A chemoselective aniline-chloropyrimidine coupling in a competing electrophilic environment

Article abstract of DOI:10.1016/j.tetlet.2007.11.009

A highly chemoselective substitution on 4-amino-6-chloropyrimidine ring system having a competing aldehyde functionality has been realized with anilines which produces 4, 6-diaminopyrimidine-5-carbaldehyde in high yield. The reaction may involve the inter

Full text of DOI:10.1016/j.tetlet.2007.11.009

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 102–105
A chemoselective aniline–chloropyrimidine coupling in a
competing electrophilic environment
Anusuya Choudhury,* Hongfeng Chen, Christopher N. Nilsen and Kirk L. Sorgi
Chemical Development, Johnson & Johnson Pharmaceutical Research & Development, L.L.C.,
1000 Route 202, Raritan, NJ 08869, USA
Received 21 September 2007; revised 30 October 2007; accepted 1 November 2007
Available online 6 November 2007
Abstract—A highly chemoselective substitution on 4-amino-6-chloropyrimidine ring system having a competing aldehyde function-
ality has been realized with anilines which produces 4, 6-diaminopyrimidine-5-carbaldehyde in high yield. The reaction may involve
the intermediacy of imines. A variety of aromatic amines participate in this reaction successfully to generate diaminopyrimidine
aldehydes in moderate to high yield.
Ó 2007 Elsevier Ltd. All rights reserved.
Ar
The chemistry of pyrimidines is one of the most widely
studied subjects because of the occurrence of such
NH2
path "A"
path "B"
Cl
H
heterocycles in biologically relevant systems such as
nucleic acids, cofactors, various toxins, and other
products.1,2 Substituted pyrimidines, particularly with
2,4-diamino or 2,4,6-triamino substitution are
known pharmacophores in several structure-based drug
design approaches in medicinal chemistry.3 Their
inherent shape and hydrogen bonding capability makes
this heterocyclic motif suitable for interaction with many
biological targets. Therefore, extensive work has been
done in this area to gain control over the sequential
introduction of amino functionalities on the pyrimidine
templates by using the readily available corresponding
halopyrimidines as electrophiles.4 It has been well esta-
blished that for both catalytic (palladium mediated)
and non-catalytic methods, the order of nucleophilic
attack on the pyrimidine template occurs in the following
order: C4/C6 > C2 ꢀ C5.1,2 However, to our knowl-
edge, there is no systematic study regarding such nucleo-
philic displacements in the presence of other appended
electrophilic centers, such as an aldehyde group (Fig. 1).
6
1
2
N
O
5
4
N
3
Figure 1. Pyrimidine with competing electrophilic sites.
3-chloro-4-benzyloxyaniline (2) and 4-amino-6-chloro-
pyrimidine-5-carbaldehyde (1). Early work within our
laboratories called for reacting the chloropyrimidine
(1) with benzyloxyaniline (2) in DMSO at 100 °C in
the presence of triethylamine (TEA) (Scheme 1).5
Cl
BnO
OBn
Cl
Cl
CHO
NH2
NH
N
TEA , 100 °C
DMSO
+ TEA.HCl
+
CHO
NH2
N
N
+ H2O
NH2
N
1
2
3
70%
Cl
We were very much interested in preparing 4-amino-6-
(4-benzyloxy-3-chloro-phenylamino)-pyrimidine-5-carb-
aldehyde (3) via a chemoselective coupling involving
BnO
OBn
Cl
H
Cl
N
OBn
H
NH
N
N
N
Cl
N
NH2
Keywords: Chloropyrimidine; Acid catalyzed coupling; Diamino-
N
NH2
pyrimidines.
4
5
(4 + 5; 25%)
*
Corresponding author. Tel.: +1 908 704 4975; fax: +1 908 231
7429; e-mail: achoudhu@prdus.jnj.com
Scheme 1.
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.11.009

