Oxidative carbon-carbon bond formation in the synthesis of bioactive spiro β-lactams

Article abstract of DOI:10.1021/ol901005x

New oxidative dearomatization procedures leading to spiro/Mactams and oxindoles were developed. By a variation of the oxidative reaction conditions, the usefulness of phenolic amides, derived from 4-aminophenol, In the synthesis of structurally different

Full text of DOI:10.1021/ol901005x

ORGANIC
LETTERS
2009
Vol. 11, No. 13
2820-2823
Oxidative Carbon-Carbon Bond
Formation in the Synthesis of Bioactive
Spiro ꢀ-Lactams
Jixuan Liang, Jingbo Chen, Fengxiang Du, Xianghui Zeng, Liang Li, and
Hongbin Zhang*
Key Laboratory of Medicinal Chemistry for Natural Resource (Yunnan UniVersity),
Ministry of Education, School of Chemical Science and Technology,
Yunnan UniVersity, Kunming, Yunnan 650091, P. R. China
zhanghb@ynu.edu.cn; zhang_hongbin@hotmail.com
Received May 1, 2009
ABSTRACT
New oxidative dearomatization procedures leading to spiro ꢀ-lactams and oxindoles were developed. By a variation of the oxidative reaction
conditions, the usefulness of phenolic amides, derived from 4-aminophenol, in the synthesis of structurally different types of molecules was
demonstrated.
In the past few years, there is growing interest in the synthesis
of azetidinones. Numerous synthetic methods exist that lead
Scheme 1. Synthesis of Spiro-ꢀ-lactams by Oxidative C-N
Coupling and Our Speculation Towards the Synthesis of Highly
Functional Azetidinones
to azetidinones,1 so why bother choosing ꢀ-lactam? The
reason was quite simple: the ꢀ-lactam substructure is not
only present in biologically active natural products but is
also incorporated into numerous pharmaceutical ingredients
such as penicillins and carbapenems.2 It is well-known that
bacterial resistance to existing antibiotics is increasing and
therefore there is great demand for the creation of a new
type of compound bearing a ꢀ-lactam motif. Of the methods
leading to azetidinones, we were especially interested in the
oxidative dearomatization strategies developed by the groups
of Kikugawa,3a Wardrop3b and Kita.3c In their methods, a
carbon-nitrogen bond was formed in the oxidation steps as
indicated in Scheme 1. Inspired by these elegant oxidative
dearomatization procedures, we began to ponder whether a
new synthetic methodology that leads to ꢀ-lactam building
(1) For reviews on synthesis and application of ꢀ-lactams, see: (a)
Alcaide, B.; Almendros, P.; Aragoncillo, C. Chem. ReV. 2007, 107, 4437.
(b) France, S.; Weatherwax, A.; Taggi, A. E.; Lectka, T. Acc. Chem. Res.
2004, 37, 594. (c) Singh, G. S. Tetrahedron 2003, 59, 7631.
(2) Butler, M. S. Nat. Prod. Rep. 2005, 22, 162.
blocks could be developed by an oxidative carbon-carbon
bond formation of phenolic amide derivatives (Scheme 1).
10.1021/ol901005x CCC: $40.75
Published on Web 06/10/2009
 2009 American Chemical Society

