Alternative synthetic route to annulated diaminopyrimidines

Article abstract of DOI:10.1016/j.tetlet.2014.05.057

We report an alternative synthetic approach to annulated diaminopyrimidines based on a one-pot process, in which a Thorpe-Ziegler cyclization of a dinitrile is followed by in situ reaction with a cyanamide under highly concentrated and basic conditions. T

Full text of DOI:10.1016/j.tetlet.2014.05.057

Tetrahedron Letters xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Alternative synthetic route to annulated diaminopyrimidines
⇑
Alexandre Erdmann, Ludovic Halby, Frédéric Cantagrel, François Sautel, Georges Massiot ,
Paola B. Arimondo
⇑
CNRS-Pierre Fabre USR No. 3388 ETaC, CRDPF, 3 Avenue Hubert Curien, 31035 Toulouse Cedex 01, France
a r t i c l e i n f o
a b s t r a c t
Article history:
We report an alternative synthetic approach to annulated diaminopyrimidines based on a one-pot pro-
cess, in which a Thorpe–Ziegler cyclization of a dinitrile is followed by in situ reaction with a cyanamide
under highly concentrated and basic conditions. The total reaction can be carried out as fast as 30 min.
The reaction mechanism is discussed. This one-pot synthetic route is universally applicable to a large
variety of cyanamides and dinitriles, even natural products, to afford complex annulated
diaminopyrimidines.
Received 13 February 2014
Revised 12 May 2014
Accepted 13 May 2014
Available online xxxx
Keywords:
Ó 2014 Elsevier Ltd. All rights reserved.
Annulated diaminopyrimidines
One-pot reaction
Cyanamides
Quinazolones
Annulated 2,4-diaminopyrimidine is a motif present in many
synthetic molecules1–5 and in a few natural products, mainly in
the 7-deazaguanosine series, including, for example, archeosine6
and queuine.7 Diaminopyrimidines are prepared by multiple step
synthesis8,9 or, when possible, in a single step from a ketone and
a dicyandiamide.10–12 We wished to develop an alternative route
to access rapidly and widely to fully substituted and annulated dia-
minopyrimidines, starting from dinitriles and cyanamide in basic
and concentrated solution. Our sequence is based on the well-
known Thorpe–Ziegler reaction,13 which offers an easy access to
enaminonitriles, recently discovered to react with non enolizable
nitriles to offer the title compounds.14 In addition, it was reported
that cyanamides behave as nitriles and originate cycloaddition
products with diynes under nickel or cobalt catalysis.15,16 Further-
more, cyanamides, even complex ones, are readily available from
the corresponding amines. Based on these findings, we have
replaced nitriles with cyanamides to react with the enaminonitr-
iles from the Thorpe–Ziegler reaction in order to form in a single
step highly functionalized annulated diaminopyrimidines.
Scheme 1. One step synthesis of annulated diaminopyrimidines 4. A is a hetero
atom or a carbon atom, n and m correspond to the number of carbon atoms (1 or 2),
when A is a cycle m = 0. tBuOK is potassium tert-butoxide.
annulated diaminopyrimidines—are obtained without the need to
isolate and purify intermediate 2 and are recovered in high yield
with few side products. The reaction is rapid, as fast as 30 min.
The reaction reported in Scheme 1 was optimized. It tolerated a
large excess of dinitrile 1 due to rapid formation of intermediate 2
that preferentially reacted with cyanamide 3 rather than with the
starting dinitrile 1. Thus, if the cyanamide is the most valuable
compound in the sequence, it is best to use a large excess of dinit-
rile to bring the reaction to completion. Table 1 summarizes the
results obtained by varying reaction time, base, and solvent. The
optimization was carried on para-methoxybenzyl (PMB) protected
dicyanoethylamine 1a and morpholine-4-carbonitrile 3a as model
substrates of 1 and 3, respectively.
