A formal synthesis of martinelline via a combination of two types of radical reactions

Article abstract of DOI:10.1016/j.tetlet.2004.02.153

A formal synthesis of martinelline has been accomplished via two types of radical reactions as the key steps. These are the radical addition-cyclization- elimination of an oxime ether carrying an unsaturated ester and a C-C bond formation through a radica

Full text of DOI:10.1016/j.tetlet.2004.02.153

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 3481–3484
A formal synthesis of martinelline via a combination of two types of
radical reactions
Yoshifumi Takeda, Toshiki Nakabayashi, Atsushi Shirai, Daisuke Fukumoto,
Toshiko Kiguchi and Takeaki Naito*
Kobe Pharmaceutical University, Motoyamakita, Higashinada, Kobe 658-8558, Japan
Received 2 February 2004; revised 27 February 2004; accepted 27 February 2004
Abstract—A formal synthesis of martinelline has been accomplished via two types of radical reactions as the key steps. These are the
radical addition–cyclization–elimination of an oxime ether carrying an unsaturated ester and a C–C bond formation through a
radical 1,5-hydrogen atom translocation.
Ó 2004 Elsevier Ltd. All rights reserved.
Martinelline and martinellic acid were isolated from an
organic extract of the Martinella iquitosensis root in
1995.1 These compounds show antibiotic activity against
Gram-positive and Gram-negative bacteria, affinity for
several G-protein receptors, and are the first nonpeptide
bradykinin receptor antagonist reported to date. The
pyrrolo[3,2-c]quinoline ring system of the martinellines
core has not been reported previously in any natural
product. Their biological activity and unique structure
have made them the subject of intense synthetic interest
(Fig. 1).2–11
HN
1
2
HN
O
N
9
8
3
RO
H
N
H
N
4
5
7
N
6
H
NH
H
N
H2N
Martinelline: R =
NH
As part of our program on radical addition–cyclization
of oxime ethers connected with a,b-unsaturated car-
bonyl group,12 we focused our efforts upon a synthesis
of martinelline and designed our strategy leading to a
key intermediate 18 for the synthesis of martinelline.9b
Our synthetic strategy includes two crucial radical
reactions; (1) a newly-found radical addition–cycliza-
tion–elimination of an oxime ether carrying an unsatu-
rated ester for the construction of the pyrroloquinoline;
(2) C–C bond formation through a 1,5-hydrogen atom
translocation13 for the stereoselective introduction of the
side chain into the C(4) position of pyrroloquinoline
(Scheme 1).
Martinellic acid: R = H
Figure 1.
We first investigated the radical addition–cyclization of
an oxime ether carrying an unsaturated ester 1 and found
an interesting radical addition–cyclization–elimination,
which has provided a novel and efficient method for the
construction of the pyrroloquinoline. Requisite oxime
ether 1 was readily prepared from commercially avail-
able methyl anthranilate 5. N-Tosylation of 5 gave 6 in
68% yield, which was reduced with LiAlH4 and then
oxidized with MnO2 to afford aldehyde 8. Condensation
of aldehyde 8 with O-benzylhydroxylamine hydrochlo-
ride in the presence of AcONa gave oxime ether 9 in 80%
yield (three steps from 6), which was then N-alkylated
with ethyl 4-bromocrotonate to afford 1 in 95% yield.
Keywords: Pyrrolo[3,2-c]quinoline; Oxime ether; Radical reaction;
Martinelline.
According to our procedure developed in the radical
addition–cyclization of oxime ethers,12 treatment of
* Corresponding author. Tel.: +81-78-441-7554; fax: +81-78-441-7556;
e-mail: taknaito@kobepharma-u.ac.jp
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.02.153

