The first total synthesis and structural determination of lagunamycin

Article abstract of DOI:10.1016/j.tetlet.2006.06.158

Lagunamycin (1) has been synthesized by using our developed remote stereoinduction, Knorr condensation, periodinane oxidation, and diazonation. This enantioselective synthesis determined the absolute configuration of lagunamycin. Existence of rotational c

Full text of DOI:10.1016/j.tetlet.2006.06.158

Tetrahedron Letters 47 (2006) 6183–6186
The first total synthesis and structural determination
of lagunamycin
Seijiro Hosokawa,* Shoichi Kuroda, Keisuke Imamura and Kuniaki Tatsuta*
Department of Applied Chemistry, School of Science and Engineering, Waseda University, 3-4-1 Ohkubo,
Shinjuku-ku, Tokyo 169-8555, Japan
Received 12 June 2006; revised 28 June 2006; accepted 29 June 2006
Abstract—Lagunamycin (1) has been synthesized by using our developed remote stereoinduction, Knorr condensation, periodinane
oxidation, and diazonation. This enantioselective synthesis determined the absolute configuration of lagunamycin. Existence of rota-
tional conformers was confirmed by NMR studies.
Ó 2006 Elsevier Ltd. All rights reserved.
Lagunamycin (1), a metabolite isolated from the culture
filtrate of Streptomyces sp. AA0310 showed inhibitory
activity against 5-lipoxygenases and antibacterial activ-
ity against Gram-positive bacteria.1a The structure of
lagunamycin (1) has been elucidated to possess the di-
azotetraoxoquinoline skeleton attaching the branched
alkyl chain as shown in Figure 1 by a combination of
NMR studies and chemical degradations.1b The com-
plexity of NMR spectra of lagunamycin (1) has been be-
lieved to be the result from the rotational isomers caused
by the bulky side chain against the 9-methyl group
attaching on quinoline plane.1b Interested in the struc-
ture and bioactivities, we embarked on the synthetic
studies on lagunamycin (1). Herein, we present the first
total synthesis and structural determination of laguna-
mycin (1).
Our retrosynthetic analysis is shown in Scheme 1. The
diazo group would be introduced in the final step and
oxygenated quinoline moiety could be synthesized from
Scheme 1. Retrosynthetic analysis of lagunamycin.
*
Corresponding authors. Tel./fax: +81 3 3200 3203; e-mail: tatsuta@ waseda.jp
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.06.158

