Synthetic study of kosinostatin aglycone: Synthesis of BCDE rings using alkoxycarbonylmethylation of diazonaphthoquinone

Article abstract of DOI:10.1016/j.tetlet.2014.01.082

A synthetic study of kosinostatin aglycone is reported. Synthesis of key intermediate lactone 3, which corresponds to the BCDE ring fragment, was accomplished, and the precursor BCD ring fragment 5 was synthesized via two routes. First, 5 was synthesized from 2,5-dimethoxybenzaldehyde 16 by the combination of typical known transformations including efficient application of non-aqueous OsO₄ oxidation in the presence of PhB(OH)₂. However the synthesis required 15 long steps, and its main difficulty was ortho-alkoxycarbonylmethylation of 1-naphthol. Next we attempted to apply our recently developed alkoxycarbonylmethylation of diazonaphthoquinone for the synthesis of 5, and 5 was successfully synthesized in 9 steps from the same starting compound 16. Finally, 5 was stereoselectively converted to lactone 3 via trifluoroacetic acid-mediated cyclization of the 3,4- epoxycylohexanecarboxylic acid derivative.

Full text of DOI:10.1016/j.tetlet.2014.01.082

Tetrahedron Letters 55 (2014) 1653–1656
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthetic study of kosinostatin aglycone: synthesis of BCDE rings
using alkoxycarbonylmethylation of diazonaphthoquinone
⇑
Mitsuru Kitamura , Kenji Kubo, Shogo Yoshinaga, Hiroki Matsuzaki, Kantaro Ezaki, Taisuke Matsuura,
Daigo Matsuura, Noriyuki Fukuzumi, Keiichiro Araki, Masafumi Narasaki
Department of Applied Chemistry, Kyushu Institute of Technology, 1-1 Sensuicho, Tobata, Kitakyushu 804-8550, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A synthetic study of kosinostatin aglycone is reported. Synthesis of key intermediate lactone 3, which
corresponds to the BCDE ring fragment, was accomplished, and the precursor BCD ring fragment 5 was
synthesized via two routes. First, 5 was synthesized from 2,5-dimethoxybenzaldehyde 16 by the combi-
nation of typical known transformations including efficient application of non-aqueous OsO4 oxidation in
the presence of PhB(OH)2. However the synthesis required 15 long steps, and its main difficulty was
ortho-alkoxycarbonylmethylation of 1-naphthol. Next we attempted to apply our recently developed
alkoxycarbonylmethylation of diazonaphthoquinone for the synthesis of 5, and 5 was successfully syn-
thesized in 9 steps from the same starting compound 16. Finally, 5 was stereoselectively converted to lac-
tone 3 via trifluoroacetic acid-mediated cyclization of the 3,4-epoxycylohexanecarboxylic acid derivative.
Ó 2014 Elsevier Ltd. All rights reserved.
Received 24 December 2013
Revised 16 January 2014
Accepted 22 January 2014
Available online 28 January 2014
Keywords:
Alkoxycarbonylmethylation
Diazonaphthoquinone
Kosinostatin
Lactone
Quinocyclin
The quinocycline/isoquinocyline antibiotics1,2 constitute a class
of natural products, and show several antibiotic and cytotoxic
activities (Fig. 1). In the 1950s, quinocyclines were first isolated
from Streptomyces aureofaciens1 and the structure of isoquinocy-
cline A was confirmed by X-ray analysis in 1968.1e Although the
structure of quinocycline A and B was unclear, quinocycline B
was also isolated from Micromonospora sp. TP-A0468 (along with
isoquinocycline B)2a,2b as kosinostatin in 2002, and the structure
was fully assigned by NMR experiments. Recently, kosinostatin
was also isolated from Streptomyces violaceusniger Strain HAL64. 2c
Quinocyclines/isoquinocyclines have a unique structure con-
sisting of a tetrahydronaphthacenequinone core (ABCD rings) hav-
ing 3-chiral centers (C7, C9, C10) connecting the unusual
pyrrolopyrrole (FG rings) via N,O-acetal, and a sugar. Stimulated
by the unique structure, we have been studying the synthesis of
kosinostatin aglycone 1. Scheme 1 outlines our retrosynthesis plan.
