
R1 ϭ OEt, R2 ϭ OMe, R3 ϭ R4 ϭ H, R5 ϭ Me
R1 ϭ R2 ϭ R3 ϭ R5 ϭ H, R4 ϭ Me
Scheme 5.
Table 3. Oxidation of chromanones
and the proximity of the oxidizing agent, find a precedent in the
stereoselective C3 hydroxylation of flavanones by biosynthetic
enzymes. Different mechanisms of oxidation are applicable
although, as an SN2 mechanism is presumably applicable in
the case of the N-sulfonyloxaziridines,[31] whereas a radical
mechanism prevails in the case of the FeII 2-ODDs[20] (which
utilize 2-oxoglutarate as co-substrate to achieve the two-electron
oxidation of the substrate).
Scrutiny of a dihydroflavonol (DHF) structure (e.g. 38 in
Scheme 9) reveals the presence of labile bonds, the cleavage of
which can have a significant effect on the stereochemistry of a
particular compound. It is conceivable that the cleavage of the
O1–C2 bond, ably assisted by the inductive effect of the 4ꢀ-OH
group (as the phenoxide), is more likely than the competing
removal of H3 (C), which, in turn, is synchronized with the
departure of the relatively good leavingA-ring (benzoyl moiety)
to afford the conjugated C2–C3 double bond. For the H3 (C) to
act as the counterpart for the A-ring for elimination of these two
groups with the formation of the α-hydroxychalcone (e.g. 42),
they must be able to form an antiperiplanar-type transition state
to accomplish the favourable orbital alignment – a prerequisite
that will not be satisfied in all the cases under discussion. In con-
trast, it is clear that the formation of the quinone methide (e.g. 39)
Entry
Substrate
Product Yield [%]
ee [%]
Absolute
configuration
1
2
3
4
5
20
21
22
23
24
25
26
27
28
29
61
74
82
23
74
20
3.4
69
62
>95
S
R
R
R
R
Furthermore, it was confirmed that there exists a preferen-
tial facial attack on the enolate by the enantiopure oxazaridine
(+)-11 and that if such a propensity is synergistically sup-
ported by the crucial and dominating directing effect of the
bulky B-ring towards a hydroxylation step to the ‘opposite’
side, the stereochemical induction is amplified to an exception-
ally high level with the overwhelming dominance of the trans
isomers. Moreover, this investigation culminated in the success-
ful C3 hydroxylation of an enantiopure flavanone to yield the
enantiopure dihydroflavonol in acceptable yield.
The influence of the ‘fit’of the substrate and oxidizing agent,
as well as the influence of the stereochemistry of the B-ring

126
Z.-M. Border et al.
Table 4. 1H NMR data of the (S)-(−)-MTPA esters of 25–29 in CDCl3 (J values in Hz)
(R) and (S) notation refers to the configuration of the newly formed chiral centre. MTPA, α-methoxy-α-(trifluoromethyl)phenyl acetic
Proton
ArH
25
26
27
28
29
7.90–6.98 (9H, m)
7.83–6.40 (8H, m)
7.68–7.64 (2H, m)
7.45–7.37 (3H, m)
6.08–6.04 (2H, m)
7.80–7.73 (2H, m;
3S-isomer)
7.69–7.62 (2H, m,
3R-isomer)
7.97–6.96 (9H, m)
7.47–7.39 (3H, m)
7.18 (1H, s)
6.39 (1H, s)
H3
5.73 (1H, s)
H3 (3S)
H3 (3R)
H2
5.92 (1H, dd, J 6.0, 12.5)
5.84 (1H, dd, J 6.0, 12.5)
5.86 (1H, dd, J 6.0, 11.5)
5.78 (1H, dd, J 6.0, 11.5)
5.74 (1H, dd, J 6.0, 10.5)
5.68 (1H, dd, J 6.0, 11.0)
5.04 (1H, d, J 10.5)
4.10 (1H, d, J 10.5)
H2eq (3R)
H2ax (3R)
H2eq (3S)
H2ax (3S)
MTPAOMe (3S)
MTPAOMe (3R)
OMe
4.62 (1H, dd, J 6.0, 11.0)
4.53 (1H, dd, J 11.0, 12.5)
4.49 (1H, dd, J 6.0, 11.0)
4.36 (1H, dd, J 11.0, 12.5)
3.69 (3H, m)
4.59 (1H, dd, J 6.0, 11.0)
4.51 (1H, dd, J 11.0, 11.5)
4.47 (1H, dd, J 6.0, 11.0)
4.34 (1H, dd, J 11.0, 11.5)
3.68 (3H, m)
4.54 (1H, dd, J 6.0, 11.0)
4.47 (1H, t, J 11.0)
4.40 (1H, dd, J 6.0, 11.0)
4.39 (1H, dd, J 10.5, 11.0)
3.70 (3H, m)
3.66 (3H, m)
3.59 (3H, m)
3.83 (3H, s)
3.67 (3H, m)
3.59 (3H, m)
3.58 (3H, m)
3.60 (3H, m)
3.83 (3H, s)
3.55 (3H, m)
3.86 (3H, s)
3.82 (3H, s)
CH2CH3
CH2CH3
2-CH3
4.11 (2H, q, J 7.0)
1.48 (3H, t, J 7.0)
1.56 (3H, s)
1.35 (3H, s)
3-CH3
2.15 (3H, s)
R
O
O
R
O
R
O
O
NHMDS
ϩ
(ϩ)-11, X ϭ Cl
OH
OH
O
2R,3R
2S,3S
30
31
32
33
R ϭ H
R ϭ OMe
Scheme 6.
