Total synthesis of (±)-streptonigrin: De novo construction of a pentasubstituted pyridine using ring-closing metathesis

Article abstract of DOI:10.1021/ja207835w

The synthesis of the potent antitumor agent (±)-streptonigrin has been achieved in 14 linear steps and 11% overall yield from ethyl glyoxalate. The synthesis features a challenging ring-closing metathesis reaction, followed by elimination and aromatization, to furnish a key pentasubstituted pyridine fragment.

Full text of DOI:10.1021/ja207835w

COMMUNICATION
pubs.acs.org/JACS
Total Synthesis of (()-Streptonigrin: De Novo Construction of a
Pentasubstituted Pyridine using Ring-Closing Metathesis
Timothy J. Donohoe,*,† Christopher R. Jones,† and Luiz C. A. Barbosa§
†Department of Chemistry, University of Oxford, Chemistry Research Laboratory, Mansfield Road, Oxford OX1 3TA, U.K.
§Department of Chemistry, Universidade Federal de Vicosa, Vicosa 36570-000, MG, Brazil
S Supporting Information
b
Scheme 1
ABSTRACT: The synthesis of the potent antitumor agent
(()-streptonigrin has been achieved in 14 linear steps and
11% overall yield from ethyl glyoxalate. The synthesis
features a challenging ring-closing metathesis reaction,
followed by elimination and aromatization, to furnish a
key pentasubstituted pyridine fragment.
treptonigrin is a highly functionalized aminoquinone anti-
tumor antibiotic that has attracted considerable attention
S
from both synthetic organic chemists and biochemists interested
in its challenging molecular framework and broad-spectrum
anticancer activity.1À3 Found to be extremely effective in the
treatment of cancer, streptonigrin reached phase II clinical trials;
however, high levels of toxicity caused side effects which neces-
sitated the cessation of these trials in the late 1970s.4
required. To this end, we devised a retrosynthetic strategy that
would see the two biaryl units constructed by metal-catalyzed
cross-coupling reactions,13 leaving the dual activated pyridine
C-ring 2 as the key fragment (Scheme 1). It was envisaged that
pentasubstituted pyridine 2 could be readily accessed through
methodology we had recently developed, namely the generation
of heteroaromatic compounds utilizing a ring-closing metathesis
(RCM) reaction as the cyclization step (see 5f4).14 Subsequent
elimination and aromatization of 4 would furnish pyridine 2. The
quinone 1 would be obtained via late-stage oxidation of a func-
tionalized quinoline component, while the D-ring precursor 3
would come from commercially available 2,3-dimethoxyphenol.
Synthesis of pyridine 2 began by condensing methoxyamine
with commercially available ethyl glyoxalate 6 (Scheme 2). Next,
a zinc-mediated crotylation of the resulting oxime furnished 7 in
96% yield over the two steps. Acylation of 7 with N-phthalimide-
protected α-amino acrylic acid 8 (obtained in one step from
DL-serine)15 was effected by in situ formation of the correspond-
ing acid chloride followed by addition of 7 to afford the RCM
precursor 9 in 90% yield.
The crucial RCM reaction represented a challenge, as it
necessitated tolerating a 1,1-disubstituted alkene with a pro-
tected amine as one of the substituents. Through extensive
optimization, we obtained the RCM product 10 in 76% yield
using 5 mol % of HoveydaÀGrubbs second-generation catalyst.
Key to this reaction was the slow addition of catalyst via a syringe
pump, combined with the use of benzoquinone16 to quench
RuÀH species formed in situ.17
Following isolation from Streptomyces flocculus in 1959 by Rao
and Cullen,1 the structure of streptonigrin was established by
Rao, Biemann, and Woodward in 1963 via spectroscopy and
chemical degradation.5 In 1975 Chiu and Lipscomb confirmed
the structure using X-ray crystallography.6 Subsequent investi-
gations7 involving truncated analogues established the cytotoxic
mode of action, believed to be free radical-mediated DNA strand
cleavage resulting from reductive activation by a metal ion
and oxygen.8 Elsewhere, circular dichroism studies revealed
the absolute stereochemistry about the configurationally stable
CÀD ring axis to be (M).9
To date there has been only one total synthesis of (()-
streptonigrin in pioneering work by Weinreb in 1980.10 Two
impressive, formal, syntheses from Kende11 in 1981 and Boger12
in 1985 stopped five and seven steps short of the target,
respectively. Weinreb’s landmark synthesis of (()-streptonigrin
was 34 steps long with considerably less that 1% overall yield.
