Synthesis of per-sulfated flavonoids using 2,2,2-trichloro ethyl protecting group and their factor Xa inhibition potential

Article abstract of DOI:10.1016/j.bmc.2004.11.060

The synthesis of per-sulfated flavonoids, organic compounds with multiple sulfate groups, is challenging. We present here a two-step synthesis of fully sulfated flavonoids in high overall yields using the 2,2,2-trichloroethyl moiety as a protecting group.

Full text of DOI:10.1016/j.bmc.2004.11.060

Bioorganic & Medicinal Chemistry 13 (2005) 1783–1789
Synthesis of per-sulfated flavonoids using 2,2,2-trichloro
ethyl protecting group and their factor Xa inhibition potential
Gunnar T. Gunnarsson, Muhammad Riaz, Joanna Adams and Umesh R. Desai*
Department of Medicinal Chemistry, Institute for Structural Biology and Drug Discovery, Virginia Commonwealth University,
8
00 East Leigh Street, Suite 212, Richmond, VA 23219, USA
Received 28 September 2004; revised 30 November 2004; accepted 30 November 2004
Available online 28 December 2004
Abstract—The synthesis of per-sulfated flavonoids, organic compounds with multiple sulfate groups, is challenging. We present here
a two-step synthesis of fully sulfated flavonoids in high overall yields using the 2,2,2-trichloroethyl moiety as a protecting group. The
two-step synthesis results in exclusive formation of the per-sulfated product in contrast to common sulfating agents that yield dif-
ferentially sulfated mixture of compounds. Most per-sulfated flavonoids studied are activators of antithrombin for accelerated inhi-
bition of factor Xa, a key enzyme of the blood coagulation cascade. As a group the per-sulfated flavonoids possess a range of factor
Xa inhibition potential.
ꢀ
2004 Elsevier Ltd. All rights reserved.
1
5,16
1. Introduction
a member of the serpin superfamily of proteins.
Antithrombin is a major regulator of the blood coagu-
lation cascade and it performs this function by inhibit-
ing thrombin and factor Xa, two key proteinases of
Organic molecules possessing sulfate moieties are
increasingly gaining importance as modulators of phy-
siological or pathological function. These functions
1
5,16
the clotting cascade.
Antithrombin inhibits these en-
1
–6
7–12
include inhibition of clotting, viral infection,
1
and
Yet, the number of multiple sulfated syn-
zymes relatively slowly under physiological conditions,
however, this inhibition can be greatly accelerated in
the presence of fully sulfated flavanoids and flavo-
3,14
cancer.
thetic, organic molecules available for investigation
remains small, primarily because of difficulties in synthe-
sis and isolation. As the structural diversity in sulfated
organic molecules increases, better modulators are
expected.
1
,3–5
noids.
A central aspect of these new designed synthetic anti-
thrombin activators and other sulfate-containing biologi-
cally active molecules is the requirement of high sulfate
content. Typically sulfation of alcoholic or phenolic
compounds is performed at the very end of a synthetic
scheme because they radically alter the physical proper-
ties of the organic molecule. Sulfate groups greatly
reduce the hydrophobic character of the molecule
thereby, eliminating organic solvent solubility and
introducing water solubility. Further, the small size of
typical organic scaffolds generates major challenges
in purification of these multiple sulfated organic
molecules.
Our work has focused on designing sulfated molecules
that activate antithrombin, a plasma glycoprotein and
Abbreviations: DMF, dimethyl formamide; DMSO, dimethyl sulfox-
ide; TCE-Cl, 2,2,2-trichloroethylsulfuryl chloride; RP-HPLC, Reverse
Phase HPLC; DCC, dicyclohexylcarbodiimide; ESI-MS, electrospray
ionization mass spectrometry; fXa, factor Xa; AT, antithrombin; D-
0
0
MAP, N ,N -4-dimethylaminopyridine; TEAST, triethylamine–sulfur
trioxide complex; TCA, trichloroacetic acid; MES, 2-(N-morpho-
lino)ethanesulfonic acid, hemisodium salt; kCAT, second-order rate
constant for antithrombin-factor Xa inhibition reaction in the presence
of activator; kUNCAT, second-order rate constant for antithrombin-
An added complication in sulfation of multiple func-
tionalized alcoholic or phenolic groups is the possibility
of numerous products resulting from incomplete sulfa-
tion. Although in principle, it should be possible to drive
the reaction to completion using a huge excess of the
factor Xa inhibition reaction; K
D
, equilibrium dissociation constant
for antithrombin–small activator complex.
*
Corresponding author. Tel.: +1 804 828 7328; fax: +1 804 827
664; e-mail: urdesai@vcu.edu
3
0
968-0896/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.11.060

1784
G. T. Gunnarsson et al. / Bioorg. Med. Chem. 13 (2005) 1783–1789
sulfating reagent, in practice a Gaussian distribution of
products is typically observed. Thus, for an organic
skeleton with five potential sulfation sites, a total of 31
mono-, di-, tri-, tetra-, and penta-sulfated species
are possible, of which tri- and tetra-substituted prod-
ucts are generally formed in greater proportion. A pri-
mary reason for this pattern is the lowering of
nucleophilicities of remaining –OH groups following
introduction of successive sulfate groups on a small
organic skeleton. Additionally, adjacent sulfate groups
are likely to create charge repulsion making further
sulfation more difficult.
