Carbon radical attack on a pyrimidine nitrogen. An unusual entry into polycyclic aminopyrimidones

Article abstract of DOI:10.1021/ol501104h

Unprecedented examples of a synthetically useful radical ring closure onto a pyrimidine nitrogen leading to novel, highly functionalized bi- and tricyclic pyrimidone structures are reported. An unexpected hydrogen atom translocation prior to the cyclization step was observed in some cases.

Full text of DOI:10.1021/ol501104h

Letter
pubs.acs.org/OrgLett
Carbon Radical Attack on a Pyrimidine Nitrogen. An Unusual Entry
into Polycyclic Aminopyrimidones
Ling Qin, Zhibo Liu, and Samir Z. Zard*
̀
Laboratoire de Synthese Organique, UMR 7652 CNRS/Ecole Polytechnique, 91128 Palaiseau, Cedex, France
S
* Supporting Information
ABSTRACT: Unprecedented examples of a synthetically
useful radical ring closure onto a pyrimidine nitrogen leading
to novel, highly functionalized bi- and tricyclic pyrimidone
structures are reported. An unexpected hydrogen atom
translocation prior to the cyclization step was observed in
some cases.
e recently observed the dramatic effect a protecting
whereas carbamate analogue 4 (R = Boc or COOMe)
undergoes attack on the pyrimidine nitrogen leading to bicyclic
pyrimidone 6 as the major product, under otherwise identical
experimental conditions. In the case of symmetrical precursor
7, the bicyclic pyrimidone 8 is obtained, despite the nature of
the protecting group on the extranuclear nitrogen. The
symmetrical intermediate carbon radical 9 is able to attack
either of the pyrimidine nitrogens to finally give the same
pyrimidone 8.
W
group exhibited on the regioselectivity and outcome of a
radical cyclization on a pyridine ring.1 We found that the radical
derived from xanthate 1, with an acetyl or a mesyl group on the
extranuclear nitrogen (R = Ac or SO2Me), cyclizes in the usual
manner to give ultimately the expected azaindoline product 2
(Scheme 1). In striking contrast, and for reasons still not clear,2
Scheme 1. Unexpected Cyclizations
Not surprisingly, the process in this last instance is
significantly more efficient in terms of yield. This tempted us
to extend the cyclization to the more difficult case of a six-
membered ring formation, by starting with the homologous
precursor 10. This would allow us to attain the more elusive
bicyclic pyrimidone 11 through cyclization of radical 12
(Scheme 1).
The synthesis of the requisite precursors 16a,b, outlined in
Scheme 2, is trivial. It simply involves the protection of
aminopyrimidine 13, itself derived by the reaction of 2,4,6-
trichloropyrimidine with 3-butenylamine, by acetylation (14a)
or by tert-butoxycarbonylation (14b). Lauroyl peroxide (DLP)
mediated radical addition of S-cyanomethyl xanthate 15 affords
the respective adducts 16a and 16b in high yield.4
Disappointingly, no cyclized product 17a or 17b was formed
upon exposure of either precursor to stoichiometric amounts of
DLP in refluxing EtOAc. A rather complex mixture of products
was formed in each case, in which the prematurely reduced
uncyclized substances could be identified as the main
components by NMR spectroscopy.
In the hope of diminishing the degrees of freedom in the side
chain and perhaps encourage cyclization on one of the two
pyrimidine nitrogens, we examined the behavior of adduct 19
derived by addition of the same xanthate 15 to alkene 18
(Scheme 2). The latter was available to us as a side product in a
previous study, in which we had found that cyclization on the
the mere replacement of the nitrogen protection by a Boc
group or any other carbamate (R = Boc or COOMe) results in
the formation of a totally unexpected pyridone product 3
through a hitherto unprecedented radical ring closure on the
nitrogen atom of the pyridine nucleus. The presence of the
chlorine (or a fluorine) atom adjacent to the nitrogen is
necessary; otherwise, the reaction is not clean.
