Synthesis of cytimidine through a one-pot copper-mediated amidation cascade

Article abstract of DOI:10.1021/ol2018196

A concise synthesis of cytimidine was developed utilizing tandem Cu-mediated N-aryl amidations followed by global deprotection. This sequence exploits a regioselective coupling of an iodobenzamide with a halopyrimidine that allows the union of three fragments in a single synthetic manipulation and will permit the efficient and rapid diversification of the cytimidine core.

Full text of DOI:10.1021/ol2018196

ORGANIC
LETTERS
2011
Vol. 13, No. 19
5000–5003
Synthesis of Cytimidine through a
One-Pot Copper-Mediated Amidation
Cascade
Catherine M. Serrano and Ryan E. Looper*
Department of Chemistry, University of Utah, 315 South 1400 East, Salt Lake City,
Utah 84112, United States
r.looper@utah.edu
Received June 17, 2011
ABSTRACT
A concise synthesis of cytimidine was developed utilizing tandem Cu-mediated N-aryl amidations followed by global deprotection. This sequence
exploits a regioselective coupling of an iodobenzamide with a halopyrimidine that allows the union of three fragments in a single synthetic
manipulation and will permit the efficient and rapid diversification of the cytimidine core.
Various classes of antibiotics including aminogly-
cosides,1 macrolides,2 and tetracyclines3 have long demon-
strated the drugability of the ribosome. These families
comprise clinically relevant compounds that inhibit protein
synthesis by binding to rRNA.4 One particular class of
ribosomal antibiotics, the aminohexopyranose nucleo-
sides, have remained significantly less explored (Figure 1).
Members of this family include amicetin,5 bamicetin,5
oxamicetin,6 plicacetin,7 and cytosaminomysin B8 among
others. Previous studies have suggested that these natural
products share a common binding site near the peptidyl
transferase center (PTC) of the ribosome.9,10
Of particular interest to us is the antibiotic amicetin,
isolated from Streptomyces plicatus in 1953, for its potent
antibacterial activity against Mycobacterium tuberculosis
H37Rv (IC100 0.5 μg/mL) and Staphylococcus aureus
FDA-209 (IC100 0.2 μg/mL).5 To date, the total synthesis
of amicetin and its analogues, bamicetin and oxamicetin,
have yet to be reported. Total synthesis of other members
of the aminohexopyranose nucleoside family are quite few
and far in between.12,13 With the resurgence of drug
resistant pathogens such as multidrug resistant (MDR)
and extensively drug resistant (XDR) strains of M.
tuberculosis14 the need for more efficient therapeutics
prompted us to develop a convergent approach to the
synthesis of amicetin and its close analogues.
(1) (a) Thomas, J. R.; Hergenrother, P. J. Chem. Rev. 2008, 108,
1171–1224. (b) Leibovici, L.; Vidal, L.; Paul, M. J. Antimicrob.
Chemother. 2009, 63, 246.
(2) Zhanel, G. G.; Dueck, M.; Hoban, D. J.; Vercaigne, L. M.; Embil,
J. M.; Gin, A. S.; Karlowsky, J. A. Drugs 2001, 61, 443.
(3) Griffin, M. O.; Fricovsky, E.; Ceballos, G.; Villeareal, F. Am. J.
Physiol. Cell Physiol. 2010, 299, C539.
(4) (a) Mankin, A. S. Mol. Biol. 2001, 35, 509. (b) Poehlsgaard, J.;
Douthwaite, S. Nat. Rev. Microbiol. 2005, 3, 870.
(5) (a) Hinman, J. W.; Caron, E. L.; DeBoer, C. J. Am. Chem. Soc.
1953, 75, 499. (b) Hinman, J. W.; Caron, E. L.; DeBoer, C. J. Am. Chem.
Soc. 1953, 75, 5864. (c) Flynn, E. H.; Hinman, J. W.; Caron, E. L.;
Woolf, D. O. J. Am. Chem. Soc. 1953, 75, 5867. (d) Smith, J. L.;
Sundaralingam, M. Acta Crystallogr. 1981, B37, 1095.