The two subsequent syntheses represented a significant improve-
ment in efficiency especially given the state of organic synthesis
at the time the work was conducted. Kende’s synthesis required
27 steps and Boger’s just 17 if both had been taken through
Weinreb’s end game; however, when completed, both routes
would afford streptonigrin in less than 1% overall yield.
Despite the termination of clinical trials, interest remains in
generating analogues of streptonigrin so as to harness the
potency of the natural product while tempering its cytotoxicity
and thereby making it amenable as a potential drug.2 In order
to enable facile preparation of a range of analogues at key sites
on streptonigrin, a more efficient and convergent synthesis is
Received: August 18, 2011
Published: September 23, 2011
r
2011 American Chemical Society
16418
dx.doi.org/10.1021/ja207835w J. Am. Chem. Soc. 2011, 133, 16418–16421
|

Journal of the American Chemical Society
COMMUNICATION
Scheme 2
Scheme 3
The next step was an elimination of the methanol leaving
group, using quinuclidine as the optimum base. This reaction was
performed in methanol so that a transesterification occurred,
generating the corresponding methyl ester and enabling our
route to converge with a late-stage intermediate from Weinreb’s
synthesis.10 Moreover, methylamine was added to the reaction
mixture to effect cleavage of the N-protecting group, resulting
in an efficient one-pot, three-transformation procedure to afford
aminopyridone 11 in 85% yield from dihydropyridone 10.
With pyridine 2 in mind, we desired a bromine at the
4-position of the pyridine and a more reactive trifluoromethane-
sulfonyl group at the 2-position, which should preferentially
react in the first of two cross-coupling reactions. As a result,
pyridone 11 was converted to pyridine 12 in 84% yield using
triflic anhydride, 2,6-di-tert-butyl-4-methyl pyridine and hexa-
fluoroisopropanol (HFIPA). Subsequent bromination with
N-bromosuccinimide furnished the penta-substituted pyridine
13 in 92% yield. The structure of 13 was confirmed by X-ray
crystallographic analysis.
yield with nitric acid (Scheme 3).21 Reduction using catalytic
Pd/C and hydrogen afforded aniline 19, which was subsequently
acylated with (E)-3-ethoxyacryloyl chloride 2022 to generate aryl
amide 21 in 72% yield over the two steps and as an incon-
sequential mixture of stereoisomers. Cyclization in sulfuric acid
established the AB-ring scaffold, generating 22 in 85% yield,
while treatment with phosphorus oxychloride effected aromati-
zation to chloroquinoline 23 in 88% yield.
From 23, we attempted to construct the corresponding
boron-, tin-, zinc-, and silicon-containing compounds. However,
standard lithiumÀhalogen exchange conditions followed by
trapping failed to furnish any metal substituted quinolines.
Following work by Padwa,20d we discovered that treatment
of 23 with hexamethylditin and tetrakis(triphenylphosphine)-
palladium gave the corresponding stannane 24, albeit contami-
nated with triphenylphosphine oxide (75% conversion by
1HNMR spectroscopy; stannane 24 was unstable to chromatog-
raphy and used without purification).
Attention then turned to the two remaining coupling partners
1 and 3. Work by Quꢀeguiner and Holzapfel has shown that the
CÀD ring axis in model systems could be constructed via a
Suzuki cross-coupling reaction, and we followed a similar strategy
to attach the D-ring fragment.18 Consequently, borate ester 16
was synthesized in three steps from commercially available 14
(see box, Scheme 3). A regioselective bromination of 14 with
N-bromosuccinimide19 afforded bromophenol 15 as the sole
regioisomer (confirmed by X-ray crystallographic analysis) in
near quantitative yield. Protection of the phenol proceeded
smoothly to generate a benzyl ether which was subsequently
treated with tert-butyllithium and 2-isopropoxy-4,4,5,5-tetra-
methyl-1,3,2-dioxaborolane to give pinacol ester 16 in 60% yield.
