4056 J . Org. Chem., Vol. 66, No. 11, 2001
Notes
phylline derivatives 14 or 15, respecively (Scheme 1). The
latter structure was discarded, however, due to the steric
hindrance caused by the proximity of the N1-CH3 and
R groups as shown by molecular models. Assignment of
the structure 14 to the products isolated from the studied
reactions rather than the possible alternative structure
15 is also based on similar results in the literature
reports documenting that N-7 of the theophylline ring
system is the site of preference for nucleophilic additions5
as well as nucleophilic substitutions.3,19,20 In addition, the
Exp er im en ta l Section
Melting points were determined in open capillaries and are
uncorrected. 1H NMR specra were recorded at 200 MHz in CDCl3
or DMSO-d6 and referenced to TMS (1H). IR spectra were
recorded in KBr. Mass spectra were measured at 70 eV using a
GC-MS. Elemental analyses were provided by the Microana-
lytical Laboratory at Cairo University, Giza, Egypt. The starting
hydrazonoyl halides 1a -o17 and 8-substituted theophyllines
2-518 were prepared as previously described.
1,3-D i m e t h y l-2,4-d i o x o -6,8-d i s u b s t i t u t e d -1,2,3,4-
tetr ah ydr o[1,2,4]tr iazolo[3,4-f]pu r in es 14. Gen er al Meth od .
To a mixture of the appropriate hydrazonoyl halide 1 (0.01 mol)
and 8-chlorotheophylline 2 (0.01 mol) in dioxane (30 mL) was
added triethylamine (1.5 mL, 0.01 mol). The mixture was then
refluxed for 5.0-10.0 h according to the halide used. Then the
excess solvent was distilled under reduced pressure. The crude
product that precipitated upon cooling was collected by filtration,
washed with ethanol, dried, and finally crystallized from the
appropriate solvent to give the corresponding 14.
When the above procedure was repeated using 8-nitrotheo-
phylline 3 in place of 2, it yielded the respective 14 identical in
all respects with that one obtained above from 1 and 2.
Furthermore, when an equimolar mixture (0.01 mol each) of
1 and 8-methylthiotheophylline 4 (or 8-mercaptotheophylline 5)
in pyridine (20 mL) was refluxed till methanethiol (or hydrogen
sulfide) ceased to evolve (4.0-5.0 h), the respective 14 was also
obatined after workup.
1
structures of 14a -o were further evidenced by their H
NMR data. For example, the 1H NMR spectra of the
isolated products revealed the signal of the N1-CH3
protons at δ 3.50-3.70. This value is very close to that
of N3-CH3 of theophylline (δ 3.59). This similarity of δ
values is consistent with structure 14. This is because
had the products structure 15, it would be expected that
the N1-CH3 group will be shielded by the neighboring
group R, and thus its δ value will be less than 3.50 and
it will vary with the degree of shielding exerted by the R
group.
To account for the formation of the products 14 from
reactions of 1 with each of 2-4, the tentative mechanism
outlined in Scheme 1 is proposed. According to this
mechanism, it is suggested that the reaction of 1 with
each of 2-4 starts with 1,3-dipolar addition of N7-H of
2 to the nitrilium imide, generated in situ by the action
of triethylamine on 1, to give the respective amidrazone
intermediates 6-9, respectively. The latter then undergo
intramolecular addition to give the cycloadducts 10-13,
respectively. Alternatively, the latter cycloadducts 10-
13 can result directly via cycloaddition of the nitrilium
imide to the CdN bond in theophylline derivatives. The
resulting cycloadducts 10-13 in turn eliminate an HX
molecule to give 14 as the end products (Scheme 1). In
our hands, all attempts to isolate the amidrazone inter-
mediates 6-9 or the cycloadducts 10-13 failed.
With respect to reaction of 1 with 5 leading to 14, it
can be suggested that the reaction in this case can start
with the formation of the thiohydrazonate ester which
undergoes in situ Smiles type rearrangement to give the
respective thiohydrazide which in turn cyclizes intramo-
lecularly to give 14 as end product via elimination of
hydrogen sulfide. Similar rearrangements of the thiohy-
drazonate esters followed by elimination of hydrogen
sulfide have been reported previously.21 However, it is
not unreasonable to conclude, on the basis of our finding
that reactions of 1 with either 4 or 5 gave the same
product 14, that both reactions proceed via similar
intermediates, namely 12 and 13, respectively, as out-
lined in Scheme 1.
The physical constants of the products 14a -o are given below.
1,3-Dim eth yl-2,4-d ioxo-6,8-d ip h en yl-1,2,3,4-tetr a h yd r o-
[1,2,4]tr ia zolo[3,4-f]p u r in e (14a ): yield 82%, mp 286-288 °C
(dioxane); IR(KBr) 1712, 1674 cm-1; 1H NMR (CDCl3) δ 3.43 (s,
3H, N3-Me), 3.70 (s, 3H, N1-Me), 7.50-8.26 (m, 10H, ArH); MS
m/z 372(M+, 100%), 373(M+ +1, 92%). Anal. Calcd for C20H16N6O2:
C, 64.51; H, 4.33; N, 22.57. Found: C, 64.4; H, 4.5; N, 22.3.
