Enhancement of the water solubility of flavone glycosides by disruption of molecular planarity of the aglycone moiety

Article abstract of DOI:10.1021/np300460a

Enhancement of the water solubility by disruption of molecular planarity has recently been reviewed as a feasible approach in small-molecule drug discovery programs. We applied this strategy to some natural flavone glycosides, especially diosmin, a highly insoluble citroflavonoid prescribed as an oral phlebotropic drug. Disruption of planarity at the aglycone moiety by 3-bromination or chlorination afforded 3-bromo- and 3-chlorodiosmin, displaying a dramatic solubility increase compared with the parent compound.

Full text of DOI:10.1021/np300460a

Article
pubs.acs.org/jnp
Enhancement of the Water Solubility of Flavone Glycosides by
Disruption of Molecular Planarity of the Aglycone Moiety
Guy Lewin,*,† Alexandre Maciuk,† Aurelien Moncomble,‡ and Jean-Paul Cornard‡
́
†Laboratoire de Pharmacognosie, Faculte
́
de Pharmacie, Universite
́
Paris-Sud BIOCIS UMR-8076 CNRS LabEx LERMIT, Avenue
́ ̂
J.B. Clement, 92296 Chatenay-Malabry Cedex, France
‡Laboratoire de Spectrochimie IR et Raman, Universite
́
Lille 1 UMR-8516 CNRS, 59655 Villeneuve d’Ascq Cedex, France
S
* Supporting Information
ABSTRACT: Enhancement of the water solubility by disruption of molecular
planarity has recently been reviewed as a feasible approach in small-molecule drug
discovery programs. We applied this strategy to some natural flavone glycosides,
especially diosmin, a highly insoluble citroflavonoid prescribed as an oral
phlebotropic drug. Disruption of planarity at the aglycone moiety by 3-bromination
or chlorination afforded 3-bromo- and 3-chlorodiosmin, displaying a dramatic
solubility increase compared with the parent compound.
he classical approach used to increase water solubility of a
T
drug consists of introducing hydrophilic groups, which
results in a concomitant decrease of hydrophobicity, measured
by a smaller log P value. However, lowering of oral
bioavailability is often observed due to a decrease of membrane
permeability related to smaller hydrophobicity. Recently,
Ishikawa and Hashimoto reported the improvement in water
solubility in small-molecule drug discovery programs by
disruption of molecular planarity and symmetry.1 The authors
reviewed such an alternative strategy for improving water
solubility that does not affect hydrophobicity. One of the
examples dealt with a series of aryl hydrocarbon receptor
agonists with a naphthoflavone structure. β-Naphthoflavone (1)
contains an aryl moiety at C-2 that can rotate around the C-2−
C-1′ bond. It was postulated that substitution at C-2′ and/or C-
6′ would increase the dihedral angle of this bond to explain the
enhanced water solubility of compounds 2, 3, and 4.
RESULTS AND DISCUSSION
■
Diosmin (5) is a citroflavonoid, prepared industrially by
dehydrogenation of the corresponding flavanone, hesperidin
(6), and is prescribed in some European countries as an oral
phlebotropic drug in the treatment of chronic venous
insufficiency and hemorrhoidal disease. In a previous study
related to Suzuki reactions at C-3 of flavones, we used diosmin
to prepare 3-bromodiosmetin (9) via intermediate octacetyl-
diosmin (7) and its 3-bromo derivative 8 by a three-step
sequence of peracetylation in Ac2O−pyridine, 48 h, rt, 3-
bromination using N-bromosuccinimide (NBS) in dichloro-
methane (DCM)−pyridine (5:1), 20 h, rt, and concomitant
acid hydrolysis of the glycosidic bond and acetyl esters in
aqueous HCl, 11 N, 2 h, 55 °C.5 Performing this last step in
THF−NaOH(aq), 1 N, 3 h, rt, afforded 3-bromodiosmin (10)
Though water solubility of the four compounds remained
weak, compounds 2−4 were 3 to 15 times more soluble than 1.
This property was correlated with an increase of the dihedral
angle, calculated to be between 17.8° and 70° for 1−4,2 and to
a lowering of the melting point via a decrease of the crystal
packing energy.3 Moreover, 2−4 displayed increased hydro-
phobicity compared to 1, as shown by calculated log P (clogP)4
and retention time data on reversed-phase HPLC. Herein, we
report a significant enhancement of water solubility of some
semisynthetic derivatives of natural flavonoids, especially
diosmin (5).
Received: July 2, 2012
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
dx.doi.org/10.1021/np300460a | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
in 82% yield (from 8) by cleavage of only ester bonds.