3482
Y. Takeda et al. / Tetrahedron Letters 45 (2004) 3481–3484
R2
Ac
X
C-C bond
radical
formation via
1,5-hydrogen
atom translocation
addition-
cyclization-
elimination
BnON
N
CO2Et
N
N
R1
Ac
N
N
CO2Me
Ts
R3
2a : X = O, R1 = R2 = H, R3 = Ts
3 : X = H2, R1 = R2 = Ac, R3 = 2-bromobenzoyl
Bz
1
4
Scheme 1.
oxime ether 1 with Bu3SnH and AIBN in refluxing
benzene gave two types of products 2a and 10. Sur-
prisingly, a major product of the reaction was an
unexpected tricyclic pyrroloquinoline 2a, which does not
have the benzyloxy group (52%, cis/trans ¼ 1:1) in the
molecule. The minor product was an expected bicyclic
tetrahydroquinoline 10 (22%, cis/trans ¼ 1:1.5). The
structures of cis-2a and trans-2a were deduced from
their spectral data.14 The debenzyloxylated major
product 2a would be formed as a result of stannyl rad-
ical addition–cyclization–elimination accompanied with
debenzyloxylation. In order to propose a reaction
pathway from 1 to 2a, we examined the interconversion
of the products. When cis-10 was further treated with
Bu3SnH and AIBN, most of the starting material was
recovered. Heating of cis-10 in MeOH in the presence of
TsOH gave cyclized N-benzyloxy cis-2b, which however
was not converted into cis-2a under the same conditions
for the radical reaction. When heated in MeOH in the
presence of TsOH or under radical reaction conditions,
trans-10 was completely recovered. The reaction path-
way of the radical addition–cyclization–elimination is
ambiguous at the moment (Scheme 2).
AcCl in the presence of Et3N to give amide 11 in 90%
yield (two steps from cis-2a). According to the known
procedure,15 we investigated the Friedel–Crafts reaction
of 11. When 11 was treated with AcCl (3 equiv) and
AlCl3 (6 equiv) in refluxing 1,2-dichloroethane, we
obtained fortunately the 1,5,8-triacetyl compound 12 in
which the tosyl group was converted into the easily
removable acetyl group.16 Treatment of 12 with 10% aq
NaOH in MeOH at room temperature caused selective
deacetylation at the vinylogous imide system to give the
1,8-diacetyl compound 13 in 47% yield (two steps from
11).
Snieckus and co-workers13e have reported a diastereo-
selective carbon–carbon bond formation of an a-carbon
adjacent to a nitrogen via a 1,5-hydrogen atom trans-
location and subsequent Michael-type radical addition
to methyl acrylate. This method would allow the intro-
duction of the C3 unit side chain into the C(4) position
adjacent to nitrogen from the less hindered convex face
to afford the desired product. Amine 13 was acylated
with 2-bromobenzoyl chloride in the presence of Et3N to
give the radical precursor 3 in 91% yield. When a solu-
tion of Bu3SnH and AIBN in benzene was slowly added
to a solution of 3 and methyl acrylate in refluxing benz-
ene using a syringe pump, the desired compound 4 was
formed as a single diastereomer in 43% yield along with
the pentacyclic product 14 (19%). Formation of 14
would involve radical cyclization of the transiently
formed aryl radical A onto the aromatic ring followed
by oxidation. The conversion of the C(8)-acetyl group
into the methoxycarbonyl group in 4 was achieved by
the conventional procedure using haloform reaction.
Treatment of 4 with Br2 in 2.5% aq NaOH followed by
esterification of the resulting carboxylic acid gave ester
15 in 76% yield (Scheme 3).
We next examined the introduction of a carbonyl group
into the C(8) position and a C3 unit into the C(4)
position of cis-2a possessing requisite stereostructure for
the synthesis of martinelline. Reduction of cis-2a with
BH3ÆMe2S followed by workup with 6 M HCl to cleave
the B–N bond of the resulting intermediate gave the
corresponding amine, which was then acetylated with
BnON
CO2Me
R1
R2
b
d
e
NHTs
NHTs
5: R1 = NH2
7: R2 = CH2OH
8: R2 = CHO
In order to convert 15 into the known intermediate 18
for martinelline synthesis, we investigated the conver-
sion of functional groups. Treatment of 15 with
Et3OÆBF4 followed by hydrolysis of the resulting imidate
with saturated aq NaHCO3 afforded amine 16 in 40%
yield along with starting material 15 (32%). Amine 16
was reacted with CbzCl in the presence of Et3N to give
the protected compound 17 in 76% yield. Finally, the
regioselective reduction of the isolated ester moiety in 17
with LiBH4 in a mixture of 1/10MeOH–THF at room
temperature gave the desired alcohol 18, which is a key
intermediate for the synthesis of martinelline. The
spectra of 18 were superimposable with those provided
by Professor Ma.9b
9
a
c
6: R1 = NHTs
O
R3
BnOHN
N
CO2Et
BnON
CO2Et
N
f
+
N
Ts
N
Ts
Ts
1
2a: R3 = H
10
2b: R3 = OBn
Scheme 2. Reagents and conditions: (a) TsCl, Et3N, CH2Cl2, rt; (b)
LiAlH4, THF, 0 °C; (c) MnO2, CH2Cl2, rt; (d) BnONH2ÆHCl, AcONa,
MeOH–CH2Cl2, rt; (e) ethyl 4-bromocrotonate, K2CO3, acetone, rt; (f)
Bu3SnH, AlBN, benzene, reflux.

