New Heterocyclic Product Space for the Castagnoli-Cushman Three-Component Reaction

Article abstract of DOI:10.1021/acs.orglett.5b02014

Significant expansion of heterocyclic product space accessible by the Castagnoli-Cushman reaction (CCR) has been achieved via the use of glutaric anhydride analogues containing endocyclic substitutions with oxygen, nitrogen, and sulfur. Incorporation of these heteroatoms in the anhydride's backbone results in enhanced reactivity and generally lower temperatures that are required for the reactions to go to completion. These findings are particularly significant in light of the CCR recently recognized as an efficient tool for lead-oriented synthesis.

Full text of DOI:10.1021/acs.orglett.5b02014

Letter
pubs.acs.org/OrgLett
New Heterocyclic Product Space for the Castagnoli−Cushman Three-
Component Reaction
Dmitry Dar’in, Olga Bakulina, Maria Chizhova, and Mikhail Krasavin*
Institute of Chemistry, Saint Petersburg State University, 26 Universitetskii Prospect, Peterhof 198504, Russian Federation
S
* Supporting Information
ABSTRACT: Significant expansion of heterocyclic product
space accessible by the Castagnoli−Cushman reaction (CCR)
has been achieved via the use of glutaric anhydride analogues
containing endocyclic substitutions with oxygen, nitrogen, and
sulfur. Incorporation of these heteroatoms in the anhydride’s
backbone results in enhanced reactivity and generally lower
temperatures that are required for the reactions to go to
completion. These findings are particularly significant in light of the CCR recently recognized as an efficient tool for lead-
oriented synthesis.
Scheme 2. CCR of Succinic Anhydrides Containing Exocyclic
Heteroatom Substituents15
eaction of alicyclic anhydrides (1 and 2) with imines 3
R(either prepared in a separate step or formed in situ)
provides convenient access to polysubstituted, predominantly
transconfigured lactams 4 via a formal cycloaddition process at
elevated temperatures (Scheme 1).1 The reaction was first
described by Castagnoli in 1969;2 its scope and applications were
thoroughly investigated in the work of Cushman (initially, in the
Castagnoli lab).3−6 The substrate scope of the reaction was
subsequently expanded in the Cushman lab to include more
reactive anhydrides such as homophthalic anhydride7,8 and 3-
phenylsuccinic anhydride,9 which required lower (ambient)
temperature for the reaction to proceed.
The usage of homopthalic anhydride is particularly notable as
it provided a facile entry into the isoquinolonic acid core. This, in
turn, gave rise to a fascinating medicinal chemistry program
aimed at the discovery and optimization of novel tetracyclic
topoisomerase I inhibitors based on indenoisoquinolone10 and
dibenzo[c,h][1,5]naphthyridinedione scaffolds.11 Compounds
belonging to the former series (LMP776 and LMP400) have
advanced into human clinical trials for cancer,12 and similar
compounds have been implicated as chemopreventive agents in
oncology.13
a
Table 1. Screening of Reaction Conditions toward 7a
solvent
35 °C (24 h) 60 °C (4 h) 80 °C (2 h) 110 °C (2 h)
1,4-dioxane
toluene
16%
24%
24%
17%
13%
1.5%
35%
52%
46%
15%
ND
ND
43%
65%
35%
ND
ND
ND
ND
52%
ND
ND
ND
ND
ethyl acetate
acetonitrile
THF
chloroform
a
Data represent NMR yield of the product (trans + cis) determined
relative to n-tetradecane as internal standard (see Supporting
Information); ND, not determined.
Due to the importance of this formal three-component
reaction, which we term “the Castagnoli−Cushman reaction”
(CCR), in medicinal chemistry and natural product synthesis,1
further expanding its scope is a worthy undertaking, particularly
with respect to the anhydride inputs. Unveiling of hetero-
Scheme 3. Preparation of Sulfone Lactam 8
Scheme 1. Reaction of Imines and Alicyclic Anhydrides
aromatic analogues of homophthalic anhydride14 represented a
significant step forward. Recent work by Shaw included in situ
Received: July 14, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.5b02014
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 2. 5-Oxothiomorpholine-, 5-Oxomorpholine-, and 5-Oxopiperazine-2-carboxylic Acids (7, 10, and 13) Obtained via the
CCR of Respective Anhydrides (6, 9, and 12)
a
b
Reactions with 9 (X = O) and 12 (X = NSO2Ph) were run at 110 °C. Diastereomeric ratio (trans/cis) in reaction mixture (determined by crude
c
d
e
f
1H NMR). The dr of isolated products is given in parentheses. Reaction conducted at 80 °C. Reaction conducted at 110 °C. Confirmed by
single-crystal X-ray analysis (see Supporting Information). CCR product was converted to the methyl ester prior to isolation (see Supporting
g
Information).