Bromination was also performed in DCM−MeOH (2:1) using
2 equiv of NBS (4 h, rt).6 Under these conditions, compound
11 was obtained in 90% yield as a 43:57 mixture of two
diastereoisomeric 3-bromo-2-methoxyflavanones as established
1
by H NMR spectroscopy. Similar alkaline hydrolysis of 11
gave 10 in 85% yield by concomitant cleavage of ester bonds
and β-elimination of MeOH from the heterocycle. Unexpect-
edly, a high water solubility was observed for compound 10,
while diosmin is known to be highly insoluble. This observation
also prompted the preparation of other 3-halogenated
analogues from diosmin. The 3-chloro derivative 13 was
obtained in the same manner as 10, by replacing NBS with N-
chlorosuccinimide (NCS) in the second step [both conditions
used for bromination were tested for the halogenation step, but
chlorination, leading to 12, was effective only in DCM−
pyridine (5:1) (5 equiv of NCS, 80 h, rt)]. 3-Chlorodiosmin
(13) was approximately as soluble in water as 10.
3-bromorhoifolin (19) was readily achieved only via the 3-
bromo-2-methoxyflavanone (18) (direct bromination to the 3-
bromoflavone in DCM−pyridine was ineffective). As expected
from results in the diosmin series, bromo analogues 17 and 19
appeared to possess high water solubility compared to their
respective parent compounds. It is noteworthy that similar
enhancement of water solubility was not observed in the
aglycone series: comparison of the water-insoluble aglycones of
diosmin, linarin, and rhoifolin, namely, diosmetin (20), acacetin
(21), and apigenin (22), with their respective 3-halo derivatives
9, 23, 26, and 27 did not show any appreciable increase in
water solubility.
In order to compare physicochemical parameters of the five
3-substituted flavone glycosides, 10, 13, 14, 17, and 19, with
those of 3-unsubstituted compounds 5, 15, and 16, we carried
out accurate measurements of water solubility, log P, and
retention time by RP-HPLC experiments (Table 1). Reversed
phase retention time can be correlated to lipophilicity and
biological distribution.11
Table 1. Physicochemical Parameters of Diosmin (5),
Linarin (15), Rhoifolin (16), and Their 3-Substituted
Derivatives
a
b
compd water solubility (g/L)
log P
HPLC retention time (min)
Diosmin (Diosmetin) Series
5
0.0012
−0.10
+0.61
+0.44
+0.06
18.58
20.87
20.57
13.31
10
13
14
214
231
0.0056
Linarin (Acacetin) Series
15
17
0.0046
13.8
+0.84
+1.17
36.95
51.51
We could not study the influence of a 3-fluoro or 3-iodo
substituent on water solubility, as efforts to synthesize the
required analogues from octacetyldiosmin (7) with N-
fluorobenzenesulfonimide7 and iodine−cerium(IV) ammonium
nitrate,8 respectively, failed. However, we obtained 3-hydrox-
ydiosmin (14) with the Algar−Flynn−Oyamada (AFO)
reaction9 from hesperidin (6), according to the method of
Pacheco et al.,10 but 14 was not more water-soluble than
diosmin (5).
As 3-bromination was easier to carry out than 3-chlorination
in the diosmin series, we applied the same functionalization to
linarin (3′-deoxydiosmin) (15) and rhoifolin (apigenin 7-
neohesperidoside) (16), two other very weakly soluble natural
7-O-flavone glycosides, previously obtained from diosmin and
naringin, respectively.5 3-Bromolinarin (17) was prepared from
15 by any of the above bromination methods, while synthesis of
Rhoifolin (Apigenin) Series
16
19
0.114
>100
+0.71
+0.93
12.57
18.65
a
b
Measured at 20 2 °C. See Experimental Section.
The results confirm the dramatic improvement in water
solubility observed for 3-bromo and chloro derivatives,
especially in the diosmin series with a factor of about 105.
Moreover these 3-halo derivatives possess increased hydro-
phobicity compared with their corresponding parent glycoside,
as demonstrated by log P and retention time values. On the
contrary, the 3-hydroxy derivative 14 remains almost insoluble
in water in spite of a reduced hydrophobicity.
To investigate the influence of structure on solubility,
quantum chemical calculations were carried out at the DFT
B
dx.doi.org/10.1021/np300460a | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
level of theory. However, flavone glycosides are conformation-
ally flexible compounds that present many rotatable groups,
leading to an excessive number of conformers. Since
optimizations of these products require excessive amounts of
time, structural analysis of the stable conformations were made
with aglycones for the following reasons: (i) the nature of the
two glycosyl units (rutinosyl or neohesperidosyl) did not affect
the observed trends, (ii) the γ-pyrone-B-ring moiety is
suspected to mainly explain the water solubility increase, and
(iii) for the derivatives of a parent compound, the glycosyl unit
could be considered as a constant modification located away
from this γ-pyrone-B-ring moiety. Consequently, the aglycones
20−22 and their 3-substituted derivatives 9 and 23−27 were
used as models for diosmin (5), linarin (15), rhoifolin (16),
and the synthesized 3-substituted glycosides 10, 13, 14, 17, and
19, respectively (Table 2). First, geometry optimizations were
hydroxy group, is weaker (2.00 Å), due to the strong geometry
constraints.
