A practical synthesis of 2-arylamino-6-alkylaminopurines from 2,6-dichloropurine

Article abstract of DOI:10.1021/op060053m

A practical synthesis of N-[4-(6-cyclobutylamino-9H-purin-2-ylamino)- phenyl]-N-methyl acetamide (QAB205, 5a), an anti-asthmatic agent, is described from 2,6-dichloropurine (1) by base-assisted substitution of the 6-chloro substituent with cyclobutylamine

Full text of DOI:10.1021/op060053m

Organic Process Research & Development 2006, 10, 799−802
A Practical Synthesis of 2-Arylamino-6-alkylaminopurines from
2,6-Dichloropurine
Lech Ciszewski, Liladhar Waykole,* Mahavir Prashad, and Oljan Repic´
Process R&D, Chemical and Analytical DeVelopment, NoVartis Pharmaceuticals Corporation, One Health Plaza,
East HanoVer, New Jersey 07936, U.S.A.
Scheme 1
Abstract:
A practical synthesis of N-[4-(6-cyclobutylamino-9H-purin-2-
ylamino)-phenyl]-N-methyl acetamide (QAB205, 5a), an anti-
asthmatic agent, is described from 2,6-dichloropurine (1) by
base-assisted substitution of the 6-chloro substituent with
cyclobutylamine (2a) followed by a new trimethylsilyl chloride-
catalyzed displacement of the 2-chloro group in intermediate
6-cycbutylamino-2-chloropurine (3a) with an aromatic amine.
Both steps can also be carried out in one pot without isolating
the intermediate 6-cyclobutylamino-2-chloropurine (3a). The
general synthetic utility of this route is demonstrated by
synthesizing several 2-arylamino-6-alkylaminopurines (5).
this paper we report a practical synthesis of 5a from 2,6-
dichloropurine (1) first by base-assisted substitution of the
6-chloro substituent with cyclobutylamine (2a), followed by
a new trimethylsilyl chloride catalyzed displacement of the
2-chloro group in intermediate 3a with 4a in n-butanol at
117 °C. Both of the steps can also be carried out sequentially
in one pot without isolating the intermediate 3a. The general
synthetic utility of this route is also demonstrated by
synthesizing several 2-arylamino-6-alkylaminopurines (Scheme
2).
Introduction
Purines and aminopurines are a pharmaceutically impor-
tant class of compounds. Recently, they have received a great
deal of attention due to their important physiological and
pharmacological properties.1 They are reported to have
potential as target nucleotide binding proteins that play a
significant role in many biological processes.2 They have
been reported to possess antirhinovirus activity.3 Also, purine
scaffolds have been shown to have antitumor activity as
kinase inhibitors.4 Therefore, a large number of purine
libraries have been synthesized using solid polymer supports
or in solution.5
N-[4-(6-Cyclobutylamino-9H-purin-2-ylamino)-phenyl]-
N-methyl acetamide (5a) is a potential antiasthmatic agent.1
While the synthesis of 5a was straightforward utilizing 2,6-
dichloropurine (1) as the starting material (Scheme 1), it was
necessary for us to optimize this synthesis for large-scale
preparation. Our main goals were to improve the displace-
ment of the 2-chloro substituent in intermediate (3a) with
4-N-methylacetanilidoaniline (4a) that required high tem-
perature (130 °C) and to possibly develop a one-pot
procedure from (1) to make the synthesis more efficient. In
Results and Discussion
Original conditions for the reaction of 1 with 2a involved
treatment of 1 (1 equiv) with 2a (2.0 equiv) in n-butanol at
55 °C for 16 h to afford 3a in 85% yield. Then, the reaction
of the resulting 3a with 4a (2.7 equiv) in NMP at 130 °C
over 6 h afforded the drug substance 5a in 81% yield as a
hemi-hydrochloride salt. These procedures suffered from
shortcomings such as irreproducibility, very high reaction
temperature, and difficult isolation and purification proce-
dures due to the use of excess amounts of 2a (2.0 equiv)
and 4a (2.7 equiv). Also the quality of the product was
unsatisfactory due to its dark color and contamination with
deacylated byproduct.