O
O
Ϫ
Ϫ
O
34
O
H
O
H
O
O
36
O
30
Ϫ
35
Scheme 7.
is accomplished in a relatively facile mode as the free rotation
of the B-ring assists the formation of a transition state in which
the obligatory orbital arrangement is possible. It is also worth
mentioning the presence of the relatively strong intramolecular
H-bond between the carbonyl functionality and the α-hydroxyl
group. The planar character of this formed five-membered ring
must have an influence on the preferred conformation of the C-
ring, hence exerting an influence on the necessary orbital overlap
to influence significantly the formation of the enol tautomer
(cf. the H3 (C) of trans- and cis-DHF). The structural features
mentioned (see above) act to abate the probable scrambling of
the formed stereogenic centres of those natural DHFs. On the
contrary, when the C-ring has opened to form the open-chain
analogue, most of the conformational restrictions mentioned
lapse, hence enabling the optimal orbital overlap to afford in
thermodynamic ratios E- and Z-α-hydroxychalcones and hence

α-Hydroxylation of Flavanones
127
all the possible stereoisomers of the DHF, whereas the same α-
hydroxychalcones may lead to the diketone 43 and hence the
coumaranone 44 (Scheme 9).
probability exists that compounds such as α-hydroxychalcones
and 2-benzyl-2-hydroxy-3-coumaranones are artefacts rather
than natural products.
These arguments are corroborated by in vitro retention of
configuration at C3[18] and the absence of deuterium incorpora-
tion at C3[38] under mild conditions, as well as the co-occurrence
of non-racemic cis- and trans-dihydroflavonol isomer pairs with
similar C3 absolute configurations in nature,[18,39] which sug-
gests the formation of a quinone methide intermediate 39 for
cis–trans isomerization[40] of 3-hydroxyflavanols under mild
conditions. 2-Benzyl-2-hydroxy-3-coumaranone 44, however,
forms only on harsh treatment,[18] which suggests the forma-
tion of an α-hydroxychalcone 42 as intermediate under those
circumstances.
The authors thus feel confident to suggest that the most prob-
able biosynthetic route towards trans- and cis-dihydroflavonols
entails the 2-ODD-catalyzed hydroxylation of flavanones,
whereas in vivo isomerization of dihydroflavonols via the
quinone methide and especially via the α-hydroxychalcone still
need to be confirmed.
Table 5. 1H NMR data of the (S)-(−)-MTPA esters of 32 and 33 in
CDCl3 (J values in Hz)
(R) and (S) notation refers to the configurations of the chiral centres at
C2 and C3. MTPA, α-methoxy-α-(trifluoromethyl)phenyl acetic
A few α-hydroxychalcones[13–17] and 2-benzyl-2-hydroxy-
3-coumaranones[14,16,41–43] have been reported to be isolated
from plants, but as the conditions under which flavonoids are
isolated (Soxhlet extraction, high temperature, acid content of
EtOAc, evaporation under reduced pressure at ∼50◦C, etc.)
are comparable with the harsh conditions considered above,
the authors feel modestly confident to suggest that a distinct
Proton
32
33
ArH
7.94–7.04 (14H, m)
7.63–7.20 (10H, m)
6.53 (1H, d, J 2.5)
6.50 (1H, d, J 2.5)
6.71 (1H, dd, J 2.5, 9.0)
6.713 (1H, dd, J 2.5, 9.0)
7.89 (1H, d, J 9.0)
7.90 (1H, d, J 9.0)
6.04 (1H, d, J 12.5)
6.13 (1H, d, J 12.5)
5.51 (1H, d, J 12.5)
5.40 (1H, d, J 12.5)
3.87 (3H, s)
H-8 (2R,3R)
H-8 (2R,3S)
H-6 (2R,3R)
H-6 (2R,3S)
H-5 (2R,3R)
H-5 (2R,3S)
H-3 (2R,3R)
H-3 (2R,3S)
H-2 (2R,3R)
H-2 (2R,3S)
7-OMe
O
O
6.04 (1H, d, J 12.5)
6.15 (1H, d, J 12.5)
5.50 (1H, d, J 12.5)
5.42 (1H, d, J 12.5)
NHMDS
(ϩ)-11, X ϭ Cl
OH
O
O
(2R,3R)-32
(2S)-37
MTPAOMe (2R,3R)
MTPAOMe (2R,3S)
3.33 (3H, m)
3.55 (3H, m)
3.38 (3H, m)
3.57 (3H, m)
Scheme 8.