A. Choudhury et al. / Tetrahedron Letters 49 (2008) 102–105
Table 1. A study toward chemoselective chloropyrimidine aldehyde–aniline couplinga
103
Cl
Cl
BnO
OBn
Cl
BnO
+
OBn
Cl
H
C
OBn
Cl
Cl
Cl
N
H
C
solvent
CHO
NH2
NH
N
NH
+
N
+
N
CHO
NH2
base, temp
N
N
N
N
N
NH2
NH2
N
NH2
5
3
1
2
4
Entry
Solvent
TEA (equiv)
1
2
4
5
3
1
2
3
4
5
6b
7
8
9
THF
2-Me-THF
ACN
Acetone
DMF
NMP
MeOH
1-Propanol
IPA
IPA
IPA/H2O
2
2
2
2
2
2
2
2
54.0
56.3
46.3
51.8
30.0
11.2
—
6.7
9.5
37.4
0.7
nd
0.4
1.17
31.3
1.3
1.4
0.6
0.8
1.8
1.6
0.9
2.3
0.6
5.4
nd
2.9
1.3
7.6
5.6
4.7
2.9
1.0
4.1
1.7
26.1
11.0
0.7
nd
nd
10.5
9.1
31.9
26.7
29.2
16.0
34.4
10.2
5.0
4.6
14.4
3.0
18.7
12.6
47.0
49.8
70.4
81.9
83.5
16.6
85.3
98.2
c
2
10
—
—
—
—
—
IPA/H2O/HCl
2-Methoxyethanol/H2O/HCl
ACN/H2O/HCl
1.0
1.6
0.63
nd
0.3
0.5
11
12
94.0d
98.0
a Reactions were carried out using 1 mmol of 1 and 1 mmol of 2. For the base mediated reactions the reaction mixture was analyzed after 2.5 h at
ꢁ65 °C by HPLC and the area% to determine the composition of the mixture at 230 nm. The geometry of the imines has not been determined; the
HPLC area% represents the mixture of imines (4 and 5).
b The reviewer wanted to know if resubmission of product mixture from entry 6 to conditions in entry 12 leads to an improved product distribution?
Accordingly the reaction was performed in NMP for the indicated time, the solvent was removed in high vacuum (1 mm/Hg), then the resulting
mixture was subjected to conditions of entry 12 as per the suggestion of the reviewer. The product distribution of entry 6 changed to 5.9, 0.5, 59.6,
0.3 and 33.6 (1, 2, 4, 5, 3). Further addition of 0.1 mL of H2O and 0.1 mL of 6 N HCl resulted in substantial hydrolysis of chloropyrimidine
(10.2%), 3.5% and 86.0% (4 and 3).
c 1 gets converted to methoxy amino pyrimidine (17.4%).
d 3.5% of the hydroxy pyrimidine resulted from the hydrolysis of the chloropyrimidinealdehyde.
The rationale for using TEA was to neutralize the HCl
generated during the displacement and thus prevent
any possible acid catalyzed imine formation producing
imines 4 and 5.6 Despite this, significant amounts of
these imine by-products were formed and were difficult
to remove. The isolated pyrimidine carbaldehyde 3
was contaminated with up to 25% of undesired imines.
We considered addressing this poor chemoselectivity
using a conventional protection/deprotection sequence,
but instead believed that careful choice and control of
the reaction conditions could lead to a selective prepara-
tion of 3.
arising from methanolysis of chloropyrimidine alde-
hyde. On the other hand, 1-propanol and IPA (entries
8 and 9) resulted in increased conversion to coupled
product 3 while producing undetectable levels of alco-
holysis by-products. Scaling-up the TEA mediated reac-
tion in IPA resulted in an incomplete reaction.
Prolonged heating or addition of more base did not
improve the conversion above ꢁ83%. The product was
isolated in only 76% yield and was contaminated with
imine impurities 4 and 5. Although IPA in the presence
of TEA resulted in an 83% conversion to product, we
found that without TEA the coupling led to a compli-
cated reaction mixture consisting of only 16% of desired
product 3 and 5% and 26% of imines 4 and 5, respec-
tively, (entry 10).
Thus we briefly screened the coupling reaction of 4-
amino-6-chloropyrimidine-5-carbaldehyde (1) with 3-
chloro-4-benzyloxyaniline (2) with and without TEA
in a variety of solvents. The results of the study are sum-
marized in Table 1.
We viewed these imine impurities as Schiffs bases arising
from the reaction of the substituted aniline with the
starting aldehyde 1 or the product aldehyde 3.8,9 We
rationalized that the nucleophilic displacement of chloro-
pyrimidine with aniline generates HCl and the Schiffs
base formation liberates H2O (Scheme 1). The imines
could exist in equilibrium with the starting aldehyde
and aniline or with the product and this equilib-
rium could be affected by the amount of HCl and
water in the reaction. We hypothesized that careful
control of this equilibrium could then lead to the exclusive
formation of the desired aniline–pyrimidine aldehyde
(Fig. 2).
The reactions run in THF, 2-methyltetrahydrofuran (2-
Me-THF), acetonitrile, and acetone (entries 1–4) in the
presence of 2 equiv of TEA returned largely starting
materials. Both DMF and NMP (entries 5 and 6) offered
a very poor reaction profile with several unidentified
small impurities. In general, the alcohols were found
to be superior compared to other solvents in the screen
leading to moderate to high conversions to product 3.
Reaction in MeOH (entry 7) was found to contain about
17% of methoxy pyrimidine aldehyde, a by-product