Aglycone 1 is simplified to tetracyclic lactone 3 by assuming the
Diels–Alder reaction with diene 23 and nucleophilic addition of
vinylamidine moiety 4. Three carbon chiral centers in D rings of
lactone 3 were assumed to be constructed by the oxidative cycliza-
tion of alkenyl carboxylic acid;4 therefore, ester 5 was set as its
precursor. The D ring in 5 was expected to be constructed by olefin
metathesis of diene 6,5 which, in turn, could be derived from
1-naphthalenol 8 by the ortho-alkoxycarbonylmethylation and
the successive introduction of a methallyl group at the activated
methylene position in 7.
Recently, Koert and co-workers reported the synthesis of CDEFG
rings 15 of isoquinocyclines (Scheme 2).6 First, lactone 12 was syn-
thesized via iodo lactonization of carboxylic acid 11 and then intro-
duction of 13 or 14 as a unit of A ring to lactone 12, which was
transformed to 15 by several steps. For the synthesis of quinocy-
clines’ aglycone based on Koert’s strategy, preparation of an
ortho-alkoxycarbonylmethyl 1-naphthol derivative [
xyl-2-naphthyl)ester] such as or 2-alkoxycarbonylmethyl
1-anthranol derivative -(1-hydroxy-2-anthryl)ester] may be
a-(1-hydro-
7
[
a
⇑
Corresponding author. Tel./fax: +81 93 884 3304.
E-mail address: kita@che.kyutech.ac.jp (M. Kitamura).
Figure 1. Quinocycline/isoquinocycline antibiotics.
http://dx.doi.org/10.1016/j.tetlet.2014.01.082
0040-4039/Ó 2014 Elsevier Ltd. All rights reserved.

1654
M. Kitamura et al. / Tetrahedron Letters 55 (2014) 1653–1656
Scheme 1. Retrosynthesis of kosinostatin aglycone 1.
Scheme 3. Synthesis of BCD ring fragment
5 (1st generation). Reagents and
conditions: (a) Dimethyl succinate, NaH, cat. MeOH, toluene, rt, 3 h; (b) NaOAc,
Ac2O, 140 °C, 3 h (2 steps 58%); (c) K2CO3, MeOH, rt, 3 h (63%); (d) CH2@CHCH2Br,
K2CO3, acetone, 60 °C, 5 h (74%); (e) DMF, 180 °C, 4 h (67%); (f) MeI, K2CO3, DMF,
80 °C, 6 h (66%); (g) 5 mol % microencapsulated OsO4,11 NMO, H2O, acetone, CH3CN,
rt, 4 days (51%); (h) 7 mol % OsO4, NMO, PhB(OH)2, Na2SO4, CH2Cl2, rt, 20 min; (i)
NaIO4, THF, H2O, 50 °C, 1.5 h (2 steps 83%); (j) NaClO2, Me2C@C(H)Me, NaH2PO4, t-
BuOH, H2O, rt, 1 h; (k) Boc2O, DMAP, t-BuOH, rt, 5 h (2 steps 62%); (l) NaH,
CH2@C(CH3)CH2Br, DMF, rt, 5.5 h (71%); (m) LiAlH4 (3 equiv), THF, À78 ? À30 °C,
1.5 h (92%); (n) SO3Ápyridine, Et3N, DMSO, rt, 8 h (90%); (o) Ph3P+CH3ÁBrÀ, t-BuOK,
THF, rt, 30 min (98%); (p) Grubbs 2nd cat., toluene, 80 °C, 6 h (96%).