OH
OH
OH
HO
O
HO
O
O
OH
OH
OH
OH
O
41
OH
38
OH
OH
O
HO
OH
O
HO
OH
O
OH
OH
OH
OH
42
OH
39
α-Hydroxychalcone
Quinone methide
OH
OH
OH
OH
HO
OH
HO
O
O
O
OH
O
43
OH
OH
40
OH
OH
HO
O
OH
O
44
OH
2-Benzyl-2-hydroxy-3-coumaranone
Scheme 9.

128
Z.-M. Border et al.
Experimental
acrylonitrile and Triton (40% benzyltrimethylammonium
hydroxide in methanol), to Hoesch reaction conditions[45]
afforded the title compound[46] 22 as an amorphous solid (3.8 g,
18 mmol, 52%). δH (C3D6O) 6.15 (1H, s, H6 or H8), 6.08 (1H,
s, H6 or H8), 4.43 (2H, t, J 6.2, H2), 3.83 (3H, s, 1 × OMe),
3.79 (3H, s, 1 × OMe), 2.58 (2H, t, J 6.2, H3). Found: [M+] m/z
208.0740. C11H12O4 requires M 208.0736.
General
1H NMR spectra were recorded on a Bruker AM-300 spec-
trometer for solutions as indicated with Me4Si as internal
standard. J values are given in Hz. Mass spectra were obtained
with a VG 70-70 instrument. TLC was performed on pre-
coated Merck plastic sheets (silica gel 60 PF254, 0.25 mm) and
the plates were sprayed with H2SO4–HCHO (40:1 v/v) after
development. Preparative plates (PLC), 20 × 20 cm, Kieselgel
PF254 (1.0 mm) were air-dried and used without prior activa-
tion. Column chromatography was on silica (Merck Kieselgel
60, 230–400 mesh) in various columns, solvent systems, and
flow rates (to be specified in each instance) under the influ-
ence of gravity or pressure from a nitrogen cylinder in the
case of flash column chromatography. Diazomethane methy-
lations were performed with an excess of diazomethane in
MeOH/diethyl ether over 48 h at −15◦C. Evaporations were
done under reduced pressure at ambient temperatures in a rotary
evaporator. 3-Methoxyphenol, 2,5-dimethoxyphenol, 2ꢀ,4ꢀ,6ꢀ-
trihydroxyacetophenone, 2ꢀ,4ꢀ-dimethoxyacetophenone, propio-
phenone 12, 4ꢀ-methoxypropiophenone 13, 4-chromanone
20, 7-ethoxy-6-methoxy-2,2-dimethyl-4-chromanone 23, 7-
methoxyflavanone 31, and flavanone 30 were obtained
from Aldrich, whereas (+)-(8,8-dichlorocamphorylsulfonyl)
oxaziridine (+)-11 was obtained from Merck. 2ꢀ,4ꢀ,6ꢀ-
Trimethoxyacetophenone[36] was prepared by the methyl-
ation of 2ꢀ,4ꢀ,6ꢀ-trihydroxyacetophenone with dimethylsulfate
in dry acetone. 2ꢀ-Hydroxy-4ꢀ-methoxyacetophenone was pre-
pared by the regioselective diazomethane methylation of 2ꢀ,4ꢀ-
dihydroxyacetophenone.
3-Methylchroman-4-one 24
Anenolateofchroman-4-one(2.0 g, 13 mmol)inanhydrousTHF
(75 mL) was prepared by transferring the chroman-4-one solu-
tion to freshly prepared LDA (1.2 equiv.) in dry THF (20 mL)
at −78◦C, whereafter MeI (1.2 equiv.) was added. After 0.5 h,
the reaction temperature was increased to −40◦C for 1 h and
finally to 0◦C for 2–6 h. Following completion of the reaction
according to TLC, NH4Cl was added and the reaction mixture
extracted three times with ethyl acetate. The organic layer was
concentrated under vacuum and the product purified by means
of column chromatography with hexane/ethyl acetate/acetone
(9:0.5:0.5, v/v) to yield the title compound 24 as an amorphous
1
solid (1.53 g, 9.43 mmol, 73%) and H NMR data consistent
with those published.[48] Found: [M+] m/z 162.0679. C10H10O2
requires M 162.0681.