6184
S. Hosokawa et al. / Tetrahedron Letters 47 (2006) 6183–6186
2. To avoid the difficulty from steric interaction between
the side chain and the quinolone moiety, we planned to
construct the quinolone 2 by Knorr condensation with
the b-ketoamide 3,2 which would be obtained by cou-
pling of the b-hydroxycarboxyacid 4 and the aniline 5.
The b-hydroxycarboxyacid 4 might be derived from
the a,b-unsaturated imide 6,3c which had already been
obtained by our developed remote stereoinduction reac-
tion with the N,O-ketene acetal 8.3
the primary alcohol 10. The allylic alcohol 10 was oxi-
dized to the aldehyde 11,4 which was submitted to aldol
reaction with the dianion derived from propionic acid.
The resulting b-hydroxycarboxylic acid 4 was used for
the further transformation as a diastereomeric mixture.
Construction of the quinolone moiety and completion of
total synthesis of lagunamycin is described in Scheme 3.
Condensation of the b-hydroxycarboxylic acid 4 and
the aniline 55 with WSCI afforded the anilide 12 in excel-
lent yield. Treatment of anilide 12 with o-iodoxybenzoic
acid (IBX) to promote Nicolaou oxidation6 smoothly
(30 min) at room temperature gave the quinone moiety
and oxidation of allylic alcohol slowly (overnight) at
40 °C gave the b-ketoamide 13. The structure of 134
was determined as o-quinone as shown in Figure 1. The
NOE between the methoxy proton and both protons
on the quinone ring was observed. Additionally, the
HMBC spectrum shows the cross peaks between the
methoxy proton and both C3 and C5 carbons. These
results support the o-quinone structure of 13. The
The actual synthesis of lagunamycin (1) was started
from construction of the side chain moiety 4 (Scheme
2). Our remote stereoinduction reaction with the ketene
N,O-acetal 8 and isobutyraldehyde (7) gave the adduct 6
in excellent yield and stereoselectivity in multi-gram
scale as expected from our previous report.3c De-oxy-
genation at C50 position was realized by hydride reduc-
tion with the sulfonate 9. Treatment of phenylsulfonate
9 with super hydride (LiBEt3H) made rapid progress of
the reductive removal of oxazolidine ring at ꢀ78 °C and
promoted de-sulfonation at room temperature to give
Scheme 2. Reagents and conditions: (a) Ref. 3c; (b) LHMDS, PhSO2Cl, THF, ꢀ78 to ꢀ40°C, 94%; (c) LiBEt3H, THF, ꢀ78 °C to rt, 12 h, 65%; (d)
MnO2, hexane, 40°C, 12 h, 92%; (e) LDA, CH3CH2CO2H, THF, ꢀ78 °C, 30 min, 70%.
Scheme 3. Reagents and conditions: (a) WSCIÆHCl, CH2Cl2, rt, 2 h, 96%; (b) IBX, PhMe–DMSO, rt, 30 min, then 40 °C, 12 h, 83%; (c) 10% H2SO4
aq, CH3CN, rt, 15 min; (d) Na2S2O4, 40 °C, 12 h; (e) OXONEÒ, rt, 15 min, 87% from 13; (f) p-acetamidobenzenesulfonyl azide, DBU, THF, rt, 4 h,
52%.

S. Hosokawa et al. / Tetrahedron Letters 47 (2006) 6183–6186
6185
spectrum of 1 at 24 °C showed two kinds of compounds.
At 80 °C, the peaks assigned as H30 and H40 got simple,
and a broad peak of H10 became doublet showing allylic
coupling with H30 ðJ1 ;3 ¼ 1:2 HzÞ. The peaks around
0.9 ppm became simpler, however, nine peaks were ob-
served yet. After cooling from 80 °C to room tempera-
ture, the resulting sample gave the same spectrum as