We focused on optimizing both the steps to achieve
operational efficiency. In the first step, use of an excess
amount (2 equiv) of the expensive cyclobutylamine 2a was
undesired. We found that the use of 1.1 equiv of 2a was
optimal when used with 1.1 equiv. of N,N-diisopropylethyl-
amine (DIEA) as an acid scavenger. Also, the reaction time
was shortened to 4.5 h when the reaction was carried out at
75 °C instead of 55 °C. Thus, to a mixture of 1 (1 equiv)
and N,N-diisopropylethylamine (1.1 equiv) in n-butanol at
75 °C was added a solution of 2a (1.1 equiv) in n-butanol
over 1 h. After the addition of 2a, the reaction mixture was
stirred at 75 °C for an additional 3.5 h. The reaction
proceeded smoothly and regioselectively as evidenced by
* Corresponding author: liladhar.waykole@novartis.com.
(1) Collingwood, S. P.; Hayler, J.; Le Grand, D. M.; Mattes, H.; Menear, K.
A.; Walker, C. V.; Cockcroft, X. L. WO 01/09134, February, 2001.
(2) Hurst, D. T. The Chemistry and Biology of Pyrimidines, Purines, and
Pteridines; Wiley: Chichester, U.K., 1980.
(3) Kelley, J. L.; Linn, J. A.; Krochamal, M. P.; Selway, J. W. T. J. Med. Chem.
1988, 31, 2001.
(4) (a) Garcia-Echeverria, C.; Taxler, P.; Evans, D. B. Med. Res. ReV. 2000,
20, 28 and references therein. (b) Schindler, T.; Bornmann, W.; Pellicena,
P.; Miller, W. T.; Claerkson, B.; Kuryan, J. Science 2000, 289, 1938 and
references therein.
(5) (a) Nugiel, D. A.; Lyndon A. M. C.; Corbett, J. W. J. Org. Chem. 1997,
62, 201-203 (b) Ding, S.; Gray, N. S.; Ding, Q.; Schultz J. Am. Chem.
Soc. 2001, 66, 8273-8276.
10.1021/op060053m CCC: $33.50 © 2006 American Chemical Society
Published on Web 06/02/2006
Vol. 10, No. 4, 2006 / Organic Process Research & Development
•
799

Scheme 2
HPLC analysis. The reaction mixture was cooled to 0-5
°C and was stirred for 1 h prior to the filtration of
intermediate 3a. Intermediate 3a was isolated in 87% yield
with >98% purity.
After having no success with base catalysis we decided
to explore acid catalysis. Our decision was based on the
rationale that the protonation of either ring nitrogen N-3 or
the 6-amino group would make the six-membered ring
electron-deficient and facilitate the displacement of the
2-chloro substituent. To the best of our knowledge there are
no reports of acid-catalyzed C-2 chlorine displacement of
2-chloropurine with an amine. The TFA (0.4-10 equiv)-
catalyzed C-2 fluorine displacement of 6-alkyloxy 2-fluoro
purine with an excess of amine in trifluoroethanol has been
recently reported.7 The use of an excess of TFA and a
halogenated solvent were not preferred for scale-up. As our
choice of solvent for both steps was n-butanol, after
completion of the first step without isolation of 3a, 0.5 equiv
of aqueous hydrochloric acid (37%) was added to the reaction
mixture at 117 °C, followed by addition of 4a (1.25 equiv).
The reaction was complete in 12 h. However, the isolated
yield of 5a was only 23%, and the filtration of product was
very slow. Also the product contained ∼2% of the deacety-
lated byproduct. When 3a was isolated and treated with 4a
in the presence of 0.1 equiv of 37% HCl, the product 5a
was obtained in 55% yield containing 1.8% of the deacylated
byproduct. The lower yield due to deacylation of 5a, slow
filtration, and formation of deacylated byproduct were
attributed to the presence of water in the reaction mixture.
Water is not detrimental to the reaction itself, but the
deacylation problem occurs in case of 5a because of water.