2. Methods
2.1. Materials
Human antithrombin and factor Xa were from Hemato-
logic Technology (Essex Junction, VT). Spectrozyme
fXa was from American Diagnostica (Greenwich, CT).
Flavonoids 1–6 were from Indofine (Somerville, NJ)
and were used as such. Palladized charcoal (10%),
2,2,2-trichloroethanol, sulfuryl chloride, ammonium
formate, and HPLC solvents were from Sigma-Aldrich
(Milwaukee, WI).
Numerous sulfation reagents and conditions are re-
ported in literature. For example, sulfur trioxide com-
plex with pyridine is extensively used for preparing
2.2. Synthesis of the protecting group
2,2,2-Trichloroethyl chlorosulfonyl chloride (TCE-Cl)
was synthesized by dropwise addition of sulfuryl chlo-
ride to 2,2,2-trichloroethanol (1:1 molar ratio) in diethyl
ether in the presence of pyridine (1:1 molar ratio) at
ꢀ78 ꢁC. Pyridine hydrochloride formed in the reaction
was filtered and the product purified using vacuum dis-
tillation (bp 32 ꢁC at 0.4 mm).
1
7–19
sulfated oligosaccharides
natural sugars, including sulfated hyaluronic acid,
and sulfated polymeric
2
0
2
1
22
23
24
chitosan, b-glucan, dextran, galactomannan, fu-
2
5
26
can, and curdlan. In contrast, sulfation of serine
and tyrosine residues in peptides has been achieved with
sulfur trioxide complex with DMF or trimethyl-
2
7,28
amine.
Sulfation of organic skeletons with phenolic
moieties, including flavanoids, flavonoids and steroids,
has been achieved with triethylamine–sulfur trioxide
2.3. Reverse phase HPLC analysis
3
–5,9,13,14,29
complex
or tetrabutylammonium hydrogen
Reverse phase HPLC of TCE protected (1p–6p) and sul-
fated (1s–6s) flavonoids was performed using a C-18
3
0
sulfate (TBAHS) in combination with DCC. While
reasonable yield of the fully sulfated products was
reported using the triethylamine–sulfur trioxide com-
TM
analytical column YMC -ODC-HQ (4.6 · 250mm,
Waters) on a Shimadzu 10Avp HPLC system. The col-
umn was pre-equilibrated with buffer A consisting of
acetonitrile–H O (50:50 v/v) solvent. Analysis was per-
3
–5
plex,
the TBAHS–DCC combination led primarily
to partially sulfated species. Likewise, sulfation of a pen-
tagallolyl derivative of glucose, containing the 3,4,5-tri-
hydroxybenzoic acid moiety, using trimethylamine–
sulfur trioxide complex resulted in sulfation of only
2
formed using a linear gradient of buffer A to buffer B
consisting of acetonitrile–H O (95:5) in 25 min at a flow
2
rate of 0.5 mL/min and monitoring UV absorbance at
279 nm.
3
1
two phenolic groups of three.
Recently, Liu et al. reported the use of trichloroethyl
sulfonyl (TCE) group, which can be removed in a selec-
tive manner using catalytic hydrogenation, to achieve
2.4. General synthesis of TCE-protected flavones 1p–6p
To a stirring solution of flavone (1 mmol), triethylamine
(2–4 mmol per –OH group), and DMAP (1 mmol per
–OH group) in anhydrous THF (15 mL), a solution of
TCE-Cl (2–4 mmol per –OH group) in anhydrous
THF (5 mL) was added dropwise at ambient tempera-
ture over 15 min. The solution was stirred overnight,
after which the reaction mixture was washed with
0.5 M HCl, 5% (w/v) K CO , brine and water, and then
2
9
mono-sulfation of an estrone derivative. To test
whether TCE protecting group paves a cleaner way for
per-sulfation, we studied sulfation of flavones contain-
ing varying number of phenolic groups. Our work shows
that TCE protection and deprotection of flavonoids is
an excellent way of generating per-sulfated products.
In addition, the method possesses several other impor-
tant advantages, including ease of purification and
identification, which are unsurpassed in comparison
to traditional sulfating methods. Finally, the sulfated
flavonoids display interesting factor Xa inhibition
properties.
2
3
dried over anhydrous Na SO . Removal of solvents in
4
2
vacuo resulted in a solid containing the TCE-protected
flavone and TCE dimer. The mixture was fractionated
on silica gel column using 1% EtOAc in petroleum ether.
1
1p: H NMR (CDCl ) d: 4.93 (s, 2H, –CH ), 4.95 (s, 2H,
3
2
_
+
OSO3 NH4
OH
OSO OCH CCl
3
2
2
2
2
-4 eq TCE-Cl,
10% Pd/C, HCOONH4
THF-MeOH, RT
4-6 hr
O
O
O
eq TEA, 1 eq DMAP
THF, RT
_
+
HO
Cl CH COO SO
NH4 O SO
O
3
2
2
O
3
O
1– 6
1p
1s–6s
–
6p
Scheme 1.

