v-Triazolo5,4-dpyrimidines (8-azapurines). Part 20. 1-Alkyl derivatives

摘要

1978 513 v-Triazolo5,4-dpyrimidines (8-Azapurines). Part 20.l 1-Alkyl Deriva-tives By Adrien Albert, Department of Pharmacological Sciences, Health Sciences Center, State University of New York at Stony Brook, New York 11794, U.S.A. 9 -Be nzy I-1-met h yI-8-azapu ri n -6 -one (2 b ) was made (a) by heating 4-amino-3-benzyI-1,2,3 -t riazoIe -5-(N-rnethylcarboxamide) (7a) with formamide, and (41). quantitatively, by methylating 9-benzyl-8-azapurin-6-one. It was debenzylated to 1 -methyl-8-azapurin-6-one (2a). which was also prepared by condensing 4-amino-l.2.3- triazole-5- (N-methylcarboxamide) (7c) with forrnamide. The i.r. spectrum of compound (2a) indicates an un- usually dipolar nature. 1 -Butyl-8-azapurin-6-one and its 9-benzyl derivative were synthesized analogously.1 -Methyl-8-azapurine-6-thione (10).prepared from the 6-oxo-analogue (2a). resisted further changes. Attempted chlorination of the oxo-compound (2a), with thionyl chloride and dimethylformamide, produced what appears to be the pyrimidine (1 2). 4-Amino-5-arninomethyI-3-benzyl-l.2.3-triazole was mono- and bis-trifluoroacetylated, and the former product (1 3c) was N-methylated then deacylated to 4-amino-3-benzyl-5-rnethylaminomethyl-l.2.3-triazole (1 3f). The latter was condensed with formamidine to 9-benzyl-1.6-dihydro-1 -methyl-8-azapurine (1 1 b). 1-Methyl-8-aza-purine (3) was made by nitrosating 5-amino-1.4-dihydro-4-imino-1 -methylpyrimidine hydrochloride (1 6). Com-pound (3) was found to be about 15 covalently hydrated at equilibrium in D,O whereas the cation was completely hydrated (n.m.r.and U.V. evidence). The physical properties in which the neutral species differs most from its known isomers are m.p.. volatility, and extractability from water, in all of which it behaves as a much more polarized substance. ALTHOUGHmany derivatives of 8-azapurine (1) methyl-ated in the 7-, 8-, or 9-position are known, very few 1- alkylated analogues have been reported. Medical interest in 8-azapurines as anticancer drugs2 and for the prevention of anaphylactic shock prompted this exploration of the synthesis and properties of l-alkyl derivatives such as (2a) and (3). The known 1-alkyl derivatives are (a) 2-amino-l- methyl-8-azapurin-6-one (' l-methyl-8-azaguanine ') (4), made by the action of nitrous acid on 2,4,5-triamino-l- methylpyrimidin-6-one, and (b) a set of five 1,g-dihydro- 6-imino-2-methyl-8-azapurines(5) (four of them have another ,V-methyl substituent in the triazole ring) pre-pared by the action of methylamine on the corresponding 5-cyano-4-e t hoxymeth yleneamino-derivat ives of 1,2,3-triazole.Unfortunately these imines proved unsuitable for the present project because they rapidly isomerized in cold alkali to 6-methylamino-8-azapurines (by a Dimroth rearrangement), and dilute acid brought about ring opening to triazole-5-carbo~amidines.~However, the simplest example (5a) still held some promise because its basic strength (pKa 3.25), more than a thousand times less5 than that of the 9-methyl homologue (5b) t In this series, the amino group of aminotriazoles is consistent-ly numbered 4 to facilitate comparisons.Part 19, A. AlbertandC. J. Lin, J.C.S. PerkinI, 1977, 1819. G. Rrult5, S. J. Eckhardt, T. C. Hall, and A. Winkler, ' DrugTherapy of Cancer,' World Health Organization, Geneva, 1973, pp. 46, 114, 132; L. L. Bennett, M. H. Vail, P. W. Allan, and W. I. Laster, Cancer