the original one obtained at 24 °C. At 90 °C (not
included in Fig. 2), eight peaks were observed around
0.9 ppm and the sample began to decompose slowly.
At 110 °C, three methyl groups of lagunamycin includ-
ing H70, H80, and H90 appeared as three doublets, which
imposed on the peaks of the unidentified degradation
products (three doublets at 0.89, 0.92, and 1.00 ppm).
Therefore, additional NMR experiments on rotation
of the side chain were practiced with the more stable
quinone 15 (Fig. 3). The 1H NMR spectrum of 15
showed two isomers at 26 °C and its behavior at differ-
ent temperatures was the same as that of lagunamycin
(1) without remarkable decomposition. The peaks
around 0.9 ppm became simpler at 100 °C, and finally,
they fixed to three doublets at 110 °C. The spectrum at
110 °C became the simple one as expected from the pla-
nar structure of 15. After cooling to room temperature,
the resulting sample gave the same spectrum as the ori-
ginal one obtained at 26 °C. The feature of temperature-
dependence of the quinone 15 was as same as that of
lagunamycin (1). Thus, the origin of complexity of
NMR spectra was confirmed due to the existence of a
rotational isomer.
0
0
Figure 1. Structure of o-quinone 13.
o-quinone 13 was transformed to the quinolone 15 by the
one-pot procedure. Subsequent manipulation of quinone
13 (red solution) including (i) hydrolysis of methyl ether
to convert to the unstable quinone 14 (yellow solution),
(ii) selective reduction of the quinone moiety with
Na2S2O4 to provide the very labile trihydroxyanilide 3
(colorless solution), (iii) Knorr condensation under the
acidic conditions2 to give quinolone 2 (pale yellow
solution), and (iv) oxidation of hydroquinone with
Oxone, delivered the quinone 154 (orange solution) in
high yield.7 Finally, treatment of 15 with p-acet-
amidobenzenesulfonyl azide8 in the presence of DBU
gave lagunamycin (1),9 whose analytical data were
consistent with those reported previously.1 Thus, total
synthesis of lagunamycin was accomplished and the abso-
lute structure of 1 was determined as 40R configuration.
To prove the existence of rotational isomers, the follow-
ing NMR experiments were carried out. H NMR spec-
Additionally, the NOE experiments made clear the ratio
of rotational isomers at room temperature (Fig. 4). The
relations between H9 and H90, H10 and H40, and be-
tween H9 and H60 were observed. The 600 Hz 1H
NMR spectrum made it possible to distinguish the three
methyl groups of rotational isomers (the major iso-
mer:the minor isomer = 1.2:1).9 The NOESY spectrum
of 1 (in CDCl3, 27 °C) determined the stereochemistry
of these isomers. The relations between H9 and H60 of
the major isomer, H9 and H70 of the major isomer,
1
tra of lagunamycin (1) in dimethyl sulfoxide-d6 were
observed at different temperatures including 24 °C,
1
40 °C, 60 °C, 80 °C, and 110 °C (Fig. 2). The H NMR
Figure 2. 1H NMR spectrum of lagunamycin (1) in DMSO-d6 at the
various temperatures: (a) H30; (b) H40; (c) H10; (d) H70, 80, and 90.
Figure 3. 1H NMR spectrum of the quinone (15) in DMSO-d6 at
various temperatures: (a) H6; (b) H70, 80, and 90.