(6) Konishi, M.; Kimeda, M.; Tsukiura, H.; Yamamoto, H.;
Hoshiya, T.; Miyaki, T.; Fujisawa, K.; Koshiyama, H.; Kawaguchi,
H. J. Antibiot. 1973, 26, 752.
(7) (a) Haskell, T. H.; Ryder, A.; Frohardt, R. p.; Fusari, S. A.;
Jakubowski, Z. L.; Bartz, Q. R. J. Am. Chem. Soc. 1958, 80, 743. (b)
Haskell, T. H. J. Am. Chem. Soc. 1958, 80, 747.
(8) Haneda, K.; Shinose, M.; Seino, A.; Tabata, N.; Tomoda, H.;
Iwai, Y.; Omura, S. J. Antibiot. 1994, 47, 774.
(9) Leviev, I. G.; Rodriguez-Fonseca, C.; Phan, H.; Garrett, R. A.;
Heilek, G.; Noller, H. F.; Mankin, A. S. EMBO J. 1994, 13, 1862.
(10) Barbacid, M.; Vazquez, D. Eur. J. Biochem. 1974, 44, 445.
Hansen, J. L.; Moore, P. B.; Steitz, T. A. J. Mol. Biol. 2003, 330, 1061.
(11) Haneda, K.; Shinose, M.; Seino, A.; Tabata, N.; Tomoda, H.; J,
L.; Moore, P. B.; Steitz, T. A. J. Mol. Biol. 2003, 330, 1061.
(12) Sugimura, H.; Watanabe, R. Chem. Lett. 2008, 37, 1038.
(13) Stevens, C. L.; Nemec, J.; Ransford, G. H. J. Am. Chem. Soc.
1979, 94, 3280.
(14) (a) Fischbach, M. A.; Walsh, C. T. Science 2009, 325, 1089. (b)
Koul, A.; Arnoult, E.; Lounis, N.; Guillemot, J.; Andries, K. Nature
2011, 469, 483.
r
10.1021/ol2018196
Published on Web 09/13/2011
2011 American Chemical Society

peptide synthesis. We anticipated that the steric differentia-
tion between 3 and 4, which bears a quaternary R-stereo-
center, would afford a significant difference in their
respective rates of amidation that could be exploited. In
concert with the fact that the carbonꢀhalogen bond in the
pyrimidine is weaker that that in the aryl halide we hypothe-
sized that 2 and 3 would undergo relatively fast coupling
followed by coupling to 4. Thus we attempted a one-pot
three-component tandem N-aryl amidation to unite these
fragments. This approach proved problematic (vide infra),
and it became apparent that the same copper-ligand set
would not be applicable for both couplings. We then sought
to study the coupling reactions individually to learn more
about the subtleties that govern these amidations, specifi-
cally with pharmacologically important scaffolds such as
4-halopyrimidines.
Figure 1. Aminohexopyranose nucleoside antibiotics.
Our initial synthetic efforts were guided by early biological
studies with amicetin and other related natural products.
Haskell and co-workers reported lower in vitro and in vivo
activity against M. tuberculosis (H37Rv) for plicacetin, the
compound lacking the R-methylserine moiety, compared to
amicetin and bamicetin.7 Degradation studies on amicetin by
Flynn and co-workers reported a significant loss of activity at
a measurable rate under basic conditions (pH = 8), but not
under acidic conditions (pH = 1.3) (Figure 2).5 Cytosamine,
the base hydrolysis product, was reported to be devoid of
antibacterial activity.7 This suggests that cytimidine, which
arises from the acid catalyzed hydrolysis, retains activity and
points to the importance to the cytosine-peptidic portion of
the molecule as being requisite for activity. To the best of our
knowledge, there have been no structureꢀactivity relation-
ship (SAR) studies on cytimidine, which appears to be a more
relevant analog to amicetin than cytosamine. Thus, our initial
efforts toward the total synthesis of amicetin were directed at
developing a highly efficient route to cytimidine. This would
allow us to evaluate a truncated version of amicetin and other
analogues for antimicrobial activity.