Research, 1973, 33, 465. C. J. Coulson, K.E. Ford, S. Marshall, J. L. Walker, K. R. H. Wooldridge, K. Rowden, and T. J. Coombes, Nature, 1977, 285, 545. (pK, 6.84), indicated a different structure, namely the 6-amino derivative of structure (3),in resonance with the zwitterion (6). However, attempts to deaminate it by standard methods and by a new procedure (pentyl nitrite in tetrahydrofuran 6) were fruitless, hence starting materials were sought among suitably sub- sti tuted 4-amino- 1,2,3-t riazoles.An authentic specimen of 9-benzyl-1-methyl-8-aza-purin-6-one (2b) was made by heating 4-amino-3-benzyl- 1,2,3- triazole-5- (N-methylcarboxamide) (7a) with formamide. This triazole was obtained in two ways: (a)by the action of methylamine on 4-amino-3-benzyl-5- (methylthio)carbonyl-1,2,3-triazole(8a)by an improved method which avoids use of a sealed tube, and (b) by heating benzyl azide with N-methylcyanoacetamide in methanolic sodium methoxide.8 It was then found that the cold methylation of 9-benzyl-8-azapurin-6-one, readily prepared in two steps from benzyl azide and cyan~acetamide,~gave the l-methyl derivative (2b) almost quantitatively and with no detectable amount of the 3-methyl isomer which was on hand from another project .9-Benzyl-l-methyl-8-azapurin-6-one(2b) was rapidly C. W. Noell, L. B. Townsend, and R. K. Robins, Synth. Proc. Nucleic Acid Chem., 1968, 1, 44. A. Albert, J.C.S. Perkin I, 1973, 2659; 1974, 2030. 6 J. Cadogan and G. Moliner, J.C.S. Perkin I, 1973, 541. 'I A. Albert, J. Chem. SOC. (C), 1969, 2379. May and Baker Ltd., Brit. Pat. 1,338, 235/1973, U.S. Pat. 3, 819,631/1974.* (a) J. R. E. Hoover and A. R. Day, J. Amer. Chern. Soc., 1966, 78, 5832; (b) A. Dornow and J. Helberg, Chem. Ber., 1960, 93, 2001; (c) A. Albert, J. Chem. SOC. (C),1969, 152. J.C.S.Perkin I decomposed by boiling aqueous alkali to 3-benzyl-4-singlet at 2.70 (5 H) and another at 4.63 (2 H) were formamido-l,2,3-triazole-5-(N-methylcarboxamide)(7b) obviously derived from a non-eliminated benzyl group.and the deformylated analogue (7a). 9-Benzyl-l- A quartet at 7.22 (3 H) (J 4.40), and a doublet at 7.33 butyl-8-azapurin-6-one, a homologue of (2b), was made (3 H) (J 4.80), both signals changing to singlets in D,O, by heating formamide and 4-amino-3-benzyl-l,2,3-were allocated to N=CH-NHMe and CONHn'Ie respect- triazole-5-(N-butylcarboxamide),which was obtained ively. This evidence combined with the i.r. spectrum 0 (2) a; R=H b; R=CHzPh (5 1 a; R=H b; R=Me N51co*NHMe'N N:CH.NHMe R CHzPh (71 (81 a; R' = CHzPh, R2 =H .a;R = CHzPh b; R' = CHzPh, RZ =CHO b; R = H from butylamine and the (methy1thio)carbonyl deriv- and elemental analysis established the decomposition ative @a).product as 3-benzyl-4-methylaminomethyleneamino-l,2,-Although some cyclic tertiary amides can be deoxygen- 3-triazole-5-(N-methylcarboxamide) '(9), TI11s was con- ated, for example N-alkylacridones are reducible to N-firmed by heating the product at 210 "C for 30 min, alkylacridans by zinc and hydrochloric acid ,lo9-benzyl-which converted it quantitatively into 9-benzyl-l-l-methyl-8-azapurin-6-one and its 1-butyl homologue methyl-8-azapurin-6-one (2b). Several dimethyl-were partly unchanged but mainly destroyed by this aminomethyleneamino-1,2,3-triazolesare known,l1 but procedure, whereas zinc in boiling 9O:h acetic acid did this is the first monoalkyl analogue to be reported.not affect them. Lithium borohydride in cold tetra- The elusive l-niethyl-8-azapurin-6-one (2a) was eventu- hydrofuran, and also sodium diethyldihydroaluminate, ally made in two ways: (a) by condensing 4-amino- were destructive. 