A novel one-pot synthesis of N,6-disubstituted 1,3,5-triazine-4,6-diamines from isothiocyanates and amidines

Article abstract of DOI:10.1016/j.tetlet.2006.11.069

A novel one-pot procedure for the preparation of N,6-disubstituted-1,3,5-triazine-2,4-diamines and its scope of application are demonstrated with a number of examples. The new procedure involves the treatment of isothiocyanates with sodium hydrogencyanami

Full text of DOI:10.1016/j.tetlet.2006.11.069

Tetrahedron Letters 48 (2007) 435–437
A novel one-pot synthesis of N,6-disubstituted
1,3,5-triazine-4,6-diamines from isothiocyanates and amidines
Chunjian Liu,* James Lin and Katerina Leftheris
Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton, NJ 08543-4000, United States
Received 26 September 2006; revised 8 November 2006; accepted 10 November 2006
Available online 1 December 2006
Abstract—A novel one-pot procedure for the preparation of N,6-disubstituted-1,3,5-triazine-2,4-diamines and its scope of
application are demonstrated with a number of examples. The new procedure involves the treatment of isothiocyanates with sodium
hydrogencyanamide, followed by amidines in the presence of 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride.
ꢀ 2006 Published by Elsevier Ltd.
N,6-Disubstituted-1,3,5-triazine-2,4-diamines have often
been found to be synthetic targets as chemotherapeutic
agents.1–5 Previous synthetic methods for such com-
pounds mainly relied on two routes, each requiring two
or three steps. One route starts with cyanuric chloride,
which can sequentially be reacted with Grignard
reagents,1,6 ammonia, and amines. The utility of this pro-
cedure is limited by the fact that Grignard reagents are
highly reactive and therefore prevent versatile function-
ality for further elaboration. The other route involves
the reaction of substituted biguanides with acid chlo-
rides,7,8 anhydrides,9 or carboxylates.3–5 This procedure
appears more general. However, substituted biguanides
usually need to be prepared from dicyandiamide and
amines. The preparation occasionally requires harsh
conditions and the products are extremely water soluble,
making product isolation difficult. In this letter, we
report a facile, one-pot procedure for the preparation
of the title compounds from isothiocyanates, sodium
hydrogencyanamide, and amidines in the presence of
1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydro-
chloride (EDC).
with amidines. Thus, 3,4,5-trimethylphenyl isothio-
cyanate (1a) was reacted with commercially available
sodium hydrogencyanamide in DMF to provide
N-cyanothiourea sodium salt 2a (Scheme 1). Without
NH
.
HCl
R2
NH2
3a-b
H
N
S
Na
NaNHCN
R1
R1NCS
N
NEt3, EDC
CN
2a-e
1a-e
1a, 2a: R1 = 3,4,5-trimethylphenyl
1b, 2b: R1 = 2,4-dichlorophenyl
1c, 2c: R1 = phenyl
1d, 2d: R1 = cyclohexyl
1e, 2e: R1 = benzyl
3a: R2 = phenyl
3b: R2 =
O
NH2
H
H
N
H
N
N
N
R2
R2
NH
R1
R1
N
N
N
CN
NH2
5a-f
We previously reported a novel, one-pot synthesis of
1,2,4-triazole-3,5-diamine derivatives from isothiocya-
nates, sodium hydrogencyanamide, and mono-substi-
tuted hydrazines in the presence of EDC.10 It was
envisioned that this methodology could be extended to
the preparation of N,6-disubstituted-1,3,5-triazine-2,4-
diamines, by replacing mono-substituted hydrazines
4a-f
4a, 5a: R1 = 3,4,5-trimethylphenyl, R2 = phenyl
4b, 5b: R1 = 2,4-dichlorophenyl, R2 = phenyl
4c, 5c: R1 = phenyl, R2 = phenyl
O
4d, 5d: R1 = phenyl, R2 =
NH2
4e, 5e: R1 = cyclohexyl, R2 = phenyl
4f, 5f: R1 = benzyl, R2 = phenyl
*
Corresponding author. Tel.: +1 609 252 3682; fax: +1 609 252
6601; e-mail: chunjian.liu@bms.com
Scheme 1.
0040-4039/$ - see front matter ꢀ 2006 Published by Elsevier Ltd.
doi:10.1016/j.tetlet.2006.11.069

436
C. Liu et al. / Tetrahedron Letters 48 (2007) 435–437
isolation, 2a was treated with benzamidine hydrochlo-
ride (3a) at rt in the presence of triethylamine and
EDC. Thirty minutes later, 2a was completely con-
sumed, giving rise to two products in a ratio of 5:1. Both
products had the desired mass. This ratio reversed to
1:8.6 after the mixture was heated at 75 ꢁC for 1 h. After
an additional 1 h of heating, the initially more polar
product was completely converted to the less polar
one. The final product was isolated by flash chromato-
graphy in 71% yield (Table 1) and confirmed to be
1,3,5-triazine-2,4-diamine 5a11 by comparison with an
authentic sample, prepared by a known procedure.1
The more polar, less stable product is believed to be
intermediate 4a, though it was not isolated due to its
instability. In a similar manner, 2,4-dichlorophenyl
isothiocyanate (1b) was reacted with sodium hydrogenc-
yanamide to generate N-cyanothiourea sodium salt 2b.