To find which major factor (steric hindrance or electronic
effects) explains the increase of the dihedral angle in the
halogenated compounds, we studied the hypothetical 3-
mercaptodiosmetin (25) on account of some interesting
characteristics of the sulfhydryl functional group for this
comparison: first, a size similar to chlorine (van der Waals
radii for chlorine and sulfur are 1.75 and 1.80 Å, respectively)
and, second, an ability to form hydrogen bonds and electronic
characteristics quite close to that of a hydroxy group.
Optimization of the structure of 25 clearly shows a similarity
with 3-chlorodiosmetin (23): the dihedral angle is 37.6° and
the hydrogen bond with the C-5 hydroxy group is 1.72 Å. As
this result allowed us to propose that the steric hindrance is the
main explanation for the increase of the dihedral angle, we then
turned to study the correlation between structure and solubility.
Solubility is known to be linked to the energy difference
between the solid phase and the liquid phase. This energy
difference between a reoptimized planar structure (that allows
packing effects in the solid phase) and the fully optimized one
Table 2. Selected Geometrical Parameters for Computed
Optimized Structures of Aglycones: Properties of their
Planar Structures
inter-ring
angle (deg)
H bonds
lengths (Å)
ΔEplan
nature of the planar
structure
compd
(kcal·mol−1
)
(more probable in solution) can be referred as ΔEplan = Eplanar
−
Eoptimized
.
14 As expected, a large dihedral angle in the optimized
Diosmetin (Diosmin) Series
a
structure of the aglycone (9, 23, 26, 27) leads to a large ΔEplan
(Table 2), which is connected to a high water solubility of the
corresponding 3-haloflavone glycoside (10, 13, 17, 19): in this
case, the rotation from the planar structure to a large dihedral
angle releases a large amount of energy during solvation that is
linked to better solubility. Inversely, a smaller dihedral angle in
the optimized structure of the aglycone (20−22, 24) leads to a
smaller ΔEplan, which is connected to low water solubility of the
corresponding flavone glycoside (10, 13, 17, 19). Lastly, the
slight improvement in water solubility of 3-hydroxydiosmin
(14) in relation to diosmin, despite a calculated planar structure
of its aglycone 24, may result from the effect of the 3-hydroxy
group, moderated, however, by the presence of an additional
intramolecular hydrogen bond.
20
9
20.5
44.9
42.6
0.0
1.71
0.18
2.39
1.88
0.00
1.41
TS
1.72
TS
23
24
25
1.72
TS
1.77; 2.00
1.72; 2.03
minimum
TS
37.6
Acacetin (Linarin) Series
21
26
19.6
44.6
1.71
1.72
0.11
2.34
TS
TS
Apigenin (rhoifolin) series
22
27
19.4
44.9
1.71
1.72
0.15
2.42
TS
TS
a
Transition state connecting the two symmetrical twisted forms.
In conclusion, this study illustrates the Ishikawa and
Hashimoto strategy, which consists in disrupting molecular
planarity within a chemical series with the aim of improving
water solubility. Our reported results are noteworthy since
particularly impressive enhancements in water solubility, in the
105 range, were noted. Such strong enhancement results from
the glycosidic nature of the starting compounds, 5, 15, and 16,
which are almost insoluble in water (especially diosmin) in
spite of a large hydrophilic glycosidic moiety bearing six
hydroxy groups. Disruption of the planarity of the aglycone
moiety by halogenation at C-3 allows the recovery of high
water solubility, more in agreement with the glycosidic nature
of the compounds. Therefore, these results unambiguously
prove the close relationship of the value of the dihedral angle
between the chromone moiety and the B ring, and
consequently the conformation of the aglycone part of these
glycosides, to the water solubility. Lastly, comparison of 10, 13,
and 14, differing in the nature of the C-3 substituent,
demonstrates that steric hindrance at C-3 is more significant
than electronic factors to increase water solubility: high water
solubility is therefore found with halogenated compounds 10
and 13, which are however more lipophilic than diosmin
(according to measured log P and retention times in HPLC),
while substitution by the hydroxy group leads to 14, which is
more polar but only slightly more soluble than diosmin.
conducted for model compounds 20−22. As expected, the
energy minima structures are not planar.12 The dihedral angle
between the chromone moiety and the B ring is close to 20° in
the three cases. However, these values are significantly lower
than the analogous ca. 45° dihedral angle calculated for
biphenyl.13 This difference can be explained by better electron
delocalization in flavone, enhanced by the para-substitution of
the B ring. A hydrogen bond is present between the C-5
hydroxy group and the C-4 carbonyl group in the three
compounds, the O---H distance being 1.71 Å in each case.