1,2,3-t riazole-5- (N-me t h ylcarboxaniide) (7c) with form- In attempted debenzylation of 9-benzyl-l-methyl-8- amide the triazole was prepared 'from the corresponding azapurin-6-one with sodium in liquefied inethylamine 5-(methylthio)carbonyltriazole(8b)l; (b) by hydrogen- (the purinone being insoluble in liquid ammonia), an olytic debenzylation of 9-benzyl-l-methyl-8-azapurin-6-unexpected product, of niolecular weight 272, was one (2b) over palladium, The latter reaction, reluctant formed in high yield.The lH n.m.r. spectrum (solvent under usual conditions, succeeded in hot butanol-acetic dimethyl sulphoxide) showed a quartet at T 1.15 (J acid. The product was soluble in cold x-hydrochloric 4.40 Hz) (singlet in deuterium oxide), assigned to N=CH lo K. Lehmstedt and H. Hundertmark, Ber., 1931, 64, 2386.coupled to CH-NH. A broad singlet at 1.90 (2 H), l1 A. Albert, J.C.S. Perkin I, 1972, 461; A. Albert and H. eliminated in D,O, was assigned to two NH groups. A Taguchi, ibid.,1973, 2037. acid and x-sodium hydroxide, in both of which slow decomposition occurred, mainly to the methylcarbox- amide (7c). The i.r. spectrum of l-methyl-8-azapurin-6-one is remarkably different from those of the 7-methy1,12 8-methyl,13 and 9-methyl l3 isomers (all Nujol mulls) in having strong absorption in the 2 595 cm-1 area (parti- cularly.evident in a hexachlorobutadiene mull), and (10) (11) 515 3 015s (NH),1685s (amide I), 1 560s (amide II), 1 345m, 1315m, 1245s (amide 111),and 1025m c11i-I. 1-hfethyl-8-azapurine-6-t hione (10) was reaclily pre- pared by the action of phosphorus pentasulpliide in boiling pyridine on 1-methyl-13-a/,apurin-6-01~~~(2a) All attempts to desulphurize it to compound (3) or (lla), with Kaney nickel, failed: mild conditions left it unchanged whereas a more vigorous attack completely CI a; R=H b; R=CHtPh N;iCH 2*NR' RZ bsol; NHR3 CH,Ph CHtPh (13) (14 1 a; R'= R2 =R3= H b; R'= CHO, RZ= R3=H C; R'=C0.CF3,R2=R3=H d; R'= R3 =CO*CF3, R2 = H e ; R1= Me, R2 =CO* CF3,R3= H f; R1.=Me,F$=R3=H N?amp;JN H two well defined peaks at 1 945 and 1 815 cm-l which, although weak, are prominent because of the lack of general absorption in this region.It is interesting that two related compounds, 9-benzyl-l-methyl- and -1-butyl-S-azapurin-6-one, show absorption near 1950 cm-l, but not the other anomalies (re-examination of the parent, 9-benzyl-8-azapurin-6-one,revealed a faint peak at 1 945 cm-l overlooked in the original report).9C This extra absorption in the spectrum of l-methyl-8-azapurin-amp;one is taken as evidence of a strongly dipolar nature. The related 1-methylpurine-6-one has strong absorption in the 2 800-2 500 cm-l area.l* In other ways, the spectrum of l-metliyl-8-azapurin-6-one closely resembles those of its isomers, having bands at 3 150m, l2 A.Albert and K. Tratt, J. Chem. Soc. (C),1968, 344. l3 A. Albert, J. Chem. SOC.