G. T. Gunnarsson et al. / Bioorg. Med. Chem. 13 (2005) 1783–1789
1785
4
(d, 1H, J = 1.8 Hz, H5). 5s: H NMR (DMSO-
1
–
CH ), 5.01 (s, 2H, –CH ), 6.75 (s, 1H, H3), 7.67 (d, 1H,
2
2
3
0
4
J = 8.1 Hz, H8), 7.70(m, 1H, H5 ), 7.75 (dd, 1H,
J = 8.1 Hz, J = 2.7 Hz, H7), 7.78 (dd, 1H, J =
d + D O) d: 7.18 (d, 1H, J = 1.8 Hz, H8), 7.19 (d,
6
2
3
4
3
4
1H, J = 1.8 Hz, H6), 7.58 (d, 1H, J = 8.7 Hz, H5 ),
3
0
4
0
3
.1 Hz, J = 1.8 Hz, H6 ), 7.87 (dd, 1H, J = 8.1 Hz,
3
4
7.97 (dd, 1H, J = 8.7 Hz, J = 2.4 Hz, H6 ), 8.06 (d,
0
8
4
0
4
J = 1.8 Hz, H4 ), 8.23 (d, 1H, J = 2.7 Hz, H5). 2p:
4
0
1
1H, J = 2.4 Hz, H2 ). 6s: H NMR (DMSO-
d + D O) d: 6.36 (s, 1H, H3), 7.21 (d, 1H,
J = 2.1 Hz, H8), 7.26 (d, 1H, J = 2.4 Hz, H6), 7.83
1
H NMR (CDCl ) d: 4.92 (s, 2H, –CH ), 4.96 (s, 2H,
3
2
6
2
4
4
–
1
7
CH ), 5.28 (s, 2H, –CH ), 6.81 (s, 1H, H3), 7.54 (d,
2
2
4
H, J = 2.1 Hz, H8), 7.58 (d, 2H, J = 6.9 Hz, H2 ),
3
0
0
(s, 2H, H2 ).
4
.71 (d, 1H, J = 2.1 Hz, H6), 8.00 (d, 2H, J = 6.9 Hz,
3
0
1
H3 ). 3p: H NMR (CDCl ) d: 4.97 (s, 2H, –CH ),
2.6. Mass spectrometric analysis of TCE protected (1p–
6p) and deprotected (1s–6s) flavonoids
3
2
5
–
7
.02 (s, 2H, –CH ), 5.04 (s, 2H, –CH ), 5.29 (s, 2H,
2 2
4
CH ), 6.83 (s, 1H, H3), 7.58 (d, 1H, J = 1.5 Hz, H8),
2
4
.71 (d, 1H, J = 1.5 Hz, H6), 7.84 (d, 1H, J = 5.4 Hz,
3
ESI-MS (positive ion mode) of TCE—protected flavo-
nes 1p–6p was performed in high-resolution mode
using Micromass QTOF-Ultima mass spectrometer
(Waters Corporation, Milford, MA). Sample was dis-
solved in acetonitrile and 10 lL was injected into the
Q-TOF. Sodium TFA was used as a calibrant. ESI-
MS (negative ion mode) of per-sulfated flavones 1s–6s
was performed using a Micromass ZMD 4000 single
quadrapole mass spectrometer (Waters Corp., Milford,
MA). Each sulfated flavones in acetonitrile–water (1:1)
was infused at 10 lL/min for a 1 min period during
which data was acquired in cumulative MCA mode.
Mass scans were obtained in the range 200–800 a.m.u
at a scan rate of 400 a.m.u./s. Ionization conditions
were optimized for each compound to maximize ioniza-
tion of each molecule. The capillary voltage was varied
between 3.0and 4.0V, while the cone voltage usually
ranged from 30to 65 V. The remaining ionization
parameters remained constant, the extractor voltage
was 4.0, the Rf lens voltage was 0.1 V, the source block
temperature was 100 ꢁC and the desolvation tempera-
ture was 120 ꢁC.
0
3
4
0
H5 ), 7.98 (dd, 1H, J = 5.4 Hz, J = 1.5 Hz, H6 ), 8.15
0
4
1
(
d, 1H, J = 1.5 Hz, H2 ). 4p: H NMR (CDCl ) d:
3
4
.95 (s, 2H, –CH ), 4.96 (s, 2H, –CH ), 4.97 (s, 2H,
2 2
3
–
CH ), 5.20(s, 2H, –CH ), 7.72 (dd, 1H, J = 4.8 Hz,
2 2
3
0
3
J = 4.8 Hz, H5 ), 7.79 (d, 1H, J = 7.2 Hz, H8), 7.80
3
dd, 1H, J = 4.8 Hz, J = 0.9 Hz, H6 ), 7.82 (dd, 1H,
4
0
(
3
4
3
J = 7.2 Hz, J = 1.8 Hz, H7), 7.97 (dd, 1H, J =
4
.8 Hz, J = 0.9 Hz, H4 ), 8.30(d, 1H,
0
4
4
J = 1.8 Hz,
1
H5). 5p: H NMR (CDCl ) d: 4.96 (s, 2H, –CH ), 4.97
3
2
(
5
s, 2H, –CH ), 5.19 (s, 2H, –CH ), 5.22 (s, 2H, –CH ),
.32 (s, 2H, –CH ), 7.40(d, 1H, J = 2.4 Hz, H8), 7.62
2
2 2 2
4
4
d, 1H, J = 2.4 Hz, H6), 7.90(d, 1H,
3
(
H5 ), 8.10(dd, 1H, J = 9.0Hz, J = 2.4 Hz, H6 ), 8.21
J = 9.0Hz,
0
0
3
4
4
0
1
(
d, 1H, J = 2.4 Hz, H2 ). 6p: H NMR (CDCl ) d :
3
4
–
1
8
.97 (s, 2H, –CH ), 5.04 (s, 4H, –CH ), 5.12 (s, 2H,
CH ), 5.26 (s, 2H, –CH ), 6.84 (s, 1H, H3), 7.58 (d,
2 2
2
2
4
H, J = 2.1 Hz, H8), 7.70(d, 1H, J = 2.4 Hz, H6),
4
0
.15 (s, 2H, H2 ).