When 2b was treated with benzamidine hydrochloride
(3a), triethylamine, and EDC at rt for 1 h, 4b and 5b
formed in a ratio of 4:1, respectively. However, when
the mixture was heated to 75 ꢁC for an additional 1 h,
5b12 was the only product detected with the desired
mass. The isolated yield of 5b was 54%. In the interest
of developing synthetic methodology, it is not impera-
tive for the second part of the one-pot procedure to
begin at rt. Elimination of the initial rt treatment gave
similar yields. For example, heating 2c, the adduct of
phenyl isothiocyanate (1c) and sodium hydrogencyana-
mide, with benzamidine (3a) in the presence of triethyl-
amine and EDC at 75 ꢁC for 3 h resulted in the
formation of 5c13 in 72% yield. Similarly, 5d14 was
obtained from 1c and 3b in 65% yield. Cycloalkyl and
alkyl isothiocyanates proved to work as demonstrated
by cyclohexyl isothiocyanate and benzyl isothiocyanate.
The products 5e and 5f were isolated in 50% and 44%
yields, respectively. The structure of 5f was again con-
firmed by comparison with an authentic sample, pre-
pared by a known procedure.1
NH
H
.
MeO
MeO
N
N
R2
HCl
R2
NH2
N
N
3c-d
NEt3, EDC
NH2
i-Pr (33% yield for 5g)
OMe
3c, 5g: R2 =
3d, 5h: R2 =
t
-Bu (44% yield for 5g)
2a
H
N
NH2
CN
MeO
MeO
NH
R2
.
HCl
N
NH2
3f-h
OMe
NEt3, EDC
6
3f: R2 = n-Pr
3g: R2 = Me
3h: R2 = H
Scheme 2.
to 75 ꢁC for 1 h, the reaction became much cleaner with
only one peak corresponding to the desired mass. Prod-
uct 5g16 was obtained in 33% yield. t-Butyl carbamidine
hydrochloride (3d) performed similarly to isobutyrami-
dine hydrochloride (3c) to afford 5h17 in 44% yield.
However, when butyramidine hydrochloride (3f), trieth-
ylamine, and EDC were mixed with 2a at rt, a very clean
reaction occurred within 40 min to give N-cyanoguani-
dine 618 in 80% yield. The structure of 6 was confirmed
by comparison with an authentic sample prepared by a
known synthetic method.19 It was further found that
acetamidine hydrochloride (3g) and formamidine hydro-
chloride (3h) also gave product 6 exclusively under the
same conditions.
The isolation of the single product 6 from the reactions
of 2a with 3f–h prompted us to re-examined the reac-
tions of 2a with 3c and 4d and the examples in Scheme
1. The formation of N-cyanoguanidine 6 or its ana-
logues as by-products appeared to be general in those
reactions, but they were minor. For example, the ratios
of 6 to 5g and 5h in the reactions of 2a with 3c and 3d
were 1:8 and 1:11, respectively, by LC–MS.15 The ratio
of 6 to 5a in the reaction of 2a with 3a was 1:12.
Table 1. Isolated yields of 5a–f
Product Yield (%) R1
R2
5a
5b
5c
71
54
72
3,4,5-Trimethylphenyl Phenyl
2,4-Dichlorophenyl
Phenyl
Phenyl
Phenyl
O
The mechanism by which the common product 6 was
formed exclusively with amidines 3f–h is not exactly
clear at this point. One possibility is that once inter-
5d
65
Phenyl
NH2
5e
5f
50
44
Cyclohexyl
Benzyl
Phenyl
Phenyl
H
N
H
R2
NH
N
MeO
MeO
Further examination of the scope of the synthetic meth-
odology revealed that aliphatic carbamidines behaved
differently in the reaction and the outcome appeared
to depend on the nature of the aliphatic groups. When
N-cyanothiourea sodium salt 2a was treated with iso-
butyramidine hydrochloride (3c) in the presence of
triethylamine and EDC at rt for 30 min (Scheme 2),
multiple peaks appeared as judged by LC–MS.15 Two
peaks had the desired mass, but they accounted for only
4.3% and 9.6% of the total UV active components,
respectively. Interestingly, when the mixture was heated
R2
N
R2
NH2
N
CN
NH
OMe
4a: R2 = Ph
4g: R2 =
4h: R2 =
4i: R2 =
4j: R2 = Me
4k: R2 = H
7a: R2 = n-Pr
7b: R2 = Me
7c: R2 = H
i
t
n
-Pr
-Bu
-Pr
Figure 1.