(C), 1968, 2076. l4 J. A. Montgomery and H. J. Thomas, J. 0i.g. Chert.,1965,30, 3235. l6 W. Traube and F. Winter, Arch. PAarm., 1906, 244, 11; W. Traube, Annalen, 1904, 331, 64. destroyed it. Equally unsuccessful were the use of hydrogen peroxide,15 nitrous acid,15 iodine,16 and aluminium amalgam. l-Methyl-8-azapurine-6-tl1ione (10) was destroyed by iodomethane in cold aqueous sodium hydroxide, a procedure that converted the 7-, 8-, and 9-isomers into the corresponding 6-~ilethylthio- compounds ;9c912913 the use of iodomethane with yotas- sium carbonate in cold dimethylformamide also failed.Conversion of 1-methyl-8-azapurin-G-one into 6-chloro-l-methyl-8-azapurineusing triphenylphosphine in carbon tetrachloride l7 proved unsuccessful, as did the use of thionyl chloride, catalysed by dimethylform- amide,ls in boiling chloroform. The latter procedure had proved useful l3 for preparing 6-chloro-S-methyl- 8-azapurine, of which the l-methyl isomer may have l6 I. L. Doerr, I. Wenipen, D. A. Clark, and J. J Fox, J. Org.Chew, 1961, 26, 3401. l7 J. B. Lee and T. J. Nolan, Canad. J. Chpin , 1966, 44, 1331. H. H. Bosshard, R. Mory, IM.Schniid, and H. Zollinger, Helv. Chim. Acta, 1959, 42, 1653. been formed here but underwent further attack.Most surprisingly, the product seems to be 6-chloro-4-formyl- imino-1-met hyl-5-dimet hylamino- 1,6-dihydropyrim-idine (12),on the basis of the elemental analysis (with its unexpectedly high C : N ratio), molecular weight, and lH n.m.r. and i.r. spectra. With one exception,lS the reported decompositions of 8-azapurines have produced 1,2,3-triazoles (summarized in ref. l), but 2-amino-8- azapurin-amp;one gave 2,4,5-triaminopyrimidin-6-onewith hot, dilute acid. In the present example, the reagents, or their product (chloromethylenedimethylammonium chloride 20), may attack at N-9; this is followed by elimination of a molecule of nitrogen from N-7 and N-8 and converted attack of Me2N on the 5-position of the pyrimidine ring. Further work is planned to test the constitution (12).A different approach to l-methyl-8-azapurines with no oxygen atom in the 6-position was then made through J.C.S. Perkin I ation, both by sodium in ammonia and by hydrogen- ation over palladium. When conditions were forced, as with hydrogenation in boiling butanol, the molecule was fragmented. In an attempt to obtain the compound (1lb) by another route, 9-benzyl-l,6-dihydro-8-aza-purine 22 was stirred with iodomethane and potassium carbonate in dimethylformamide. Even when less than one molecular proportion of iodomethane was used, the sole product was a dimethyl derivative, 9-benzyl- 1,6-dihydro-x,y-dimethy1-8-azapuriniumiodide (pending the results of X-ray crystallography, x and y are tenta- tively assigned as 1 and 3).l-Methyl-8-azapurine (3) proved more accessible from pyrimidine intermediates than from triazoles. 5-Amino-1,4-dihydr0-4-imino-l-methylpyrimidine hydrochloride (16), obtained by methylating 4,5-diarnin0pyrimidine,~3 reacted with propyl nitrite in propanol to give the hydrochloride of 1,6-di hydro-1 -met hyl-6-propoxy-8-aza- derivatives of 4-amino-5-aminomethyl-3-benzyl-l,2,3-purine (17), which is the propanol adduct of the cation triazole (13a). The latter can be cleanly monoformyl- ated on the 5-aminomethyl group without affecting the less basic 4-amino group.21 Unfortunately, the 5-formamido proton in the product (13b) proved in-sufficiently activated for replacement by a methyl group.In a search for the more reactive 5-trifluoro- acetamido analogue (13c), the initial difficulty in finding conditions to favour monoacylation were overcome by the use of trifluoroacetic anhydride in cold trifluoroacetic acid. The deliberate preparation of the diacyl deri- vative (13d) is also described in the Experimental section. The insolubility of the monoacyl compound in cold N-hydrochloric acid confirmed the position of acylation, and the solubility in cold h:-sodium hydroxide bore witness to