2.5. General procedure for synthesis of per-sulfated
flavonoids 1s–6s
The protected flavone (1 mmol) was dissolved in 2 mL
anhydrous THF to which was added 2 mL MeOH. To
this stirring solution under nitrogen was added
2.7. Accelerated inhibition of factor Xa by sulfated
flavonoids 1s–6s
1
–
0wt.% of 1 0% Pd/C and NH HCO (6 mmol per
4
OH group). The reaction was continuously monitored
The antithrombin inhibition of factor Xa in the presence
of sulfated flavonoids 1s–6s was determined using a sin-
gle time point method. The inhibition reactions were
carried out in PEG20K-coated cuvettes at 25 ꢁC. Activa-
tor 1s–6s (10or 50 lM) was incubated at 25 ꢁC with
antithrombin (1 lM) in 20mM sodium phosphate buffer
containing 20mM NaCl, 0.1 mM EDTA, and 0. 1%
PEG8000 at pH 6.0. Factor Xa (30 nM) in MES buffer,
pH 6.0, was then added. The inhibition reaction was al-
lowed to proceed for 600 s, following which the residual
enzyme activity was determined by following the amid-
olysis of substrate Spectrozyme fXa (100 lM) in
20mM sodium phosphate buffer containing 1 00 mM
NaCl, 0.1 mM EDTA, and 0.1% PEG8000 at pH 7.4.
The initial slope of absorbance at 405 nm in the presence
of the activator was compared with that in its absence to
obtain the percent inhibition of factor Xa. Each experi-
ment was done in duplicate.
2
on RP-HPLC. Following completion of reaction, the
mixture was centrifuged and the supernatant filtered
on Whatman filter paper containing Celite. Evaporation
of solvent gave a solid to which was added ethanol and
the solution allowed to stand overnight at room temper-
ature. The precipitated per-sulfated product was col-
1
lected by filtration. 1s: H NMR (DMSO-d + D O) d:
6
2
0
6
.61 (s, 1H, H3), 7.27 (m, 1H, H5 ), 7.45 (dd, 1H,
0
3
4
J = 7.8 Hz, J = 1.5 Hz, H6 ), 7.57 (dd, 1H,
J = 9.3 Hz, J = 2.7 Hz, H7), 7.65 (d, 1H, J = 9.3 Hz,
3
4
3
3
4
J = 8.25 Hz, J = 1.5 Hz, H4 ),
0
H8), 7.70(dd, 1H,
7
4
.81 (d, 1H, J = 2.7 Hz, H5). 2s: H NMR (DMSO-
1
d + D O) d: 6.68 (s, 1H, H3), 7.24 (d, 1H,
6
2
4
4
J = 2.4 Hz, H8), 7.33 (d, 1H, J = 2.7 Hz, H6), 7.34
3
d, 2H, J = 9.0Hz, H2 ), 7.97 (d, 2H, J = 9.0Hz,
0
3
(
0
1
H3 ). 3s: H NMR (DMSO-d + D O) d: 6.52 (s, 1H,
6
2
4
H3), 7.22 (d, 1H, J = 2.4 Hz, H8), 7.30(d, 1H,
4
3
J = 2.4 Hz, H6), 7.66 (d, 1H, J = 9.0Hz, H5 ), 7.68
0
3
dd, 1H, J = 9.0Hz, J = 2.4 Hz, H6 ), 8.10(d, 1H,
4
0
(
3. Results and discussion
4
0
1
J = 2.4 Hz, H2 ). 4s: H NMR (DMSO-d + D O) d:
6
2
3
.12 (dd, 1H, J = 7.8 Hz, J = 7.8 Hz, H5 ), 7.42 (dd,
3
0
7
We selected flavones 1–6 (Fig. 1) that possess three to
five phenolic groups. Flavones 1 and 2 contain three
–OH groups, while flavones 3 and 4, and 5 and 6 possess
3
H, J = 7.8 Hz, J = 1.8 Hz, H6 ), 7.48 (2H, H7 and
4
0
1
H8), 7.72 (dd, 1H, J = 7.8 Hz, J = 1.8 Hz, H4 ), 7.83
3
4
0

1786
G. T. Gunnarsson et al. / Bioorg. Med. Chem. 13 (2005) 1783–1789
R3’
R2’
R4’
R5’
1
1
1
–6
:: G = -H
R7
O
p
s
– 6p :: G = -SO OCH CCl
2
2
3
R6
R3
_
–6s :: G = -SO3
O
R5
R7
R6
R5
R3
R ’
R ’
R ’
R ’
5
4
3
2
1
2
3
4
5
6
H
OG
H
H
H
H
H
H
H
H
OG OG
OG
OG
H
OG
OG
H
OG
OG
H
H
H
H
H
H
OG
OG
H
H
OG
H
OG OG
OG
OG
OG OG
OG
H
OG
OG
OG
OG
H
H
H
OG
Figure 1. Structure of flavonoids: Per-sulfation was studied in two steps—TCE protection of flavonoids 1–6 to give 1p–6p, followed by
0
hydrogenation to yield 1s–6s. Substituents R
OSO OCH CCl
2
through R , R
for TCE-protected flavonoids 1p–6p, or is –SO
5
0
3
, and R
ꢀ
5
through R
+
4
7
are either –H or –OG, where G is –H for native flavonoid, is
–
2
2
3
3
NH
for per-sulfated flavonoids 1s–6s.
four and five –OH groups, respectively. As a group they
represent small organic molecules with the end-to-end
Newly synthesized 2,2,2-trichloroethyl chlorosulfuryl
chloride (TCE-Cl) was reacted with the flavone in anhy-
drous THF at room temperature in the presence of
DMAP and triethylamine as base (Scheme 1). At a
TCE-Cl molar excess of less than 1.5-fold per available
hydroxyl group, the reaction resulted in multiple prod-
ucts as assayed using reverse phase HPLC (not shown).