generation of 3-substituted succinic anhydride by reacting a thiol
with maleic anhydride and not only enabled a true multi-
component format for this reaction and facilitated the reaction
but also allowed manipulation of the stereochemistry and
increased skeletal complexity of the final products.15−18
Surprisingly, succinic anhydrides containing exocyclic oxygen
and nitrogen substituents had been found to be ineffective
(Scheme 2).15
B
DOI: 10.1021/acs.orglett.5b02014
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 4. tert-Butyl Group Removal from Lactam 10g
Scheme 6. N-Acyl Enamine Formation from Imine 21
Similar examples reported by Shaw include the use of
sulfone-19 and cyano-substituted20 succinic anhydrides. How-
ever, no examples of heteroatom-containing cyclic anhydrides
had been reported at the time we initiated this study.21 Such an
extension of the reaction scope would be particularly desired, in
light of the recently demonstrated22,23 propensity of the CCR to
deliver lead-like compounds. Herein, we fill this void and report
our recent results obtained using oxa, aza, and thia analogues of
glutaric anhydride in the Castagnoli−Cushman reaction.
Prompted by the findings from the Shaw group,15 we first
investigated the CCR of thiodiglycolic anhydride (6), which was
conveniently prepared on multigram scale using a modified
literature protocol.24 To our delight, it reacted in toluene, rapidly
and cleanly, with imines 3 to furnish 5-oxothiomorpholin-2-
carboxylic acid products 7a−h. The temperature required to
complete the reaction was markedly lower compared to that of
the CCR of unsubstituted glutaric anhydride, which is consistent
with the expected influence of the sulfur atom on the α-CH
acidity of the anhydride and the resulting facilitation of the
overall CCR.15 The yields obtained were good to excellent, in
addition to the satisfactory diastereoselectivity. We also screened
several other solvents and temperature regimens in preparation
of 7a as a model reaction. This confirmed toluene at 80−110 °C
to be the optimal conditions for thiodiglycolic anhydride (Table
1).
Compounds 7a−h contain a divalent sulfur atom, which
introduces a metabolic liability from a drug design prospective.25
Sulfone or sulfoxide analogues of 6 are unknown, but compounds
7 can be oxidized to the respective sulfones, as was demonstrated
by preparation of compound 8 (Scheme 3). The latter represents
an intriguing polyfunctional building block, the synthetic
potential of which remains to be unveiled.
Encouraged by these initial findings (and also by the new
literature data21), we proceeded to investigate the applicability of
diglycolic anhydride, 9, in the CCR. With this reagent, the
reactions with imines 3 were found to be high yielding and
displayed clear preference for the transconfigured products 10a−
h (Table 2). The reactions proceeded to completion within 4−19
Scheme 7. Isoquinoline in the CCR with 6
h in refluxing toluene, with the conditions being much less
forceful than those reported by Burdzhiev.21
Product 10g provides a notable (and rare9) example of
incorporating a hindered amine into the CCR product structure.
It can also be easily converted into lactam 11 on treatment with
TFA (Scheme 4).
Having explored the CCR for the thia and oxa analogues of
glutaric anhydride, we were curious to see if a similar endocyclic
placement of a nitrogen atom within the structure of glutaric
anhydride (as in 12) would deliver an effective anhydride partner
for the CCR (cf. the respective exocyclic substitution of succinic
anhydride with formylamino group, which completely disabled
the subsequent reaction with an imine partner, Scheme 2).
Reaction with 12 was complete within 1−2.5 h in refluxing
toluene and delivered the expected lactams 13a−j in good to
excellent yields, with a significant preference for the trans-isomer
(Table 2). Compounds 13 represent a substantial advancement
in terms of structural diversity of heterocycles that are accessible
by the CCR. They contain the privileged piperazine core26 and
an α-amino acid motif and may be of interest for peptidomimetic
design27 and drug lead optimization.28
This mandated orthogonal protection of the piperazinone
scaffold. The initial attempt to introduce the acid-labile Boc
group at N1 in combination with a PMP group at N4 (which could
be removed later under oxidative conditions, e.g., on treatment
with CAN29) was unsuccessful because Boc-protected morpho-
lin-2,6-dione (prepared as described in the literature30),
surprisingly, failed to furnish a CCR product.