the mobility of the neighbouring proton, The trifluoroacetamido compound (13c) gave a good of the goal (3), and has a main U.V. absorption peak displaced from that of l-methyl-8-azapurines, as ex-pected, to a shorter wavelength near to that of the cor- responding peak of the (hydrated) cation of the parent (3)* Silver carbonate converted this adduct into l-methyl- 8-azapurine which differed little from its 7-niethy1,12 8-methy1,13 and 9-methy1,12*24 isomers in pK, and U.V.or n.m.r. spectrum. The n.m.r. spectrum (in D20) showed weak, highfield signals which denote equilibrium with about 15 of the covalent hydrate as in methyl- 8-azapurine.13 The cation has a U.V. spectrum displaced to much shorter wavelengths, showing that it is entirely hydrated, as are the cations of the 7-and methyl isomers. Nevertheless 1-methyl-8-azapurine has a much more polar character than its isomers, as indicated by the yield of 4-amino-3-benzyl-5-(N-methyltrifluoroacet-higher m.p.(235"; cf. amidomethyl)-l,2,3-triazole (13e) with iodomethane, and this product was cleanly deacylated to 4-amino-3- benzyl-5-methylaminomethyl-1,2,3-triazole(13f) (puri- fied as the phosphate) by brief heating in aqueous sodium hydroxide. In an application of the new general method22 for synthesizing 1,6-dihydro-8-azapurines,the methyl-aminomethyltriazole (13f) was condensed with form- amidinium acetate to give an excellent yield of 9-benzyl-l,6-dihydro-l-methyl-8-azapurine(11 b). The re- tention of the N-methyl group gives the first indication that this reaction proceeds through a tetrahedral intermediate (14) which, by acquiring a proton and eject- ing an ammonium ion, produces the final intermediate (15).The new dihydro-8-azapurine (1lb) was stable to aerial oxidation.Unfortunately, it resisted debenzyl- 19 Y. Hirata, K. Iwashita, and K. Teshima, Nagoya SangjtoKuguku, 1957, No. 9, 83 (Chem.Abs., 1957 1957, 51, 12074). 2o H. H. Bosshard and H. Zollinger, Helv. Chim. Acta, 1959, 42, 1659. 21 A. Albert, J.C.S. Perkin I, 1973, 1634. 22 A. Albert, J.C.S. Perkin I, 1976, 291. 23 D. J. Brown and N. W. Jacobsen, J. Chem. SOC.,1962, 3172. 167, 153, and 88" for the 7-, 8-, and 9-methyl isomers respectively), and by its in- ability to be sublimed (even at 200 "C and 0.03 mmHg), or extracted from water by dichloromethane, whereas the three known isomers are easily sublimed and extract- able.25 Moreover, it is much less stable than the three isomers.The polar characteristics are attributed to a high proportion of the charged canonical form (18) in the resonance hybrid. Comparison with 1-methylpurine 26 (vis-a-vis the latter's 7- and 9-methyl isomers 27) shows few parallels. There is a similar increase in m.p. for the l-methylpurine isomer, but a U.V. spectral shift of about 10 nm to longer wavelengths, in both neutral species and cation, signifies a change in conjugation but no covalent hydration. No marked differences in solubility are evident from the literature. l-Ethyl-purine is significantly (about 2.6 pK units) more basic 24 A. Albert, J. Chem. SOC.(B),1966, 427. 25 A. Albert, W. Pfleiderer, and D. Thacker, J. Chem. SOC.(C),1969, 1084. 26 L. B. Townsend and R. K.Robins, J. Org. Chem., 1962, 27, 990. 27 A. Bendich, P. J. Russell, and J. J. Fox, J. Amer. Chem. SOC.,1954, 76,6073; Fischer, E., Bey., 1898, 31, 2550; A. Albert and D. J. Brown, J. Chem. SOC.,1954, 2060. 