However, when the molar excess of TCE-Cl was in-
creased to two-fold per available hydroxyl group, the
reaction resulted in a single compound, the most non-
polar product formed. ESI mass spectrometric analysi+s
of the product showed molecular ion peaks (M + H)
between 900.681 and 908.667 m/z in positive mode with
an isotopic pattern characteristic for nine chlorine atoms
˚
length of ꢁ7–8 A. Thus, per-sulfation of these mole-
cules is expected to generate high charge density species.
Although the placement of –OH groups is varied, there
are five instances of adjacent hydroxyls (flavones 1, 3–6).
Flavone 6 with three consecutive phenolic groups at
0
0
0
positions 3 , 4 5 is extremely interesting because it rep-
resents a structure with highest charge density. In addi-
tion, sulfation of such moieties is nearly impossible with
3
1
sulfur trioxide complexes.
0
0
6,2 ,3 -Trihydroxy flavone (1) was used as test molecule
to study the applicability of TCE as a sulfating agent.
Table 1. Reaction yield and molecular ion peaks in positive ion ESI-MS spectrum of TCE-protected flavonoids 1p–6p
+
Flavonoid
Yield in protection (%)
Positive ion ESI-MS profile
Found (rel int) (m/z) Calculated (m/z)
[M + 1]
Isotopes
35
37
Cl
Cl
X
Y
9
02.671 (71)
904.668 (100)
902.671
904.665
906.662
7
2
3
4
2
3
4
2
3
4
2
3
4
1
2
3
4
3
3
3
3
5
5
5
5
37
37
37
37
1p
2p
3p
4p
90
95
95
57
C
C
C
C
21
21
23
23
H
H
H
H
13
13
14
14
O
O
O
O
14
14
18
18
S
S
S
S
3
3
4
4
Cl
Cl
Cl
Cl
X
X
X
X
Cl
Cl
Cl
Cl
Y
Y
Y
Y
6
5
9
9
06.666 (83)
02.676 (52)
902.671
904.665
906.662
7
904.667 (100)
06.669 (58)
130.539 (84)
1132.538 (100)
6
9
5
1
1130.535
1132.532
1134.529
1130.535
1132.532
1134.529
1376.386
1378.383
1380.380
1382.377
1384.374
1356.401
1358.398
1360.395
1362.392
1364.389
10
9
1
1
134.528 (70)
130.535 (84)
8
10
9
1132.532 (100)
1
1
1
134.529 (79)
376.392 (20)
378.386 (52)
8
14
13
12
11
105
13
12
11
105
9
a
35
37
Cl
5
p
p
82
92
1380.384 (95)
C
C
H
25 15
O
22
S
5
Cl
X
Y
1
1
1
1
382.382 (100)
384.379 (62)
356.404 (51)
358.399 (97)
2
3
4
35
37
Cl
6
1360.397 (100)
H
25 15
O
22
S
5
Cl
X
Y
1
1
362.396 (62)
364.395 (30)
6
a
35
5
37
Cl
Sodium adducts, rather than free ions, were detected. The adducts correspond to mono-sodiated ions, for example, C25
H O
14 22
S
Cl
X
Y
Na.

G. T. Gunnarsson et al. / Bioorg. Med. Chem. 13 (2005) 1783–1789
1787
(
Table 1). This isotopic pattern is the most distinguish-
The idea of using ammonium formate as a mild hydro-
gen source to selectively hydrogenolyse the SO –O bond
instead of the ArO–SO bond is appealing for two rea-
2
sons; (i) it eliminates HCl formed during the reaction,
thereby preventing strongly acidic conditions in which
ing feature of the ESI-MS spectrum of 1p, thus facilitat-
ing rapid analysis of the number of protecting groups
charged on the flavonoid.
2
3
2,33
TCE protection of flavones 2–6 worked well in a similar
manner, except for the observation that nearly three- to
four-fold molar excess of TCE-Cl per available –OH
group is required to fully protect pentahydroxy flavones
per-sulfated flavonoids are unstable
, and (ii) it intro-
+
duces NH4 as a counter-ion, which is known to ease the
MS analyses of highly sulfated molecules using electro-
3
4–36
+
In contrast, Na counter-ions typi-
spray ionization.
5
and 6 (Table 1). The reason for this high equivalence
cally introduced in reactions of SO3 complexes give
highly complex sodium adducts of polysulfated mole-
requirement is not clear, although it is likely that mois-
ture, tightly bound to flavonoids, causes hydrolysis of
TCE-Cl and its subsequent dimerization to Cl CCH O-
3
7,38
cules that are difficult to analyze.