Scheme 5. Protecting Group Manipulation on the Piperazin-2-one Scaffold
C
DOI: 10.1021/acs.orglett.5b02014
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(5) Cushman, M.; Castagnoli, N. J. Org. Chem. 1973, 38, 440.
Our next choice of protecting group was (o-nitrophenyl)-
sulfonyl (ONPS) present in 14 (see Supporting Information for
preparation). The respective CCR product 15 was converted
into methyl ester 16 and then into 17 via a facile ONPS removal.
However, the PMP group was found to be resistant to removal
with CAN and had to be replaced with PMB in 18. The latter was
conveniently and orthogonally deprotected to give 19 and 20 in
sequential fashion (Scheme 5).
(6) Cushman, M.; Castagnoli, N. J. Org. Chem. 1974, 39, 1546.
(7) Cushman, M.; Gentry, J.; Dekow, F. W. J. Org. Chem. 1977, 42,
1111.
(8) An independent report on the usage of homophthalic anhydride:
Haimova, M. A.; Mollov, N. M.; Ivanova, S. C.; Dimitrova, A. I.;
Ognyanov, V. I. Tetrahedron 1977, 33, 331.
(9) Cushman, M.; Madaj, E. J. J. Org. Chem. 1987, 52, 907−915.
(10) Pommier, Y.; Cushman, M. Mol. Cancer Ther. 2009, 8, 1008.
(11) (a) Kiselev, E.; Dexheimer, T. S.; Pommier, Y.; Cushman, M. J.
Med. Chem. 2010, 53, 8716. (b) Kiselev, E.; Empey, N.; Agama, K.;
Pommier, Y.; Cushman, M. J. Org. Chem. 2012, 77, 5167.
(12) Chen, A. P. A. Phase I Study of Indenoisoquinolines LMP400 and
LMP776 in Adults with Relapsed Solid Tumors and Lymphomas, 2015.
National Institutes of Health, ClinicalTrials.gov. https://clinicaltrials.
gov/ct2/show/NCT01051635 (accessed July 2015).
(13) Conda-Sheridan, M.; Park, E.-J.; Beck, D. E.; Reddy, P. V. N.;
Nguyen, T. X.; Hu, B.; Chen, L.; White, J. J.; van Breemen, R. B.;
Pezzuto, J. M.; Cushman, M. J. Med. Chem. 2013, 56, 2581.
(14) Kita, Y.; Mohri, S.; Tsugoshi, T.; Maeda, H.; Tamura, Y. Chem.
Pharm. Bull. 1985, 33, 4723.
With the exception of a few modestly yielding reactions,1
aromatic (or non-enolizable aliphatic) aldehydes are common-
place carbonyl components used in the CCR. With enolizable
aldehydes, the anhydride component acts primarily as an
acylating agent and the reaction exclusively yields N-acyl
enamines. We also observed the formation of such compounds
(22−24) in the reaction of imine 21 with thia, oxa, and aza
analogues of glutaric anhydride (Scheme 6).
The use of cyclic imines (such as dihydroisoquinoline) in the
CCR is quite common7,31 but was found to be poorly workable
for the anhydrides studied here. To our surprise, however,
thiadiglycolic anhydride (but not its oxygen- and nitrogen-
containing counterparts) reacted with isoquinoline to give
tricyclic compound 25, with modest yield and diastereoselectiv-
ity (Scheme 7). To our knowledge, this is the first example of
unsubstituted isoquinoline taking part in the CCR.
In conclusion, the successful employment of heteroatom-
containing analogues of glutaric anhydride in the CCR described
in this paper significantly broadens the heterocyclic product
space accessible by this reaction. This is likely to result in the
development of new focused libraries for drug discovery. Such
efforts are underway in our laboratories and will be reported in
due course.
(15) Ng, P. Y.; Masse, C. E.; Shaw, J. T. Org. Lett. 2006, 8, 3999.
(16) Wei, J.; Shaw, J. T. Org. Lett. 2007, 9, 4077.
(17) Younai, A.; Fettinger, J. C.; Shaw, J. T. Tetrahedron 2012, 68,
4320.
(18) Biggs-Houck, J. E.; Davis, R. L.; Wei, J.; Mercado, B. Q.;
Olmstead, M. M.; Tantillo, D. J.; Shaw, J. T. J. Org. Chem. 2012, 77, 160.