Scheme 1. Retrosynthetic Analysis
We began our synthesis with the formation of 4-halopyr-
imidines 2aꢀc (Scheme 2). These were prepared by
halide exchange of the commercially available 4-chloro-
2-methylthiopyrimidine 5a with TMSBr or TMSI to give
the corresponding halides 5b and 5c.15 The halopyrimidines
5aꢀc were then oxidized to the sulfone,16 followed by
nucleophilic aromatic substitution17 with the potassium
salt of allyl alcohol to give 2aꢀc, in good to excellent yields.
The allyl group was introduced to offer the flexibility to be
unmasked separately from R-methylserine, for glycosyla-
tion, or to be deprotected concurrently en route to cytimi-
dine and its analogues.
We then proceeded with the synthesis of the oxazolidine
carboxamide 4 which corresponds to the R-methylserine
moiety cytimidine (Scheme 3). Oxazolidine ester 7 was
formedfrom L-serine methylester hydrochlorideasa single
diastereomer based on published procedures.18 It was then
alkylated employing Seebach’s memory of chirality,19
giving the R-methylated oxazolidine ester 8 in excellent yield
and >95:5 dr. The ester was then saponified to the acid 9
and converted to the carboxamide 4 via the acid chloride.
Figure 2. Base and acid hydrolysis products of amicetin.
ꢀ
ꢀ
(15) Hockova, D.; Antonın, H.; Masojıdkova, K. D. T.; de Jersey, J.;
´ ´
We envisioned that cytimidine could be prepared quite
efficiently if we could exploit a regioselective coupling of this
halopyrimidine 2to 4-iodobenzamide 3, and the oxazolidine
carboxamide 4, via copper catalyzed N-aryl amidation
reactions (Scheme 1). This strategy would avoid a series of
protectionꢀdeprotection chemistries normally employed in
Guaddat, L. W. W. Bioorgan. Med. Chem. 2009, 17, 6218.
(16) Baudin, J. B.; Julia, S. A.; Lorne, R.; Ruel, O. Bull. Chim. Soc.
Fr. 1993, 130, 856.
(17) Falk-Pedersen, M. L.; Benneche, T.; Undheim, K. Acta Chim.
Scan. 1993, 47, 72.
(18) Brunner, M.; Saakrento, P.; Straub, T.; Rissanen, K.; Koshinen,
A. M. P. K. Eur. J. Org. Chem. 2004, 3879.
(19) Seebach, D.; Aebi, J. D. Tetrahedron Lett. 1984, 25, 2545.
Org. Lett., Vol. 13, No. 19, 2011
5001

Table 1. Optimization of the 1st N-Aryl Amidationa
Scheme 2. Synthesis of 4-Halopyrimidines
entry
X
ligand
dmeda
solvent
base
time yield
1
Cl
toluene
toluene
toluene
K3PO4
K3PO4
K3PO4
24 h
24 h
24 h
ꢀ
2
Br dmeda
ꢀ
3
I
dmeda
ꢀ
4
I
diamine 2
dioxane Cs2CO3 24 h
ꢀ
5
Cl
Br
I
ꢀ
dioxane K3PO4
dioxane K3PO4
dioxane K3PO4
24 h
24 h
24 h
18 h
18 h
24 h
24 h
24 h
24 h
24 h
18 h
ꢀ
6
ꢀ
<1%
<1%
ꢀ
7
ꢀ
Scheme 3. Synthesis of Oxazolidine Carboxamide
8
Cl
1,10-phen
toluene
toluene
toluene
toluene
toluene
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
9
Br 1,10-phen
69%
71%
37%
14%
14%
12%
72%
10
11
12
13
14
15
I
I
I
I
I
I
1,10-phen
2,20-bipy
8-hq
2,4,6-collidine toluene
β-ketoester
1,10-phen
toluene
dioxane K3PO4
a Performed with 10 mol % CuI, 20 mol % ligand, 1.0 equiv of 2aꢀc,
1.0 of equiv 3, and 2.0 equiv of base; dmeda = N,N0-Dimethylethyle-
nediamine; diamine 2 = trans-N,N-dimethylcyclohexyldiamine; 1,
10-phen = 1,10-phenanthroline; β-ketoester = Ethyl 2-cyclohexanone
carboxylate; 8-hq = 8-hydroxyquinoline; 2,20-bipy = 2,20-bipyridine.
