An expedient synthesis of (+)-quinolactacin A2

Article abstract of DOI:10.1016/j.tetlet.2004.10.001

An expedient synthesis of quinolactacin A2 from N-methylisatoic anhydride and N-Boc-(2S,3S)-isoleucine has been achieved. The key step involves the Friedl?nder-type annulation of isatoic anhydride and β-ketoester derived from isoleucine.

Full text of DOI:10.1016/j.tetlet.2004.10.001

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 8793–8795
An expedient synthesis of (+)-quinolactacin A2
a
Su-Jin Park, Kwang-Nym Cho, Won-Gon Kim and Kee-In Lee
a
b
a,
*
a
Bio-Organic Science Division, Korea Research Institute of Chemical Technology, PO Box 107, Yusong, Taejon 305-600, Korea
b
National Research Laboratory of Antioxidant, Korea Research Institute of Bioscience and Biotechnology, PO Box 115,
Yusong, Taejon 305-600, Korea
Received 1September 2004; accepted 1October 2004
Abstract—An expedient synthesis of quinolactacin A2 from N-methylisatoic anhydride and N-Boc-(2S,3S)-isoleucine has been
achieved. The key step involves the Friedl a¨ nder-type annulation of isatoic anhydride and b-ketoester derived from isoleucine.
ꢀ
2004 Elsevier Ltd. All rights reserved.
Quinolactacin A, B, and C containing a novel pyr-
rolo[3,4-b]quinolone skeleton were isolated from the fer-
mentation broth of the fungal strain, Penicillium sp.
The first biomimetic synthesis of quinolactacin B sug-
gests that quinolactacins might be biologically synthe-
sized from three components, amino acids, anthranilic
1
3
EPF-6. The structure of quinolactacins is unique in that
acid, and acetic acid. This synthesis enlightens a general
a quinolone skeleton is conjugated with a c-lactam ring.
While a series of quinolactacins exhibited very weak
anti-microbial activity, quinolactacin A showed inhibi-
tory activity against tumor necrosis factor production
induced by murine peritoneal macrophage. More
recently, two diastereomeric quinolactacin A1and A2,
as shown in Figure 1, were isolated from solid-state fer-
route to the many quinolactacin analogs. Recently, the
enantioselective syntheses of (+)-quinolactacin A2 and
(+)-quinolactacin B have been made through Winter-
feldt oxidation of the Pictet–Spengler adducts starting
4
from tryptamine and the corresponding aldehydes.
It has been known that isatoic anhydride 1 is susceptible
5
2
mentation of Penicillium citrinum 90684. It has been
to nucleophilic attack, normally at the ÔacidÕ 4-carbon.
reported that quinolactacin A2 exhibited 14 times higher
anti-acetylcholinesterase activity than its diastereomer
quinolactacin A1, and selective inhibitory activity for
acetylcholinesterase against butyrylcholinesterase.
In case of malonates, the reaction proceeds with concomi-
tant removal of alcohol to afford quinoline-2,4-dione 2A
(route A, Scheme 1). Whereas, b-ketoesters react with 1
and further undergo ring closure by enamine formation
with the b-keto group to produce 4-oxoquinoline 2B
(
route B). We envisioned that a Friedl a¨ nder-type annu-
O
O
O
O
lation of isatoic anhydride and b-ketoesters derived
from amino acids would offer the most concise synthetic
route to quinolactacins. Here we would like to report an
expedient synthesis of quinolactacin A2.
N H
R2
N H
R2
N
N
Me
Me
R1
H
The required isoleucine derivative 3 was prepared from
N-Boc-(2S,3S)-isoleucine according to the reference
procedure. After activation of its carboxyl group to
the corresponding imidazolide (CDI, THF, 2h), fol-
1
2
Quinolactacin A : R = Me, R = H
6
1
2
2
Quinolactacin B : R = H, R = H
Quinolactacin A1 : R = β-H
Quinolactacin A2 : R = α-H
1
2
2
Quinolactacin C : R = Me, R = OH
lowed by the treatment with ethyl lithioacetate (EtOAc,
7
Figure 1. Structures of quinolactacin A–C and diastereomeric quino-
lactacin A1and A2.
LiHMDS, ꢀ78ꢁC, THF)
afforded the known
8
b-ketoester 3 in 81% yield. Next, we evaluated the
Friedl a¨ nder-type annulation of the readily available
N-methylisatoic anhydride 1 with the b-ketoester 3 to
secure the 4-oxoquinoline 4, as shown in Scheme 2.
After examining a number of bases commonly employed
Keywords: Isatoic anhydride; b-ketoester; Friedl a¨ nder-type annulation;
(
*
+)-Quinolactacin A2.
Corresponding author. Tel.: +82 42 860 7186; fax: +82 42 860
160; e-mail: kilee@krict.re.kr
9
7
in such cyclization, we found that DBU was the base of
0
040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.10.001