If an oxidative dearomatization followed by a concomitant
carbon-carbon formation occurs, ꢀ-lactams with a rich
functionality could be readily provided and thus diverse
manipulation is possible. The success of such a procedure
may rely on the oxidative dearomatization of phenol deriva-
tives with proper oxidants. After an intensive survey of the
literature,4 we decided to use the readily available hyperva-
lent iodine reagents,5 namely, iodobenzene diacetate (IBD).
A program with the aim of synthesizing highly functional
spiro ꢀ-lactams, as shown in Scheme 1, was then initiated.
Having failed to effect the desired transformation, an
alternative way was thus sought. Because copper salts have
been widely employed in coupling reactions,6 we decided
to examine the combination of copper salts and IBD first. A
number of copper salts were tested, and the results are
summarized in Table 1. To our surprise, the addition of
Table 1. Attempts Towards the Synthesis of Spiro-ꢀ-lactamsa
Scheme 2. Initial Attempts Towards Oxidative Coupling
entry
copper salts
ligand (base)
solvent
yieldb
1
2
3
4
5
6
7
8
CuI
CuCl2
Cu(acac)2
Cu(OAc)2
CuSO4·5H2O
CuSO4·5H2O
CuSO4·5H2O
CuSO4·5H2O
CuSO4·5H2O
CuSO4·5H2O
CuSO4·5H2O
CuSO4·5H2O
CuSO4·5H2O
-
-
-
-
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
CH3CN
MeOH
20%
25%
12%
24%
35%
<5%
<5%
<5%
51%
85%
57%c
30%
0%d
-
PPh3
Proline
TMEDA
Pyridine
DMAP
DMAP
DMAP
DMAP
9
The first amide (compound 1 in Scheme 2) was obtained
by the direct reaction of 4-(benzylamino)phenol with ethyl
3-chloro-3-oxopropanate in 92% yield. Although oxidation
of phenols with polyvalent iodine reagents to enone deriva-
tives has been well documented in the literature,8 no
oxidative reaction has ever been conducted using amide 1.
The initial oxidative coupling of amide 1 was then attempted
with IBD in methanol at room temperature. To our disap-
pointment, only cyclohexadienone 2 and 2a were detected
and no desired coupling reaction occurred. Using PIFA
(PhI(OCOCF3)2) as an oxidizing agent as well as using a
base (Et3N, DBU (1,8-diazabicyclo[5.4.0]undec-7-ene), NaH-
CO3 and K2CO3) as an additive to the reaction system did
not alter the reaction pathway, with compound 2 and
compound 2a being the isolated products. Varying the
solvents in the reaction system (dichloromethane, trifluoro-
ethanol, acetonitrile, ethyl acetate, DMF) was also fruitless.
10
11
12
13
a Reactions, except entry 11 and 13, were carried out at 0 °C with amide
1 (0.5 mmol) and IBD (0.6 mmol, 1.2 equiv) in a solvent (10 mL) for 1-2
h in the presence of a copper salt (0.5 mmol, 1.0 equiv). Entries 6-12
were carried out with a ligand (0.5-0.6 mmol). b Yields represent isolated
yields of 3. c Reaction was conducted at 0 °C with amide 1 (0.5 mmol),
IBD (0.6 mmol, 1.2 equiv), catalytic amount of CuSO4·5H2O (0.1 mmol)
and DMAP (0.12 mmol) for 2 h. d Reaction was conducted at 0 °C with
amide 1 (0.5 mmol), CuSO4·5H2O (1.2 mmol, 2.4 equiv) and DMAP (1.2
mmol) for 2 h in methanol (10 mL) in the absence of IBD.
copper salts resulted in the desired azetidinone 3. Charac-
teristic NMR peaks that were observed for compound 3 are
a singlet signal at 4.14 ppm in the 1H NMR spectrum and a
quaternary carbon resonance at 57.7 ppm in the 13C NMR
spectrum. The best result (Table 1, entry 10) was obtained
upon the addition of copper(II) sulfate pentahydrate and
4-dimethylaminopyridine (DMAP) in methanol. The desired
azetidinone (3) was obtained in 85% yield together with a
trace amount of byproduct 2a. In the absence of IBD,
however, copper(II) sulfate pentahydrate did not promote the
formation of the spiro-ꢀ-lactam (Table 1, entry 13).
To get further insights toward the generality of this
process, a number of phenol derived amides (1-1i) were
prepared and evaluated under optimized conditions. Good
yields were obtained, and the results are summarized in Table
2. After careful recrystallization, an X-ray crystal structure
for compound 3i was obtained and the spiro ꢀ-lactam
structure was confirmed. All azetidinone compounds listed
in Table 2 are bioactive against a number of tumor cell lines
with IC50 potencies ranging from 17 to 0.3 µM.7
(3) (a) Miyazawa, E.; Sakamoto, T.; Kikugawa, Y. Heterocycles 2003,
53, 149. (b) Wardrop, D. J.; Burge, M. S. J. Org. Chem. 2005, 70, 10271.
(c) Dohi, T.; Maruyama, A.; Minamitsuji, Y.; Takenaga, N.; Kita, Y. Chem.
Commun. 2007, 1224.
(4) For recent reviews on oxidative coupling, see: (a) Quideau, S.;
Pouysegu, L.; Deffieux, D. Synlett 2008, 467. (b) Ciufolini, M. A.; Braun,
N. A.; Canesi, S.; Ousmer, M.; Chang, J.; Chai, D. Synthesis 2007, 3759.
(c) Magdziak, D.; Meek, S. J.; Pettus, T. R. R. Chem. ReV. 2004, 104,
1383. (d) Rodr´ıgues, S.; Wipf, P. Synthesis 2004, 2767. (e) Van De Water,
R. W.; Pettus, T. R. R. Tetrahedron 2002, 58, 5367.
(5) For reviews on polyvalent iodine chemistry, see: (a) Zhdankin, V. V.;
Stang, P. J. Chem. ReV. 2008, 108, 5299. (b) Richardson, R. D.; Wirth, T.
Angew. Chem., Int. Ed. 2006, 45, 4402. (c) Wirth, T. Angew. Chem., Int.
Ed. 2005, 44, 3656. (d) Moriarty, R. M. J. Org. Chem. 2005, 70, 2893. (e)
Zhdankin, V. V.; Stang, P. J. Chem. ReV. 2002, 102, 2523, and references
cited therein.
(6) Evano, G.; Blanchard, N.; Toumi, M. Chem. ReV. 2008, 108, 3054.
(7) The bioassay was carried out with the MTT method. In vitro studies
demonstrated that the azetidinones listed in Table 2are active against A431,
HepG-2 and Skov-3 tumor cells.
(8) (a) Peng, H. M.; Wester, R. D. J. Org. Chem. 2008, 73, 2169. (b)
Eickhoff, H.; Jung, G.; Rieker, A. Tetrahedron 2001, 57, 353, references
cited therein.
The reaction pathway might consist of a radical coupling
reactions between the para-position of a phenol unit and the
Org. Lett., Vol. 11, No. 13, 2009
2821