Geometry optimizations were then carried out for structures of
the four 3-haloflavones 9, 23, 26, and 27: while all structures
still display the hydrogen bond without any significant change,
a strong effect of the 3-substituent on the dihedral angle is
observed with calculated values close to 45° for brominated
compounds 9, 26, and 27 and 42.6° for 3-chlorodiosmetin
(23). A different result was obtained with 3-hydroxydiosmetin
(24), for which a planar structure (Cs symmetry) was deduced
from computation. For 24, when fixing the angle between the
chromone part and the B ring to 20.5° (as in diosmetin), a
slightly higher energy (0.2 kcal·mol−1) relative to the planar
structure was calculated. This planar structure displays a
carbonyl group that is involved in two hydrogen bonds: the first
one, with the C-5 hydroxy group, is longer (1.77 Å) than in
previous studied aglycones; the second one, with the C-3
C
dx.doi.org/10.1021/np300460a | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Evaporation under vacuum of the remaining n-BuOH as an azeotropic
mixture with H2O gave 10 as a pale yellow powder (0.56 g, 82%).
Compound 10: [α]D = −53.5 (c 0.2, H2O); UV (EtOH) λmax nm (log
EXPERIMENTAL SECTION
■
General Experimental Procedures. Optical rotations were
measured on an Optical Activity PolAAr 32 polarimeter. NMR
spectra, including NOESY and 1H−13C (HMQC and HMBC)
experiments, were recorded on Bruker AC-300 (300 MHz) or Bruker
AM-400 (400 MHz) spectrometers. ESIMS were recorded on an
Agilent HP 1100 MSD spectrometer (ESI source). HRMS were
obtained from the ILV-UVSQ on a Xevo-Q Tof Waters. HPLC
analyses were performed on a Waters system equipped with a 600E
controller, a 717 autosampler, and a 996 PDA detector (retention
times) or an Agilent HP 1260 Infinity system (solubilities and log P),
using a Waters Spherisorb ODS1 5 μm, 4.6 × 100 mm column.
Physicochemical Measurements. Chromatographic retention
times were measured under isocratic elution using a mobile phase
consisting of H2O−MeOH (60:40). Measurements of purities and
concentrations for solubilities and log P were done under gradient
elution using H2O−MeOH, from 10% to 100% MeOH in 10 min.
Detection and integration were performed at wavelengths close to the
λmax of the compounds (255 nm for 10, 13, and 14, 267 nm for 17 and
19, 284 nm for 6, and 340 nm for 5, 15, and 16). Quantification was
done using calibration curves obtained with compounds dissolved in
DMSO. Solubilities were measured following published guidelines.15
Briefly, for poorly soluble compounds, 200 μL of deionized H2O was
added to 2 mg of the compound, whereas for highly soluble
compounds, 100 μL of deionized H2O was added to 30 mg. Vials were
vortexed and left to settle for 2 days, then centrifuged for 15 min at
4000 rpm to allow injection of 1 or 10 μL of supernatant for HPLC
analysis. Log P values were determined by adding 500 μL of H2O and
500 μL of octanol to 0.5 mg of compound, which was briefly warmed
to 40 °C, vortexed, and left to settle for 1 day. Then 20 μL of each
phase was injected in HPLC. Log P determination measurements were
done in duplicates.
Computational Details. All calculations (for diosmetin, acacetin,
apigenin, and their derivatives) were carried out at the DFT level using
the Gaussian 09 package.16 The PBE0 exchange−correlation func-
tional was used throughout.17,18 All atoms were described using the 6-
311+G(d) Pople basis set.19−22 The 6-311++G(d,p) basis set was also
used in the case of 3-chlorodiosmetin and 3-hydroxydiosmetin to
check the limitation due to the size of the basis set. No significant
change in optimized structures was observed, so the smallest basis set
was used in the study. Structures were optimized under vacuum
without any symmetry constraint. The nature of the stationary points
(minima and transition states) was assigned by vibrational analysis (no
and one negative eigenvalue in the Hessian matrix, respectively).
Planar structures were optimized fixing only the inter-ring dihedral
angle to 0° without any symmetry constraint.