C. Liu et al. / Tetrahedron Letters 48 (2007) 435–437
437
3. Hajiduk, P. J.; Dinges, J.; Schkeryantz, J. M.; Janowick,
D.; Kaminski, M.; Tufano, M.; Augeri, D. J.; Petros, A.;
Nienaber, V.; Zhong, P.; Hammond, R.; Coen, M.; Beutel,
B.; Katz, L.; Fesik, S. W. J. Med. Chem. 1999, 42, 3852–
3859.
mediate 4i–k formed, they underwent electrophilic reac-
tions with unreacted amidines 3f–h at the position de-
noted with an arrow before intramolecular cyclizations
could occur (Fig. 1). The by-products from such reac-
tions would be 7a–c that could also participate in the
conversion of 4i–k to 6 in the same manner. If by-prod-
ucts 7a–c were not stable, they would likely decompose
back to the starting amidines 3f–g. In the case of 4a, 4g,
and 4h, the steric hindrance provided by phenyl, i-pro-
pyl, and t-butyl might have impeded such electrophilic
reactions significantly, and therefore, 1,3,5-triazine-2,4-
diamine derivatives 5a, 5g, and 5h were generated as
the major products. Benzamidine is also less nucleo-
philic than amidines 3f–h to initiate the side reaction,
and this may also play a role in determining the extent
of by-product 6 in the case of 4a.
´
´
4. Kosary, J.; Kasztreiner, E.; Rabloczky, G.; Kurthy, M.
¨
Eur. J. Med. Chem. 1989, 24, 97–99.
5. Irikura, T.; Abe, Y.; Okamura, K.; Higo, K.; Maeda, A.;
Morinaga, F.; Shirai, G.; Hatae, S. J. Med. Chem. 1970,
13, 1081–1089.
6. Hirt, R.; Nidecker, H.; Berchtold, R. Helv. Chim. Acta
1950, 33, 1365–1369.
7. Yuki, Y.; Sakurai, S.; Kakurai, T.; Noguchi, T. Bull.
Chem. Soc. Jpn. 1970, 43, 2130–2134.
8. Shapiro, S. L.; Isaacs, E. S.; Parrino, V. A.; Freedman, L.
J. Org. Chem. 1961, 26, 68–74.
9. Alkalay, D.; Volk, J.; Bartlett, M. F. J. Pharm. Sci. 1976,
65, 525–529.
10. Liu, C.; Iwanowicz, E. J. Tetrahedron Lett. 2003, 44, 1409–
1411.
In summary, a convenient synthetic procedure for
the preparation of N,6-disubstituted-1,3,5-triazine-2,4-
diamines from isothiocyanates, sodium hydrogencyan-
amide, and amidines has been reported in this letter.
This procedure possesses the advantages of a one-pot
operation that requires only mild conditions. The new
protocol appears to be general with isothiocyanates
(both aromatic and aliphatic) and aromatic carbami-
dines. With aliphatic carbamidines, the outcome of the
procedure is apparently determined by steric hindrance.
More hindered aliphatic carbamidines give the desired
product, while less hindered aliphatic carbamidines lead
to N-cyanoguanidines.
11. Compound 5a: 1H NMR (400 MHz, DMSO-d6) d: 9.41
(1H, s), 8.33 (2H, d, J = 6.9 Hz), 7.56–7.49 (3H, m), 7.28
(2H, s), 7.18 (2H, br s), 3.80 (6H, s), 3.63 (3H, s); 13C
NMR (100 MHz, DMSO-d6) d: 170.5, 167.4, 164.8, 152.9,
137.1, 136.4, 132.9, 131.8, 128.6, 128.1, 98.0, 60.4, 56.1.
12. Compound 5b: 1H NMR (400 MHz, DMSO-d6) d: 8.91
(1H, s), 8.25 (2H, d, J = 6.9 Hz), 7.83 (1H, d, J = 8.7 Hz),
7.69 (1H, s), 7.57–7.44 (4H, m), 7.15 (2H, br s); 13C NMR
(100 MHz, DMSO-d6) d: 170.7, 167.6, 136.9, 135.4, 131.8,
129.6, 129.5, 129.2, 128.9, 128.7, 128.2, 127.8.