Scheme 2. Koert’s synthesis of CDEFG rings of isoquinocyclins.
needed. In our synthetic plan, ortho-alkoxycarbonylmethylation of
1-naphthol is required. However, there are few efficient methods
for the synthesis of a-(ortho-hydroxyaryl)esters from phenol deriv-
atives. In this study, we describe the synthesis of the key interme-
diate lactone 3 for the synthesis of kosinostatin aglycone via two
routes, including study of the ortho-alkoxycarbonylmethylation
reaction to 1-naphthols.
mainly. After the transformation of aldehyde 19 to tert-butyl ester
7a, a methallyl group was introduced at the active methylene part
to afford diester 21. Selective reduction of the methyl ester in 21
was appropriately achieved with LiAlH4 at À30 °C without touch-
ing the tert-butyl ester to give alcohol 22. Oxidation of the alcohol
followed by the Wittig reaction gave diene 6a, whose intramolec-
ular olefin metathesis using Grubbs catalyst (2nd generation)5
smoothly proceeded to afford BCD ring fragment 5.
Although the BCD ring fragment 5 was synthesized from
commercially available aldehyde 16, it required 15 long steps.
The following two are the difficult transformations in Scheme 3:
(i) introduction of the alkoxycarbonylmethyl group to 1-naphthol
8a, and (ii) transformation of the methyl ester in 8a to an alkenyl
moiety, whose ester’s origin is dimethyl succinate on the Stobbe
reaction. The former process was expected to be improved by
alkoxycarbonylmethylation using diazonaphthoquinone, which
Synthesis of the BCD ring fragment 5 started with the readily
available known 1-naphthol 8a, which was synthesized from 2,5-
dimethoxybenzaldehyde (16) in 3 steps via the Stobbe reaction
with dimethyl succinate (Scheme 3).7 For the introduction of an
alkoxycarbonylmethyl group at the ortho position in 1-naphthol
8a, the allyl group was selected as its equivalent.8 After O-allyla-
tion of naphthol 8a, Claisen rearrangement followed by O-methyl-
ation of the thus-formed naphthol gave 2-allyl naphthalene 17.
Although ozonolysis of alkene 17 was unsuccessful, 17 was
efficiently transformed to aldehyde 19 via non-aqueous OsO4 oxi-
dation in the presence of PhB(OH)2 and the successive NaIO4 oxida-
tion following Narasaka’s procedure.9–11 When 17 was subjected to
general aqueous OsO4 reaction conditions, lactone 20 was obtained

M. Kitamura et al. / Tetrahedron Letters 55 (2014) 1653–1656
1655
naphthyl acetate, which was hydrolyzed to give naphthol 8b. Dia-
zo-transfer reaction for 8b by 2-azido-1,3-dimethylimidazolinium
chloride (ADMC, 29) formed by the reaction of 2-chloro-1,3-dime-
thylimidazolinium chloride (DMC) and sodium azide smoothly
proceeded to afford corresponding diazonaphthoquinone 30.16
Rhodium-catalyzed cyclization of 30 with ketene silyl acetal fol-
lowed by a ring opening reaction successfully proceeded to afford
alkoxycarbonylmethylated naphthalene 7b in 36% yield in 3 steps
from 1-naphthol 8b.12 Introduction of a methallyl group to the
active methylene part in 7b proceeded to afford diene 6b, whose
intramolecular olefin metathesis proceeded to afford the DEF ring
fragment 5. By the new synthetic route for 5, the number of reac-
tion steps was reduced by 6, from 15 to 9 steps, and the total yield
was increased to 8.6% yield from 3.4% yield compared to the former
synthetic route (Scheme 3).
Next we tried the transformation of 5 to lactone 3 (Scheme 6).
Several reports have appeared about the oxidative cyclization of
cyclohex-3-encarbonic acid to the corresponding hydroxy lactone
via epoxide.4 However, epoxidation of the alkene in the D ring of
ester 5 or the corresponding carboxylic acid 32 was unsuccessful
because of over oxidation of aromatic rings. Selective oxidation
of the electron-rich B ring in 5 was performed by ceric ammonium
nitrate (CAN) to give quinone 33, whose D ring’s epoxidation was
accomplished by the treatment of m-CPBA to afford 34 as a ca
1/1 diastereomer mixture. The products were rather unstable,
and the crude material was treated with trifluoroacetic acid to
obtain lactone 3.17 The structure/stereochemistry of 3 was identi-
fied by NMR experiments (1H NMR, NOE), as shown in Figure 2.