1978 than the 7-and 9-ethyl isomers.28 Calculation of charge distribution (CNDOmethod) 29 indicated that the pyrimi- dine ring is positively charged in l-methylpurine, but negative in the 7-and 9-methyl isomers. EXPERIMENTAL Determinations of physical constants and establishment of chemical identity were made essentially as in Part 19.l The mass spectra were obtained with a Hewlett-Packard 5983 A instrument. Most of the elemental analyses were performed by Galbraith Laboratories, Tennessee, and the others by the Australian National University's Analytical Service, in Canberra.9-Benzyl-l-methyl-8-azapurin-6-one(2b) (3-Benzyl-3,6-di- hydro-6-metl~yl-v-triazolo4,5-dpyrimidin-7-one).-(a) 4-Amino-3-benzyl- 1,2,3-triazole-5- (N-methylcarboxamide) (see following) (2.3 g, 0.01 mol) and formamide (20 ml) were heated at 190 "C (bath temp.) in an open vessel for 1 h. The volatile materials were removed at 150 "C and 25 mmHg. The solid residue was rubbed with water (10 ml), filtered off, dried, and recrystallized from 60 parts of benzene-ethanol ( 1 : l), yielding 62 of 9-benzyl- l-methyl- 8-azu~urin-6-one, m.p. 221.5". It also crystallized well from 13 parts of 2-methoxyethanol but was poorly soluble in boiling ethanol or water (Found: C, 59.9; H, 4.5; N, 29.2.C12H11N,0 requires C, 59.7; H, 4.6; N, 29.0y0), T (CD,),SO 1.47 (1 H, H-2), 2.70 (5 H, Ph), 4.27 (2 H, CH,), and 6.53 (3 H, CH,), vmax. 1 950w, 1 700br s (C : 0 str), 1555m, 1 325m, 1 270m, and 1200m cm-l. (b) Preferred method. 9-Benzyl-8-azapurin-6-one (2.27 g, 0.01 rnol), dimethylformamide (28 ml), potassium carbonate (Aanie-dried and finely powdered; 2.8 g, 4 equiv.), and iodomethane (4.3 g, 3 equiv.) were stirred at 24 "C for 24 11. Volatile materials were removed at 110 "C and 25 mmHg. Water (20 ml) was added to the residue and 9-benzyl-l-methyl-8-azapurin-6-one,m.p. 22 lo, was filtered off in 94 yield after washing with ethanol and drying at 110 "C (identical with authentic material). 4-Anzino-3-benzyl-1,2,3-triazole-5-(N-methylcarboxarnide)-(7a).-4-Amino-3-benzyl-5-(methylthio)carbonyl-1,2,3-triazole (8a) (1.0 g, 0.004 mol), in fine powder, was stirred with ethanolic 35 methylamine (20 ml; Fluka) for 45 h.The solution was taken to dryness at 50 "C, giving 94 of this amide, ni.p. 155" (lit.,7 155"), from 8 parts of ethanol. Hydrolysis of 9-Benzyl- l-methyl-8-azapurin- 6-one.-This azapurinone (0.241 g, 0.001 mol) and N-potassium hydroxide (2.5 nil) were heated under reflux for 5 min; the mixture was then refrigerated and filtered. The precipitate was pure 4-amino-3-benzyl- 1,2,3-triazole-5- (N-methylcarboxamide) (7a) (50y0),m.p. 155" (see fore- going). The filtrate, adjusted to pH 2.5 with sulphuric acid gave a white precipitate of 3-benzyl-4- formamido- 1,2,3-triazole-5-(N-methylcarboxamide) (7b), m.p.138", from 19 parts of water and 8 parts of 95 ethanol (40 yield) (Found : C, 55.6; H, 5.1; N, 27.0. Cl,Hl,N,O, requires C, 55.6; 5.05; X, 27.0y0),M 231 other prominent signals at m/e 200, 199, 173, 171, 145, and 91 (benzyl) this spectrum is identical with that for the deformylated analogue (7a)l. 4-Amino-3-benzyl-1,2,3-triazole-5-(N-butylcarboxamide).-4-Amino- 3-benzyl-5- (methylthio) carbonyll- 1,2,3-triazole (8a) (0.62 g, 0.0025 mol) and butylamine (5 ml, 20 equiv.) were refluxed for 2 h. Excess of amine was removed (oil- bath at 100 "C and 25 mmHg). The residue, recrystallized from a little ethanol (2 crops) gave the butylcarboxamide almost quantitatively, m.p. 150" (Found: C, 61.2; H, 6.8; N, 25.7. C,,H,,N,O requires C, 61.5; HI 7.0; N, 25.6), v,,,,.3 350, 3 290, 3 230, 3 180 (all m, NH str.), 1 645 1630s (amide I band, free and assoc.), and 1540s cm-l (amide 11), insoluble in cold N-NaOH and -KOH (test confirmation that no Dimroth rearrangement of benzyl group has occurred). 