Y. Takeda et al. / Tetrahedron Letters 45 (2004) 3481–3484
3483
Ac
Ac
O
O
Ac
N
N
HN
N
Ac
Ac
R1
a, b
f
e
N
Br
N
N
Ts
N
R2
O
11: R1 = H, R2 = Ts
12: R1 = Ac, R2 = Ac
13: R1 = Ac, R2 = H
A
3
cis-2a
c
d
Ac
N
Ac
Ac
N
N
Ac
Ac
R3
+
N
N
Bz
N
Bz
CO2Me
O
B
4: R3 = Ac
14
g, h
15: R3 = CO2Me
i
R4
N
Cbz
N
MeO2C
MeO2C
k
OH
N
Bz
CO2Me
N
H
16: R4 = H
17: R4 = Cbz
18
j
Scheme 3. Reagents and conditions: (a) BH3ÆMe2S, THF, reflux; (b) AcCl, Et3N, CH2Cl2, 0 °C; (c) AcCl, AlCl3, ClCH2CH2Cl, reflux; (d) 10%
NaOH, MeOH, rt; (e) 2-bromobenzoyl chloride, Et3N, CH2Cl2, rt; (f) methyl acrylate, Bu3SnH, AlBN, benzene, reflux; (g) Br2, 2.5% NaOH, 0 °C;
(h) concd H2SO4, MeOH, reflux; (i) Et3OÆBF4, NaHCO3, rt, then satd NaHCO3, rt; (j) CbzCl, Et3N, CH2Cl2, 0 °C; (k) LiBH4, MeOH–THF, rt.
Anderson, P. S.; Pitzenberger, S. M.; Varga, S. L. J.
Am. Chem. Soc. 1995, 117, 6682–6685.
2. Ho, T. C. T.; Jones, K. Tetrahedron 1997, 53, 8287–8294.
3. Gurjar, M. K.; Pal, S.; Rao, A. V. R. Heterocycles 1997,
45, 231–234.
In conclusion, we have developed a novel and efficient
strategy for the synthesis of pyrroloquinoline and suc-
ceeded in a formal synthesis of martinelline via the route
involving two radical reactions of which the newly-
found radical addition–cyclization–elimination con-
structed the martinelline skeleton in one procedure.
4. (a) Snider, B. B.; Ahn, Y.; Foxman, B. M. Tetrahedron
Lett. 1999, 40, 3339–3342; (b) Snider, B. B.; O’Hare, S. M.
Tetrahedron Lett. 2001, 42, 2455–2458; (c) Snider, B. B.;
Ahn, Y.; O’Hare, S. M. Org. Lett. 2001, 3, 4217–4220.
5. (a) Batey, R. A.; Simoncic, P. D.; Lin, D.; Smyj, R. P.;
Lough, A. J. Chem. Commun. 1999, 651–652; (b) Batey, R.
A.; Powell, D. A. Chem. Commun. 2001, 2362–2363; (c)
Powell, D. A.; Batey, R. A. Org. Lett. 2002, 4, 2913–2916.
6. (a) Hadden, M.; Stevenson, P. J. Tetrahedron Lett. 1999,
40, 1215–1218; (b) Hadden, M.; Nieuwenhuyzen, M.;
Osborne, D.; Stevenson, P. J.; Thompson, N. Tetrahedron
Lett. 2001, 42, 6417–6419; (c) Hadden, M.; Nieuwenhuy-
zen, M.; Potts, D.; Stevenson, P. J.; Thompson, N.
Tetrahedron 2001, 57, 5615–5624.
Acknowledgements
We are grateful to Prof. D. Ma, Shanghai Institute of
Organic Chemistry, Chinese Academy of Sciences, for
kindly providing the spectra of the authentic sample of
18, and the Ministry of Education, Culture, Sports,
Science, and Technology of Japan for Grant-in-Aid for
Young Scientists (B) (Y.T.) and the Science Research
Promotion Fund of the Japan Private School Promotion
Foundation for research grants. We are also grateful to
Prof. I. Ryu, Osaka Prefecture University, for his
helpful discussions on reaction pathway of radical
addition–cyclization–elimination of oxime ethers.
7. (a) Lovely, C. J.; Mahmud, H. Tetrahedron Lett. 1999, 40,
2079–2082; (b) Mahmud, H.; Lovely, C. J.; Dias, H. V. R.
Tetrahedron 2001, 57, 4095–4105.
ꢀ
8. Frank, K. E.; Aube, J. J. Org. Chem. 2000, 65, 655–666.
9. (a) Ma, D.; Xia, C.; Jiang, J.; Zhang, J. Org. Lett. 2001, 3,
2189–2191; (b) Xia, C.; Heng, L.; Ma, D. Tetrahedron
Lett. 2002, 43, 9405–9409; (c) Ma, D.; Xia, C.; Jiang, J.;
Zhang, J.; Tang, W. J. Org. Chem. 2003, 68, 442–451.
10. Nieman, J. A.; Ennis, M. D. Org. Lett. 2002, 2, 1395–
1397.
References and notes
1. Witherup, K. M.; Ransom, R. W.; Graham, A. C.;
Bernard, A. M.; Salvatore, M. J.; Lumma, W. C.;