1
ε) 258 (4.39), 274 sh (4.36), 345 (4.16); H NMR (DMSO-d6) δ
[aglycone moiety] 3.87 (3H, s, OCH3-4′), 6.53 (1H, d, J = 2 Hz, H-6),
6.71 (1H, d, J = 2 Hz, H-8), 7.15 (1H, d, J = 8.5 Hz, H-5′), 7.32−7.33
(2H, m, H-2′ and 6′), 9.5 (1H, br s, OH-3′), 12.4 (1H, s, OH-5);
[sugar moiety: inner glucose (″) and terminal rhamnose (‴)] 1.08
(3H, d, J = 6.4 Hz, H-6‴), 3.15 (2H, m, H-4″ and H-4‴), 3.28 (2H, m,
H-2″ and H-3″), 3.44 (3H, m, 1H-6″, H-3‴ and H-5‴), 3.61 (1H, m,
H-5″), 3.65 (1H, m, H-2‴), 3.81 (1H, m, 1H-6″), 4.54 (1H, s, H-1‴),
5.09 (1H, d, J = 7.2 Hz, H-1″), 4.54, 4.61, 4.70, 5.19 (×2), 5.43 (6H,
sugar hydroxyls); 13C NMR (DMSO-d6) δ [aglycone moiety] 55.7
(CH3, OCH3-4′), 94.4 (CH, C-8), 99.9 (CH, C-6), 104.2 (C, C-10),
105.0 (C, C-3), 111.7 (CH, C-5′), 116.2 (CH-2′), 121.5 (CH-6′),
124.0 (C, C-1′), 145.9 (C, C-3′), 150.4 (C, C-4′), 156.5 (C, C-9),
160.5 (C, C-5), 162.7 (C, C-7), 163.1 (C, C-2), 176.7 (C, C-4); [sugar
moiety]: inner glucose 66.0 (CH2, C-6″), 69.6 (CH, C-4″), 73.0a (CH,
C-2″), 75.6 (CH, C-5″), 76.2a (CH, C-3″), 99.6 (CH, C-1″); terminal
rhamnose 17.7 (CH3, C-6‴), 68.2 (CH, C-5‴), 70.2a (CH, C-2‴),
70.7a (CH, C-3‴), 72.0 (CH, C-4‴), 100.5 (CH, C-1‴), ainterchange-
able; ESIMS m/z 709−711 [M + Na]+; anal. C 46.69, H 4.77%, calcd
for C28H31O15Br (5% H2O), C 46.59, H 4.86%.
3-Chlorooctacetyldiosmin (12). Chlorination of 7 (1.42 g, 1.5
mmol) to compound 12 was performed under the same conditions as
previously described for bromination to 8, by using NCS instead of
NBS. Chlorination, slower than bromination (80 instead of 20 h),
afforded 12 as a yellowish, amorphous powder (1.29 g, 88% yield).
Compound 12: 1H NMR (CDCl3) δ [aglycone moiety] 2.33 and 2.44
(6H, 2s, OAc-5 and 3′), 3.91 (3H, s, OMe-4′), 6.69 (1H, d, J = 2 Hz,
H-6), 6.89 (1H, d, J = 2 Hz, H-8), 7.08 (1H, d, J = 8.8 Hz, H-5′), 7.58
(1H, d, J = 2.2 Hz, H-2′), 7.77 (1H, dd, J = 8.8 and 2.2 Hz, H-6′);
[sugar moiety: inner glucose (″) and terminal rhamnose (‴)] 1.13
(3H, d, J = 6.4 Hz, H-6‴), 1.92−2.06 (18H, 6s, 6 sugar acetyls), 3.66
(1H, H-6″), 3.80 (2H, H-6″and H-5‴), 3.94 (1H, H-5″), 4.67 (1H, s,
H-1‴), 4.99 (1H, H-4‴), 5.13−5.28 (5H, H-2″, 3″, 4″, 2‴, and 3‴),
5.27 (1H, H-1″).
