Synthesis of flavonol derivatives as probes of biological processes

Article abstract of DOI:10.1016/S0040-4039(00)01767-6

Two synthetic derivatives of the flavonol kaempferol were prepared in order to probe the biological mechanism of action of this natural product. Efficient synthetic routes to both a benzophenone-containing derivative and an amine-containing derivative are reported herein. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01767-6

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 9735–9739
Synthesis of flavonol derivatives as probes of biological
processes
Hiroko Tanaka,a Michelle M. Stohlmeyer,a Thomas J. Wandlessa,* and
Loverine P. Taylorb
aDepartment of Chemistry, Stanford University, Stanford, CA 94305, USA
bSchool of Molecular Biosciences, Washington State University, Pullman, WA 99164, USA
Received 25 August 2000; accepted 18 September 2000
Abstract
Two synthetic derivatives of the flavonol kaempferol were prepared in order to probe the biological
mechanism of action of this natural product. Efficient synthetic routes to both a benzophenone-containing
derivative and an amine-containing derivative are reported herein. © 2000 Elsevier Science Ltd. All rights
reserved.
Flavonoids are phenolic secondary metabolites that are widely distributed throughout the
plant kingdom. The average Western diet includes up to 2 g of flavonoids per day, and they
have been identified as antitumor agents, antioxidants, and free radical scavengers.1 In plants,
flavonoids function as UV-protectants, pollinator attractants, and as signaling molecules
between roots and nitrogen-fixing bacteria in the soil. An essential role in plant reproduction has
been established for one class of flavonoid, the flavonols. Pollen from maize and petunia
mutants lacking flavonols is unable to germinate. However, the mutant pollen is viable and the
defect can be biochemically complemented by adding kaempferol at the time of pollination.2
In an effort to identify the biologically relevant receptors of the flavonols, we synthesized two
derivatives of kaempferol at a site that is tolerant of modification.3 The first flavonol derivative
(1) possesses a benzophenone chromophore for use as a photoaffinity reagent.4 The second
derivative (2) substitutes an amino group for the kaempferol hydroxyl group at C4% in order to
facilitate the preparation of a solid-phase matrix for affinity chromatography. Both molecules
are proving to be useful probes for understanding the biochemical basis for pollen germination.
OH
O
OH
O
OH
O
OH
OH
OH
O
HO
O
HO
O
HO
O
Kaempferol
1
2
OH
NH2
* Corresponding author. Tel/fax: 650-723-4005; e-mail: wandless@stanford.edu
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01767-6

9736
Phloroglucinol and benzyloxyacetonitrile were dissolved in ether and treated with HCl to
provide, after hydrolytic workup, the ketone product 3 (Scheme 1).5 Treatment of 3 with
tert-butylchlorodiphenylsilane and triethylamine provided the bis-protected phenol 4 in good
yield. This intermediate may be converted into either 1 or 2, depending on the carboxylic acid
used for the subsequent acylation. Previous flavonol syntheses often did not involve protection
of the phenolic hydroxyl groups, which resulted in multiple acylations in subsequent steps.6 We
selectively protected two of the three available phenols of 3 to ensure that only one hydroxyl
group is available for acylation.
OH
O
TBDPSO
O
TBDPS-Cl
Et3N
phloroglucinol
HCl, ether
OBn
OBn
NC
O
91%
HO
OH
TBDPSO
OH
CH2Cl2
85%
3
4
Scheme 1.
To prepare the benzophenone flavonol derivative, commercially available 3-benzoylbenzoic
acid and phenol 4 were coupled using 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydro-
chloride in the presence of catalytic 4-dimethylaminopyridine and para-toluenesulfonic acid as
shown in Scheme 2. These coupling conditions provide a good yield of ester 5, which is suitably
functionalized for ring closure. Intramolecular dehydrative condensation to provide flavonol 6
was achieved by treating ester 5 with potassium carbonate in refluxing pyridine.7 Partial loss of
the TBDPS protecting groups was observed, however these conditions consistently provide
moderate to good yields of a mixture of 6 and the product lacking both TBDPS groups (9). The
benzyl ether was removed by hydrogenolysis with care taken to prevent overreduction, and
deprotection of the remaining TBDPS ether was achieved with tetrabutylammonium fluoride to
provide the desired benzophenone-substituted flavonol 1 in 57% yield for the final two
transformations.
TBDPSO
O
EDCI
DMAP, TsOH
OBn
+
TBDPSO
OH
O
HO2C
CH2Cl2
84%
4
O
TBDPSO
OBn
K2CO3, pyridine, reflux
66%
TBDPSO
O
5
O
O
TBDPSO
OH
O
O
OH
O
1) H2, Pd/C
2) TBAF, THF
OBn
OH
O
O
HO
O
57%
for 2 steps
6
1
Scheme 2.