The effect of the reagents’ concentration was investigated. Dilu-
tion of the reaction gave lower yields of 6-(4-methoxybenzyl)-2-
morpholino-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidin-4-amine 4a
(Table 1, compare entry I to entry III), therefore concentrated con-
ditions must be chosen. A range of reaction times, from 30 min to
4 h, was tested. The reaction yield did not significantly change with
The one-pot reaction is depicted in Scheme 1. The cyclization of
dinitrile 1 to afford b-cyanoenamine 2 as the intermediate is com-
bined in situ, without isolation, with cyanamide 3 to give the
desired product 4. This method ensures access to a large panel of
annulated diaminopyrimidine derivatives upon use of different
dinitriles and cyanamides. In addition, the desired structures—
⇑
Corresponding authors. Tel.: +33 534506517; fax: +33 534503517 (G.M.); tel.:
+33 534506492; fax: +33 534503492 (P.B.A.).
E-mail addresses: georges.massiot.externe@pierre-fabre.com (G. Massiot), paola.
arimondo@etac.cnrs.fr (P.B. Arimondo).
http://dx.doi.org/10.1016/j.tetlet.2014.05.057
0040-4039/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Erdmann, A.; et al. Tetrahedron Lett. (2014), http://dx.doi.org/10.1016/j.tetlet.2014.05.057

2
A. Erdmann et al. / Tetrahedron Letters xxx (2014) xxx–xxx
conditions to yield 49% of the unprotected desired product N2,N2-
dimethyl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidine-2,4-diamine
4g, thus opening the way to further derivatives.
Table 1
Optimization of the reaction conditions on model substrates 1a and 3a (at 1.5 equiv
and 1 equiv, respectively)
O
All results are in full agreement with the previous results pub-
lished for the monoaminopyrimidine motif.14
N
Next we investigated the reaction mechanism. It involves two
steps: an intramolecular cyclization, known as the Thorpe–Ziegler
reaction, followed by the condensation of the intermediate enam-
ine on the cyanamide to obtain the annulated diaminopyrimidines
as depicted in Scheme 2. When intermediate 2a was placed under
the same reaction conditions with 3a, the annulated diaminopyr-
imidine 4a was obtained in lower yield than in the one-pot process.
This second step also requires a base since simple mixing and heat-
ing of 2a and 3a in the absence of the base lead to the recovery of
the starting materials.
N
N
N
NC
CN
O
N
NH2
base
+
130 °C
N
CN
OMe
OMe
4a
3a
1a
Entry
Base
Time (h)
Solvent
Yield (%)
In an effort of exemplification, the above one-pot reaction was
applied to the synthesis of various complex products by using dif-
ferent dinitriles 1 that were reacted with 3a to provide a large
range of diaminopyrimidine derivatives (Table 2). Due to the pres-
ence of side-products arising from self-condensation of 1b and 1c,
as described in the literature,17 compounds 4b and 4c were
obtained in moderate yield (45% and 50%, respectively). Surpris-
ingly, compound 1d failed to undergo the reaction presumably
because seven atom ring formation is penalized under these condi-
tions. In the case of compound 1e, product 4e was not isolated
because a retro-Michael reaction occurred thus forming the acrylo-
nitrile that polymerized. In contrast, an interesting scaffold was
isolated starting from phenylenediacetonitrile 1f in high yield.
I
II
tBuOK
tBuOK
tBuOK
tBuOK
tBuOK
tBuOK
NaH
tBuOK
NaH
tBuOK
NaH
2
4
2
1
0.5
0.5
2
2
2
Xylene (0.054 M)
Xylene (1.8 M)
Xylene (1.8 M)
Xylene (1.8 M)
Xylene (1.8 M)
Xylene (1.8 M)
Xylene (1.8 M)
Pentanol (1.8 M)
Pentanol (1.8 M)
DMF (1.8 M)
53
50
81
78
81
90a
82
0
III
IV
V
VI
VII
VIII
IX
X
0
72
70
2
2
XI
DMF (1.8 M)
The yield was calculated compared to the starting cyanamide.
a
This yield was obtained with 4 equiv of 1a.
The presence of strongly acidic
a-protons can explain the high
time between 2 h, 1 h, and 30 min (Table 1, entries II, III, IV, and V).
Since reaction for 4 at high temperature led to the degradation of
the reaction products, an average time of 2 h was chosen for the
subsequent reactions.