We have been able to extend this chemistry to the pyrimidine
series, where two possibilities arise, depending on the position
of the nitrogen substituent.3 In the case of derivatives 4, the
acetyl-protected substrate 4 (R = Ac) furnishes diazaindoline 5,
Received: April 16, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ol501104h | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
The equilibrium almost certainly favors the latter in a nonpolar
medium with a low dielectric constant. Hydrolysis of 26 upon
workup furnishes the desired pyrimidone 21. Capture of
cationic species 25 and 27 by adventitious water also gives
pyrimidone 21.
Scheme 2. Synthesis of a Bicyclic Pyrimidone
Removal of the HCl from the medium would limit the
formation of cationic species 27 from either precursor 25 or 26
and consequently hinder the undesired generation of chloride
22. Since both the desired pyrimidone 21 and chloride 22 arise
from the same precursors, suppression of the latter would
increase the yield of the former to synthetically respectable
levels.
We therefore repeated the cyclization in the presence of
calcium carbonate as a harmless solid scavenger of HCl
Unfortunately, the yield of pyrimidone 21 improved only
marginally to 34%. In contrast, the addition of 2,6-lutidine, a
sterically hindered base we knew from past experience to be
generally compatible with the radical chemistry of xanthates,
had an almost dramatic effect: the yield nearly doubled to 53%
and became almost comparable to the yields we had observed
in the formation of five-membered rings. Clearly, the soluble,
hindered lutidine is a faster scavenger for HCl and thus
efficiently blocks the formation of unwanted chloride 22.
With a potential route to an interesting family of hetero-
cycles, we next briefly explored its scope. Using the same
alkene, we first varied the xanthate component. The results are
presented in Scheme 4. In addition to the nitrile present in
pyrimidone 21, an ester (29), a ketone (31), a trifluoromethyl
group (33), and a phthalimido-protected amine (35) could also
be introduced. While the yields for the intermolecular addition
were generally high, those of the cyclization step were fair to
modest, reflecting the inherently difficult ring-closure process.
In the case of the trifluoromethyl substituted product 33,6 the
slow cyclization is further aggravated by a polarity mismatch
between the intermediate carbon radical, the SOMO of which
is lowered by the presence of the strongly electron-withdrawing
CF3-group, and the electrophilic dichloropyrimidine ring.7 We
found that lowering the concentration in this case from 0.05 to
0.01 M increased the yield to 50%.
carbon of pyridines and pyrimidines was possible even with a
naked extranuclear nitrogen.5
We therefore subjected compound 19 to the action of
stoichiometric amounts of DLP in refluxing ethyl acetate and
isolated three main products from the reaction mixture
(Scheme 2). The minor component (7%) was the reduced,
uncyclized compound 20, but the next main product turned out
to be the desired bicyclic pyrimidone 21 (26%). More
surprisingly, the third product proved to be the totally
unexpected chloride 22 (30%).
This last compound presumably arises from the ring opening
of intermediate 27 by the chloride anion, as depicted in Scheme
3. This reactive intermediate would result from the action of
hydrogen chloride on ion-pair 25, which has a laurate
counteranion. This ion pair is produced upon oxidation of
the cyclized radical 24 by electron transfer to the peroxide and
combines rapidly but also reversibly into neutral structure 26.
The piperidine could be replaced without loss of efficiency by
a simple cyclohexyl (38) or a cyclopentyl ring (41 and 43), or
even by a geminal diethyl motif (46). However, when substrate
48, derived from alkene 47 and possessing a cycloheptane ring,
was examined, we were surprised to find that the reaction gave
two products in high combined yield (Scheme 5). The major
product was the expected pyrimidone 50, formed in a typical
yield of 62%, whereas the minor product turned out to be
bridged tricyclic compound 52, isolated in 19% yield.