3484
Y. Takeda et al. / Tetrahedron Letters 45 (2004) 3481–3484
11. Makino, K.; Hara, O.; Takiguchi, Y.; Katano, T.;
Asakawa, Y.; Hatano, K.; Hamada, Y. Tetrahedron Lett.
2003, 44, 8925–8929.
12. Naito, T.; Fukumoto, D.; Takabayashi, K.; Kiguchi, T.
Heterocycles 1999, 51, 489–492.
m), 7.31 (1H, td, J ¼ 7:5, 1 Hz), 7.52 (2H, br d, J ¼ 8 Hz),
7.71 (1H, dd, J ¼ 8, 1 Hz). HRMS m=z: calcd
C18H18N2O3S (Mþ) 342.1037. Found: 342.1042. trans-2a:
1
IR mmax cmÀ1 (CHCl3): 3429, 1727, 1356, 1168. H NMR
(CDCl3, 500 MHz) d: 2.10(1H, m), 2.20(1H, dd, J ¼ 15:5,
12.5 Hz), 2.39 (3H, s), 2.55 (1H, dd, J ¼ 15:5, 6.5 Hz), 3.23
(1H, d, J ¼ 10:5 Hz), 3.58 (1H, br t, J ¼ 11 Hz), 3.99
(1H, dd, J ¼ 11, 6 Hz), 6.99 (1H, br d, J ¼ 7:5 Hz), 7.17–
7.21 (3H, m), 7.30–7.35 (1H, br t, J ¼ 8 Hz), 7.40–
7.46 (3H, m), 7.82 (1H, dd, J ¼ 8, 0.5 Hz). HRMS m=z:
calcd C18H18N2O3S (Mþ) 342.1037. Found: 342.1047.
The stereostructures of cis-2a and trans-2a were
deduced on the basis of their J values of 9b-H (6.5 Hz
for cis-2a and 10.5 Hz for trans-2a) and confirmed by
chemical conversion of cis-2a into the known intermediate
18.9b
13. For a review of radical translocation reactions, see:
Robertson, J.; Pillai, J.; Lush, R. K. Chem. Soc. Rev.
2001, 30, 94–103; For examples of substitution at a-carbon
adjacent to nitrogen via 1,5-hydrogen atom translocation,
see: (a) Snieckus, V.; Cuevas, J.-C.; Sloan, C. P.; Liu, H.;
Curran, D. P. J. Am. Chem. Soc. 1990, 112, 896–898; (b)
Murakami, M.; Hayashi, M.; Ito, Y. J. Org. Chem. 1992,
57, 793–794; (c) Undheim, K.; Williams, L. J. Chem. Soc.,
Chem. Commun. 1994, 883–884; (d) Williams, L.; Booth,
S. E.; Undheim, K. Tetrahedron 1994, 50, 13697–13708; (e)
Beulieu, F.; Arora, J.; Veith, U.; Taylor, N. J.; Chapell, B.
J.; Snieckus, V. J. Am. Chem. Soc. 1996, 118, 8727–8728.
15. Ishihara, Y.; Tanaka, T.; Goto, G. J. Chem. Soc., Perkin
Trans. 1 1992, 3401–3406.
1
14. cis-2a: IR mmax cmÀ1 (CHCl3): 3433, 1701, 1354, 1166. H
NMR (CDCl3, 500 MHz) d: 2.02 (1H, dd, J ¼ 17:5, 2 Hz),
2.40(3H, s), 2.54 (1H, m), 2.62 (1H, dd, J ¼ 17:5, 9 Hz),
3.15 (1H, dd, J ¼ 14, 12 Hz), 4.20(1H, dd, J ¼ 14, 5 Hz),
4.37 (1H, d, J ¼ 6:5 Hz), 6.60(1H, br s), 7.16–7.28 (4H,
16. A similar conversion of the N-Ts group into the N-acyl
group under Friedel–Crafts reaction conditions has been
reported: Jiang, Y.; Ma, D. Tetrahedron Lett. 2002, 43,
7013–7015.