104
A. Choudhury et al. / Tetrahedron Letters 49 (2008) 102–105
Table 2. Acid catalyzed chemoselective aniline–pyrimidine coupling
Cl
BnO
Cl
Ar
OBn
Cl
NH•HCl
CHO
H
C
CHO
NH2
NH
HCl/H2O
N
+
Ar-NH2
Cl
N
N
N
ACN
70-75 °C
N
BnO
N
NH2
N
NH2
OBn
Cl
H
C
Cl
4
NH•HCl
CHO
N
N
Ar-NH2
Product
Isolated
yielda (%)
N
N
NH2
N
NH2
3
5
"B"
N
"B"
N
H
N
Cl
N
93
87
•HCl
CHO
"A"
Cl
BnO
Cl
NH2
OMe
NH
NH2
Cl
CHO
H2N
CHO
NH2
CHO
NH2
H+/H2O
N
"A"
N
+
OMe
OMe
N
H
N
Cl
N
NH2
N
H
N
•HCl
OBn
1
N
H
2
MeO
O2N
NH2
NH2
H2N
CHO
Figure 2. Possible mechanism for the H2O/H+ catalyzed aniline–
chloropyrimidine coupling.
N
H
N
•HCl
N
NO2
88
52
To test this idea we added a small amount of H2O to this
complex mixture (entry 10/IPA–H2O) and heated to
ꢁ75 °C. This operation increased the product to 85%
and lowered the imine by-product to 11%. The resulting
mixture was heated for several hours but the composi-
tion of the reaction mixture did not change. We then
added a small amount of aqueous 6 N HCl to the reac-
tion mixture and heated the reaction to 75 °C. We were
pleased to find that this addition led to 98% conversion
to product.
CHO
H2N
H2N
NH2
NH2
N
N
H
N
•HCl
N
H2N
CHO
H
N
•HCl
PhO
OPh
N
85
74
88
85
88
NH2
H2N
CHO
Encouraged by this result we scaled the IPA/H2O/HCl
reaction to 200 g, but the mixture became viscous and
stirring was extremely difficult. We tried several varia-
tions of the solvent composition but encountered the
same problem. We then tested 2-methoxyethanol and
acetonitrile (Table 1, entries 11 and 12) under these
aqueous acid conditions and found a similar result
(ꢁ94–98%). It is remarkable that the TEA mediated
coupling reaction in acetonitrile (entry 3) largely
returned the starting material while the aqueous
acid mediated reaction (entry 12) offered exclusively
product 3.
N
Et
H
N
•HCl
Et
N
NH2
CHO
H2N
F
N
N
N
H
N
•HCl
N
CHO
F
H2N
NH2
F
H
N
•HCl
Br
F
N
Br
F
CHO
F
H2N
NH2
Intrigued by these results, we evaluated the scope of this
reaction using a variety of anilines in acetonitrile, the
results of which are summarized in Table 2. Several ani-
lines, both electron poor and electron rich, successfully
underwent coupling under the described acid catalyzed
conditions to offer coupling products in moderate to
high yields. The reaction is easily performed by heating
the mixture of chloropyrimidine aldehyde 1 and an ani-
line in acetonitrile with a catalytic amount of aqueous
acid for 4–6 h at 70–75 °C. Upon cooling the reaction
mixture, the product precipitated and was simply iso-
lated by filtration.10
Cl
H
N
•HCl
N
H2N
CHO
Cl
NH2
Cl
N
H
N
•HCl
BnO
OBn
N
96
CHO
Cl
H2N
NH2
a All compounds were characterized by 1H NMR and mass spectral
analysis.
The mechanism of chloropyrimidine displacement has
been studied.7 The leading hypothesis calls for the
activation of the chloropyrimidine ring via protonation
followed by the addition of a non-protonated amine,
which is presumably in equilibrium with its major
protonated form. At this point we can only hypothesize
what is the reaction pathway of this chemoselective
coupling. The product formation is believed to be an
intricate interplay of a series of complex equilibria involv-
ing reaction pathways ‘A and/or B’ as shown in
Figure 2.