8794
S.-J. Park et al. / Tetrahedron Letters 45 (2004) 8793–8795
O
O
OH
O
O
R1
R1
O
R1
R1
R2
OR2
R2
route B
_
_
O
route A
N
N
O
N
Me
Me
Me
2
A
1
2B
Scheme 1.
O
H
O
O
O
O
OEt
OEt
a
H
H
b
N H
H
1
+
NHBoc
NHBoc
O
N
N
Me
Me
H
H
3
4
5
˚
Scheme 2. Reagents and conditions: (a) DBU, molecular sieves 3A, CH
2
Cl
2
, 20h, 42%; (b) Et
2
O/H O/TFA, 2h, 87%.
2
O-
O
O
O
N H
N H
H
Quinolactacin A2
N
N
Me
Me
H
H
Quinolactacin A1
Figure 2. Epimerization of quinolactacin A2 into quinolactacin A1.
choice and the yield was improved by adding molecular
˚
sieves 3A as a dehydrating agent. In this way, the
In summary, we have achieved the expedient synthesis
of (+)-quinolactacin A2 from isatoic anhydride and
N-Boc-(2S,3S)-isoleucine utilizing Friedl a¨ nder-type
annulation. This route described herein may offer highly
efficient approach to the total synthesis of quinolactacin
family and analogs. Currently, a similar chemistry is
ongoing to establish the absolute configurations of
quinolactacins, which have not been clearly determined.
1
0
desired 4-oxoquinoline
4
was realized in 42%
yield without the formation of quinoline-2,4-dione.
Unfortunately, we could not avoid the regeneration of
N-methylanthranilic acid during the reaction. Finally,
N-Boc amino ester 4 was subjected to deprotection of
N-Boc group under acidic conditions (Et O/H O/TFA)
2
2
and which in turn spontaneously cyclized to give the
tricyclic compound 5 (87%) and its diastereomer at the
1
1
ratio of ꢁ8:1. The compound 5 was matched with
References and notes
(
+)-quinolactacin A2 in the spectral and optical data
21
reported in the literatures; ½aꢂ +20.5 (c 0.26, DMSO)
D
2
5
1b
20
1. (a) Kakinuma, N.; Iwai, H.; Takahashi, S.; Hamano, K.;
Yanagisawa, T.; Nagai, K.; Tanaka, K.; Suzuki, K.;
Kirikae, F.; Kirikae, T.; Nakagawa, A. J. Antibiot. 2000,
[
0
lit. ½aꢂ +17.9 (c 0.13, DMSO) and ½aꢂD +19.5 (c
D
.7, DMSO) ].
4
5
3, 1247; (b) Takahashi, S.; Kakinuma, N.; Iwai, H.;
While an expedient synthesis of (+)-quinolactacin A2
from N-Boc-(2S,3S)-isoleucine had been achieved, it
was prone to epimerized to its diastereomer with pro-
longed contact time with silica gel, protic solvents, and
acidic media and vice versa (Fig. 2). To evaluate the epi-
merization of (+)-quinolactacin A2, the diastereomeric
mixture was isolated by chiral HPLC (Chirex (R)-
Yanagisawa, T.; Nagai, K.; Suzuki, K.; Tokunaga, T.;
Nakagawa, A. J. Antibiot. 2000, 53, 1252.
2. Kim, W.-G.; Song, N.-K.; Yoo, I.-D. J. Antibiot. 2001, 54,
831.
3
. Tatsuta, K.; Misawa, H.; Chikauchi, K. J. Antibiot. 2001,
4, 109.
. Zhang, X.; Jiang, W.; Sui, Z. J. Org. Chem. 2003, 68, 4523.
5
4
1
2
5. (a) Coppola, G. M. Synthesis 1980, 505; (b) Coppola, G.
M.; Hardtmann, G. E. J. Heterocycl. Chem. 1979, 16,
NGLY and DNB (4 · 250mm)). The epimer was
1
13
totally matched with H NMR and C NMR data of
quinolactacin A1and thus identified as C-3 diastereo-
mer to (+)-quinolactacin A2. This result clearly
suggested that quinolactacin A2 and A1are C-3
diastereomers each other.
1
605.
6
. (a) Hamada, Y.; Kondo, Y.; Shibata, M.; Shioiri, T. J.
Am. Chem. Soc. 1989, 111, 669; (b) Harris, B. D.; Joulli e´ ,
M. M. Tetrahedron 1988, 44, 3489.
7. Rathke, M. W. J. Am. Chem. Soc. 1970, 92, 3222.