Reaction conditions for the copper catalyzed N-aryl
amidation of 2aꢀc with 4-iodobenzamide 3 were then
investigated (Table 1). Surprisingly, there was little pre-
cedence for the Cu-mediated amidation of halopy-
rimidines.20,21 Buchwald has shown a single example of
the N-amidation of cyclohexanecarboxamide with 5-bro-
mopyrimidine, utilizing the trans-N,N0-dimethylcyclohex-
ane-1,2-diamine ligand. Legraverend has reported the
synthesis of purines from the one-pot amidationꢀ
condensation of 5-amino-4-iodopyrimidines with benza-
mides using the same ligand set. Given this precedence, we
first applied these conditions to our first coupling reaction
(Table 1, entries 1ꢀ4). However, instead of giving the
desired pyrimidylbenzamide 10, NMR analysis showed
that these halopyrimidines had reacted with the diamine
ligands to give the aminopyrimidines22 via the copper-
mediated N-aryl amination with the diamine ligand, as
similarly described by Fukuyama with the amination of
2-iodopyridine.23 We also tried the ligandless Goldberg
protocol24 (entries 5ꢀ7) which gave only a trace of the
product as detected by LCMS. Other ligands typically used
in copper-mediated chemistry were also screened (entries
8ꢀ14).25ꢀ29 The most promising ligand for the reaction
was found to be 1,10-phenanthroline giving pyrimidylben-
zamide10in goodyields. 4-Chloropyrimidine 2a, however,
was unreactive in any of the reaction conditions (entry 8),
while 4-bromo- and 4-iodopyrimidines 2b and 2c exhibited
similar yields (entries 9 and 10). This suggests that the
coupling process does not occur via a nucleophilic aro-
matic substitution mechanism, wherein chlorides have
been reported to react faster than bromides and iodides.30
This rather supports the oxidative addition of the aryl
halide to the copper complex as proposed earlier.31
We then proceeded with the coupling of pyrimidylben-
zamide 10 and oxazolidine carboxamide 4, for which 1,
10-phenanthroline was found to be a less competent ligand
(Table 2, entry 1). The steric demands of the amide
oxazolidine 4 in the copper complex may not be well suited
for coupling when a very rigid Cu-phen complex31 is
involved, as compared to a more flexible Cu-diamine
(20) Klapars, A.; Huang, X.; Buchwald, S. L. J. Am. Chem. Soc.
2002, 124, 7421.
(21) Ibrahim, N.; Legraverend, M. J. Org. Chem. 2009, 74, 463.
(22) Aminopyrimidines byproduct:
(26) Lai, C.; Gum, R. J.; Daly, M.; Fry, E. H.; Hutchins, C.; Abad-
Zapatero, C.; von Geldern, T. W. Biorg. Med. Chem. Lett. 2006, 16,
1807.
(27) Guzzo, P. (AMR Technology, Inc. USA). U.S. Patent
20090082359, 2009; SciFinder Scholar AN 2009:116350.
(28) Maruyama, T.; Fujiwara, K.; Fukuhara, M. J. Chem. Soc.,
Perkin Trans. 1 1995, 733.
(23) Okano, K.; Tokuyama, H.; Fukuyama, T. Org. Lett. 2003, 5,
4987.
(24) Goldberg, I. Ber. Dtsch. Chem. Ges. 1906, 39, 1691.
(25) Lin, J.; Elangovan, A.; Ho, T. J. Org. Chem. 2005, 70, 7397.