It was envisaged that 2-chloro-5,6,8-trimethoxy quinoline
would provide access to a range of nucleophilic coupling partners
that could potentially establish the ABÀC ring linkage.20 To
this end, commercially available 2,4,5-trimethoxybenzaldehyde 17
was converted to the corresponding nitro compound 18 in 80%
With the three main fragments in hand, we began to con-
struct the tetracyclic scaffold of streptonigrin. A Stille reaction
with tetrakis(triphenylphosphine)palladium efficiently coupled
stannylquinoline 24 with pyridine 13, generating the ABÀC
framework 25 in 77% yield (Scheme 4).23 Pleasingly, no product
resulting from coupling to the 4-bromo position on the pyridine
was observed.24 Next, a Suzuki reaction involving bromopyridine
25, aryl borate ester 16, and tetrakis(triphenylphosphine)
palladium established the CÀD ring axis 26 in 74% yield.25
Finally, a highly regioselective CAN-mediated oxidation26 of
quinoline 26 to quinone 27 resulted in the production of a late-
stage intermediate in Weinreb’s total synthesis and the final
compound described in Kende’s work. In our case this synthesis
has been achieved in 10 linear steps with an overall yield of 22%
from ethyl glyoxalate, which is less than half the number of steps
originally required.10
All that remained to convert quinone 27 into streptonigrin
was the introduction of the quinone amino group, removal of the
16419
dx.doi.org/10.1021/ja207835w |J. Am. Chem. Soc. 2011, 133, 16418–16421

Journal of the American Chemical Society
COMMUNICATION
Scheme 4
Taking Weinreb’s conditions as a starting point we sought to
introduce iodine selectively at the 7-position of quinone 27 with
IN3. However, in our hands we were not able to reproduce this
transformation28 so an alternative activating group was chosen.
Having studied a range of brominating reagents we found that
competing bromination on the D-ring was unavoidable; yet in
the presence of excess bromine and pyridine we achieved
quantitative conversion of quinone 27 to the dibrominated
compound 28 (Scheme 5). This was found to be relatively
unstable to chromatography ꢀand so was treated immediately
with sodium azide in DMF at 0C to furnish azidoquinone 29. As
29 also proved to be unstable, the crude material was taken on
and subjected to catalytic Pd/C under an atmosphere of hydro-
gen. To our satisfaction, these reaction conditions effected the 3
desired manipulations: azide reduction, benzyl ether hydroge-
nolysis and reduction of the unwanted aryl bromide. Pleasingly,
streptonigrin methyl ester 30 was obtained in 57% overall yield
from the quinone formal intermediate 27. Finally, treatment with
potassium carbonate in aqueous methanol revealed the free
C-ring carboxylic acid, affording streptonigrin in 88% yield;
the spectroscopic data of which matched those reported in
the literature29 and that of an authentic sample of the natural
product.
To conclude, we have completed a concise and efficient total
synthesis of the antitumor agent (()-streptonigrin, in 14 linear
steps from ethyl glyoxalate; this sequence has an overall yield of
11%. The synthesis showcases recent methodology in the rapid
generation of a complex pyridine and revises the long-standing
end game for streptonigrin, providing a shorter and more robust
set of tactics. This represents the shortest and highest yielding
synthesis to date. Our synthesis will allow the rapid assembly of
a comprehensive range of analogues, whose potential as viable
drug candidates could be readily assessed.
Scheme 5
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures and
b
spectroscopic data for all new compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
Timothy.donohoe@chem.ox.ac.uk
’ ACKNOWLEDGMENT
We thank St. John’s College, Oxford (C.R.J.) for supporting
this project and the Brazillian Research Council (CNPq) for a
Fellowship (L.C.A.B.). Akshat H. Rathi and David B. Baker are
thanked for X-ray crystallographic analysis. Matthew R. Tatton
and Louise J. Walport are thanked for experimental assistance.