General Oxidation Procedure[36]
The appropriate base (LDA or 0.88 M N-sodiohexamethyldisi-
lazane (NHMDS)) in freshly distilled dry THF was cooled to
−78◦C, whereafter a solution of the ketone in THF was slowly
transferred to the base and the mixture stirred for ∼30 min in
an argon atmosphere. The mixture was warmed to −40◦C, kept
at this temperature for 0.5 h, and cooled to −78◦C, whereafter
a solution of (+)-(8,8-dichlorocamphorylsulfonyl)oxaziridine
(+)-11 (1.2 equiv. propiophenones, 1.5 equiv. chromanones and
7-methoxyflavanone 31, or 1.7 equiv. flavanone 30) in dry THF
was added slowly over 20 min. The reaction was subsequently
quenched by the addition of a saturated NH4Cl solution, diluted
with diethyl ether at −78◦C, and allowed to reach room tem-
perature. The aqueous layer was extracted with diethyl ether
(3 × 25 mL) and the combined organic layers were washed
with saturated aq. Na2S2O3 (2 × 20 mL) and brine solutions
(2 × 20 mL). The organic layer was dried with Na2SO4, filtered,
and the filtrate concentrated under reduced pressure.
Preparation of Propiophenones 14 and 15
LDA was prepared by standard procedure from equivalent
amounts of dry diisopropylamine and butyllithium (BuLi) in
freshly distilled anhydrous tetrahydrofuran (THF) at 0◦C over
∼10 min under argon and subsequently cooled to −78◦C, where-
after a solution of the appropriate acetophenone, in dry THF,
was slowly transferred to the base and stirred for ∼30 min in
an argon atmosphere to form an enolate. MeI was added and
the temperature kept at −78◦C for 0.5 h, then at −40◦C for
1 h and finally at 0◦C for 2–6 h. Following completion of the
reaction according to TLC, NH4Cl was added and the reac-
tion mixture extracted with diethyl ether (×3). The organic
layer was concentrated under vacuum and the products puri-
fied to afford 2ꢀ,4ꢀ-dimethoxypropiophenone 14 with NMR
data entirely consistent with those published[44] and 2ꢀ,4ꢀ,6ꢀ-
trimethoxypropiophenone 15. δH (CDCl3) 6.08 (2H, s, H3ꢀ and
H5ꢀ), 3.80 (3H, s, 1 × OMe), 3.75 (6H, s, 2 × OMe), 2.72 (2H,
q, J 7, H2) and 1.11 (3H, t, J 7, H3) in acceptable yields.
Silylation and Desilylation[49,50]
To facilitate the separation of the substrate chromanone (except
25) and 7-methoxyflavanone 31 from the respective reaction
products, silylation was performed after workup of the oxidation
mixture. The residue of the oxidation mixture was dried in ben-
zene, after which 4-(dimethylamino)pyridine (DMAP; 5 mg),
imidazole (2 equiv.), tert-butyldimethylsilyl chloride (TBDM-
SCl; 1.2 equiv. for the chromanones and 1.5 for the flavanone),
and dry THF (2 mL) were added. The reaction mixture was
stirred for 12 h, after which ether (50 mL) was added. Filtration,
followed by concentration under vacuum and flash chromato-
graphywithhexane/acetone(9:1, v/v)yieldedthepuresilylether.
This derivative was dissolved in a minimum volume of dry THF
(∼2 mL), whereafter tetrabutylammonium fluoride (TBAF; 1.5
equiv.) was added. After 5 min, a saturated aq. NH4Cl solu-
tion was added and the reaction mixture extracted with ether
(3 × 20 mL). The combined organic layers were dried (Na2SO4)
and concentrated under vacuum to afford the α-hydroxy ketone.
7-Methoxychroman-4-one 21
Exposure of 3-(3ꢀ-methoxyphenoxy)propionitrile (5.0 g,
28 mmol), prepared in 66% yield from 3-methoxyphenol, acry-
lonitrile, andTriton (40% benzyltrimethylammonium hydroxide
in methanol), to Hoesch reaction conditions[45] afforded the title
compound[46] 21 as an amorphous solid (3.6 g, 24 mmol, 70%),
the 1H NMR data completely consistent with those published.[47]
Found: [M+] m/z 178.0633. Calc. for C10H10O3 178.0630.
5,7-Dimethoxychroman-4-one 22
Exposure of 3-(3ꢀ,5ꢀ-dimethoxyphenoxy)propionitrile (7.4 g,
36 mmol), prepared in 70% yield from 3,5-dimethoxyphenol,

α-Hydroxylation of Flavanones
129
Absolute Configuration and ee Determinations
(R)-3-Hydroxy-7-methoxychroman-4-one 26
The(S)-(−)-α-methoxy-α-(trifluoromethyl)phenylaceticderiva-
tives ((S)-(−)-MTPA) of the α-hydroxy compounds were pre-
pared according to the standard procedure[51] and separated.