That is, the 1H NMR signal of Ha of the hydroxyl group was
observed as a doublet (J = 7.5 Hz) at 6.50 ppm in DMSO-d6, which
was eliminated by the addition of D2O, and NOE was detected
between Ha and Hb (2.14 ppm). These data suggested that the
Scheme 4. Synthesis of phosphonate 27. Reagents and conditions: (a) SOCl2
(1.05 equiv); (b) P(OEt)3 (1 equiv), 150 °C, 12 h; (c) m-CPBA (1.7 equiv), NaHCO3
(0.2 equiv), CH2Cl2, rt, 72 h (97%); (d) NaOMe, MeOH, 0 °C, 6.5 h (76%); (e)
CH3C(OEt)3 (5 equiv), CH3CH2CO2H, 150 °C, 8 h (65%).
was recently developed in our laboratory.12 For the latter problem,
we planned to study the Horner–Wittig reaction of aldehyde 16
with phosphonate 27 that is ready arranged alkenyl moiety for
olefin metathesis. Then, we tried the synthesis of BCD rings 5 by
the new route.
Phosphonate 27 was synthesized from alcohol 24 by reference
to the synthesis of similar phosphonates (Scheme 4).13
Transformation of alcohol 24 to the corresponding chloride
followed by the Arbuzov reaction gave allyl phosphonate 25.14
m-Chloroperoxybenzoic acid (m-CPBA) oxidation of 25 gave the
corresponding epoxide, which was treated with sodium methox-
ide, giving allylic alcohol 26. The Johnson–Claisen rearrangement
of 26 with CH3C(OEt)3 proceeded to afford phosphonate 27 having
a styryl group.15
The Wittig–Horner reaction of aldehyde 16 with phosphonate
27 gave the corresponding unsaturated ester, whose ester was
hydrolyzed to give unsaturated acid 28 (Scheme 5). Treatment of
28 with acetic anhydride in the presence of sodium acetate gave
product from 34 was not d-lactone but c-lactone, and oxygen func-
tional groups attached at C8 and C9 in the D ring are placed in the
anti position, as shown for compound 3.
Scheme 6. Synthesis of BCDE ring fragment 3. Reagents and conditions: (a)
CF3CO2H, CH2Cl2, rt, 1.5 h (quant.); (b) Ce(NH4)2(NO3)6, CH3CN, H2O, 0 °C, 10 min;
(c) m-CPBA, NaHCO3, CH2Cl2, rt, 2.5 h; (d) CF3CO2H, CH2Cl2, 0 °C, 40 min (3 steps
31%).
Scheme 5. Synthesis of BCD ring fragment
5 (2nd generation). Reagents and
conditions: (a) 27 (1,2 equiv), NaH (3 equiv), THF, À78 °C to rt, 3.5 h (75%); (b) 10 M
KOH aq n-Bu4NBr (0.1 equiv), 1,4-dioxane, 90 °C, 24 h; (c) NaOAc (2 equiv), Ac2O,
140 °C, 7 h; (d) K2CO3, MeOH, 1,4-dioxane, rt, 4.5 h (3 steps 42%); (e) DMC (5 equiv),
NaN3 (5 equiv), CH3CN, À25 °C, 1.5 h: then 8b, Et3N (2.5 equiv), THF, À25 °C to rt,
3 h (89%); (f) CH2@C(OTBS)(Ot-Bu) (2 equiv), 1 mol % Rh2(OAc)4, 40 °C, 1 h (45%);
(g) n-Bu4NF (2 equiv), MeI (10 equiv), Na2S2O4 (0.1 equiv), THF, rt, 1.5 h (91%); (h)
CH2@C(CH3)CH2Br, NaH, 45 °C, 3 h (91%); (i) 5 mol % Grubbs 2nd cat., toluene, 80 °C,
12 h (82%).
Figure 2. NOE experiment of lactone 3.