3-Chlorodiosmin (13). Saponification of 12 (0.98 g, 1 mmol) to
give 13 as a yellow powder (0.565 g, 88%) was done under the same
conditions as for the hydrolysis of 8 to 10. Compound 13: [α]D
=
−53.9 (c 0.2, H2O); UV (EtOH) λmax nm (log ε) 257 (4.33), 271 sh
1
(4.30), 347 (4.14); H NMR (DMSO-d6) δ [aglycone moiety] 3.87
(3H, s, OCH3-4′), 6.52 (1H, d, J = 2 Hz, H-6), 6.73 (1H, d, J = 2 Hz,
H-8), 7.16 (1H, d, J = 8.5 Hz, H-5′), 7.39−7.40 (2H, m, H-2′ and 6′),
9.5 (1H, br s, OH-3′), 12.3 (1H, s, OH-5); [sugar moiety: inner
glucose (″) and terminal rhamnose (‴)] 1.08 (3H, d, J = 6.4 Hz, H-
6‴), 3.15 (2H, m, H-4″ and H-4‴), 3.28 (2H, m, H-2″ and H-3″),
3.44 (3H, m, 1H-6″, H-3‴ and H-5‴), 3.61 (1H, m, H-5″), 3.65 (1H,
m, H-2‴), 3.82 (1H, m, 1H-6″), 4.54 (1H, s, H-1‴), 5.10 (1H, d, J =
7.2 Hz, H-1″), 4.54, 4.61, 4.70, 5.20 (×2), 5.44 (6H, sugar hydroxyls);
13C NMR (DMSO-d6) δ [aglycone moiety] 55.7 (CH3, OCH3-4′),
94.6 (CH, C-8), 100.0 (CH, C-6), 104.6 (C, C-10), 111.9 (CH, C-5′),
113.6 (C, C-3), 116.0 (CH-2′), 121.5 (CH-6′), 122.6 (C, C-1′),
146.1(C, C-3′), 150.6 (C, C-4′), 156.4 (C, C-9), 160.5 (C, C-5), 161.3
(C, C-2), 163.2 (C, C-7), 176.3 (C, C-4); [sugar moiety]: inner
glucose 66.0 (CH2, C-6″), 69.6 (CH, C-4″), 73.0a (CH, C-2″), 75.6
(CH, C-5″), 76.2a (CH, C-3″), 99.7 (CH, C-1″); terminal rhamnose
17.7 (CH3, C-6‴), 68.3 (CH, C-5‴), 70.3a (CH, C-2‴), 70.7a (CH, C-
3-Bromooctacetyldiosmin (8). Compound 8 was prepared from
diosmin (5) in a two-step sequence via octacetyldiosmin (7), as
previously described (85% from diosmin).5
3-Bromo-2-methoxyoctacetylhesperidin (11). NBS (0.72 g, 4
mmol) was added to a solution of 7 (1.89 g, 2 mmol) in CH2Cl2−
MeOH (2:1) (50 mL). After 4 h at rt, the reaction mixture was taken
up in CH2Cl2 and washed with 0.1 M Na2S2O3 (60 mL) and H2O (2 ×
80 mL). Standard workup of the organic phase led to a mixture of two
diastereoisomers of 11 as a pale yellow, amorphous powder (1.90 g,
1
90%). Compound 11: H NMR (CDCl3) characteristic signals at δ
1.12 and 1.15 (3H, 2d, J = 6.2 Hz, H-6‴ rhamnosyl), 2.33 and 2.38
(6H, 2s, OAc-5 and 3′), 3.07 (3H, s, OMe-2), 3.88 (3H, s, OMe-4′),
4.19 and 4.20 (1H, 2s, H-3), 4.69 and 4.73 (1H, 2s, H-1‴ rhamnosyl),
6.43 and 6.44 (1H, 2d, J = 2.3 Hz, H-6), 6.64 and 6.65 (1H, 2d, J = 2.3
Hz, H-8), 7.04 (1H, d, J = 8.5 Hz, H-5′), 7.29 (1H, d, J = 2.4 Hz, H-
2′), 7.42 (1H, dd, J = 8.5 and 2.4 Hz, H-6′); ESIMS (+) m/z 1077−
1079 [M + Na]+, 997 [M + Na − HBr]+.
3-Bromodiosmin (10). NaOH, 1 N (50 mL), was added to a
solution of 8 (1.02 g, 1 mmol) in THF (30 mL); then the reaction was
stirred for 3 h at room temperature. The mixture was diluted with H2O
(300 mL), adjusted to pH 6 with 5 N HCl, and then extracted with n-
BuOH (2 × 100 mL). The organic phase was evaporated to a yellow
dried residue, which was dissolved in deionized H2O (30 mL).
a
3‴), 72.0 (CH, C-4‴), 100.5 (CH, C-1‴), interchangeable; ESIMS
m/z 665−667 [M + Na]+; anal. C 50.14, H 5.05%, calcd for
C28H31O15Cl (4% H2O), C 50.29, H 5.10%.
3-Hydroxydiosmin (14). Aqueous 30% H2O2 (1 mL) was added
to a cooled (0 °C) solution of 6 (1 g, 1.64 mmol) in aqueous 2 N
NaOH (20 mL). After 24 h at 0 °C, a further aliquot of aqueous H2O2
30% (1 mL) was added and the reaction left for 24 h at 0 °C. The
mixture was adjusted to pH 6 with HOAc and kept again for 24 h at 0
°C. Sodium metabisulfite (2.4 g) was added, and the mixture was
heated and stirred under reflux for 2 h. The yellow precipitate was
recovered by filtration, washed with H2O, and dried with P2O5 under
vacuum to afford crude 7-O-rutinosyltamarixetin (14), which was
D
dx.doi.org/10.1021/np300460a | J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
purified by successive precipitations from MeOH (192 mg, 19%).