Table 2. ꢀ-Lactams Prepared under Optimized Conditionsa
Scheme 3
.
Further Attempts Towards Oxidative Coupling under
Optimized Conditions
pling under our oxidizing condition as indicated in Scheme
3. A complex mixture was produced in the absence of a
substituent (6, Scheme 3) in the nitrogen atom of amide 6.
To extend the utility of our oxidative conditions (IBD
coupled with CuSO4·5H2O), we then synthesized an benzyl
amide derivative 6a and attempted an oxidative coupling
under optimized reaction conditions. To our delight, a spiro
pyrrolidone, compound 7, was obtained in 72% yield. By
the way, spiro pyrrolidone 7, which bears a full carbon
quarternary carbon center as shown in Scheme 4, could not
a For reaction conditions, see the Supporting Information. b Yields
represent isolated yields (average of two runs).
R-position of the two carbonyl groups and both radical
species might be generated through oxidation reactions of
IBD in the presence of copper(II) sulfate pentahydrate (Table
1, entry 10). To better understand the mechanism of this
coupling process, amide 1j was prepared from 2-(benzy-
lamino)phenol and subjected to the oxidative coupling
(Scheme 3). In this reaction, a new carbon-oxygen bond
rather than a carbon-carbon bond formed by a direct
coupling of the ortho-phenolic oxygen with the methylene
carbon adjacent to the two carbonyl groups. Although we
are unable to exclude the possibility of an ionic pathway8
(as shown in Scheme 1) between the phenoxonium cation
generated by the oxidation of phenol and the enolate
produced by the copper salt in the presence of a base
(DMAP), more evidence needed, we favor a radical
mechanism.8b The outcome of the oxidative coupling of
amide 1j could be better explained by a radical pathway.
The role of the copper salts in this typical coupling reaction
has not been determined and it was speculated that it might
act as a co-oxidizing agent and electron acceptors.9 With a
substituent (allyl group, amide 1k) present at the R-position
of the two carbonyl groups, no azetidinone was observed,
with compound 5 being isolated in 91% yield. This was
possibly due to the steric hindrance imposed by the substi-
tutent which sheltered the R-position adjacent to the two
carbonyl groups against further carbon-carbon bond cou-
Scheme 4. Attempts Towards the Synthesis of Spiro Pyrrolidone
be prepared by C-N coupling methods reported in the
literature (ref 3).
Inspired by the results presented in Scheme 3 (1k to 5),
we next set our goal toward the synthesis of oxindole units
using phenolic amides derived from 4-aminophenol. Oxin-
dole substructures are also widely found in bioactive natural
products as well as in pharmaceuticals.10 We considered that
the oxindole structure might also be synthesized from the
same phenolic amides through an oxidative dearomatization
followed by an intramolecular Michael addition and an acid
(9) Baran, P. S.; DeMartino, M. P. Angew. Chem., Int. Ed. 2006, 45,
7083.
2822
Org. Lett., Vol. 11, No. 13, 2009

