Palladium-catalyzed substitution of unsaturated lactones. Application to the synthesis of carbocyclic polyoxins and nikkomycins

Full text of DOI:10.1021/jo952187w

1192
J . Org. Chem. 1996, 61, 1192-1193
Sch em e 1
P a lla d iu m -Ca ta lyzed Su bstitu tion of
Un sa tu r a ted La cton es. Ap p lica tion to th e
Syn th esis of Ca r bocyclic P olyoxin s a n d
Nik k om ycin s
Varinder K. Aggarwal,*,† Nuno Monteiro,†
Gary J . Tarver,† and Steven D. Lindell‡
Department of Chemistry, University of Sheffield,
Sheffield S3 7HF, U.K., and AgroEvo UK Ltd.,
Chesterford Park, Saffron Walden, Essex CB10 1XL, U.K.
Received December 12, 1995
Initial attempts at opening the model lactone 2 using
Pd(0) with either uracil anion or the more nucleophilic
NaN3 met with failure.
Carbocyclic analogues of nucleosides display a wide
range of biological activity and have attracted particular
attention as antitumor1 and antiviral agents.2 The
potential advantage in the therapeutic use of these
compounds include greater metabolic stability and in
some cases increased biological activity compared to
nucleosides. These features have fueled an explosion of
interest in the design and synthesis of further analogues
of this class of compounds.3 However, little attention has
been given to the carbocyclic analogues of the nikkomy-
cins or polyoxins, a class of nucleosides in which the 5′
position is substituted by an amino acid.4 The natural
compounds exhibit selective activity against certain
pathogenic fungi by preventing fungal cell wall growth
through inhibition of the enzyme chitin synthase, but are
nontoxic to bacteria, plants, and mammals.5 As this
enzyme is also found in insects, these compounds are also
effective insecticides. In this paper we describe the first
stereospecific total synthesis of the carbocyclic analogue
1 of uracil polyoxin C and significant findings related to
palladium-catalyzed substitution reactions.
Unlike the situation with allylic acetates or carbonates,
with unsaturated lactones the intermediate π-allyl pal-
ladium species has an internal nucleophile, the carboxy-
late, which competes with the external nucleophile
during the substitution step. There is, therefore, an
equilibrium between the lactone and the intermediate
π-allyl palladium species. With unstrained lactones this
equilibrium is very much on the side of the lactone and
only exceptionally good nucleophiles, like dimethyl ma-
lonate,6 can react with the very small amount of the
intermediate π-allyl palladium species present. With
strained lactones the equilibrium is shifted toward the
intermediate π-allyl palladium species, and they can now
react with less reactive nucleophiles.7 We therefore
needed to influence the position of this equilibrium. We
reasoned that an additional substituent R to the carbonyl
group should achieve this as the endo isomer should shift
the equilibrium toward the intermediate π-allyl pal-
ladium species relative to the exo isomer due to the
additional strain introduced upon ring closure back to
the lactone (Scheme 2). We initially prepared the two
hydroxy lactones8 2a and 3a in racemic form and reacted
them with sodium diethyl malonate, and we observed a
dramatic difference in reactivity of the two isomers. The
endo isomer 3a gave a complete reaction after only 1 h
at room temperature while the exo isomer 2a gave
essentially no product even after several days in refluxing
THF. The previously reported unsubstituted lactone
reacts at an intermediate rate.6
Our strategy toward this class of molecules involved a
palladium-catalyzed substitution reaction of an unsatur-
ated lactone as the key step as shown in Scheme 1. There
appears to be only one example of an unstrained lactone
being opened by a nucleophile under palladium catalysis,
and this lack of precedence provided further impetus for
us to explore this class of reactions.6
* To whom correspondence should be addressed. E-mail: V.
Aggarwal@Sheffield.ac.uk.
The two bromo lactones were also prepared9 and
reacted with sodium diethyl malonate, but this time both
the endo and exo isomers reacted although the endo
isomer 3b reacted more quickly than the exo isomer 2b.
However, both reactions led to the same diastereomer 5b.
This indicated that the exo isomer 2b epimerized slowly
to give some of the more strained endo lactone 3b, and
this species then underwent palladium-catalyzed allylic
substitution. The scope of the reaction was determined
† University of Sheffield.
‡ AgroEvo UK Ltd.
(1) Kaneko, M. J . Synth. Org. Chem. J pn. 1991, 49, 989-1001.
(2) Boehme, R. E.; Borthwick, A. D.; Wyatt, P. G. Annu. Rep. Med.
Chem. 1994, 29, 145-154.
(3) For reviews see: Borthwick, A. D.; Biggadike, K. Tetrahedron
1992, 48, 571-623. Agrofoglio, L.; Suhas, E.; Farese, A.; Condom, R.;
Challand, S. R.; Earl, R. A.; Guedj, R. Tetrahedron 1994, 50, 10611-
10670.
(4) The natural compounds have been the focus of recent attention:
Dondoni, A.; J unquera, F.; Mercha´n, F. L.; Merino, P.; Tejero, T.
Tetrahedron. Lett. 1994, 35, 9439-9442 and references cited therein.
A partial synthesis of racemic carbocyclic nikkomycin Z has been
reported: Baumgartner, H.; Marschner, C.; Pucher, R.; Singer, M.;
Griengl, H. Tetrahedron. Lett. 1992, 33, 6443-6444.
(5) Cabib, E. Antimicrob. Agents. Chemother. 1991, 35, 170-173.
Milewski, S.; Mignini, F.; Borowski, E. J . Gen. Microbiol. 1991, 137,
2155-2161. Chapman, T.; Kinsman, O.; Houston, J . Antimicrob.
Agents. Chemother. 1992, 36, 1909-1914.
(7) Matsushita, H.; Negishi, E. J . Chem. Soc., Chem. Commun. 1982,
160-161. Bystrom, S. E.; Aslanian, R.; Ba¨ckvall, J . E. Tetrahedron.
Lett. 1985, 26, 1749-1752. Trost, B. M.; Murphy, D. J . Organometallics
1985, 4, 1143-1145. Murahashi, S. I.; Taniguchi, Y.; Imada, Y.;
Tanigawa, Y. J . Org. Chem. 1989, 54, 3292-3303.
(8) Lubineau, A.; Auge´, J .; Grand, E.; Lubin, N. Tetrahedron 1994,
50, 10265-10276.
(9) Ho, P. T.; Davies, N. J . Org. Chem. 1984, 49, 3027-3029.
(6) Trost, B. M.; Verhoeven, T. R. J . Am. Chem. Soc. 1980, 102,
4730-4743.
0022-3263/96/1961-1192$12.00/0 © 1996 American Chemical Society