9737
Synthesis of the amino-substituted kaempferol 2 is shown in Scheme 3 and proceeds via a
similar strategy to the benzophenone derivative outlined above. Phenol 4 was acylated with
4-azidobenzoic acid under the same conditions used for the benzophenone derivative to provide
ester 7 in 71% yield. Potassium carbonate in refluxing pyridine transformed 7 into the protected
flavonol with concomitant partial deprotection of the TBDPS ethers. Treatment of the resulting
mixture with tetrabutylammonium fluoride in THF provided intermediate 8 in 42% yield
through two steps. The benzyl ether was hydrogenolyzed and the azide group was reduced using
5% palladium on carbon under a hydrogen balloon to provide the desired amino-substituted
kaempferol derivative 2 in 88% yield.
TBDPSO
O
EDCI
DMAP, TsOH
OBn
N3
+
TBDPSO
OH
O
HO2C
CH2Cl2
71%
4
TBDPSO
OBn
1) K2CO3, pyridine
N3
TBDPSO
O
2) TBAF, THF
42% for 2 steps
7
O
OH
O
OH
O
OBn
OH
H2, Pd/C
88%
HO
O
HO
O
8
2
N3
NH2
Scheme 3.
The syntheses outlined above provided probe reagents 1 and 2 for biological characterization,
and preliminary studies show that both kaempferol variants are biologically active. The
benzophenone- and amino-substituted flavonols are both substrates for a pollen-specific flavonol
galactosyltransferase (F3GalTase).8 Further, the benzophenone derivative 1 can be photoacti-
vated and covalently crosslinked to the F3GalTase protein making it a useful probe for catalytic
residues. The amino-derivative induces flavonol-deficient pollen to germinate as efficiently as the
endogenous kaempferol. An affinity matrix prepared from immobilized 2 is being used to
identify kaempferol-binding proteins in germinating pollen. Further studies that utilize these
kaempferol derivatives to probe the biology of pollination are currently in progress and will be
reported in due course.
Selected experimental procedures
Transformation of 3 to 4: Ketone 3 (1.0 g, 3.65 mmol; Rf=0.4, 15% ethyl acetate in hexanes)
was dissolved in 50 mL dichloromethane. Triethylamine (1.27 mL, 9.13 mmol) and tert-
butylchlorodiphenylsilane (2.38 mL, 9.13 mmol) were added over 4 h, and the reaction stirred
overnight at room temperature. Upon completion, the reaction solution was washed twice with
1% HCl. The organic layer was dried over MgSO4, concentrated, and the product was purified
from silica gel eluting with 6% ethyl acetate in hexanes to isolate 2.33 g (3.11 mmol, 85% yield)