Scheme 3. An Unexpected Chloride Formation
Clearly, because of the greater flexibility of the seven-
membered ring and the sluggish cyclization step, intermediate
radical 49 is able to abstract a hydrogen to give isomeric radical
51, which can also ring-close on the nitrogen of the pyrimidine
to give rearranged product 52. While the translocation step is in
principle reversible, in view of the similar stabilities of radicals
49 and 51, it is not clear if an equilibrium is indeed established
under the reaction conditions.8
Interestingly, we found that the addition of S-trifluoromethyl-
O-octadecyl xanthate to alkene 47 furnished, in addition to the
expected adduct 53, a small amount of the translocated isomer
54. Further exposure to lauroyl peroxide gave rise to the linear
and bridged tricyclic derivatives 55 (27%) and 56 (26%),
respectively (Scheme 5). Because of the strongly electron-
withdrawing nature of the CF3-group, the hydrogen abstraction
B
dx.doi.org/10.1021/ol501104h | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 4. Synthesis of Bicyclic Pyrimidones
Scheme 5. Unexpected Hydrogen Atom Transfer and
Cyclization
deoxysugars and to accomplish deiodination of iodolactones
using cyclohexane as the hydrogen atom source.9
In the case of xanthate 58, derived from isobutyl-substituted
alkene 57, the reaction furnished rearranged bicyclic
pyrimidone 62, accompanied by some reduced, uncyclized
material 60 and, surprisingly, by the radical Smiles rearranged
product 65 (Scheme 6). No significant “normal” cyclized
product was observed. Presumably, with only one substituent
next to the extranuclear nitrogen atom, the Thorpe−Ingold
effect is not as effective in promoting the ring closure of
intermediate radical 59, resulting in a competition between
hydrogen abstraction from the solvent and intramolecular
abstraction of the tertiary hydrogen in the isobutyl side chain.
The latter pathway culminates in the formation of rearranged
bicyclic pyrimidone 62. A third competing pathway involves
formation of spiro radical 63 followed by fragmentation to give
aminyl radical 64, which abstracts a hydrogen from the solvent
and then reacts ionically with lauroyl peroxide to give the final
rearranged lauramide 65. While well-documented, radical
Smiles rearrangements leading to the direct formation of
nitrogen centered radicals are relatively uncommon.10
The present study reports several examples of the hitherto
exceedingly rare radical attacks on a nitrogen atom leading to
the formation of six-membered rings, and the first involving a
heteroaromatic nitrogen. Much remains to be done to better
delineate the scope of the process, in particular as regards the
effect of substituents and possible extension to other nitrogen
heteroaromatic systems. Nevertheless, the compounds pres-
ently accessible by this expedient approach would be very
difficult to obtain by other routes. Indeed, the bicyclic
aminopyrimidone core skeleton has rarely been described in
becomes significantly faster, allowing the observation of
uncyclized isomeric xanthate 54. We have exploited in the
past these accelerating polar effects to cheaply prepare various
C
dx.doi.org/10.1021/ol501104h | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(d) Zard, S. Z. Aust. J. Chem. 2006, 59, 663. (e) Quiclet-Sire, B.; Zard,
S. Z. Pure Appl. Chem. 2011, 83, 519. For a review on the use of
xanthates for the synthesis of heteroaromatics, see: (f) El Qacemi, M.;
Petit, L.; Quiclet-Sire, B.; Zard, S. Z. Org. Biomol. Chem. 2012, 10,
5707.
Scheme 6. Further Examples of Hydrogen Atom Transfers
and Cyclizations
(5) Liu, Z.; Qin, L.; Zard, S. Z. Org. Lett. 2012, 14, 5976.
(6) The use of an octadecyl group instead of the ubiquitous ethyl in
this case is for practical reasons related to the specific route used to
obtain the S-CF3-xanthate: Li, S.; Zard, S. Z. Org. Lett. 2013, 15, 5898.
(7) (a) Tedder, J. M. Angew. Chem., Int. Ed. Engl. 1982, 21, 401.