Analysis of the NMR spectra of these MTPA esters not only
establishedtheirstructures, butalsogaveunambiguouslytheper-
centage ee as well as the absolute configuration of the dominant
stereoisomer.[52]
Application of the oxidation procedure to 7-methoxychroman-4-
one 21 (219.1 mg, 1.230 mmol) in anhydrousTHF (2.5 mL) with
0.88 M NHMDS (1.5 equiv.) in dryTHF afforded, after silylation
and desilylation purification, the title compound 26 as colourless
needles (176.7 mg, 0.910 mmol, 73.98% yield and 3.4% ee). mp
104–106◦C. δH (CDCl3) 7.78 (1H, d, J 9.0, H5), 6.60 (1H, dd,
J 9.0 and 2.8, H6), 6.39 (1H, d, J 2.8, H8), 4.62 (1H, dd, J 10.0
and 6.5, H3), 4.52 (1H, dd, J 13.0 and 6.5, H2eq), 4.08 (1H, dd, J
13.0 and 10.0, H2ax), 3.82 (3H, s, 1 × OMe). Found: [M+] m/z
194.0584. C10H10O4 requires M 194.0579.
(S)-α-Hydroxypropiophenone 16
Application of the oxidation procedure to propiophenone 12
(201.3 mg, 1.50 mmol), in anhydrous THF (8 mL), with LDA
(1.2 equiv.) in dry THF (2 mL) afforded, after flash chromato-
graphy purification with hexane/ethyl acetate/acetone (8:1:1,
v/v), the title compound 16 as an amorphous solid (200.5 mg,
(R)-5,7-Dimethoxy-3-hydroxychroman-4-one 27
Application of the oxidation procedure to 5,7-dimethoxy-
chroman-4-one 22 (256.0 mg, 1.230 mmol) in anhydrous THF
(2.5 mL) with 0.88 M NHMDS (1.5 equiv.) in dry THF afforded,
after silylation and desilylation purification, the title compound
27 as colourless needles (226.4 mg, 1.01 mmol, 82.11% yield
and 69% ee). mp 121–123◦C. δH (CDCl3) 6.04 (1H, d, J 2.0, H6
or H8), 6.02 (1H, d, J 2.0, H6 or H8), 4.56 (1H, dd, J 7.1 and
4.8, H3), 4.39 (1H, dd, J 9.8 and 4.8, H2eq), 3.99 (1H, dd, J 9.8
and 7.1, H2ax), 3.86 (3H, s, 1 × OMe), 3.80 (3H, s, 1 × OMe).
Found: [M+] m/z 224.0684. C11H12O5 requires M 224.0685.
1
1.335 mmol, 89.00% yield and 88% ee), with H NMR data
consistent with those published.[53] Found: [M+] m/z 150.0682.
C9H10O2 requires M 150.0681.
(S)-α-Hydroxy-4ꢀ-methoxypropiophenone 17
Application of the oxidation procedure to 4ꢀ-methoxypropio-
phenone 13 (98.9 mg, 0.602 mmol) in anhydrous THF (5 mL)
with LDA (1.2 equiv.) in dry THF (2 mL) afforded, after flash
chromatography purification with hexane/ethyl acetate/acetone
(8:1:1, v/v), the title compound 17 as an amorphous solid
(103.3 mg, 0.5732 mmol, 95.2% yield and 91% ee). δH (CDCl3)
7.90 (2H, d, J 7.0, H2ꢀ), 6.95 (2H, d, J 7.0, H3ꢀ), 5.09 (1H, q, J
7.0, H2), 3.83 (3H, s, 1 × OMe), 1.42 (3H, d, J 7.0, H3). Found:
[M+] m/z 180.0782. C10H12O3 requires M 180.0786.
(R)-7-Ethoxy-3-hydroxy-6-methoxy-2,2-dimethylchroman-
4-one 28
Application of the oxidation procedure to 7-ethoxy-6-methoxy-
2,2-dimethylchroman-4-one 23 (307.7 mg, 1.229 mmol) in
anhydrous THF (2.5 mL) with 0.88 M NHMDS (1.5 equiv.)
in dry THF afforded, after silylation and desilylation purifi-
cation, the title compound 28 as colourless needles (75.3 mg,
0.283 mmol, 23.0% yield and 62% ee). mp 119–121◦C. δH
(CDCl3) 7.15 (1H, s, H5 or H8), 6.35 (1H, s, H5 or H8), 4.33 (1H,
s, H3), 4.10 (2H, q, J 7.0, CH3CH2O), 3.84 (3H, s, 1 × OMe),
1.60 (3H, s, 2-CH3), 1.47 (3H, t, J 7.0, CH3CH2O), 1.19 (3H,
s, 2-CH3). Found: [M+] m/z 266.1152. C14H18O5 requires M
266.1154.