’ REFERENCES
(1) Rao, K. V.; Cullen, W. P. Antibiot. Annu. 1959À1960, 950.
(2) Bringmann, G.; Reichert, Y.; Kane, V. V. Tetrahedron 2004,
60, 3539.
(3) Bolzꢀan, A. D.; Bianchi, M. S. Mutat. Res. 2001, 488, 25.
(4) Von Hoff, D. D.; Rozencwieg, M.; Soper, W. T.; Hellman, J. J.;
Penta, J. S.; Davis, H. L.; Muggia, F. M. Cancer Treat. Rep. 1977, 61, 759.
(5) Rao, K. V.; Biemann, K.; Woodward, R. B. J. Am. Chem. Soc.
1963, 85, 2532.
D-ring benzyl ether, and hydrolysis of the pyridine methyl ester.
Weinreb achieved this sequence in five steps and 17% overall
yield.10,27 The end game has received virtually no attention since
the original publication, and we believed that there was now the
possibility to improve some of the requisite transformations.
16420
dx.doi.org/10.1021/ja207835w |J. Am. Chem. Soc. 2011, 133, 16418–16421

Journal of the American Chemical Society
COMMUNICATION
(6) Chiu, Y.-Y. K.; Lipscomb, W. N. J. Am. Chem. Soc. 1975, 97, 2525.
(7) SAR Reviews:(a) Hadju, J. Met. Ions Biol. Syst. 1985, 19, 53.
(b) Gould, S. J.; Weinreb, S. M. Fortschr. Chem. Org. Natur. 1982, 41, 77.
(c) Harding, M. M.; Long, G. V. Curr. Med. Chem. 1997, 4, 405.
(8) (a) White, J. R. Biochem. Biophys. Res. Commun. 1977, 77, 387.
(b) Cone, R.; Hasan, S. K.; Lown, J. W.; Morgan, A. R. Can. J. Biochem.
1976, 54, 219. (c) Hassett, D. J.; Britigan, B. E.; Svendsen, T.; Rosen,
G. M.; Cohen, M. S. J. Biol. Chem. 1987, 262, 13404. (d) Gutteridge,
J. M. Biochem. Pharmacol. 1984, 33, 3059. (e) Boger, D. L.; Yasuda, M.;
Mitscher, L. A.; Drake, S. D.; Kitos, P. A.; Collins Thompson, S. J. Med.
Chem. 1987, 30, 1918.
(9) Tennant, S.; Rickards, R. W. Tetrahedron 1997, 53, 15101.
(10) (a) Basha, F. Z.; Hibino, S.; Kim, D.; Pye, W. E.; Wu, T.-T.;
Weinreb, S. M. J. Am. Chem. Soc. 1980, 102, 3962. (b) Weinreb, S. M.;
Basha, F. Z.; Hibino, S.; Khatri, N. A.; Kim, D.; Pye, W. E.; Wu, T.-T.
J. Am. Chem. Soc. 1982, 104, 536.
(11) Kende, A. S.; Lorah, D. P.; Boatman, R. J. J. Am. Chem. Soc.
1981, 103, 1271.
(12) Boger, D. L.; Panek, J. S. J. Am. Chem. Soc. 1985, 107, 5745.
(13) For examples of streptonigrin analogues via Pd-mediated cross-
couplings see: Godard, A.; Marsais, F.; Plꢀe, N.; Trꢀecourt, F.; Turck, A.;
Quꢀeguiner, G. Heterocycles 1995, 40, 1055 and references therein.