1,3-Dim eth yl-2,4-dioxo-6-ph en yl-8-(p-n itr oph en yl)-1,2,3,4-
tetr a h yd r o[1,2,4]tr ia zolo[3,4-f]p u r in e (14b): yield 80%, mp
> 340 °C (DMF), IR(KBr) 1716, 1682 cm-1 1H NMR (DMSO-
;
d6) δ 3.26 (s, 3H, N3-Me), 3.59 (s, 3H, N1-Me), 7.60-8.60 (m,
9H, ArH); MS m/z 417(M+, 100%), 418(M+ +1, 30%). Anal. Calcd
for C20H15N7O4: C, 57.55; H, 3.62; N, 23.49. Found: C,57.5; H,
3.7; N, 23.7.
1,3-Dim eth yl-2,4-dioxo-6-m eth yl-8-(p-n itr oph en yl)-1,2,3,4-
tetr a h yd r o[1,2,4]tr ia zolo[3,4-f]p u r in e (14c): yield 80%, mp
278-280 °C (dioxane /EtOH), IR(KBr) 1711, 1680 cm-1; 1H NMR
(CDCl3) δ 2.96 (s, 3H, CH3), 3.46 (s, 3H, N3-Me), 3.59 (s, 3H,
N1-Me), 7.80-8.70 (m, 4H, ArH); MS m/z 355 (M+, 26%), 356-
(M+ +1, 17%). Anal. Calcd for C15H13N7O4: C, 50.71; H, 3.69; N,
27.59. Found: C,50.4; H, 3.4; N, 27.3.
1,3-Dim et h yl-2,4-d ioxo-6-(2-p h en ylet h en yl)-8-p h en yl-
1,2,3,4-t et r a h yd r o[1,2,4]t r ia zolo[3,4-f]p u r in e (14d ): yield
78%, mp 296-298 °C (dioxane), IR(KBr) 1707, 1665 cm-1 1H
,
NMR (DMSO-d6) δ 3.31 (s, 3H, N3-Me), 3.51 (s, 3H, N1-Me),
7.10-8.15 (m, 12H, ArH, CHdCH); MS m/z 398 (M+, 78%), 399-
(M+ +1, 77.8%). Anal. Calcd for C22H18N6O2: C, 66.32; H, 4.55;
N, 21.09. Found: C,66.6; H, 4.3; N, 21.3.
1,3-D i m e t h y l-2,4-d i o x o -6-a c e t y l-8-p h e n y l-1,2,3,4-
tetr a h yd r o[1,2,4]tr ia zolo[3,4-f]p u r in e (14e): yield 75%, mp
300-302 °C (dioxane), IR(KBr) 1717, 1697, 1662 cm-1, 1H NMR
(CDCl3) δ 2.75 (s, 3H, CH3), 3.39 (s, 3H, N3-Me), 3.68 (s, 3H,
N1-Me), 7.20-8.20 (m, 5H, ArH); MS m/z 338 (M+, 85.9%), 339
(M+ +1, 78.5%). Anal. Calcd for C16H14N6O3: C, 56.80; H, 4.17;
N, 24.84. Found: C, 56.8; H, 4.0; N, 24.6.
Compounds 14a -o represent important extensions in
the chemistry ring-fused purines. The availability of
various functional groups for further reactions offers the
potential for novel biologically active materials. Further
studies to shed light on the affinity of the compounds
prepared above for A1- and A2-adenosine receptors and
their inhibitory effects on phosphodiesterase are still in
progress.
1,3-D im e t h y l-2,4-d io x o -6-b e n zo y l-8-p h e n y l-1,2,3,4-
tetr a h yd r o[1,2,4]tr ia zolo[3,4-f]p u r in e (14f): yield 80%, mp
298-300 °C (dioxane), IR(KBr) 1712, 1670 cm-1 1H NMR
,
(DMSO-d6) δ 3.38 (s, 3H, N3-Me), 3.58 (s, 3H, N1-Me), 7.35-
8.30 (m, 10H, ArH); MS m/z 400 (M+, 66.1%), 401 (M+ +1,
44.6%). Anal. Calcd for C21H16N6O3: C, 63.00; H, 4.03; N, 20.99.
Found: C, 63.0; H, 4.1; N, 21.1.
1,3-Dim et h yl-2,4-d ioxo-6-b en zoyl-8-(p -m et h ylp h en yl)-
1,2,3,4-tetr a h yd r o[1,2,4]tr ia zolo[3,4-f]p u r in e (14g): yield
80%, mp 236-238 °C (AcOH/H2O), IR(KBr) 1713,1666 cm-1, 1H
NMR (CDCl3) δ 2.44 (s, 3H, CH3), 3.46 (s, 3H, N3-CH3), 3.71 (s,
3H, N1-CH3), 7.20-8.40 (m, 9H, ArH); MS m/z 413 (M+-1,
0.4%), 415 (M+ + 2, 34%). Anal. Calcd for C22H18N6O3: C, 63.76;
H, 4.38; N, 20.28. Found: C, 64.0; H, 4.1; N, 20.1.
(19) Ueda, T.; Oh, R.; Nagai, S.; Sakakibora, J . J . Heterocycl. Chem.
1998, 35, 135.
(20) DaSettimo, A.; Primofiore, G.; Cristina, B. M.; DaSettimo, F.;
Marini, A. M. J . Heterocycl. Chem. 1995, 32, 941.
(21) Shawali, A. S.; Gomha, S. M. J . Prakt. Chem. 2000, 342, 599
and the references therein.