S.-J. Park et al. / Tetrahedron Letters 45 (2004) 8793–8795
8795
8
9
. (a) Paris, M.; Fehrentz, J.-A.; Heitz, A.; Loffet, A.;
HREI-MS Calcd for C23
416.2309.
11. To a solution of 10mL of Et O/H O/TFA (1/1/3) was
H
32
N
2
O
5
: 416.2311, Found:
Martinez, Z. Tetrahedron Lett. 1996, 47, 8489; (b) Shioiri,
T.; Hayashi, K.; Hamada, Y. Tetrahedron 1993, 49, 1913.
. (a) Dorma, P. G.; Eng, K. K.; Farr, R. N.; Humphrey, G.
R.; McWilliams, J. C.; Reider, P. J.; Sager, J. W.; Volante,
R. P. J. Org. Chem. 2003, 68, 467; (b) Vasse, J.-L.;
Levacher, V.; Bourguignon, J.; Dupas, G. Tetrahedron
2
2
added 4 (658mg, 1.57mmol) at room temperature. The
reaction mixture was stirred for 2h. Saturated NaHCO
3
was added, and the aqueous layer was extracted three
times with EtOAc. The combined organic layer was
2
003, 59, 4911; (c) Camps, P.; G o´ mez, E.; Mu n˜ oz-Torrero,
D.; Arn o´ , M. Tetrahedron: Asymmetry 2001, 12,
909.
0. Satisfactory analytical data were obtained for the com-
4
washed with brine, dried over anhydrous MgSO , filtered,
and concentrated to give the tricyclic compound 5 and its
diastereomer at the ratio of 8:1. The diastereomeric
mixture was isolated by preparative TLC (silica gel,
2
1
1
pound 4: H NMR (CDCl
1
3
) d 8.44 (1H, dd, J = 7.5,
3
THF/CHCl = 6:1) to give 5 as a white solid (369mg,
1 13
87%). The H NMR and C NMR of 5 identical with
.4Hz), 7.68 (1H, ddd, J = 8.2, 7.5, 1.4Hz), 7.56 (1H, d,
J = 8.1Hz), 7.39 (1H, dd, J = 7.7, 7.5Hz), 6.23 (1H, d,
J = 9.3Hz), 4.95 (1H, dd, J = 10.1, 2.3Hz), 4.42 (2H, q,
J = 7.2Hz), 4.01 (3H, s), 1.82–1.75 (2H, m), 1.45 (3H, t,
J = 7.1Hz), 1.43 (9H, s), 0.94 (3H, t, J = 7.4Hz), 0.81(3H,
the data of (+)-quinolactacin A2 reported in Ref. 1b:
2
1
25
D
½aꢂ +20.5 (c 0.26, DMSO) [lit. ½aꢂ +17.9 (c 0.13,
D
1
b
20
4
DMSO) and ½aꢂ +19.5 (c 0.7, DMSO) ].
D
12. HPLC conditions: column, Chirex (R)-NGLY and DNB
1
3
d, J = 6.7Hz); C NMR (CDCl ) d 175.3, 169.8, 155.7,
(4 · 250mm), Phenomenex; eluent, MeOH–H O (45:50);
3
2
1
53.8, 141.7, 132.8, 126.7, 126.6, 124.2, 116.8, 116.5, 80.0,
1.9, 66.0, 38.8, 36.5, 28.3, 26.4, 15.7, 14.1, 11.4; EIMS m/z
flow rate, 1mL/min; UV absorbance at 315nm; retention
time, (+)-quinolactacin A1(9.8min) and (+)-quinolactacin
6
+
rel intensity) 416 (M , 75), 387 (18), 360 (20), 343 (15),
21
(
A2 (10.7min). (+)-quinolactacin A1: ½aꢂD +10.8 (c 0.39,
3
31(52), 303 (27), 271(45), 259 (30), 241(56), 2 13 ( 10 0);
DMSO).