(S)-α-Hydroxy-2ꢀ,4ꢀ-dimethoxypropiophenone 18
Applicationoftheoxidationprocedureto2ꢀ,4ꢀ-dimethoxypropio-
phenone 14 (100.0 mg, 0.5149 mmol) in anhydrous THF (5 mL)
with LDA (1.2 equiv.) in dry THF (2 mL) afforded, after flash
chromatography purification with hexane/ethyl acetate/acetone
(7:2:1, v/v), the title compound 18 as an amorphous solid
(98.7 mg, 0.469 mmol, 91.1% yield and 70% ee). δH (CDCl3)
7.89 (1H, d, J 8.9, H6ꢀ), 6.56 (1H, dd, J 8.9 and 2.8, H5ꢀ), 6.44
(1H, d, J 2.8, H3ꢀ), 5.10 (1H, q, J 7.0, H2), 3.88 (3H, s, 1 × OMe),
3.85 (3H, s, 1 × OMe), 1.30 (3H, d, J 7.0, H3). Found: [M+] m/z
210.0889. C11H14O4 requires M 210.0892.
(R)-3-Hydroxy-3-methylchroman-4-one 29
Application of the oxidation procedure to 3-methylchroman-4-
one[55] 24 (216.7 mg, 1.336 mmol) in anhydrous THF (2.5 mL)
with 0.88 M NHMDS (1.5 equiv.) in dry THF afforded, after
silylation and desilylation purification, the title compound 29
as an amorphous solid (177 mg, 0.995 mmol, 74.48% yield and
>95% ee). δH (CDCl3) 7.90–7.84 (1H, m, ArH), 7.55–7.47 (1H,
m, ArH), 7.09–6.95 (2H, m, ArH), 4.29 (2H, d, J 11.2, H2), 4.18
(2H, d, J 11.2, H2), 2.15 (3H, s, 1 × OMe). Found: [M+] m/z
178.0636. C10H10O3 requires M 178.0630.
(R)-α-Hydroxy-2ꢀ,4ꢀ,6ꢀ-trimethoxypropiophenone 19
Application of the oxidation procedure to 2ꢀ,4ꢀ,6ꢀ-trimethoxy-
propiophenone 15 (84.4 mg, 0.376 mmol) in anhydrous THF
(5 mL) with LDA (1.2 equiv.) in dry THF (2 mL) afforded,
after flash chromatography purification with hexane/ethyl
acetate/acetone (7:2:1, v/v), the title compound 19 as an amor-
phous solid (40.5 mg, 0.169 mmol, 44.9% yield and 28% ee). δH
(CDCl3) 6.12 (2H, s, H3ꢀ and H5ꢀ), 4.79 (1H, q, J 7.0, H2), 3.84
(3H, s, 1 × OMe), 3.78 (3H, s, 1 × OMe), 1.28 (3H, d, J 7.0, H3).
Found: [M+] m/z 240.1001. C12H16O5 requires M 240.0998.
(2R,3R)-Dihydroflavonol 32
Application of the oxidation procedure to flavanone 30
(200.0 mg, 0.8918 mmol) in anhydrousTHF (5 mL) with 0.88 M
NHMDS (1.2 equiv.) in dry THF afforded, after flash chromato-
graphy purification with hexane/ethyl acetate/acetone (90:5:5,
v/v), the title compound 32 as an amorphous solid (55.6 mg,
0.231 mmol, 25.9% yield, 57% ee) and 1H NMR data consistent
with those published.[56] Found: [M+] m/z 240.0786. C15H12O3
requires M 240.0787.
(R)-3-Hydroxychroman-4-one 25
Application of the oxidation procedure to chroman-4-one 20
(100.0 mg, 0.6749 mmol) in anhydrous THF (2.5 mL) with
0.88 M NHMDS (1.5 equiv.), afforded, after flash chromato-
graphy purification with hexane/acetone (9:1, v/v), the title
compound 25 as colourless needles (67.3 mg, 0.410 mmol,
60.7% yield and 20% ee). mp 58–60◦C (lit.[54] 57–58◦C) with
1H NMR data consistent with those published.[54] Found: [M+]
m/z 164.0476. C9H8O3 requires M 164.0474.
In a similar procedure, oxidation of (2S)-flavanone 37[57]
(200.0 mg, 0.8918 mmol) afforded the title compound 32
as white needles[56] (119.7 mg, 0.4982 mmol, 55.86% yield,
100% ee).

130
Z.-M. Border et al.
(2R,3R)-7-Methoxydihydroflavonol 33
[25] S. Öksüz, G.Topçu, Phytochemistry 1992, 31, 195. doi:10.1016/0031-
9422(91)83034-I
[26] L. Y. Foo, J. Chem. Soc. Chem. Commun. 1986, 675. doi:10.1039/
C39860000675
[27] L. Y. Foo, Phytochemistry 1987, 26, 813. doi:10.1016/S0031-
9422(00)84793-7
[28] F. Bohlmann, C. Zdero, Phytochemistry 1983, 22, 2877. doi:10.1016/
S0031-9422(00)97725-2
[29] X. A. Dominguez, J. Nat. Prod. 1986, 49, 143. doi:10.1021/
NP50043A019
[30] Md. Tofazzal Islam, S. Tahara, Phytochemistry 2000, 54, 901.
doi:10.1016/S0031-9422(00)00048-0
[31] F. A. Davis, B.-C. Chen, Chem. Rev. 1992, 92, 919. doi:10.1021/
CR00013A008
[32] G. M. Rubottom, M. A. Vazquez, D. R. Pelegrina, Tetrahedron Lett.