We also found that at least 0.1 equiv of aqueous hydrochloric
acid (37%) was needed for completion of the reaction within
14 h. To improve the yield, minimize the byproduct
formation, and improve the filtration, anhydrous hydrochloric
With a practical preparation of 3a in hand we decided to
investigate the displacement of relatively unreactive chlorine
at the C-2 position of 3a with 4a. The unreactive nature
generally shown by a C-2 chlorine in this case is reinforced
by the inductive effect of the amino group introduced at C-6.
The weaker nucleophilicity of the amino group of 4a made
this transformation even more challenging (higher temper-
ature was needed). First, we lowered the amount of 4a to
1.1 equiv from 2.7 equiv since the excess was complicating
the isolation and purification of the drug substance 5a. Use
of DMF or DMSO instead of NMP at 130 °C for 12-14 h
gave a dark-colored reaction mixture, whereas the use of
n-butanol at 117 °C and n-pentanol at 130 °C gave a
complete conversion in 30 and 15 h respectively, and the
product 5a was obtained as a beige solid. To have an
optimum filtration rate of the product, n-butanol was selected
as the solvent. However, the longer reaction time needed to
be shortened. To this end we decided to explore catalytic
conditions. Literature procedures either used the amines as
solvent and/or elevated temperature.6 Since 4a is a solid it
could not be used as a solvent, and elevated temperature was
undesired for the scale-up. Use of sodium bicarbonate (1.2
equiv) and sodium hydroxide (1.2 equiv) in DMF at 130 °C
gave no conversion over 10-12 h. Also, no reaction occurred
in the presence of 1.1 equiv of N,N-diisopropylethylamine
at 117 °C over 40 h.
(6) (a) Parker, C.; Entsch, B.; Letham, D. Phytochemistry 1986, 25, 303-310.
(b) Hocart, C.; Letham, D.; Parker, C. Phytochemistry 1991, 30, 2477-
2786.
(7) Whitfield, H. J.; Griffin, R. J.; Hardcastle, A. H.; Meneyrol, J.; Mesguiche,
V.; Sayle, K. L.; Golding, B. T. Chem. Commun. 2003, 2802-2803.
800
•
Vol. 10, No. 4, 2006 / Organic Process Research & Development

Table 1. Reaction of 1 with 2 and 3 with 4a
time (h)
with 1
time (h)
with 3
TMS-Cl (equiv)
one-pot process
product and
entry
R-NH2
Ar-NH2
isolated yieldb (%)
1
2
3
4
5
6
7
8
9
2a
2a
2b
2b
2c
2c
2d
2d
2e
2e
4
4
2
2
7
7
4
4
4
4
4a
4a
4b
4b
4b
4b
4b
4b
4b
4b
12
12
9
10
10
8
8
8
12
13
0.01 ( two-step process)
0
1.3
0
1.3
0
1.3
0
1.3
0
5a 62
5a 0
5b 69
5b 0
5c 61
5c 0
5d 73
5d 0
5e 81
5e 0
10
a General procedure: A mixture of (1) (1 equiv) and DIEA (1.1 equiv) in n-butanol was heated to 75 °C during 25 min with stirring. To the resulting solution was
added a solution of 2 (1.1 equiv) in n-butanol maintaining the temperature at 75-77 °C. The reaction was monitored for completion (TLC or HPLC). After completion
of the reaction, the mixture was cooled to 5 ( 3 °C over a period of 1 h. The suspension was filtered, and the filter cake was washed with n-butanol. The filter cake
of 3 was collected and dried under vacuum. Then, to a stirred mixture of 3 (1 equiv) and 4 (1.1 equiv) in n-butanol at 25 °C TMSCl (0.02 equiv) was added, and the
mixture was heated to 117 °C over 1 h. It was stirred at 117 °C until completion of the reaction monitored by TLC or HPLC (12-16 h). The reaction mixture was
cooled to 50 ( 5 °C over a period of 2 h and stirred at this temperature for 1 h. The hot suspension was filtered, and the filter cake was washed three times, with warm
(∼40 °C) n-butanol. The filter cake was collected and dried at 50 °C under vacuum to afford 5. In a one-pot procedure after completion of the first step, to the
suspension of 3 in n-butanol a solution of 4 (1.1 equiv) in n-butanol and TMSCl (0.5-1.3 equiv) was added. The mixture was heated to 117 °C and was stirred at 117
°C until completion of the reaction monitored by TLC or HPLC (12-16 h). b All the compounds gave satisfactory spectroscopy data.