Concerning the choice of the base, NaH and tBuOK gave similar
yields when xylene or DMF was used as the solvent (Table 1, com-
pare entries III to VI and X to XI). Noteworthy, the reaction toler-
ates DMF, thus allowing to run the method on compounds that
are not soluble in xylene. Interestingly, when 1-pentanol was used
as the solvent, the reaction stopped at intermediate 2 and no dia-
minopyrimidine could be detected whatever base was used
(entries VIII and IX). Probably the pentanoate is not basic enough
for the reaction to proceed further.
Under optimized conditions, a 90% yield of the annulated diami-
nopyrimidine 4a was obtained upon use of a large excess (4 equiv)
of starting material, PMB-protected 3,30-azanediyldipropanenitrile
1a (2.1 mmol), in 0.3 mL xylene at 130 °C for 30 min (entry VI),
compared to 81% of yield for entry V of Table 1. Last, unprotected
biscyanoethylamine 1g was also engaged under these reaction
yield of the reaction. Under microwave conditions, the identical
reaction afforded compound 4f in 24% yield in only 2 min of reac-
tion time.
To further exemplify the reaction, we applied it to different
cyanamides. Table 3 reports the results of the one-step syn-
thetic pathway starting from 1a in the presence of various cya-
namides 3. The best yield was obtained with morpholine-4-
carbonitrile 3a to produce compound 4a in 81% yield. Under
the above optimized conditions, the reaction on piperidine-1-
carbonitrile 3h provided compound 4h in moderate yield
(40%) together with side products identified as due to the olig-
omerization of 3h.18 The isolation of compounds 4i and 4j
showed that various cyclic and non-cyclic cyanamides can react
in the one-pot reaction.
Finally, we applied the reaction to a natural product, cytisine, a
tricyclic alkaloid described as the reference inhibitor of the
a4b2
nicotinic receptor subtype,19 recently reviewed.20 Cyanocytisine
3k, obtained by reaction of cyanogen bromide on cytisine, was
engaged in our previously defined conditions and the desired prod-
Scheme 2. Reaction mechanism between PMB-protected dinitrile and a cyanamide.
Please cite this article in press as: Erdmann, A.; et al. Tetrahedron Lett. (2014), http://dx.doi.org/10.1016/j.tetlet.2014.05.057

A. Erdmann et al. / Tetrahedron Letters xxx (2014) xxx–xxx
3
Table 2
Table 3
Yields of the reaction between morpholine-4-carbonitrile 3a and various dinitrile
Yield of the reaction of compound 1a with various cyanamides 3
derivatives
NR2R1
N
O
N
NC
CN
tBuOK
O
N
NH2
N
tBuOK
R1R2N CN
+
NC
A
+
xylene
130 °C
2h
CN
N
N
N
xylene
130 °C
2h
n
m
N
CN
NH2
OMe
n
A
m
OMe
3a
1
4
1a
3
4
Compd 1
Product 4
Yield (%)
45
Compd 1
Product 4
Yield (%)
81
CN
O
N
N
b
NC
a
N
N
N
N
N
N
N
N
O
N
N
N
N
N
O
N
CN
H2N
H2N
H2N
H2N
H2N
H2N
MeO
MeO
MeO
MeO
MeO
c
NC
CN
N
h
N
O
N
N
50
0
N
N
CN
H2N
40
65
71
CN
d
NC
N
N
O
O
i
N
N
N
N
H2N
H2N
CN
O
e
f
NC
CN
O
N
N
N
0
N
CN
j
N
N
CN
CN
N
N
O
96
49
N
H2N
O
N
N
O
k
NC
CN
N
N
g
N
H
HN
N
N
O
40
N
H2N
N
CN
A is a hetero atom or a carbon atom, n and m correspond to the number of carbons
(1 or 2), when A is a cycle m = 0.
under the same experimental conditions reacted with morpholine-
4-carbonitrile 3a to afford quinazolone 8. Therefore our pathway
can also be applied to synthetize quinazoline derivatives. In the lit-
uct 4k was isolated in a moderate but unoptimized 40% yield
(Table 3). This finding indicates that our one-step synthetic route
can be applied to a wide range of natural products containing
amino groups, to rapidly yield complex molecules and introduce
the diaminopyrimidine scaffold in pharmacologically promising
structures.