9738
1
of compound 4 as a clear oil (Rf=0.65, 20% ethyl acetate in hexanes). H NMR (400 MHz,
CDCl3) l 7.7–7.2 (m, 25H), 5.81 (d, 1H, J=2.2 Hz), 5.55 (d, 1H, J=2.2 Hz), 4.92 (s, 2H), 4.66
(s, 2H), 0.90 (s, 9H), 0.78 (s, 9H). 13C NMR (100 MHz, CDCl3) l 201.3, 166.1, 161.9, 158.8,
137.7, 135.2, 135.1, 135.0, 131.8, 131.1, 130.2, 129.9, 128.4, 128.1, 127.8, 127.7, 106.9, 104.5,
102.4, 75.8, 73.0, 26.6, 26.2, 19.3, 19.1.
Transformation of 4 to 5: Phenol 4 (1.1 g, 1.47 mmol) was dissolved in 15 mL
dichloromethane. 3-Benzoylbenzoic acid (452 mg, 2.0 mmol), DMAP (61 mg, 0.5 mmol), TsOH
(95 mg, 0.5 mmol) and EDCI (478 mg, 2.2 mmol) were added in that order. Additional EDCI
(956 mg, 4.4 mmol) was added over the course of 12 h. After 24 h, the reaction was diluted with
dichloromethane, washed twice with 0.1 M HCl then once with 0.01 M HCl. The organic layer
was dried over MgSO4, concentrated, and the product was purified from silica gel eluting with
10% ethyl acetate in hexanes to provide 1.17 g of compound 5 (1.22 mmol, 84% yield) as a pale
1
yellow oil (Rf=0.4, 20% ethyl acetate in hexanes). H NMR (400 MHz, CDCl3) l 8.48 (s, 1H),
8.30 (dd, 1H, J=7.8 Hz, J=1.4 Hz), 8.07 (dd, 1H, J=7.8 Hz, J=1.4 Hz), 7.81 (dd, 2H, J=8.3
Hz, J=1.2 Hz), 7.62, (t, 2H, J=8.0 Hz), 7.6–7.1 (m, 27 H), 6.31, (d, 1H, J=2.1 Hz), 5.87 (d,
1H, J=2.1 Hz), 4.60 (s, 2H), 4.54 (s, 2H), 0.92 (s, 9H), 0.85 (s, 9H). 13C NMR (100 MHz,
CDCl3) l 199.7, 195.5, 163.8, 157.6, 154.5, 149.0, 138.1, 137.5, 136.9, 135.2, 135.1, 134.8, 133.8,
132.9, 131.7, 131.6, 131.1, 130.1, 130.0, 129.8, 129.3, 128.8, 128.5, 128.3, 127.9, 127.8, 127.7,
117.2, 109.2, 108.1, 76.0, 73.2, 26.3, 26.2, 19.3, 19.2.
Transformation of 5 to 6: Anhydrous potassium carbonate (72.2 mg, 0.522 mmol) was placed
in a 5-mL round bottom flask and compound 5 (94.8 mg, 0.101 mmol) was added in 1 mL
pyridine. The mixture was refluxed for 1 h. After cooling, saturated aqueous ammonium
chloride was added and the organic products were extracted with ethyl acetate. The organic
layer was washed with brine and dried over sodium sulfate. The product was purified from silica
gel (eluting with a gradient of 5% ethyl acetate to 17% ethyl acetate in hexanes) to afford 21.3
mg (0.030 mmol, 30% yield) of partially deprotected compound 6 (Rf=0.6, 25% ethyl acetate in
hexanes) and 17.0 mg (0.037 mmol, 36% yield) of compound 9 (Rf=0.2, 25% ethyl acetate in
1
hexanes). H NMR of 6 (400 MHz, CDCl3) l 8.24 (s, 1H), 8.07 (d, 1H, J=8 Hz), 7.87 (d, 1H,
J=8 Hz), 7.77 (d, 2H, J=8 Hz), 7.71 (d, 4H, J=8 Hz), 7.59 (t, 1H, J=8 Hz), 7.53 (t, 1H, J=8
Hz), 7.47–7.38 (m, 8H), 7.20 (br, 5H), 6.31 (d, 1H, J=2 Hz), 6.26 (d, 1H, J=2 Hz), 5.18 (s, 2H),
1
1.11 (s, 9H). H NMR of 9 (400 MHz, CDCl3+DMSO-d6) l 8.17 (s, 1H), 8.02 (d, 1H), 7.74 (d,
1H), 7.64 (d, 2H), 7.48–7.41 (m, 2H), 7.34–7.30 (m, 2H), 7.08 (br, 5H), 6.25 (s, 1H), 6.18 (s, 1H).
Transformation of 6 to 1: Compound 6 (251 mg, 0.367 mmol) was taken up in 16 mL ethyl
acetate and 36.8 mg of 5% palladium on carbon was added. The mixture was placed under a
hydrogen balloon for 2 h. Filtration and concentration afforded 247 mg of the crude product
(Rf=0.7, 33% ethyl acetate in hexanes). This product was dissolved in 3 mL THF, and
tetrabutylammonium fluoride (0.4 mL, 0.4 mmol, 1 M solution in THF) was added at 0°C. The
reaction was complete within 30 min, and the reaction mixture was purified using silica gel
(eluent: 7% ethyl acetate in hexanes, then 17% ethyl acetate in hexanes, then 17% ethyl acetate
in hexanes with 1% TFA, then 33% ethyl acetate in hexanes with 1% TFA). Compound 1 (138
mg, 0.369 mmol, 57% yield over two steps) was obtained as a yellow solid (Rf=0.3, 33% ethyl
1
acetate in hexanes). H NMR of 1 (400 MHz, DMSO-d6) l 11.01 (s, 1H), 10.02 (s, 1H), 8.66 (s,
1H), 8.51 (d, 1H, J=8.5 Hz), 7.94–7.80 (m, 6H), 7.70 (t, 2H, J=7.6 Hz), 6.55 (d, 1H, J=1.9
Hz), 6.32 (d, 1H, J=1.9 Hz). 13C NMR (100 MHz, acetone-d6) l 177.3, 165.8, 162.7, 158.3,
139.1, 138.8, 138.5, 133.9, 132.7, 132.2, 132.1, 131.1, 130.0, 130.0, 129.7, 104.7, 99.7, 94.9.