acid was needed. Since the reaction solvent was n-butanol,
this was easily achieved by adding TMS-Cl (1.5 equiv) to
the reaction mixture at 117 °C instead of aqueous hydro-
chloric acid. However, this led to the formation of 18% of
the deacetytlated byproduct. The amount of TMS-Cl needed
was optimized to a minimum of 1 mol %. We found that
the treatment of an isolated 3a with 4a (1.25 equiv) in
n-butanol at 117 °C using 1 mol % of TMS-Cl in 12 h gave
5a in 62% yield with >98% purity (0.8% byproduct). A two-
step process was the preferred method to achieve higher yield
and purity (minimum of deacylated byproduct) of drug
substance 5a. Thus, the process was carried out at a lower
temperature, lowering the safety risks in these steps. Both
of the steps worked equally well, yielding 5a in 54% overall
yield.
substitution at the 2-position in 3 is found to be an acid-
catalyzed process. No regeoisomers were detected at any step
by HPLC analysis. The presence of aqueous hydrochloric
acid (0.1-0.5 equiv) or trimethylsilyl chloride (0.01-0.5
equiv) is essential for the efficiency of second step, which
is limited to aromatic amines. The use of aqueous hydro-
chloric acid was not preferred for the preparation of 5a as it
led to the formation of deacylated byproduct. The 2,6-
disubstitution process in 1 with alkylamine and arylamine,
respectively, can be carried out in two separate steps or
sequentially in one-pot. The two-step procedure was applied
for the preparation of 5a to minimize the formation of
deacylated impurity.
Experimental Section
1H and 13C NMR spectra were measured on a Bruker
AVANCE DPX-300 MHz spectrometer in either CDCl3 or
DMSO-d6. Analytical high performance liquid chromatog-
raphy (HPLC) was carried out using Waters Alliance 2690
Separations module, a Waters 996 Photodiode ARRAY
detector, and a C18 steel column, 5 µm particle size (4.6 mm
× 250 mm). UV detection and mobile phase were varied,
depending upon substrate. In-process control (TLC) was run
using silica gel precoated plates with a 250-µm layer
thickness. Plates were visualized with either 5% ethanolic
phosphomolybdic acid solution and heat or 254 UV light,
depending upon substrate.
To explore the general applicability of these conditions,
several 2-arylamino-6-alkylaminopurines and 2,6-diarylami-
no purines were synthesized (Scheme 2, Table 1).
All of the above preparations were carried out in n-
butanol. The first step was carried out in the presence of
N,N-diisopropylethylamine (1.1 equiv) at 75 °C and the
second step at 117 °C either in the presence or absence of
TMS-Cl. The amount of TMSCl required for the preparation
of 5b-5e in a one-pot process was 0.5-1.3 equiv and a
minimum of 0.01 equiv in the two-step process. We found
that either alkyl or arylamine (2-2e) could be used for the
displacement of C-6 chlorine in 1. In most cases the reaction
was completed in 2-7 h. For the displacement of chlorine
at C-2 in 3, the presence of acid was essential, without which
no displacement occurred. However, in the presence of acid
the highly basic and nucleophilic alkylamine (2-2c) could
not be used in this step as the protonated alkylamine failed
to react. Thus, the TMS-Cl or acid-catalyzed C-2 substitution
in 3 is limited to anilines. The 2,6-diaminopurines (5a-5e)
were obtained in moderate to a good yields.
Materials. The starting material (1) was purchased from
a vendor (Borregaard, UK), and (4a) was purchased from
L&L Technologies. All other reagents and solvents were
purchased from various commercial sources and used without
further purification. All of the isolated intermediates were
analyzed as indicated above and used directly in subsequent
steps. All reactions were run in an inert nitrogen atmosphere
unless otherwise indicated.