1656
M. Kitamura et al. / Tetrahedron Letters 55 (2014) 1653–1656
Higuchi, T.; Hirobe, M. J. Chem. Soc., Perkin Trans. 1 1996, 2309; (f) Tran, C. H.;
In conclusion, toward the synthesis of kosinostatin aglycone 1,
Grout, D. H. G. J. Chem. Soc., Perkin Trans. 1 1998, 1065; (g) Baker, R. K.;
Rupprecht, K. M.; Armistead, D. M.; Boger, J.; Frankshun, R. A.; Hodges, P. J.;
Hoogsteen, K.; Pisano, J. M.; Witzel, B. E. Tetrahedron Lett. 1998, 39, 229; (h)
Wipf, P.; Wang, X. Tetrahedron Lett. 2000, 41, 8747; (i) Kiesman, W. F.; Zhao, J.;
Conlon, P. R.; Petter, R. C.; Jin, X.; Smits, G.; Lutterodt, F.; Sullivan, G. W.; Linden,
J. Bioorg. Med. Chem. 2006, 14, 3654; (j) Hamamoto, H.; Suzuki, Y.; Takahashi,
H.; Ikegami, S. Adv. Synth. Catal. 2007, 349, 2685.
we synthesized the key intermediate lactone 3, which corresponds
to the BCDE ring fragment. The precursor BCD ring fragment 5 was
synthesized via two routes, and the improved 2nd generation
synthesis includes efficient application of our developed
alkoxycarbonylmethylation of diazonaphthoquinone. Currently,
we are studying the construction of A and FG rings, aiming at the
total synthesis of aglycone 1.
5. Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999, 1, 953.
6. (a) Cordes, J.; Harms, K.; Koert, U. Org. Lett. 2010, 12, 3808; (b) Breuning, M. A.;
Harms, K.; Koert, U. Org. Lett. 2011, 13, 1402.
7. Bloomer, J. L.; Stagliano, K. W.; Gazzillo, J. A. J. Org. Chem. 1993, 58, 7906.
8. Similar alkoxycarbonylmethylation to phenol has been reported, see, Yoshida,
M.; Higuchi, M.; Shishido, K. Org. Lett. 2009, 11, 4752.
Acknowledgments
9. Yamada, T.; Narasaka, K. Chem. Lett. 1986, 1, 131.
We thank Prof. Tatsuo Okauchi for his helpful discussion. This
work was supported by a Grant-in-Aid from the Ministry of Educa-
tion, Culture, Sports, Science, and Technology of Japan, the Naito
Foundation, JGC-S Scholarship Foundation, Nagase Science and
Technology Foundation, and The Novartis Foundation (Japan) for
the Promotion of Science.
10. Although oxidative transformation of alkene 17 to aldehyde 19 proceeded by
using microencapsulated OsO4 (MC OsO4),11 the yield was lower (62% yield in 2
steps) than that of the reaction with OsO4 [(i) cat. MC OsO4, NMO, PhB(OH)2,
Na2SO4, acetone, rt, 13 h; (ii) NaIO4, THF, H2O, 50 °C, 1 h].
11. Nagayama, S.; Endo, M.; Kobayashi, S. J. Org. Chem. 1998, 63, 6094.
12. Kitamura, M.; Araki, K.; Matsuzaki, H.; Okauchi, T. Eur. J. Org. Chem. 2013, 5045.
13. Koszuk, J. F. Synth. Commun. 1995, 25, 2533.
14. Yamagishi, T.; Fujii, K.; Shibuya, S.; Yokomatsu, T. Tetrahedron 2006, 62, 54.
15. Spectral data of 27: 1H NMR (400 MHz, CDCl3) 7.36 (d, 2H, J = 7.4 Hz), 7.29 (dd,
2H, J = 7.4, 7.3 Hz), 7.22 (t, 1H, J = 7.3 Hz), 6.58 (dd, 1H, J = 15.8, 4.9 Hz), 6.11
(ddd, 1H, J = 15.9, 9.0, 6.5 Hz), 4.11–4.01 (m, 6H), 3.30–3.18 (m, 1H), 2.85 (ddd,
1H, J = 4.1, 10.7, 15.8 Hz), 2.61 (ddd, 1H, J = 9.6, 10.3, 15.8 Hz), 1.27 (t, 3H,
J = 7.1 Hz), 1.24 (t, 3H, J = 7.1 Hz), 1.16 (t, 3H, J = 7.1 Hz); 13C NMR (101 MHz,
CDCl3) 170.7 (d, J = 19.2 Hz), 136.4 (d, J = 3.1 Hz), 134.1 (d, J = 13.5 Hz), 128.3,
127.5, 126.2, 122.8 (d, J = 10.2 Hz), 62.4 (d, J = 6.9 Hz), 62.1 (d, J = 7.0 Hz), 60.6,
38.2 (d, J = 142.1 Hz), 34.0 (d, J = 3.0 Hz), 16.2 (d, J = 5.2 Hz), 14.0 ppm; IR (KBr)
3464, 2983, 1733, 1647, 1600, 1577, 1496, 1477, 1448, 1392, 1371, 1228, 1161,
1097, 981 cmÀ1; HRMS (FAB+) m/z [M+H]+ calcd for C17H26O5P 341.3518.