9-Benzyl- 1 -butyl-8-azapurin- 6-one .-The foregoing amide (0.274 g, 0.001 mol) and formamide (2 ml) were heated at 225 "C (bath) for 1 h in an open vessel. Addition of water (4 ml), chilling, and filtering produced 9-benzyl- l-butyZ-8- uzapurin-6-one (goyo),m.p. 99", from 65 parts of 33 ethanol (Found, for material dried at 80" in air: C, 63.4; H, 5.9; N, 25.0. C1,H1,N,O requires C, 63.6; H, 6.05; N, 24.7), vmax. 1960w, 1695s (C:O), 1560m, 1 Om, 1270m, 1 185m, and 800 m cm-l, T (CD,),SO 1.38 (1 HI H-2), 2.66 (5 H, Ph), 4.26 (2 H, CH,), and 5.97 (2 HI t), 8.5 (4H, m), and 9.06 (3 H, t) (all centres, Bu), A,, (EtOH) 259 nm (log E 3.90).3-Benzyl-4-methylaminomethyleneamino-1,2,3-triazole-5-(N-methylcarboxamide) (9) (with A. M. TROTTER).-9-Benzyl-l-niethyl-8-azapurin-6-one(0.24 g, 0.001 mol) was stirred with liquefied methylamine (10 ml) while sodium (0.046 g) was added. Evaporation of the methylamine, and recrystallization of the residue from 17 parts of ethanol, gave the carboxamide (75y0), m.p. 186" (Found: C, 57.2; H, 5.9; N, 30.7. C13H1,N60 requires C, 57.3; H, 5.9; N, 30.9y0), vmax. 3 225m (NH), 1635br, s (CO), 1540br, s, 1 410m, 1 260m, and 1 200m cm-l, M+272 other prominent signals at m/e 243, 212, 186, 171, 91 (benzyl), and 691; lH n.m.r.data in main text; insoluble in cold N-sodium hydroxide and N-formic acid ; soluble in N-hydrochloric acid. l-Methyl-8-azapuvin-6-one (2a) (3,6-Dihydro-6-methyZ-v-triazolo4,5-dpyrirnidin-7-one).-(a)By debenzylation (with A. M. TROTTER). 9-Benzyl- l-methyl-8-azapurin-6-one (0.482 g, 0.002 mol), dissolved in butanol (20 ml) and acetic acid (2 ml), was hydrogenated over pre-reduced palladium-carbon (10; 0.08 g) at 117 "C and atmo-spheric pressure for 1.5 h. Without prior filtration, the suspension was dried at 90 "C and 25 mmHg. The residue, well cooled, was stirred with 0.25~-sodium hydroxide (12 ml) and kieselguhr (0.1 g), and rapidly filtered.The filtrate, adjusted to pH 5.5 with acetic acid, and concentrated at 35 "C (to 4 ml), was cooled, and acidified to pH 3.5 (with ~N-H,SO,). Refrigeration yielded l-methyl-8-azapurin-6-one (72y0), m.p. 253" (with slight effervescence), from 6 parts of water or 65 parts of 90 ethanol (Found: C, 39.8; H, 3.4; N, 46.2. C,H,N,O requires C, 39.75; H. 3.3; N, 46.3y0), T (CD,),SO 1.59 (1 H, H-2) and 6.49 (3 H, Me). (b) By ring closure of a triazole. 4-Amino-1,2,3-triazole-5-(N-methylcarboxamide) (7c) (3.53 g, 0.025 mol) and formamide (50 nil) were heated in an open vessel at 195 "C (bath) for 45 min. Excess of reagent was removed at 160 "C and 25 mmHg, and the residue, recrystallized from a little water, gave l-methyl-8-azapurin-6-one (2a) (78y0), m.p.253". l-~Wethyl-8-azapurine-6-thione( 10) .-Phosphorus penta-sulphide (0.22 g) was added to a hot solution of l-methyl- 8-azapurin-6-one (0.15 g, 0.001 mol) in dried pyridine (2 ml) and the whole was heated under reflux for 4 h. Water 28 R. W. Balsiger, A. L. Fikes, T. P. Johnston, and J. A. Mont-gomery, J. Org. Chcna., 1961, 26, 3446. 2n 2.Neiman, Experientia. 1975, 31,996. 518 (1.5 nil) xis ::{'~l~l,and the velatile materials were removed in 7incuo at 50 ^C. ITT:tter(1.5ml) w-as again added and tlie pH arljiiitctl. bsol;bsol;,?ic~nnecessary, to 3.5--.4,5,Chilling ant1 filtration ::iibsol;-' I -methyZ S-nza~.uvi?ze-6-tiiione (83?