2-Cyclobutylamino-6-chloropurine(3a).A12-L,4-necked,
round-bottomed flask, equipped with a condenser, thermom-
In summary, a highly efficient method for the preparation
of 2,6-diaminopurine derivatives 5 has been developed.8
Substitution at the 6-position of 2,6-dichloropurine (1) is a
base-catalyzed and completely regiospecific process while
(8) Ciszewski, L. A.; Waykole, L. M. (Novartis). U.S. Patent 6,897,307, May
24, 2005.
Vol. 10, No. 4, 2006 / Organic Process Research & Development
•
801

eter, and addition funnel under nitrogen atmosphere was
charged with 492.9 g (2.61 mol) of 2,6-dichloro-9H-purine
(1), 497.7 mL (2.87 mol) of N,N-diisopropylethylamine
(DIEA), and 3.65 L of anhydrous n-butanol. The mixture
was stirred and warmed to 75 °C during about 25 min. To
the solution was added 244.9 mL (2.87 mol) of cyclobuty-
lamine (2a) in 1.25 L of anhydrous n-butanol over 1 h at
75-77 °C. The reaction mixture was stirred at 75 °C for
4.5 h. The reaction was monitored by HPLC for completion.
The mixture was cooled to 5 ( 3 °C over a period of 1 h.
The suspension was filtered, and the filter cake was washed
three times, with a total of 0.9 L of (∼5 °C) n-butanol. The
filter cake was collected and dried at 75 °C/20 mm of Hg
for 24 h to give 506.6 g of 3a as a light-yellow crystalline
solid: yield: 86.9% with purity 98%; mp 216-8 °C; 1H NMR
(300 MHz, DMSO-d6) δ 13.03 (bs, 1H), 8.39 (bs, 1H), 8.12
(bs, 1 H), 4.60 (bs,1H), 2.24 (bs, 2H), 2.11 (b, s, 2 H), 1.6
(bs, 2H).
stirred at room temperature, and 2.72 mL (0.02 mol) of
chlorotrimethylsilane was added to it. The reaction mixture
was warmed to 117 °C over 50 min. It was stirred at 117 °C
for 12.5 h. The reaction mixture solution was cooled to 50
( 5 °C over a period of 2 h and stirred at this temperature
for 1 h. The hot suspension was filtered, and the filter cake
was washed three times, with a total of 1.2 L of warm (∼40
°C) n-butanol. The filter cake was collected and dried at 50
°C/20 mmHg for 24 h to yield 514.3 g of light-yellow
crystalline solid 5a; yield 61.8% with purity >98%; mp
270-2 °C; 1H NMR (300 MHz, DMSO-d6) δ 9.92 (bs, 1H),
9.36 (bs, 1H), 8.72 (bs, 1H), 7.78-7.75 (bd, 2H), 7.29-
7.26 (bd, 2H), 4.57 (bs, 1H), 3.12 (s, 3 H), 2.35 (m, 2H),
2.14-2.08 (m, 2 H), 2.08 (bs, 5H); 13C NMR (75 MHz,
DMSO-d6) δ 169.5, 150.5, 149.07, 138.9, 138.67, 128.4,
128.2, 127.6, 122.1, 120. 6, 105.9, 60.7, 46.21, 36.9, 30.2,
22.5, 15.32; MS m/z 351.3 (M+); Anal. Calcd for C18H22-
ClN7O: C, 55.74; H, 5.72; N, 25.28. Found: C, 55.74; H,
5.86; N, 24.91; IR: 3422, 3279, 3123, 2985, 29847, 2870,
1560, 1476, 1431, 1394, 1337, 1318, 1250, 1176, 1145, 844,
782, 767, 727, 605, 562.
N-[4-(Cyclobutylamino-9H-purin-2-ylamino)-phenyl]-
N-methyl-acetamide (5a). A 5-L, 4-necked, round-bottomed
flask, equipped with a condenser, thermometer, and addition
funnel under nitrogen atmosphere was charged with 479.8
g (2.15 mol) of 2-chloro-N-cyclobutyl-9H-purin-6-amine
(3a), 387.5 g (2.36 mol) of N-(4-aminophenyl)-N-methyl-
acetamide (4a), and 3.6 L of dry n-butanol. The mixture was
Received for review March 8, 2006.
OP060053M
802
•
Vol. 10, No. 4, 2006 / Organic Process Research & Development