9739
1H NMR of 2 (400 MHz, DMSO-d6) l 7.92 (s, 1H), 7.87 (d, 2H, J=8.9 Hz), 6.59 (d, 2H,
J=8.9 Hz), 6.25 (d, 1H, J=2.1 Hz), 6.11 (d, 1H, J=2.1 Hz), 2.55 (br, 2H). 13C NMR (100
MHz, DMSO-d6) l 173.7, 162.0, 159.0, 154.3, 149.3, 146.3, 133.0, 127.5, 115.8, 111.5, 101.1,
96.4, 91.6.
Acknowledgements
We would like to thank the Beckman Foundation for a Young Investigator Award to T.J.W.
and the National Science Foundation for a predoctoral fellowship to M.M.S. and an NSF
Career Advancement Award to L.P.T. for this sabbatical project (MCB-9722774).
References
1. Flavonoids in the Living System; Manthy, J. A.; Buslig, B. S., Eds.; Advances in Experimental Medicine and
Biology 439; Plenum: New York, 1998.
2. Mo, Y.; Nagel, C.; Taylor, L. P. Proc. Natl. Acad. Sci. USA 1992, 89, 7213–7217.
3. (a) Mo, Y.; Nagel, C.; Taylor, L. P. Phytochemistry 1995, 38, 589–592. (b) Xu, P.; Vogt, T.; Taylor, L. P. Planta
1996, 202, 257–265.
4. (a) Dorman, G.; Prestwich, G. D. Biochemistry 1994, 33, 5661–5673. (b) Fleming, S. A. Tetrahedron 1995, 51,
12479–12520.
5. (a) Hoesch, K. Ber. Dtsch. Chem. Ges. 1915, 48, 1122–1133. (b) Gulati, K. C.; Seth, S. R.; Venkataraman, K. In
Org. Synth. Coll.; Blatt, A. H., Ed.; John Wiley: New York, 1943; Vol. II, pp. 522–525.
6. (a) Allan, J.; Robinson, R. J. Chem. Soc. 1926, 2334–2336. (b) Robinson, R.; Venkataraman, K. J. Chem. Soc.
1926, 2344–2348. (c) Looker, J. H.; Hanneman, W. W. J. Org. Chem. 1962, 27, 3261
7. Lynch, H. M.; O’Toole, T. M.; Wheeler, T. S. J. Chem. Soc. 1952, 2063.
8. Miller, K. D.; Guyon, V.; Evans, J. N. S.; Shuttleworth, W. A.; Taylor, L. P. J. Biol. Chem. 1999, 274,
34011–34019.
.