Chemistry Letters Vol.35, No.1 (2006)
129
NH2R2
O
mesylation at the 3-position of 2 with a limited amount of mesyl
chloride in the presence of pyridine was also unsuccessful. Next,
we tried to convert rutin into 3-aminoluteorin as shown in
Scheme 4. Rutin 12 was initially benzylated and subsequently
hydrolyzed to give the benzylated compound 13 which was
selectively mesylated in pyridine at the 3-position of 13 to
give the 3-mesyloxyflavone 14. The mesyloxy compound 13
was treated with liquid ammonia under pressure to give the 3-
amino compound 15 which was debenzylated to give 3-ammino-
luteorin5 4 in 55% yield (2 steps). However, in this route, some
problems in purification process of 4 occurred due to the similar
physicochemical properties between 4 and impurities. Then, the
3-mesyloxyflavone 14 was firstly reacted with benzylamine and
the resulting benzylated compound 16 was hydrogenated to give
the desired 4 in quantitative yield without any difficulty of
purification.
O
O
OR1
NHR2
O
5 or 6
11
R2
H
O
O
O
O
R2
N
H
N
H
- R1OH
OR1
9
10
Scheme 3. Plausible reaction mechanism.
OH
OBn
OBn
HO
O
OH
OH
BnO
O
O
Thus the present method provides an easy way to make a
number of polyhydroxy-3-aminoflavones in a large quantity to
test their biological activities.
a), b)
O
OH
OH
OH
OH
O
O
OH
OH
O
OH
OH
12
13
O
c)
References and Notes
1
2
Report appears in the patent literature: K. Kanazawa, M.
Sasaki, Jpn. Kokai Tokkyo Koho 2004123728, 2004; Chem.
Abstr. 2004, 140, 350536.
OBn
OBn
OBn
OBn
BnO
O
BnO
O
f)
NHBn
OMs
a) D. Dauzonne, B. Folleas, L. Martines, G. G. Chabot, Eur.
OH
O
OH
O
´
J. Med. Chem. 1997, 32, 71. b) T. Patonay, M. Rakosi, G.
Litkei, R. Bognar, Liebigs Ann. Chem. 1979, 161. c) R.
14
16
e)
´
d)
´
´
Bognar, M. Rakosi, Liebigs Ann. Chem. 1966, 225. d) C.
O’Brien, E. M. Philbin, S. Ushioda, T. S. Wheeler, Tetra-
hedron 1963, 19, 373. e) A. Kasahara, Nippon Kagaku
Zasshi, 1959, 80, 416; Chem. Abstr. 1961, 55, 27860.
3-Aminoflavone 7: yellow solid, mp: 158–160 ꢂC (differs
from that of lit. 2: 136–138 ꢂC, however, the following spec-
tral data as well as those of 8 support the chemical structure
of 7); 1H NMR (300 MHz, CDCl3): ꢀ 7.17 (1H, ddd, J ¼ 7:5,
7.2, 0.6 Hz), 7.25 (1H, d, J ¼ 8:4 Hz), 7.48–7.55 (4H, m),
7.83–7.86 (3H, m); 13C NMR (75 MHz, CDCl3): ꢀ 112.6,
121.9, 123.2, 124.1, 128.5, 128.8, 131.0, 131.5, 132.7,
133.0, 146.6, 161.9, 180.3; HRTOFMS (ESI): m=z
238.0866 (calcd for C15H12NO2 [M + H]þ 238.0862).
3-N-Acetylaminoflavone 8: yellow solid, mp: 156–157 ꢂC;
1H NMR (300 MHz, CDCl3): ꢀ 2.25 (3H, s, CH3), 7.19–
7.26 (2H, m), 7.47–7.49 (3H, m), 7.58–7.64 (3H, m), 7.82
(1H, dd, J ¼ 7:5, 0.9 Hz), 10.83 (1H, brs, NH); 13C NMR
(75 MHz, CDCl3): ꢀ 24.9, 113.1, 122.6, 123.0, 124.2,
128.1, 129.2, 130.1, 130.6, 136.0, 136.1, 136.5, 164.7,
169.5, 184.3; HRTOFMS (ESI): m=z 280.0979 (calcd for
C17H14NO3 ½M þ Hꢃþ 280.0968).
OH
OH
OBn
OBn
HO
O
O
BnO
O
e)
NH2
NH2
OH
OH
O
3
15
4
Scheme 4. Reagents and conditions: a) BnCl, DBU, DMF, re-
flux and then b) HCl aq, EtOH, reflux, 30% (2 steps); c) MsCl,
pyridine, rt, 94%; d) NH3 liq., THF, ꢁ5 kgf/cm2, and then e)
H2, cat. Pd(OH)2, THF/EtOH, rt, 55% (2 steps); f) BnNH2,
THF, rt, 82% and then e) H2, cat. Pd(OH)2, THF/EtOH, rt,
quant.
give the aziridine 10 with loss of toluenesulfonic or methanesul-
fonic acid. The aziridine ring of 10 may open to afford the 3-
aminoflavones 11. In this mechanism, secondary amines like
diethylamine can not form the aziridine ring. At the moment
we can not have any evidence of the intermediates 9 and 10 in
terms of IR and NMR spectroscopies despite that the reaction
proceeds slowly. Probably, the initial Micheal addition is in
the rate-determining step.
We applied this method to the synthesis of 3-aminoluteorin
4 from commercially available quercetin 2 via the mesylated
quercetin. Quercetin 2 was reacted with an excess of mesyl chlo-
ride in pyridine to give the compound mesylated at the 3, 5, 7, 30,
and 40 positions of 2 which was treated with liquid ammonia at
room temperature in DMF under pressure, yielding the corre-
sponding 3-amino compound. However, the complete deprotec-
tion of the mesyloxy to hydroxyl groups in the 3-amino com-
pound under alkaline condition was failed. Attempted selective
4
5
3-Aminoluteorin 4: yellow resinous material; 1H NMR
(300 MHz, CD3OD) ꢀ 5.96 (1H, d, J ¼ 1:3 Hz), 6.08 (1H,
d, J ¼ 1:3 Hz), 6.84 (1H, d, J ¼ 8:4 Hz), 7.19 (1H, dd,
J ¼ 8:4, 1.6 Hz), 7.30 (1H, d, J ¼ 1:6 Hz); 13C NMR (75
MHz, CD3OD) ꢀ 90.7, 97.6, 107.4, 116.3, 117.2, 122.2,
125.3, 131.7, 146.3, 148.8, 149.5, 157.8, 165.2, 165.6,
179.1; HRTOFMS (ESI): m=z 302.0667 (calcd for C15H12-
NO6 ½M þ Hꢃþ 302.0659).