(14) (a) Donohoe, T. J.; Bower, J. F.; Basutto, J. A.; Fishlock, L. P.;
Procopiou, P. A.; Callens, C. K. A. Tetrahedron 2009, 65, 8969. (b)
Donohoe, T. J.; Fishlock, L. P.; Basutto, J. A.; Bower, J. F.; Procopiou,
P. A.; Thompson, A. L. Chem. Commun. 2009, 65, 3008. (c) Donohoe,
T. J.; Fishlock, L. P.; Procopiou, P. A. Org. Lett. 2008, 10, 285.
(15) Yim, A.-M.; Vidal, Y.; Viallefont, P.; Martinez, J. Tetrahedron:
Asymmetry 2002, 13, 503.
(16) Hong, S. H.; Sanders, D. P.; Lee, C. W.; Grubbs, R. H. J. Am.
Chem. Soc. 2005, 127, 17160.
(17) Schmidt, B. Eur. J. Org. Chem. 2004, 1865.
(18) For examples of analogues via Suzuki reactions see: (a) Marsais,
F.; Rovera, J. C.; Turck, A.; Godard, A.; Quꢀeguiner, G. J. Chem. Soc.,
Perkin Trans. 1 1990, 2611. (b) Godard, A.; Rocca, P.; Pomel, V.;
Thomas-dit-Dumont, L.; Rovera, J. C.; Thaburet, J. F.; Marsais, F.;
Quꢀeguiner, G. J. Organomet. Chem. 1996, 517, 25. (c) Crous, R.; Dwyer,
C.; Holzapfel, C. W. Heterocycles 1999, 51, 721. For examples using
silicon see: (d) Movery, M. E.; Deshong, P. Org. Lett. 1999, 1, 2137.
(e) McElroy, W. T.; Deshong, P. Org. Lett. 2003, 5, 4779.
ꢀ
(19) Pla, D.; Albericio, F.; Alvarez, M. Eur. J. Org. Chem. 2007, 1921.
(20) Forexamplesofstreptonigrinanaloguesvia Stille reactions see: (a)
Godard, A.; Fourquez, J. M.; Tamion, R.; Marsais, F.; Quꢀeguiner, G. Synlett
1994, 235. (b) Kimber, M.; Anderberg, P. I.; Harding, M. M. Tetrahedron
2000, 56, 3575. (c) Fryatt, T.; Goroski, D. T.; Nilson, Z. D.; Moody, C. J.;
Beall, H. D. Bioorg. Med. Chem. Lett. 1999, 9, 2195. (d) Verniest, G.; Wang,
X.; De Kimpe, N.; Padwa, A. J. Org. Chem. 2010, 75, 424.
(21) Downer, N. K.; Jackson, Y. A. Org. Biomol. Chem. 2004, 2, 3039.
(22) Fernꢀandez, F.; García-Mera, X.; Morales, M.; Rodríguez-
Borges, J. E. Synthesis 2001, 239.
(23) Mee, S. P. H.; Lee, V.; Baldwin, J. E. Angew. Chem., Int. Ed. 2004,
43, 1132.
(24) Pyridines exert a preference for coupling at the C-2 position: see
Reimann, S.; Erfle, S.; Villinger, A.; Langer, P. Synlett 2010, 1528.
(25) Asymmetric Suzuki reactions with a variety of commercially
available enantiopure ligands gave the product 26 with up to 22% ee
using BINAP.
(26) Chan, B. K.; Ciufolini, M. A. J. Org. Chem. 2007, 72, 8489.
(27) This end game had also been developed independently: Kende,
A. S.; Naegely, P. C. Tetrahedron Lett. 1978, 48, 4775.
(28) This transformation has been reproduced in a related system,26
where the authors comment on the instability of the intermediates. For
information on IN3 see: Hassner, A. Encyclopedia of Reagents for Organic
Synthesis; Paquette, L. A., Ed.; John Wiley and Sons: New York, 1995:
Vol. 4, p 2802.
(29) Streptonigrin NMR spectroscopic data: Wang, H.; Yeo, S. L.;
Xu, J.; Xu, X.; He, H.; Ronca, F.; Ting, A. E.; Wang, Y.; Yu, V. C.; Sim,
M. M. J. Nat. Prod. 2002, 65, 721 and references therein.
16421
dx.doi.org/10.1021/ja207835w |J. Am. Chem. Soc. 2011, 133, 16418–16421