A. Choudhury et al. / Tetrahedron Letters 49 (2008) 102–105
105
Yu, H.; Kohlhaas, K.; Jarvis, M. F.; Bhagwat, S. S. J.
Med. Chem. 2002, 45, 3639–3648.
In summary, we have successfully accomplished a chemo-
selective nucleophilic substitution on a chloropyrimidine
aldehyde template to generate diaminopyrimidines while
keeping the aldehyde functional group intact and avail-
able for further functionalization. A variety of aromatic
amines, electron poor and rich, participate in this reac-
tion with almost exclusive chemoselectivity producing
coupled products in moderate to high yields. The opera-
tional procedure is extremely simple and could be applied
to prepare multi-kilo quantities of material.11
5. Xu, Guozhang; Lee, Lily; Connolly, Peter J.; Middleton,
Steven A.; Emanuel, Stuart L.; Hughes, Terry V.; Abad,
Marta C.; Karnachi, Prabha S.; Wetter, Steven K. Chem.
Abstr. 2007, 147, 189189 (WO 07081630, July 19, 2007).
6. Personal communication from drug discovery chemists.
Reaction of 1 and 2 in DMSO at 100 °C (without using
TEA) resulted in 6 min: 12% 3, 39% 4 and 2% 5. In 60 min:
10% 3, 46% 4, 20% 5 were generated.
7. Successful nucleophilic substitution on pyrimidine ring has
been done in alcohols as the solvents (a) Hartung, C. G.;
Backes, A. C.; Felber, B.; Missio, A.; Philipp, A. Tetra-
hedron 2006, 62, 10055–10064; (b) Maggiolo, A.; Phillips, A.
P.; Hitchings, G. H. J. Am. Chem. Soc. 1951, 73, 106–107.
8. 2,4,6-Trimethoxy-methyl-pyrimidine-5-carbaldehyde reacts
with PhNH2 under refluxing ethanol condition to generate
the anil (a) Delia, T. J.; Wilcox, T. M.; Otteman, R. R.
J. Heterocycl. Chem 1979, 16, 1647; (b) Benzylimino-
methyl 2,4,6-(1H,3H,5H)-pyrimidinetrione is formed from
the corresponding pyrimidine aldehyde using PhNH2,
EtOH H2O, TEA, 65 °C, 5 min, 84% yield: Hoffmann-La
Roche, Br. Pat. 953,876, 1964; Chem. Abstr. 1942, 61,
29110 a.
Acknowledgements
We thank Keith Sampino and Dawei Xu for analytical
support. We thank Dr. Terry Hughes (Ref. 6), Dr.
Xun Li, Scott Youells and Drs. Ronald Russell and
Peter Connolly for helpful discussions.
References and notes
9. (a) Chapman, N. B.; Rees, C. W. J. Chem. Soc. 1954,
1190–1196; (b) Vanderplas, H. C. Acc. Chem. Res. 1978,
11, 462–468.
1. (a) Brown, D. J. In The Pyrimidines; Interscience Publish-
ers: New York, 1994; (b) Eicher, T.; Hauptmann, S. The
Chemistry of Heterocycles; Wiley-VCH: Weinheim, 2003.
2. (a) Hurst, D. T. In An Introduction to Chemistry and
Biochemistry of Pyrimidines, Purines and Pteridines; Wiley:
Chichester, 1980; (b) Brown, D. J. In Comprehensive
Heterocyclic Chemistry; Katritzky, A. R., Rees, C. W.,
Eds.; Pergamon Press: Oxford, 1984; Vol. 3, Chapter 2.13.
3. Murray, M. C.; Perkins, M. E. Annu. Rep. Med. Chem.
1996, 31, 141–150, and references cited therein.
10. General experimental: To a mixture of aniline (1 mmol)
and chloropyrimidine aldehyde (1, 1 mmol) in acetonitrile
(1.4 mL) are added 6 N HCl (15 lL) and H2O (0.1 mL).
The mixture is heated to 70–75 °C and is held at that
temperature for ꢁ4–6 h. After cooling naturally to ambi-
ent temperature, the product is filtered. The reaction flask
is rinsed with 0.5 mL of acetonitrile twice and the rinse is
used to wash the filter cake. The filter cake is washed with
0.5 mL of fresh acetonitrile. It is sucked dry under vacuum
with nitrogen for 15 min. The product is dried under lab
vacuum (25 mm/Hg) at ꢁ50 °C to a constant weight.
11. This result will be communicated separately.
4. (a) Peng, Z.; Journet, Michel; Humphrey, G. Org. Lett.
2006, 8, 395–398, and references cited therein; (b) Gom-
tsyan, A.; Didomenico, S.; Lee, C-Hung; Matulenko, M.
A.; Kim, K.; Kowaluk, E. A.; Wismer, C. T.; Mikusa, J.;