Found 341.1518.
16. Kitamura, M.; Tashiro, N.; Sakata, R.; Okauchi, T. Synlett 2010, 2503.
17. Spectral data of 3: 1H NMR (500 MHz, DMSO-d6) 7.95 (s, 1H), 7.02 (d, 1H,
J = 10.2 Hz), 6.97 (d, 1H, J = 10.2 Hz), 6.50 (d, 1H, J = 7.5 Hz), 4.46 (d, 1H,
J = 7.5 Hz), 4.17 (d, 1H, J = 4.7 Hz), 3.85 (s, 3H), 2.41 (dd, 1H, J = 12.0, 4.9 Hz),
2.14 (d, 1H, J = 12.0 Hz), 1.58 (s, 3H); 13C NMR (126 MHz, CDCl3) 184.2, 183.6,
174.8, 156.1, 143.7, 140.6, 137.4, 136.8, 133.4, 124.8, 123.3, 85.6, 70.9, 63.4,
39.4, 34.9, 21.8 ppm; IR (KBr) 3448, 2923, 2851, 1779, 1664, 1570, 1459, 1338,
1296, 1228, 1106, 1038, 930 cmÀ1; HRMS (EI+) m/z [M]+ calcd for C17H14O6
314.0790. Found 314.0801.
References and notes
1. (a) Martin, J. H.; Shay, A. J.; Pruess, L. M.; Porter, J. N.; Mowat, J. H.; Bohonos, N.
Antibiot. Annu. 1954–1955, 1020; (b) Celmer, W. D.; Murai, K.; Rao, K. V.;
Tanner, F. W., Jr.; Marsh, W. S. Antibiot. Annu. 1957–1958, 484; (c) McBride, T. J.;
English, A. R. Antibiot. Annu. 1957–1958, 493; (d) Cosulich, D. B.; Mowat, J. H.;
Broschard, R. W.; Patrick, J. B.; Meyer, W. E. Tetrahedron Lett. 1963, 453.
Erratum Tetrahedron Lett. 1964, 750; (e) Tulinsky, A. J. Am. Chem. Soc. 1964, 86,
5368.
2. (a) Furumai, T.; Igarashi, Y.; Higuchi, H.; Saito, N.; Oki, T. J. Antibiot. 2002, 55,
128; (b) Igarashi, Y.; Higuchi, H.; Oki, T.; Furumai, T. J. Antibiot. 2002, 55, 134;
(c) El-Naggar, M. Y. J. Microbiol. 2007, 45, 262.
3. Rozeboom, M. D.; Tagmo-Larsson, I.-M.; Houk, K. N. J. Org. Chem. 1981, 46,
2338.
4. (a) Bartlett, P. A.; McQuaid, L. A. J. Am. Chem. Soc. 1984, 106, 7854; (b) Shah, M.;
Taschner, M. J.; Koser, G. F.; Rach, N. L.; Jenkins, T. E.; Cyr, P.; Powers, D.
Tetrahedron Lett. 1986, 27, 5437; (c) Jones, T. K.; Reamer, R. A.; Desmond, R.;
Mills, S. G. J. Am. Chem. Soc. 1990, 112, 2998; (d) Ireland, R. E.; Highsmith, T. K.;
Gegnas, L. D.; Gleason, J. L. J. Org. Chem. 1992, 57, 5071; (e) Komuro, M.;