6186
S. Hosokawa et al. / Tetrahedron Letters 47 (2006) 6183–6186
Oka, M.; Yamasaki, T.; Konishi, M.; Oki, T. J. Antibiot.
1993, 46, 1031–1033.
2. Kelly, T. R.; Field, J. A.; Li, Q. Tetrahedron Lett. 1988, 29,
3536–3545.
3. (a) Tatsuta, K.; Hosokawa, S. Chem. Rev. 2005, 105, 4707–
4729; (b) Hosokawa, S.; Ogura, T.; Togashi, H.; Tatsuta,
K. Tetrahedron Lett. 2005, 46, 333–337; (c) Shirokawa, S.;
Kamiyama, M.; Nakamura, T.; Okada, M.; Nakazaki, A.;
Hosokawa, S.; Kobayashi, S. J. Am. Chem. Soc. 2004, 126,
13604–13605; (d) Crystallographic data (excluding struc-
ture factors) for the structures of 8 have been deposited
with the Cambridge Crystallographic Data Center as
supplementary publication numbers CCDC 609888 for 8.
4. See Supplementary data.
Figure 4. The NOE relation of rotational isomers in CDCl3.
5. Wendt, M. D.; Rockway, T. W.; Geyer, A.; McClellan, W.;
Weitzberg, M.; Zhao, X.; Mantei, R.; Nienaber, V. L.;
Stewart, K.; Klinghofer, V.; Girnda, V. L. J. Med. Chem.
2004, 47, 303–324.
and between H9 and H90 of the minor isomer were
observed. These results revealed the ratio of two isomers
at room temperature to be 1M:1P = 1:1.2.
6. (a) Nicolaou, K. C.; Baran, P.; Kranich, R.; Zhong, Y.-Li.;
Sugita, K.; Zou, N. Angew. Chem., Int. Ed. 2001, 40, 202–
206; (b) Nicolaou, K. C.; Zhong, Y.-Li.; Baran, P. Angew.
Chem., Int. Ed. 2000, 39, 622–625.
In conclusion, we achieved the first total synthesis of
lagunamycin to determine the absolute structure and
confirmed the existence of rotational isomers.
7. One-pot transformation of o-quinone 13 to p-quinone 15:
Quinone 13 (13.4 mg, 37.0 lmol) was dissolved into a
mixture of acetonitril and 10% H2SO4 aq and the mixture
was stirred at room temperature for 15 min. Na2S2O4
(25.8 mg, 148 lmol) was then added and the resulting
mixture was heated to 40 °C for 12 h. After cooling to room
temperature, Oxone (182 mg, 297 lmol) was added to the
mixture, which was stirred at room temperature for 15 min.
The reaction mixture was extracted with ethyl acetate.
Evaporation and purification by silica gel column chroma-
tography (hexane:ethyl acetate = 2:1, containing 1% triflu-
oroacetic acid) gave orange solid 15 (10.6 mg, 87%).
8. (a) Baum, J. S.; Shook, D. A.; Davis, H. M. L.; Smith, H.
D. Synth. Commun. 1987, 17, 1709–1716; (b) Gleiter, R.;
Dobler, W. Chem. Ber. 1985, 118, 4725–4742.
Acknowledgments
K.I. thanks to JSPS Research Fellowships for Young
Scientists. The authors are grateful for financial support
to 21 COE ‘Center for Practical Nano-Chemistry’, Con-
solidated Research Institute for Advanced Science and
Medical Care, and Grant-in-Aid for Scientific Research
(A), Scientific Research (C), and Scientific Research on
Priority Areas 16073220 from The Ministry of Educa-
tion, Culture, Sports, Science and Technology (MEXT).
9. Lagunamycin (1) (the value in bracket is data of the
Supplementary data
26
isomer): (synthetic): ½aꢁD ꢀ33.0 (c 0.20, MeOH) {natural
26
½aꢁD ꢀ33 (c 0.20, MeOH)}; 1H NMR (600 MHz, CDCl3,
1
The spectrum data of compounds 11, 13, and 15, H
27 °C) d 0.92 [0.89] (3H, d, J = 6.7 Hz), 0.93 [0.96] (3H, d,
J = 6.7 Hz), 1.01 [1.02] (3H, d, J = 6.8 Hz), 1.11–1.27 (2H,
m), 1.56–1.67 (1H, m), 1.903 [1.898] (1H, d, J = 1.3 Hz),
2.190 [2.186] (3H, s), 2.63–2.73 (1H, m), 4.87 [4.85] (1H, dd,
J = 9.4, 1.3 Hz); 13C NMR (150 MHz, CDCl3, 27 °C) d
14.1 [14.0], 16.83 [16.80], 20.1 [20.6], 22.2 [22.5], 23.3 [23.2],
26.0 [25.6], 30.4 [30.3], 46.7 [46.5], 87.6 [87.7], 116.4 [116.3],
129.9 [129.8], 134.8 [135.3], 137.2 [137.1], 138.5 [138.7],
151.6 [151.5], 161.30 [161.35], 168.6 [168.7], 172.53 [172.4],
172.54 [172.6]; MS (FAB+) m/z 356 [M+H]+, 328 [M+H–
N2]+; HRMS (FAB+) calcd for C19H21N3O4 [M+H]+
356.1610, found 356.1589; IR (KBr) 3437 (br), 2954,
2926, 2870, 2146, 1716, 1682, 1651, 1632, 1387, 1369,
1321, 1286, 1178.
NMR spectra of synthetic lagunamycin (400 MHz and
600 MHz in CDCl3), 13C NMR spectrum of synthetic
lagunamycin (100 MHz in CDCl3) and NMR spectra
of natural lagunamycin reprinted from Ref. 1a are pro-
vided as supplementary data, which can be found in the
online version, at doi:10.1016/j.tetlet.2006.06.158.
References and notes
1. (a) Nihei, Y.; Hasegawa, M.; Suzuki, K.; Yamamoto, S.;
Hanada, M.; Furumai, T.; Fukagawa, Y.; Oki, T.
J. Antibiot. 1993, 46, 900–907; (b) Imae, K.; Nihei, Y.;