catalyzed rearomatization. A number of amides (1, 1k-1n)
were then examined for the synthesis of oxindoles (Scheme
5). In the absence of copper salts, oxidation with IBD in
oxidation with IBD and subsequent Michael addition in the
presence of DBU, did not give an oxindole after treatment
of 2a with p-toluenesulfonic acid (Scheme 5). Treatment of
compound 8a with methyl iodide in the presence of potas-
sium carbonate and in acetone provided an advanced
intermediate for the total synthesis of Horsfiline.12
Scheme 5
.
Synthesis of Oxindoles by Oxidative
Dearomatization
In summary, we have introduced an alternative reaction
for the synthesis of highly functional spiro-ꢀ-lactams, a
supplement to the Staudinger reaction. To the best of our
knowledge, this is the first oxidative carbon-carbon bond
forming procedure that has led to spiro-ꢀ-lactams and spiro-
pyrrolidone.3,13 We have also developed a useful synthetic
procedure that leads to oxindoles. An especially attractive
feature of this research is that both methods are based on an
oxidative dearomatization from similar phenolic amide
derivatives. By varying the oxidative reaction conditions, we
demonstrate the usefulness of phenolic amides, derived from
4-aminophenol, in the synthesis of structurally different
molecules. Syntheses of ꢀ-lactams and oxindoles are ef-
ficient, have high yield and are easily handled. Further
biological studies of ꢀ-lactams as well as the total synthesis
of oxoindole related natural products are currently underway
in our laboratories and will be reported in due course.
Acknowledgment. This work was supported by grants
from the Chinese Ministry of Education for fostering
important scientific research program (706052), the Yunnan
Provincial Science & Technology Department (2007B0006Z)
and the National Natural Science Foudation of China
(20832005).
methanol resulted in 4-methoxyl cyclohexadienenones in
nearly quantitative yields. After removal of the solvent,
Michael additions were achieved by the addition of DBU in
dichloromethane. After treatment of the crude products with
TsOH, oxindoles were obtained in 81-89% yields. Although
oxindole and dihydroindoles have been realized by using
oxidative dearomatization as the key step,10g,h,11 our oxindole
procedure produced a full carbon quarternary carbon center
adjacent to the aromatic ring, therefore afforded useful
building blocks for the synthesis of natural alkaloids such
as Horsfiline. It is noteworthy that amide 1 (R ) H), after
Supporting Information Available: 1H NMR, 13C NMR
spectra of all key intermediates and experimental details. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL901005X
(10) For recent synthesis of oxindoles, see: (a) Felpin, F.-X.; Ibarguren,
O.; Nassar-Hardy, L.; Fouquet, E. J. Org. Chem. 2009, 74, 1349. (b) Luan,
X.; Mariz, R.; Robert, C.; Gatti, M.; Blumentritt, S.; Linden, A.; Dorta, R.
Org. Lett. 2008, 10, 5569. (c) Shanmugam, P.; Vaithiyanathan, V.
Tetrahedron 2008, 64, 3322. (d) Hillgren, J. M.; Marsden, S. P. J. Org.
Chem. 2008, 73, 6459. (e) Ku¨ndig, E. P.; Seidel, T. M.; Jia, Y.-X.;
Bernardinelli, G. Angew. Chem., Int. Ed. 2007, 46, 8484. (f) Teichert, A.;
Jantos, K.; Harms, K.; Studer, A. Org. Lett. 2004, 6, 3477, and references
cited thereinFor recent synthesis of oxindoles with an oxidative dearoma-
tization strategy, see: (g) Pouyse´gu, L.; Avellan, A.-V.; Quideau, S. J. Org.
Chem. 2002, 67, 3425. (h) Quideau, S.; Pouyse´gu, L.; Avellan, A.-V.;
(11) In the literature, Quideau and Clive had elegantly demonstrated
the synthesis of 2,3-dihydroindoles with an oxidative dearomatization
strategy, see: (a) Quideau, S.; Pouyse´gu, L.; Oxoby, M.; Looney, M. A.
Tetrahedron 2001, 57, 319. (b) Fletcher, S. P.; Clive, D. L. J.; Peng, J.;
Wingert, D. A. Org. Lett. 2005, 7, 23.
(12) Trost, B. M.; Brennan, M. K. Org. Lett. 2006, 8, 2027.
(13) Formation of spirolactams by oxidative carbon-nitrogen bond
coupling had been elegantly demonstrated by ref 3 and: (a) Ousmer, M.;
Braun, N. A.; Bavoux, C.; Perrin, M.; Ciufolini, M. A. J. Am. Chem. Soc.
2001, 123, 7534, and references cites therein. (b) Ousmer, M.; Braun, N. A.;
Whelligan, D. K.; Looney, M. A. Tetrahedron Lett. 2001, 42, 7393
.
Ciufolini, M. A. Org. Lett. 2001, 3, 765, and references cites therein
.
Org. Lett., Vol. 11, No. 13, 2009
2823