Communications
J . Org. Chem., Vol. 61, No. 4, 1996 1193
Sch em e 2
50% yield. In the absence of TMSCl no coupling occurred.
The penultimate step in the synthesis was stereocon-
trolled dihydroxylation of the cyclopentene, and a variety
of reaction conditions were examined. The highest yield
was obtained using 4-methylmorpholine N-oxide (NMO)14
with catalytic OsO4 in THF (90%) but as a 1:1 mixture
of 13:14.15 14 resulted from lactonization of the unde-
sired diol, and this transformation aided separation of
the two isomeric products and confirmed the stereochem-
istry of the dihydroxylation. Finally, hydrogenolysis of
both benzyl protecting groups was achieved in essentially
quantitative yield.
for the endo hydroxy lactone 3a which was found to react
with a broad range of nucleophiles (Scheme 3).
Sch em e 3
Having solved the problem of lactone opening, we
embarked on our synthesis of the carbocyclic analogue
of the nikkomycins. Enantiomerically pure lactone (-)-
3a 10 was converted to the amino compound 10 with
overall retention of configuration. This was reacted with
the anion of uracil in the presence of Pd(0) but no
coupling occurred under a variety of conditions. While
we were confident from our earlier studies that 10 could
react with NaN3 under palladium catalysis and that we
could then build up the pyrimidine ring using standard
procedures,11 the additional steps and linearity of the
synthesis would make it less efficient and less flexible.
Nucleoside bases have successfully been used in Pd(0)-
catalyzed coupling reactions with simple allylic acetates12
but are known to be rather poor nucleophiles and
evidently are not sufficiently potent to react even with
endo substituted lactones. Assuming that the problem
still lay in the position of the equilibrium favoring the
closed lactone rather than the π-allyl palladium species,
we sought to drive the reaction over by trapping the
carboxylate released with TMSCl. We were pleased to
find that treatment of 10 with O,O-bis(trimethylsilyl)-
uracil13 (11) in the presence of TMSCl gave clean coupling
and following benzylation of the acid, 12 was isolated in
In summary, we have described a short, practical,
asymmetric route to carbocyclic analogues of the nikko-
mycins and polyoxins. In addition, we have discovered
the factors that are important for achieving palladium-
catalyzed substitution reactions of unsaturated lactones,
and as many of these compounds are now available in
enantiomerically pure form this greatly expands their
synthetic utility.
Ack n ow led gm en t. We thank the European Com-
mission for a Fellowship (N.M.) (No. ERB4001GT933280)
and Chiroscience and Sheffield University for a stu-
dentship (G.T.). We thank Chiroscience for a generous
supply of (-)-3a . We thank Dr. Paul Smith (Glaxo-
Wellcome) and Dr. Ray McCague (Chiroscience) for
valuable discussions.
Su p p or tin g In for m a tion Ava ila ble: Syntheses and spec-
tral data for 1, 2b, 3b, 5a ,b, 6-10, 12, and 13 (6 pages).
J O952187W
(10) Mackeith, R. A.; McCague, R.; Olivo, H. F.; Roberts, S. M.;
Taylor, S. J . C.; Xiong, H. Bioorg. Med. Chem. 1994, 2, 387-394.
(11) Shealy, Y. F.; O’Dell, C. A. J . Heterocycl. Chem. 1976, 13, 1015-
1020.
(12) Sigismondi, S.; Sinou, D.; Pe´rez, M.; Moreno-Man˜as, M.; Pleix-
ats, R.; Villarroya, M. Tetrahedron. Lett. 1994, 35, 7085-7088 and
references cited therein.
(14) VanRheenan, V.; Kelly, R. C.; Cha, D. Y. Tetrahedron. Lett.
1976, 19, 1973-1976.
(15) Higher ratios (13:14; 2:1) were obtained using NaIO4 with
catalytic RuO4 but in only 59% yield. See: Shing, T. K. M.; Tai, V. W.
F.; Tam, E. K. W. Angew. Chem., Int. Ed. Engl. 1994, 33, 2312-2313.
Details of the dihydroxylation studies will be discussed in the full
paper.
(13) Nishimura, T.; Iwai, I. Chem. Pharm. Bull. 1964, 12, 352-356.