To enlarge the scope of the reaction, we replaced one nitrile of
the starting material 1 by an ethyl ester function (Scheme 3).
Under our experimental conditions, the desired aminopyrimidone
6 was not isolated (Scheme 3). We then chose to use an aromatic
ester analog of intermediate 2, the aminobenzoate derivative 7 that
erature, quinazolones are mainly obtained under acidic catalysis21
,
in quite restrictive conditions or upon sodium methoxide cataly-
sis.22 Here compound 8 was obtained in xylene in a yield of 36%
under simple basic conditions. Altogether these results confirm
the versatility of our synthetic route.
In conclusion, we set up and optimized a one-pot reaction to
obtain annulated diaminopyrimidine scaffolds. High concentration
conditions, elevated temperature (130 °C), and strong bases need
to be used. Interestingly, by using different solvents, this reaction
Scheme 3. Quinazolone 6 was obtained from aminobenzoate 5 (0.54 mmol) and morpholine-4-carbonitrile 3a in the presence of sodium hydride NaH at 130 °C.
Please cite this article in press as: Erdmann, A.; et al. Tetrahedron Lett. (2014), http://dx.doi.org/10.1016/j.tetlet.2014.05.057

4
A. Erdmann et al. / Tetrahedron Letters xxx (2014) xxx–xxx
Scheme 4. Comparison between Modest’s process10 and the synthetic route here described.
can be applied to a panel of dinitriles and cyanamides providing a
wide diversity of diaminopyrimidine derivatives. Finally, it can also
be applied to obtain quinazolone derivatives.
Reference and notes
1. Bergeron, P.; Cohen, F.; Estrada, A.; Koehler, M. F. T.; Lau, K. H. L.; Ly, C.;
Lyssikatos, J. P.; Ortwine, D. F.; Pei, Z.; Zhao, X. In Pyrimidine compounds,
compositions and methods of use, Patent WO2010014939A1, 2010.
2. Asano, S.; Komiya, M.; Koike, N.; Koga, E.; Nakatani, S.; Isobe, Y. Bioorg. Med.
Chem. Lett. 2010, 20, 6696.
3. Capelli, A.; Marchionni, C.; Micheli, F.; Pasquarello, A.; Perini, B.; St-Denis, Y. In
Fused pyrimidines as antagonists of the corticotropinreleasing factor (CRF), Patent
WO02088095A1, 2002.
4. Hiranum, H.; Mizogami, S.; Mori, M.; Sekiya, T.; Hanatsuka, M.; Kanayama, T. In
5,6 Alkylenepyrimidine derivatives, processes for preparing the same and
pharmaceutical compositions, Patent EP0022481A1, 1981.
5. Sekiya, T.; Hata, S.; Yamada, S. Chem. Pharm. Bull. 1983, 31, 2432.
6. Gregson, J. M.; Crain, P. F.; Edmonds, C. G.; Gupta, R.; Hashizume, T.; Phillipson,
D. W.; McCloskey, J. A. J. Biol. Chem. 1993, 268, 10076.
7. Kasai, H.; Nakanishi, K.; Macfarlane, R. D.; Torgerson, D. F.; Ohashi, Z.;
McCloskey, J. A.; Gross, H. J.; Nishimura, S. J. Am. Chem. Soc. 1976, 98, 5044.
8. Cowart, M. D.; Altenbach, R. J.; Liu, H.; Hsieh, G. C.; Drizin, I.; Milicic, I.; Miller,
T. R.; Witte, D. G.; Wishart, N.; Fix-Stenzel, S. R.; McPherson, M. J.; Adair, R. M.;
Wetter, J. M.; Bettencourt, B. M.; Marsh, K. C.; Sullivan, J. P.; Honore, P.;
Esbenshade, T. A.; Brioni, J. D. J. Med. Chem. 2008, 51, 6547.