Thus, ESI-MS
analyses following TCE-deprotection of flavonoid 1s
gave ions at 510.9, 526.8, 543.9, and 560.7 m/z values
corresponding to molecular ions with three –OSO
3
2
SO OCH CCl , a product that was identified in nearly
2
2
3
all reaction mixtures. All TCE–protected flavones 2p–
p were obtained in excellent yields following a simple
3
+
6
groups and either 0, 1, 2 or 3 NH4 counter-ions, respec-
tively (Table 2). Likewise, peaks corresponding to
purification procedure. The ESI-MS profile of these
flavones was unique due to the presence of numerous
chlorine atoms, as also noted for 1p. In addition, RP-
HPLC profile of 1p–6p (not shown) is dependent on
the number of TCE groups and the base skeleton, thus
enhancing the ease of identification.
ꢀ
[M ꢀ nH + (n ꢀ 1)NH ] , where M corresponds to the
4
free acid (–OSO H form), were detected for each of
3
the TCE-deprotected flavonoid 2s–6s, except for 5s for
which mixed molecular ions containing both Na and
+
+
NH4 were observed (see Table 2). This suggested exclu-
sive formation of per-sulfated flavonoids. The isolated
yield of the sulfated flavonoids 1s–6s range from 48%–
Selective hydrogenation of the TCE group was per-
formed using Pd/C charcoal and ammonium formate
at room temperature, essentially following Liu et al.
78%, which is 1.5- to 2-fold greater than typical isolated
3–5
yields obtained using SO complexes with amines.
2
9
3
All TCE-protected flavonoids required 3–5 h for depro-
tection, although 1p required nearly 15 h. The reactions
were continuously monitored on RP-HPLC, as for
example for 1p, the peak at ꢁ21 min is gradually re-
placed with a single peak at ꢁ5 min (Fig. 2). Peaks cor-
responding to partially deprotected species between 10
and 18 min are observed, however their concentrations
remain very small at all times. The simplicity, ease and
unambiguity of monitoring this selective hydrogenation
reaction is a particularly attractive feature of the overall
protocol.
Most importantly, the ability to sulfate three adjacent
–OH groups, as in 6, greatly enhances the applicability
of this methodology in synthesis of this important class
of biologically active compounds.
The acceleration in antithrombin inhibition of factor
Xa was studied in 20mM sodium phosphate buffer,
pH 6.0, at 25 ꢁC in the presence of fixed concentration
3
–5
of sulfated activators 1s–6s, as previously reported.
A single time point method was used, rather than an
extended determination of second-rate order con-
3
9
stant, to rapidly assess the activation potential of
these novel polysulfates. Small activator-dependent fac-
tor Xa inhibition is a function of the amount of anti-
thrombin–small activator in the conformationally
activated state, which is further dependent on the bind-
1
1
1
9
5
1
7
3
1
ing affinity (K ) and activation potential (kCAT) of the
D
small molecule activator. Assuming that the activation
potential remains nearly equivalent for this series of
structurally related compounds, as observed for sul-
Reactant 1p
Product 1s
4
fated flavanoids earlier, the single-time point method
primarily reflects the binding affinity of the sulfated acti-
vator–antithrombin complex. Thus, greater inhibition of
15 h
1
h
In the single time point method, the concentration of residual enzyme
remaining at time t is given by the pseudo-first order equation
-
0
5
10
15
20
25
Elution Time (m)
E
t
= E
O
exp(ꢀkOBS · t), E
O
= initial activity of enzyme, for single
turnover kinetics and kOBS is the pseudo-first order rate constant for
reaction between antithrombin and factor Xa. In the presence of the
activator (ACT), the observed rate constant kOBS is given by kUNCAT
[AT]O + kCAT [AT:ACT]O and in its absence kOBS is equal to kUNCAT
Figure 2. A representative RP–HPLC profile for deprotection of 1 as a
function of time: At 1 h, the starting material, 1p, at ꢁ21 min has most
remained unreacted, while at 15 h nearly quantitative conversion to a
highly polar peak at ꢁ5 min, corresponding to 1s, is observed. Note
the absence of peaks of intermediate polarity corresponding to
partially hydrogenated compounds. The trace of 15 h has been
artificially moved up for clarity.
[AT] . Thus, the ratio of the activity of the enzyme in the presence of
O
activator to its absence (E /E ) turns out to be k
CAT
2
1
· [AT:ACT]O.
Thus, in the single time point method, percent inhibition of factor Xa
is a function of both kCAT and [AT:ACT]O.

1788
G. T. Gunnarsson et al. / Bioorg. Med. Chem. 13 (2005) 1783–1789
Table 2. Reaction yields, negative ion ESI-MS spectrum profile and
factor Xa inhibition of per-sulfated flavonoids 1s–6s
(Table 2). Thus, factor Xa inhibition potential appears
to be a function of the number of sulfate groups charged
on the small skeleton. Although for these subgroups the
inhibition appears to be approximately constant, inter-
esting fine structural distinctions exist. For example, 1s
and 4s show less potency in comparison to 2s and 3s
Flavonoid
Yield in
deprotection
Negative ion ESI-
a
MS profile
Factor Xa
inhibition
(%)
(%)
Found
m/z)
Calculated
(m/z)
(
0
suggesting a preference for the 5,7,4 -substitution. Fur-
ther, introduction of 3-OSO in 4s and 5s, in compari-
510.9
526.8
509.4
526.4
543.5
ꢀ
c
1
s
s
55
40
0± 2
3
son to 1s and 3s, significantly enhances the inhibition
potential suggesting the importance of this position. In
addition, although both 5s and 6s show nearly equiva-
5
43.9
5
5
5
5
08.8
25.9
509.4
526.4
2
12 ± 2
lent activity at 50 lM, experiments at five-fold lower
concentration show that 5s, with 3-OSO group, is bet-
43.0543.5
60.7
ꢀ
560.5
3
ter than 6s. It is likely that either 3-OSO interacts di-
ꢀ
3
6
6
07.4
21.8
605.5
622.5
± 62 0
rectly with antithrombin or that the presence of this
group introduces structural constraints on the flexibility
of the unicyclic ring, thus enhancing the binding affinity.