(b) Laird, E. L.; Jorgensen, W. L. J. Org. Chem. 1990, 55, 9.
(8) Robertson, J.; Pillai, J.; Lush, R. K. Chem. Soc. Rev. 2001, 30, 94.
(9) (a) Quiclet-Sire, B.; Zard, S. Z. J. Am. Chem. Soc. 1996, 118, 9190.
(b) Boivin, J.; Quiclet-Sire, B.; Ramos, L.; Zard, S. Z. Chem. Commun.
1997, 353. (c) Trost, B. M.; Fandrick, D. R.; Brodmann, T.; Stiles, D.
T. Angew. Chem., Int. Ed. 2007, 46, 6123.
(10) (a) Studer, A. In Radicals in Organic Synthesis; Renaud, P., Sibi,
M. P., Eds.; Wiley-VCH: Weinheim, 2001; Vol. 2, pp 44−60.
́
(b) Studer, A.; Bossart, M. Tetrahedon 2001, 57, 9649. (c) Bacque, E.;
El Qacemi, M.; Zard, S. Z. Org. Lett. 2005, 7, 3817.
(11) (a) Angier, R. B.; Curran, W. V. J. Org. Chem. 1961, 26, 1891.
(b) Doerr, I. L.; Cushley, R. J.; Fox, J. J. J. Org. Chem. 1968, 33, 1592.
(c) Sasaki, T.; Minamoto, K.; Sugiura, T. J. Org. Chem. 1975, 40, 3498.
(d) Robins, M. J.; Tadashi. Kanai, T. J. Org. Chem. 1976, 41, 1886.
(e) Minamoto, K.; Tanaka, T.; Azuma, K.; Suzuki, N.; Eguchi, S.;
Kadoya, S.; Hirota, T. J. Org. Chem. 1986, 51, 4417. (f) Minamoto, K.;
Azuma, K.; Fujiwara, N.; Eguchi, S.; Hirota, T.; Kadoya, S. J. Org.
Chem. 1989, 54, 4543. (g) Font, D.; Heras, M.; Villalgordo, J. M.;
Linden, A.; Villalgordo, J. M. Tetrahedron 2006, 62, 1433. (h) Paulini,
R.; Beyer, C.; Bastiaans, H. M. M.; Rankl, N. B. (BASF SE). WO2012/
136751 A1, 2012.
the literature, with most examples being derived from a
modification of uridine and thymidine.11 Recently, plant
protection chemists at BASF patented the parent structure
with a simple benzyl group on the extranuclear nitrogen as an
agent for pest control.11h The present synthesis is modular,
metal-free, and tolerant of various functional groups and, thus,
offers numerous opportunities for postmodification and for the
construction of diverse libraries for biological testing.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, full spectroscopic data, and copies of
1H and 13C NMR spectra for all new compounds. This material
is available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: samir.zard@polytechnique.edu.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank Ecole Polytechnique for scholarships for two of us
(L.Q. and Z.L.). This paper is dedicated with respect to the
■
́
memory of Prof. Andre Dreiding (ETH, Switzerland).
REFERENCES
■
(1) El Qacemi, M.; Ricard, L.; Zard, S. Z. Chem. Commun. 2006, 42,
4422.
(2) We acknowledge an ongoing collaboration on this question with
Prof. Michelle Coote.
(3) Laot, Y.; Petit, L.; Zard, S. Z. Chem. Commun. 2010, 46, 5784.
(4) For reviews on the xanthate radical addition−transfer process,
see: (a) Zard, S. Z. Angew. Chem., Int. Ed. Engl. 1997, 36, 672.
(b) Quiclet-Sire, B.; Zard, S. Z. Chem.Eur. J. 2006, 12, 6002.
(c) Quiclet-Sire, B.; Zard, S. Z. Top. Curr. Chem. 2006, 264, 201.
D
dx.doi.org/10.1021/ol501104h | Org. Lett. XXXX, XXX, XXX−XXX