Compound 14: H NMR (DMSO-d6) characteristic signals at δ 3.87
(3H, s, OCH3-4′), 6.52 (1H, d, J = 2 Hz, H-6), 6.73 (1H, d, J = 2 Hz,
H-8), 7.16 (1H, d, J = 8.5 Hz, H-5′), 7.39−7.40 (2H, m, H-2′ and 6′),
9.5 (1H, br s, OH-3′), 12.3 (1H, s, OH-5); ESIMS m/z 647 [M +
Na]+.
ACKNOWLEDGMENTS
■
1
We thank J.-C. Jullian for NMR spectra registration, K. Leblanc
for ESIMS spectra and elemental analysis, and A. Pearson for
English corrections of the manuscript.
REFERENCES
3-Bromolinarin (17). 17 was prepared from linarin (15) (60 mg,
0.1 mmol) under the same conditions as described for 3-
bromodiosmin (yellow powder, 41 mg, 70%). Compound 17: [α]D
= −45.9 (c 0.2, H2O); 1H NMR (DMSO-d6) characteristic signals at δ
1.08 (3H, d, J = 6.4 Hz, H-6‴), 3.86 (3H, s, OCH3-4′), 4.55 (1H, s, H-
1‴), 5.08 (1H, d, J = 7.2 Hz, H-1″), 6.51 (1H, d, J = 2 Hz, H-6), 6.70
(1H, d, J = 2 Hz, H-8), 7.16 (2H, d, J = 8.5 Hz, H-3′ and 5′), 7.87
(2H, d, J = 8.5 Hz, H-2′ and 6); 13C NMR (DMSO-d6) δ [aglycone
moiety] 55.6 (CH3, OCH3-4′), 94.3 (CH, C-8), 99.6a (CH, C-6),
104.2 (C, C-10), 105.3 (C, C-3), 113.9 (CH, C-3′ and C-5′), 124.0
(C, C-1′), 131.4 (CH, C-2′ and C-6′), 156.7 (C, C-9), 161.6 (C, C-5),
162.7b (C, C-4′), 163.2b (C, C-2 and C-7), 176.6 (C, C-4); [sugar
moiety]: inner glucose 66.1 (CH2, C-6″), 69.8c (CH, C-4″), 73.0c
(CH, C-2″), 75.8 (CH, C-5″), 76.3c (CH, C-3″), 100.2a (CH, C-1″);
terminal rhamnose 17.8 (CH3, C-6‴), 68.3 (CH, C-5‴), 70.3 (CH, C-
2‴), 70.8 (CH, C-3‴), 72.1 (CH, C-4‴), 100.6 (CH, C-1‴),
a,b,cinterchangeable; ESIMS m/z 693−695 [M + Na]+.
■
(1) Ishikawa, I.; Hashimoto, Y. J. Med. Chem. 2011, 54, 1539−1554.
(2) Calculated dihedral angles were estimated by means of density
functional theory (DFT) calculations (B3LYP/6-31G*) with Gaussian
03.
(3) Jain, N.; Yalkowsky, S. H. J. Pharm. Sci. 2001, 90, 234−252.
(4) clogP values were estimated with ChemDraw Ultra, version 10.0.
(5) Quintin, J.; Roullier, C.; Thoret, S.; Lewin, G. Tetrahedron 2006,
62, 4038−4051.
(6) Bird, T. G.; Brown, B. R.; Stuart, I. A.; Tyrell, A. W. R. J. Chem.
Soc., Perkin Trans. 1 1983, 1831−1846.
(7) A direct (electrochemical) 3-fluorination of flavones with this
mild electrophilic fluorinating agent has been reported: Hou, Y.;
Higashiya, S.; Fuchigami, T. Synlett 1998, 973−974.
(8) Pal, M.; Subramanian, V.; Parasuraman, K.; Yeleswarapu, K. R.
Tetrahedron 2003, 59, 9563−9570.
(9) Cummins, B.; Donnelly, D. M. X.; Eades, J. F.; Fletcher, H.;
Cinneide, F. O’.; Philbin, E. M.; Swirski, J.; Wheeler, T. S.; Wilson, R.
K. Tetrahedron 1963, 19, 499−512.
(10) (a) Pacheco, H.; Grouiller, A. Bull. Soc. Chim. Fr. 1966, 3212−
́
3217. (b) French Patent (Brevet Special de Medicament) 3.794 M,
3-Bromo-2-methoxyoctacetylnaringin (18). The mixture of
two diastereoisomers (49:51) was prepared from octacetylrhoifolin
(100 mg, 0.11 mmol) under conditions described for 3-bromo-2-
methoxyoctacetylhesperidin (11), but with a longer reaction time (24
h) and was obtained as a pale yellow powder, 87 mg, 77%. Compound
1965.