(19) Sorto, N. A.; Di Maso, M. J.; Munoz, M. A.; Dougherty, R. J.;
̃
Fettinger, J. C.; Shaw, J. T. J. Org. Chem. 2014, 79, 2601.
(20) Tan, D. Q.; Younai, A.; Pattawong, O.; Fettinger, J. C.; Cheong, P.
H.-Y.; Shaw, J. T. Org. Lett. 2013, 15, 5126. (b) Pattawong, O.; Tan, D.
Q.; Fettinger, J. C.; Shaw, J. T.; Cheong, P. H.-Y. Org. Lett. 2013, 15,
5130.
(21) An isolated report on the usage of diglycolic anhydride in the
CCR appeared when this study was underway: Burdzhiev, N.; Stanoeva,
E.; Shivachev, B.; Nikolova, R. C. R. Chim. 2014, 17, 420.
(22) Martin, K. S.; Di Maso, M. J.; Fettinger, J. C.; Shaw, J. T. ACS
Comb. Sci. 2013, 15, 356.
ASSOCIATED CONTENT
* Supporting Information
■
S
(23) Ryabukhin, S. V.; Panov, D. M.; Granat, D. S.; Ostapchuk, E. N.;
Kryvoruchko, D. V.; Grygorenko, O. O. ACS Comb. Sci. 2014, 16, 146.
(24) Miyata, T.; Yamaoka, N.; Kodama, H.; Murano, K. PCT Int. Appl.
WO2009123241A1; Chem. Abstr. 2009, 151, 425544.
(25) Shukla, K.; Ferraris, D. V.; Thomas, A. G.; Stathis, M.; Duvall, B.;
Delahanty, G.; Alt, J.; Rais, R.; Rojas, C.; Gao, P.; Xiang, Y.; Dang, C. V.;
Slusher, B. S.; Tsukamoto, T. J. Med. Chem. 2012, 55, 10551.
(26) Welsch, M. E.; Snyder, S. E.; Stockwell, B. R. Curr. Opin. Chem.
Biol. 2010, 14, 347.
(27) Stevenazzi, A.; Marchini, M.; Sandrone, G.; Vergani, B.; Lattanzio,
M. Bioorg. Med. Chem. Lett. 2014, 24, 5349.
(28) Pickett, S. D.; McLay, I. M.; Clark, D. E. J. Chem. Inf. Model. 2000,
40, 263.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.5b02014.
Experimental procedures and characterization data for
compounds 6, 7a−h, 8, 9, 10a−h, 11, 12, 13a−j, 14−20,
and 22−25. Crystal data and full crystallographic
information for compounds 7b, 10g, and 13a,i (PDF)
AUTHOR INFORMATION
Corresponding Author
■
*Phone: + 7 931 3617872. Fax: +7 812 428 6939. E-mail: m.
krasavin@spbu.ru.
́
(29) Fustero, S.; Bartolome, A.; Sanz-Cervera, J. F.; Sanchez-Rosello,
M.; Soler, J. G.; Ramirez de Arellano, C.; Fuentes, A. S. Org. Lett. 2003, 5,
2523.
Notes
The authors declare no competing financial interest.
(30) Chen, C.; Tran, J. A.; Tucci, F. C.; Chen, W.-C. C.; Jiang, W.;
Marinkovic, D.; Arellano, M.; White, N. PCT Int. Appl.
WO2005040109; Chem. Abstr. 2005, 142, 463753.
(31) Liu, H.; Wang, J.; Zhang, R.; Cairns, N.; Liu, J. PCT Int. Appl.
WO2009002873; Chem. Abstr. 2009, 150, 98173.
ACKNOWLEDGMENTS
■
This research was supported by the Russian Scientific Fund
(Project Grant 14-50-00069). NMR, mass spectrometry, and X-
ray diffraction studies were performed at the Research Centre for
Magnetic Resonance, the Centre for Chemical Analysis and
Materials Research, and the Center for X-ray Diffraction
Methods of Research Park of Saint Petersburg State University.
REFERENCES
(1) Gonzalez-Lopez, M.; Shaw, J. T. Chem. Rev. 2009, 109, 164.
(2) Castagnoli, N. J. Org. Chem. 1969, 34, 3187.
■
(3) Castagnoli, N.; Cushman, M. J. Org. Chem. 1971, 36, 3404.
(4) Cushman, M.; Castagnoli, N. J. Org. Chem. 1972, 37, 1268.
D
DOI: 10.1021/acs.orglett.5b02014
Org. Lett. XXXX, XXX, XXX−XXX