Detailed quantitative aspects of sulfated flavonoids 3s,
5s, and 6s interacting with antithrombin are being fur-
ther investigated.
3
4
s
s
52
78
639.0639.5
6
6
56.6
73.7
656.6
673.6
6
6
56.7
73.0673.6
656.6
40± 8
7
7
7
7
17.7
57.5
62.9
74.6
718.7
757.6
762.5
774.7
In conclusion, synthesis of per-sulfated flavonoids was
achieved in two simple steps using the TCE protecting
group. Overall, good yields were obtained in compari-
son to traditional direct sulfation methods involving sul-
fur trioxide complexes. The two-step approach avoided
the formation of partially sulfated flavonoid products,
thereby obviated tedious purification procedures.
Further, the unique ESI-MS profile of TCE-protected
intermediate greatly aided rapid characterization of
products. Finally, the antithrombin–dependent factor
Xa inhibition potential of the sulfated flavonoids is var-
ied and highly structure dependent.
b
5
s
48
65
89 ± 4
7
7
01.0
35.6
701.5
735.6
ꢀ
6
s
82 ± 4
a
b
c
ꢀ
A formula [M ꢀ nH] or [M ꢀ nH + (n ꢀ 1)NH
4
]
is used to cal-
culate the expected mass, where M corresponds to per-sulfated fla-
vonoid in acid form (–OSO H-form).
For per-sulfated flavonoid 5, mixed molecular ions containing both
3
+
+
4
+
+
Na and NH
+
Na + 1NH
4
were observed. These include the 1Na + 2NH ,
+
, and 1Na + 3NH
+
+
2
4
4
ions.
Error in two independent measurements.
factor Xa suggests better binding affinity of the sulfated
activator.
Acknowledgments
We thank Prof. Michael Hindle of Virginia Common-
wealth University for making his mass spectrometry
facility generously available to us. In addition, mass
spectrometry support was also provided by the Wash-
ington University Mass Spectrometry Resource, an
NIH Research Resource (P41RR0954). This work was
supported by American Heart Association—Mid-Atlan-
Figure 3 shows the profile of factor Xa inhibition poten-
tial of sulfated flavonoids 1s–6s (see also Table 2). As a
group the sulfated flavonoids exhibit a full range of inhi-
bition potential. Whereas 1s and 2s, with three sulfate
groups each, exhibit less than 10% inhibition, 5s and
6
s with five sulfate groups exhibit ꢁ90% inhibition
1
00
8
6
4
2
0
0
0
0
0
1
s
2s
3s
4s
5s
6s
Figure 3. Antithrombin-dependent factor Xa inhibition property of sulfated flavonoids 1s–6s: Factor Xa inhibition was assessed in 20mM sodium
phosphate buffer, pH 6.0, at 25 ꢁC using the single time point method in the presence and absence of fixed concentration of sulfated flavonoids and
antithrombin. Error bars represent the variation in percent inhibition observed in duplicate measurements. See Methods for additional details.

G. T. Gunnarsson et al. / Bioorg. Med. Chem. 13 (2005) 1783–1789
1789
tic affiliate (0256286U), and the National Heart, Lung
and Blood Institute (RO1 HL069975). GTG thanks
AHA-Mid-Atlantic Affiliate for the award of post-doc-
toral fellowship (0325621U).
17. Petitou, M.; van Boeckel, C. A. Angew. Chem., Int. Ed.
004, 43, 3118–3133.
2
1
1
2
2
8. Koshida, S.; Suda, Y.; Sobel, M.; Kusumoto, S. Tetrahe-
dron Lett. 2001, 42, 1289–1292.
9. Ikeda, K.; Asahara, T.; Achiwa, K.; Hoshino, H. Chem.
Pharm. Bull. 1997, 45(2), 402–405.
0. Barbucci, R.; Lamponi, S.; Magnani, A.; Renier, D.
Biomaterials 1998, 19, 801–806.
1. Drozd, N. N.; Sher, A. I.; Makarov, V. A.; Galbraikh, L.
S.; Vikhoreva, G. A.; Gorbachiova, I. N. Thromb. Res.
2001, 102, 445–455.
References and notes
1
. Desai, U. R. Med. Res. Rev. 2004, 24, 151–181.
2
. Rimbach, G.; Weinberg, P. D.; de Pascual-Teresa, S.;
Alonso, M. G.; Ewins, B. A.; Turner, R.; Minihane, A.
M.; Botting, N.; Fairley, B.; Matsugo, S.; Uchida, Y.;
Cassidy, A. Biochim. Biophys. Acta 2004, 1670, 229–237.
. Gunnarsson, G. T.; Desai, U. R. Bioorg. Med. Chem. Lett.
22. Alban, S.; Jeske, W.; Welzel, D.; Franz, G.; Fareed, J.
Thromb. Res. 1995, 78, 201–210.