(11) Valko,
1
́
K. J. Chromatogr. A 2004, 1037, 299−310.
18: H NMR (CDCl3) characteristic signals at δ 1.19 and 1.24 (3H,
(12) Aparicio, S. Int. J. Mol. Sci. 2010, 11, 2017−2038.
2d, J = 6.4 Hz, H-6‴ rhamnosyl), 2.35 and 2.40 (6H, 2s, OAc-5 and
3′), 3.09 and 3.14 (3H, 2s, OMe-2), 4.24 and 4.25 (1H, 2s, H-3), 6.48
and 6.49 (1H, 2d, J = 2 Hz, H-6), 6.69 and 6.70 (1H, 2d, J = 2 Hz, H-
8), 7.23 and 7.26 (2H, d, J = 8.8 Hz, H-3′ and H-5′), 7.62 and 7.66
(2H, 2d, J = 8.8 Hz, H-2′ and H-6′).
(13) Grein, F. J. Phys. Chem. A 2002, 106, 3823−3827.
(14) Lima, C. F. R. A. C.; Rocha, M. A. A.; Schroder, B.; Gomes, L.
̈
R.; Low, J. N.; Santos, L. M. N. B. F. J. Phys. Chem. B 2012, 116,
3557−3570.
(15) Sugano, K.; Okazaki, A.; Sugimoto, S.; Tavornvipas, S.; Omura,
A.; Mano, T. Drug Metab. Pharmacokinet. 2007, 22, 225−254.
(16) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
3-Bromorhoifolin (19). Compound 19 was prepared from 18 (77
mg, 0.075 mmol) under the same conditions as described for 3-
bromodiosmin (yellow powder, 38 mg, 77%). Compound 19: 1H
NMR (DMSO-d6) δ [aglycone moiety] 6.42 (1H, d, J = 1.8 Hz, H-6),
6.71 (1H, d, J = 1.8 Hz, H-8), 6.95 (2H, d, J = 8.7 Hz, H- 3′ and 5′),
7.74 (2H, d, J = 8.7 Hz, H-2′ and 6′); [characteristic peaks of the sugar
moiety: inner glucose (″) and terminal rhamnose (‴)] 1.18 (3H, d, J =
6.2 Hz, H-6‴), 5.13 (1H, s, H-1‴), 5.23 (1H, d, J = 7.2 Hz, H-1″); 13C
NMR (DMSO-d6) δ [aglycone moiety] 94.3 (CH, C-8), 99.7 (CH, C-
6), 104.2 (C, C-10), 104.7 (C, C-3), 115.1 (CH, C-3′ and C-5′), 122.2
(C, C-1′), 131.4 (CH, C-2′ and 6′), 156.6 (C, C-9), 160.4 (C, C-5 and
C-4′), 162.8 (C, C-2 and C-7), 176.7 (C, C-4); [sugar moiety]: inner
glucose 60.4 (CH2, C-6″), 70.4a (CH, C-4″), 76.2b (CH, C-2″), 77.0b
(CH, C-3″), 77.0b (CH, C-5″), 97.7 (CH, C-1″); terminal rhamnose
18.0 (CH3, C-6‴), 68.3 (CH, C-5‴), 69.6a (CH, C-2‴), 70.3a (CH, C-
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
a
b
3‴), 71.8 (CH, C-4‴), 100.4 (CH, C-1‴), interchangeable, inter-
changeable; ESIMS m/z 679−681 [M + Na]+.
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
revision B.01; Gaussian, Inc.: Wallingford, CT, 2009.
(17) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77,
3865−3868.
(18) Adamo, C.; Barone, V. J. Chem. Phys. 1999, 110, 6158−6169.
(19) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys.
1980, 72, 650−654.
(20) McLean, A. D.; Chandler, G. S. J. Chem. Phys. 1980, 72, 5639−
5648.
(21) Curtiss, L. A.; McGrath, M. P.; Blandeau, J. P.; Davis, N. E.;
Binning, R. C., Jr.; Radom, L. J. Chem. Phys. 1995, 103, 6104−6113.
(22) Clark, T.; Chandrasekhar, J.; Spitznagel, G. W.; Schleyer, P. V.
R. J. Comput. Chem. 1983, 4, 294−301.
ASSOCIATED CONTENT
■
S
* Supporting Information
1H and 13C NMR spectra of diosmin (5) and compounds 10,
13, 17, 19; 1H NMR spectrum of epimeric mixture 11; HRMS
data of 10, 13, 17, 19. This material is available free of charge
via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*Tel: 33 1 46 83 55 93. Fax: 33 1 46 83 53 99. E-mail: guy.
lewin@u-psud.fr.
Notes
The authors declare no competing financial interest.
E
dx.doi.org/10.1021/np300460a | J. Nat. Prod. XXXX, XXX, XXX−XXX