13. Compound 5c: 1H NMR (400 MHz, DMSO-d6) d: 9.55
(1H, s), 8.33 (2H, d, J = 6.7 Hz), 7.85 (2H, d, J = 7.9 Hz),
7.58–7.49 (3H, m), 7.31 (2H, dd, J = 7.8, 7.8 Hz), 7.16
(2H, br s), 7.00 (1H, t, J = 7.3 Hz); 13C NMR (100 MHz,
DMSO-d6) d: 170.6, 167.5, 165.0, 140.3, 137.1, 131.8,
128.8, 128.6, 128.1, 122.3, 120.3.
A representative procedure demonstrated by the prepara-
tion of 5b: To a solution of 2,4-dichlorophenyl isothio-
cyanate (0.215 g, 1.00 mmol) in dry DMF (5 mL) was
added sodium hydrogencyanamide (68.6 mg, 1.05 mmol)
at room temperature in one portion. The mixture
was heated at 60 ꢁC for 50 min before triethylamine
(0.31 mL, 2.22 mmol), benzamidine hydrochloride
(0.235 g, 1.50 mmol), and EDC (0.240 g, 1.25 mmol)
were added at room temperature. The mixture was stir-
red at rt for 30 min and then at 75 ꢁC for 1 h. On cooling
to rt, the mixture was diluted with ethyl acetate (80 mL),
washed sequentially with water (25 mL) and 10% LiCl
solution (25 mL), and dried over anhydrous MgSO4.
The solution was concentrated under vacuum, and the
residue was subjected to flash chromatography (silica
gel, 30% ethyl acetate/hexane) to afford 5b (0.180 g,
54% yield) as a pale yellow solid.
14. Compound 5d: 1H NMR (400 MHz, DMSO-d6) d: 9.53
(1H, s), 8.28 (2H, d, J = 8.4 Hz), 8.00 (1H, s), 7.93 (2H, d,
J = 8.4 Hz), 7.77 (2H, d, J = 7.9 Hz), 7.41 (1H, s), 7.24
(2H, dd, J = 7.8, 7.8 Hz), 7.16 (2H, br s), 6.93 (1H, t,
J = 7.4 Hz); 13C NMR (100 MHz, DMSO-d6) d: 169.9,
167.8, 167.5, 164.9, 140.2, 139.6, 137.1, 128.8, 127.9, 127.8,
122.4, 120.3.
15. LC–MS conditions: Column: Phenomenex 5u C18
4.6 · 50 mm; Solvent A: 10% MeOH—90% H2O—0.1%
TFA; Solvent B: 90% MeOH—10% H2O—0.1% TFA;
Gradient time: 4 min; Detecting wavelength: 254 nm.
16. Compound 5g: 1H NMR (400 MHz, DMSO-d6) d: 9.22
(1H, s), 7.23 (2H, s), 6.97 (2H, br s), 3.75 (6H, s), 3.60 (3H,
s), 2.66 (1H, m), 1.20 (6H, d, J = 6.9 Hz); 13C NMR
(125 MHz, DMSO-d6) d: 182.4, 167.2, 164.8, 153.0, 136.6,
132.8, 98.0, 60.6, 56.2, 39.8, 21.3.
17. Compound 5h: 1H NMR (400 MHz, DMSO-d6) d: 9.14
(1H, s), 7.24 (2H, s), 6.90 (2H, br s), 3.75 (6H, s), 3.61 (3H,
s), 1.26 (9H, s); 13C NMR (125 MHz, DMSO-d6) d: 184.4,
167.1, 164.6, 152.8, 136.6, 132.6, 97.7, 60.4, 56.0, 38.8,
29.2.
References and notes
18. Compound 6: 1H NMR (500 MHz, DMSO-d6) d: 8.99
(1H, s), 6.95 (2H, s), 6.64 (2H, s), 3.76 (6H, s), 3.64 (3H, s);
13C NMR (100 MHz, DMSO-d6) d: 159.9, 153.1, 134.4,
134.0, 117.7, 100.1, 60.4, 56.1.
1. Pitts, W. J.; Guo, J.; Dhar, T. G. M.; Shen, Z.; Gu, H.;
Watterson, S. H.; Bednarz, M. S.; Chen, B.-C.; Barrish, J.
C.; Bassolino, D.; Cheney, D.; Fleener, C. A.; Rouleau, K.
A.; Hollenbaugh, D. L.; Iwanowicz, E. J. Bioorg. Med.
Chem. Lett. 2002, 12, 2137–2140.
19. Iwanowicz, E. J.; Watterson, S. H.; Liu, C.; Gu, H.; Mitt,
T.; Leftheris, K.; Barrish, J. C.; Fleener, C. A.; Rouleau,
K.; Sherbina, N. Z.; Hollenbaugh, D. L. Bioorg. Med.
Chem. Lett. 2002, 12, 2931–2934.
2. Hanessian, S.; Sgarbi, P. W. M. Bioorg. Med. Chem. Lett.
2000, 10, 437–443.