23. Logeart-Avramoglou, D.; Jozefonvicz, J. J. Biomed.
Mater. Res. 1999, 48, 578–590.
24. Pires, L.; Gorin, P. A. J.; Reicher, F.; Sierakowski, M.-R.
Carbohydr. Polymers 2001, 46, 165–169.
25. Nishino, T.; Nagumo, T. Carbohydr. Res. 1992, 229, 355–
362.
26. Alban, S.; Franz, G. Thromb. Res. 2000, 99, 377–388.
27. Durieux, P.; Fernandez-Carneado, J.; Tuchscherer, G.
Tetrahedron Lett. 2001, 42, 2297–2299.
28. Ueki, M.; Watanabe, S.; Saitoh, T.; Nakashima, H.
Bioorg. Med. Chem. 2001, 9, 487–492.
29. Liu, Y.; Lien, I. F. F.; Ruttgaizer, S.; Dove, P.; Taylor, S.
D. Org. Lett. 2004, 6, 209–212.
30. Barron, D.; Ibrahim, R. K. Tetrahedron 1987, 43, 5197–
5202.
31. Nakasima, H.; Ichiyama, K.; Hirayama, F.; Uchino, K.;
Ito, M.; Saitoh, T.; Ueki, M.; Yamamoto, N.; Ogawara,
H. Antivir. Res. 1996, 30, 95–108.
32. Kice, J. L.; Anderson, J. M. J. Am. Chem. Soc. 1966, 88,
5242–5245.
33. Roy, A. B. In Sulfation of drugs and related Compounds:
The chemistry of sulfate esters and related compounds;
Mulder, G. J., Ed.; CRC: Boca Raton, Fl, 1981, pp 5–30.
34. Chai, W. G.; Luo, J. L.; Lim, C. K.; Lawson, A. M. Anal.
Chem. 1998, 70, 2060–2066.
35. Kuberan, B.; Lech, M.; Zhang, L.; Wu, Z. L.; Beeler, D.
L.; Rosenberg, R. D. J. Am. Chem. Soc. 2002, 124, 8707–
8718.
3
4
5
6
7
2
003, 13, 679–683.
. Gunnarsson, G. T.; Desai, U. R. J. Med. Chem. 2002, 45,
460–4470.
. Gunnarsson, G. T.; Desai, U. R. J. Med. Chem. 2002, 45,
233–1243.
4
1
. Guglielmone, H. A.; Agnese, A. M.; Nunez Montoya, S.
C.; Cabrera, J. L. Thromb. Res. 2002, 105, 183–188.
. Cheshenko, N.; Keller, M. J.; MasCasullo, V.; Jarvis, G.
A.; Cheng, H.; John, M.; Li, J. H.; Hogarty, K.;
Anderson, R. A.; Waller, D. P.; Zaneveld, L. J.; Profy,
A. T.; Klotman, M. E.; Herold, B. C. Antimicrob. Agents
Chemother. 2004, 48, 2025–2036.
8
. Nyberg, K.; Ekblad, M.; Bergstrom, T.; Freeman, C.;
Parish, C. R.; Ferro, V.; Trybala, E. Antiviral. Res. 2004,
63, 15–24.
. Santos, G. A.; Murray, A. P.; Pujol, C. A.; Damonte, E.
B.; Maier, M. S. Steroids 2003, 68, 125–132.
9
1
1
0. Ibrahim, A. R. S.; Galal, A. M.; Ahmed, M. S.; Mossa, G.
S. Chem. Pharm. Bull. (Tokyo) 2003, 51, 203–506.
1. Ying, C.; Van Pelt, J. F.; Van Lommel, A.; Van Ranst, M.;
Leyssen, P.; De Clercq, E.; Neyts, J. Antivir. Chem.
Chemother. 2002, 13, 157–164.
1
1
2. Arthan, D.; Svasti, J.; Kittakoop, P.; Pittayakhachonwut,
D.; Tanticharoen, M.; Thebtaranonth, Y. Phytochemistry
2
002, 59, 459–463.
3. Howarth, N. M.; Purohit, A.; Robinson, J. J.; Vicker, N.;
Reed, M. J.; Potter, B. V. Biochemistry 2002, 41, 14801–
14814.
4. Billich, A.; Nussbaumer, P.; Lehr, P. J. Steroid Biochem.
Mol. Biol. 2000, 73, 225–235.
5. Bjork, I.; Olson, S. T. Adv. Exp. Med. Biol. 1997, 425, 17–
36. Gunay, N. S.; Tadano-Aritomi, K.; Toida, T.; Ishizuka, I.;
Linhardt, R. J. Anal. Chem. 2003, 75, 3226–3231.
37. Chai, W.; Hounsell, E. F.; Bauer, C. J.; Lawson, A. M.
Carbohydr. Res. 1995, 269, 139–156.
38. Siegel, M. M.; Tabei, K.; Kagan, M. Z.; Vlahov, I. R.;
Hileman, R. E.; Linhardt, R. J. J. Mass Spectrom. 1997,
32, 760–772.
1
1
1
3
3.
6. Gettins, P. G. W.; Patston, P. A.; Olson, S. T. Serpins:
Structure, Function and Biology; R.G. Landes and Com-
pany: Austin, TX, 1996.
39. Olson, S. T.; Bjork, I.; Shore, J. D. Methods Enzymol.
1993, 222, 525–559.