9. Hawryluk, N. A.; Merit, J. E.; Lebsack, A. D.; Branstetter, B. J.; Hack, M. D.;
Swanson, N.; Ao, H.; Maher, M. P.; Bhattacharya, A.; Wang, Q.; Freedman, J. M.;
Scott, B. P.; Wickenden, A. D.; Chaplan, S. R.; Breitenbucher, J. G. Bioorg. Med.
Chem. Lett. 2010, 20, 7137.
10. Modest, E. J.; Chatterjee, S.; Protopapa, H. K. J. Org. Chem. 1965, 30, 1837.
11. Rosowsky, A.; Papoulis, A. T.; Forsch, R. A.; Queener, S. F. J. Med. Chem. 1999, 42,
1007.
12. Sengupta, S. K.; Chatterjee, S.; Protopapa, H. K.; Modest, E. J. J. Org. Chem. 1972,
37, 1323.
A similar process was described by Modest et al.10 and a com-
parison of the two synthetic routes is shown in Scheme 4. From
a fundamental point of view, it is worth noting that most cyclic
ketones used in the Modest process are obtained by Dieckmann
or Thorpe–Ziegler condensations, which is a convergence point
between the two approaches. Whereas, Modest makes use of a
ketone coming from a Dieckmann or Thorpe–Ziegler condensation,
we directly transform the dinitrile through a cyclic intermediate
that is not isolated. The use of cyanoguanidine by Modest limits
the preparation to primary amines, while the use of cyanamide
confers more versatility to our approach. Time and reaction condi-
tions are similar, even if Modest’s synthetic pathway requires
higher temperatures (180–200 °C) compared to 130 °C in the pres-
ent work. The exemplification on cytisine constitutes a case where
it would have been difficult to apply the older routes and it repre-
sents a novel and promising opportunity to derive natural products
with annulated diaminopyrimidine scaffold. Thus, the reaction
optimized in this Letter can afford complex bicyclic structures,
with a larger chemical diversity.
In short, the alternative synthetic route presented here is more
convenient since it is performed in one-pot, it is fast (30 min) and it
can be applied universally to most of the cyanamide derivatives to
obtain complex structures.
13. Wang, Z. In Thorpe–Ziegler Cyclization; John Wiley & Sons, 2010; p 2766.
14. Chercheja, S.; Simpson, K. J.; Lam, H. W. Tetrahedron 2011, 67, 3839.
15. Bonaga, L. V. R.; Zhang, H. C.; Maryanoff, B. E. Chem. Commun. 2004, 2394.
16. Kumar, P.; Prescher, S.; Louie, J. Angew. Chem., Int. Ed. 2011, 50, 10694.
17. Thompson, Q. E. J. Am. Chem. Soc. 1958, 80, 5483.
Acknowledgments
18. Herrera, A.; Martinez-Alvarez, R.; Ramiro, P.; Chioua, M.; Chioua, R. Synthesis
2004, 4, 503.
This study was funded by the regional council of Midi-Pyrénées
(France)—Accueil Equipe d’Excellence, FEDER and ATIP CNRS INC
Jeunes Chercheurs.
19. Tasso, B.; Novelli, F.; Sparatore, F.; Fasoli, F.; Gotti, C. J. Nat. Prod. 2013, 76, 727.
20. Rouden, J.; Lasne, M. C.; Blanchet, J.; Baudoux, J. Chem. Rev. 2014, 114, 712.
21. Hua, F.; Xuhu, H.; Haijun, J.; Yufen, Z.; Renzhong, Q. Synthesis 2009, 16, 2679.
22. Li, H. Z.; He, H. Y.; Han, Y. Y.; Gu, X.; He, L.; Qi, Q. R.; Zhao, Y. L.; Yang, L.
Molecules 2010, 15, 9473.
Supplementary data
Supplementary data (experimental procedures and full spectro-
scopic data) associated with this article can be found, in the online
version, at http://dx.doi.org/10.1016/j.tetlet.2014.05.057.
Please cite this article in press as: Erdmann, A.; et al. Tetrahedron Lett. (2014), http://dx.doi.org/10.1016/j.tetlet.2014.05.057