(29) Lv, X.; Bao, W. J. Org. Chem. 2007, 72, 3863.
(30) Ritchie, C. D.; Sawada, M. J. Am. Chem. Soc. 1977, 99, 3754.
(31) Tye., J. W.; Johns, A. M.; Incarvito, C. D.; Hartwig, J. F. J. Am.
Chem. Soc. 2008, 130, 9971.
5002
Org. Lett., Vol. 13, No. 19, 2011

complex, which, in this case, has shown to be highly
effective in mediating the amidation reaction.
Scheme 5. Synthesis of Cytimidine Analoguesa
Table 2. Optimization of the 2nd N-Aryl Amidationa
entry
ligand
solvent
base
temp
time yield
1
2
1,10-phen dioxane K3PO4 120 °C
dmeda dioxane K3PO4 100 °C
48 h
60 h
20%
93%
a Performed with 10 mol % CuI, 20 mol % ligand, 1.0 equiv 10, 1.0
equiv 4, and 2.0 equiv base.
The differences in reactivities of the Cu(phen)2 or Cu-
(dmeda)2 ligand systems with respect to the first and
second N-aryl amidations did not favor our initial goal
of presenting a three-component tandem route toward
cytimidine as described earlier (Scheme 1). The competing
amination of 4-halopyrimidines 2aꢀc with the diamine
ligands in the first amidation and the incompatibility of the
Cu(phen)2 in the second amidation reaction essentially
lead us to consider an alternative strategy.
We then investigated the feasibility of a one-pot sequen-
tial amidation (Scheme 4). The sequential N-aryl amida-
tion of bromopyrimidine 2b with 4-iodobenzamide 3 and
oxazolidine carboxamide 4 gave an overall 53% yield.
Masked cytimidine 11 was then subjected to the final
deprotection sequences32 to give cytimidine in 34% overall
yield in five steps from the commercially avaiable 4 chloro-
2-methylthiopyrimidine.
a Overall yields for the two-pot sequence with intermediate purifica-
tion are given in parentheses.
these unoptimized couplings are slightly lower than those of
our optimized coupling process for cytimidine but still
represent ∼60% for each step. We have observed that under
these prolonged reaction times each substate’s performance
is dependent on competetive decomposition. In these reac-
tions we were also unable to detect any of the homocoupled
benzamide or the Ullmann33 product by LCMS. We are
currently trying to identify reaction conditions that obviate
these issues. Nevertheless, this suggests that both bromo-
and iodobenzamides can be chemoselectively cross-coupled
with heterocyclic systems bearing a halogen adjacent to the
heteroatom without competetive homocoupling.
In summary, we have developed an efficient route to
cytimidine via a one-pot sequential Cu-mediated N-ami-
dation cascade. This sequence exploits a uniquely regiose-
lective coupling of a halopyrimidine and an iodo-
benzamide. Further investigation is underway to tailor
the amidation sequences to generate medicinally relevant
analogues of this natural product fragment.
Scheme 4. One-Pot Cu-Mediated N-Aryl Amidation and Com-
pletion of the Synthesis of Cytimidine
Acknowledgment. We are grateful for financial support
from the University of Utah Research Foundation and the
Department of Chemistry, University of Utah.
Supporting Information Available. Experimental de-
tails and 1H, 13C, and 13C DEPT NMR spectra for all new
compounds. This material is free of charge via the Inter-
net at http://pubs.acs.org.
(32) Doi, T.; Kinbara, A.; Inoue, H.; Takahasi, T. Chem.;Eur. J.
2007, 188.
(33) Ullmann, F. Ber. Dtsch. Chem. Ges. 1901, 34, 2174.
(34) Cossu, S.; Conti, S.; Giacomelli, G.; Falorni, M. Synthesis 1994,
12, 1429.
The method was then applied to a number of masked
cytimidine analogues (Scheme 5). The yields obtained for
Org. Lett., Vol. 13, No. 19, 2011
5003