1974, 15, 4319. doi:10.1016/S0040-4039(01)92153-7
[33] L. A. Paquette, H.-S. Lin, M. J. Coglan, Tetrahedron Lett. 1987, 28,
5017. doi:10.1016/S0040-4039(00)96684-X
Application of the oxidation procedure to 7-methoxyflavanone
31 (254.3 mg, 1.00 mmol) in anhydrousTHF (5 mL) with 0.88 M
NHMDS (1.5 equiv.) in dry THF, afforded, after silylation and
desilylation purification, the title compound 33 as an amor-
phous solid (35.1 mg, 0.130 mmol, 13.0% yield, 63% ee), with
1H NMR data completely consistent with those published.[58]
Found: [M+] m/z 270.0893. C16H14O4 requires M 270.0892.
Acknowledgements
Support from the Sentrale Navorsingsfonds, of this University, and the Foun-
dationforResearchandDevelopment, Pretoria, isacknowledged.Wearealso
grateful to Professor D. Ferreira, formerly from this University, for financial
support to Z.-M.B. and fruitful discussions during the execution of the work
leading to the present manuscript.
References
[1] J. B. Harborne, Comparative Biochemistry in Flavonoids 1967, p. 251
(Academic Press: London).
[34] B. Plietker, Tetrahedron Asymmetry 2005, 16, 3453. doi:10.1016/
J.TETASY.2005.08.057
[2] H. Pacheco, Bull. Soc. France Physiol. Vegetale 1969, 15, 3.
[3] H. Grisebach, W. Barz, Naturwissenschaften 1969, 56, 538.
doi:10.1007/BF00597256
[4] W. Heller, G. Forkmann, The Flavonoids. Advances in Research since
1986 (Ed. J. B. Harborne) 1994, pp. 499–535 (Chapman & Hall:
London).
[5] E. Wong, Chemistry and Biochemistry of Plant Pigments (Ed.
T. W. Goodwin) 1976, pp. 464–526 (Academic Press: NewYork, NY).
[6] J. A. N. Augustyn, B. C. B. Bezuidenhoudt, D. Ferreira, Tetrahedron
1990, 46, 2651. doi:10.1016/S0040-4020(01)82043-3
[7] H. van Rensburg, P. S. van Heerden, B. C. B. Bezuidenhoudt,
D. Ferreira, Tetrahedron 1997, 53, 14141. doi:10.1016/S0040-
4020(97)00916-2
[8] D. Enders, S. Nakai, Chem. Ber. 1991, 124, 219. doi:10.1002/CBER.
19911240133
[9] W.Adam, J. Bialas, L. Hadjiarapoglou,T. Patonay, Synthesis 1992, 49.
doi:10.1055/S-1992-34184
[10] T. Patonay, G. Tóth, W. Adam, Tetrahedron Lett. 1993, 34, 5055.
doi:10.1016/S0040-4039(00)60674-3
[35] M. Engqvist, J. Casas, H. Sundén, I. Ibrahem,A. Córdova,Tetrahedron
Lett. 2005, 46, 2053. doi:10.1016/J.TETLET.2005.01.167
[36] F. A. Davis, A. C. Sheppard, B.-C. Chen, M. S. Haque, J. Am. Chem.
Soc. 1990, 112, 6679. doi:10.1021/JA00174A035
[37] E. A. González, M. A. Nazareno, C. D. Borsarelli, J. Chem. Soc.,
Perkin Trans. 2 2002, 2052. doi:10.1039/B207663B
[38] R. W. D. Welford, J. J. Turnbull, T. D. W. Claridge, A. G. Prescott,
C. J. Schofield, Chem. Commun. 2001, 1828. doi:10.1039/B105576N
[39] J. L. Ingham, S. Tahara, S. Z. Dziedzic, J. Nat. Prod. 1986, 49, 631.
doi:10.1021/NP50046A012
[40] L. N. Lundgren, O. Theander, Phytochemistry 1988, 27, 829.
doi:10.1016/0031-9422(88)84101-3
[41] J. Mathew, A. V. S. Rao, Phytochemistry 1983, 22, 794. doi:10.1016/
S0031-9422(00)86994-0
[42] T. G. Fourie, I. C. du Preez, D. G. Roux, Phytochemistry 1972, 11,
1763. doi:10.1016/0031-9422(72)85034-9
[43] K.Yoshikawa, K. Eiko, N. Mimura,Y. Kondo, S.Arihara, J. Nat. Prod.
1998, 61, 786. doi:10.1021/NP9800396
[44] X. Chen, M. Yu, M. Wang, J. Chem. Res. 2005, 2, 80.
[45] E. A. Jeffery, D. P. N. Satchell, J. Chem. Soc. B 1966, 579.
doi:10.1039/J29660000579
[46] D. M. X. Donnelly, J. P. Finet, B. A. Rattigan, J. Chem. Soc., Perkin
Trans. 1993, 1I, 1729. doi:10.1039/P19930001729
[47] N. Srikanth, O. L. Kon, S. C. Ng, K. Y. Sim, J. Chem. Res. 1997, 6,
1412.
[11] K. M. Mohamed, H. A. Hassanean, K. Ohtani, R. Kasai, K.Yamasaki,
Phytochemistry 2000, 53, 401. doi:10.1016/S0031-9422(99)00547-6
[12] R. J. J. Nel, P. S. van Heerden, H. van Rensburg, D. Ferreira, Tetra-
hedron Lett. 1998, 39, 5623. doi:10.1016/S0040-4039(98)01064-8
[13] D. G. Roux, D. Ferreira, Phytochemistry 1974, 13, 2039.
doi:10.1016/0031-9422(74)85001-6
[14] J. P. van der Merwe, D. Ferreira, E.V. Brandt, D. G. Roux, J.C.S. Chem.
Comm. 1972, 521. doi:10.1039/C39720000521
[48] D. Crich, X. Hao, J. Org. Chem. 1997, 62, 5982. doi:10.1021/
JO970500J
[15] E. Malan, D. G. Roux, Phytochemistry 1974, 13, 1575. doi:10.1016/
0031-9422(74)80331-6
[16] F. du R. Volsteedt, G. J. H. Rall, D. G. Roux, Tetrahedron Lett. 1973,
14, 1001. doi:10.1016/S0040-4039(00)72473-7
[49] S. Banerjee, A. Nayek, S. Sinha, T. Bhaumik, S. Ghosh, J. Mol. Catal.
Chem. 2006, 254, 85. doi:10.1016/J.MOLCATA.2006.03.023
[50] R. F. de la Pradilla, M. V. Buergo, C. Montero, A. Viso, Tetrahedron
2006, 62, 2684. doi:10.1016/J.TET.2005.12.024
[17] Z. P. Pakudina, Khim. Prir. Soedin. 1969, 5, 126.
[18] E. Kiehlmann, E. P. M. Li, J. Nat. Prod. 1995, 58, 450.
doi:10.1021/NP50117A018
[19] Y. Gebhardt, S. White, G. Forkmann, R. Lukacˇin, U. Matern,
S. Martens, Phytochemistry 2005, 66, 1273.
[51] J. A. Dale, H. S. Mosher, J. Am. Chem. Soc. 1973, 95, 512.
doi:10.1021/JA00783A034
[52] A. F. Hundt, J. F. W. Burger, J. P. Steynberg, J. A. Steenkamp,
D. Ferreira, Tetrahedron Lett. 1990, 31, 5073. doi:10.1016/S0040-
4039(00)97809-2
[20] R. W. D. Welford, I. J. Clifton, J. J. Turnbull, S. C. Wilson,
C. J. Schofield, Org. Biomol. Chem. 2005, 3, 3117. doi:10.1039/
B507153D
[21] J. J. Turnbull, J. Nakajima, R. W. D. Welford, M. Yamazaki, K. Saito,
C. J. Schofield, J. Biol. Chem. 2004, 279, 1206. doi:10.1074/
JBC.M309228200
[53] K. Harada, S. Shiono, Bull. Chem. Soc. Jpn. 1984, 57, 1040.
doi:10.1246/BCSJ.57.1040
[54] A. S. Demir, A. Aybey, O. Sesenoglu, F. Polat, TetrahedronAsymmetry
2003, 14, 1489. doi:10.1016/S0957-4166(03)00214-3
[55] F. A. Davis, M. C. Weismiller, J. Org. Chem. 1990, 55, 3715.
doi:10.1021/JO00299A007
[22] A. G. Prescott, N. P. J. Stamford, G. Wheeler, J. L. Firmin, Phyto-
chemistry 2002, 60, 589. doi:10.1016/S0031-9422(02)00155-3
[23] F. Wellmann, M. Griesser, W. Schwab, S. Martens, W. Eisenreich,
U. Matern, R. Lukacˇin, FEBS Lett. 2006, 580, 1642. doi:10.1016/
J.FEBSLET.2006.02.004
[56] T. Todoroki, A. Saito, A. Tanaka, Biosci. Biotechnol. Biochem. 2002,
66, 1772. doi:10.1271/BBB.66.1772
[57] This optically pure compound was obtained via chiral derivatization
and subsequent restoration of the carbonyl functionality. The details
of this protocol will be published elsewhere.
[24] D. Slade, D. Ferreira, J. P. J. Marais, Phytochemistry 2005, 66, 2177.
doi:10.1016/J.PHYTOCHEM.2005.02.002
[58] J. W. Clark-Lewis, P. G. Tucker, Aust. J. Chem. 1976, 29, 1397.
doi:10.1071/CH9761397