/,), n1.p. about 23-0' {bI~~ci-x?~s)~-1ienintroduced at 230" (from 140 parts of boilin;: water or i5 parts of 900,amp;ethanol); soluble in col(l s-sdiii!?i Ir;+droside(I~ountl:C, 35.9; H, 3.1; N, 41.6. C,EI,S,S requires C, 35.9; Id, 3.0; N, 41.90;).It was not cbsol;-en partly isomerized to a tliiadiazolopyrimidirle (the ('hristmns r-carr~.ngement,given by analogues gc) when heated undcr rttflus nTith hutanol for 1 11. Chlori II niton c,f I -,II ethyl-8-nzaputin-6-one ( 2a).-Tli ionq'1 ct!oriLtc? (3 8 nil, 0.05 mol), climeth~lforriianiide(1 .O ml, 0.014 niol), an(1 1 -1iietli~l-8-azapurin-G-one(1.51 g, 0.01 niol), sii~pcn~l(~~1i?? c!ilorofor:n (40 ml), were heated under reflux for 4 I1 (the niistiiri: became clear after 20 niin and bega.11 to rlciposit material 10 niin later. The suspension was left ;it ---LO 'Y.overnight, and filtered. 'The solid, dissolbsol;-eil ii? u.:it('r (i iiil), was adjusted to pH 10 with ammonia nnrl s1inl;en out-n-ith chloroform (2 x 15 nil). r.I he lobsol;ver labsol;-crs bsol;bsol;-(:re bulked, dried (K,CO,), and evaporated -.in i'nruo. I lie residue, recrystallized from 10 parts of benzene---c!.clolicsane ( 1 : 1), gave presumed 6-chloro-4-forin-vliiniiio- 1 -methyl- ;',--clintet/llllas12ilzo-1,6-dihydrop:,wimi-dine (12) (410,6), m.p. 125.5". Sodium carbonate coiild replace the ammonia without much loss of yield (Found : C, 45.0; H, 5.3; C1, 16.7; 9,26.3. C,H,,ClN,O requires C, 44.8; H, 5.2; C1, 16.5; S, 26.10.:)); Afr 214 ("Cl) and 216 (37Cl) (other prominent peaks at tn/e 199, 186, 170, 172, 170, 163, 138, 5i, and 42), -(D,O) 2.01 (1 H, CHO), 219 (1 H, H-2), 6.49 (3 H, l-RIe), and 6.98 (6 H, NVle,j, v,,,:~~.(Xujol) I ti50l)r,s (CO str), 1 585br.s, 1 410m, 1 330m, 1 lOOni, 955s, a:icl 780111 c~ii-~(Cl-C str.) 4-'4 iizino-3-benz.vl-5-frifluovoacetanzidonlethyl-1,2,3-triazole (1 3c). -4-:~niiiio-5-aniinonietliyl-3-benzyl-1,2,3-triazole (13a) lY(2.30 g, 0.01 niol) was dissolved in trifluoroacetic acid (15 nil) 7.vith cooling to 24 "C. Trifliioroacetic anhy- dride (2.4 g; 15"; excess) was added, and the solution set aside at 24 "C for 8 11. The volatile portion was removed in vcccuo at 35 "C, aid tlie residue recrystallized twice from 2,576 ethanol (I5 in!, then 80 nil) giving the title compound (679;), 1n.p. 188" (Fcxmd: C, 48.3; H, 4.0; I:, 19.0. X, 23.2. C,,H,,F,S,O requires C, 48.15; H, 4.0; I?, 19.0; N, 23.4(;/;).It can also be recrystallized from 300 parts of water or 190 parts of benzene. 3-tZepzz),l-4-t~i~t1c,voaceta~~~icZo-5-tr~~uoroaceta~~aidornetliyl-1,2,3-tviazok ( 13tl).---4-Amino-5-aniinoniethyl-3-benzy-1,2,3-triazolc (0.203 g, 0.001 mol) was rubbed with trifluoroacetic anhydride (1.6 g) until dissolved, then tlie solution was set aside at 21 "C for 24 11. Tlie thick paste was taken to dryness at 40 "C gibsol;.ing the product (850,:,), m.p. 149" from 4.5 parts of benzene (Founcl: C, 42.6; H, 2.8; euro;7, 28.9; Zi, 17.4. Cl4HllF6K5O2requires C, 42.5; H, 2.8; I;, 28.8; S, 17.7(:~). J.C.S. Perkin I 4.5;S, 28.4. Cl3Hl4F3X5O requires C, 49.8; H, 4.5; N, 22.4o.b). 4--4.ritiiin~-3-ben~~~1-5-)ltetliyla~~ii~ioi1~etliyl-1,2,3-triazoZe (1 3f).----Thet rifluoroacetamido-deribsol;-atibsol;.-e (13e) (0.42 g, 0.001 34 mol) and s-sodiuni liq'droxide (2.1 nil, 1.5 equiv.) were boiled for 30 s; tlie mixture was then quickly cooled and sli;iken out u-ith cliloroform (2 x 7 nil). The chloroform lajw was dried (K,C03) and taken to dryness, in vacuo, at eventually 55 "C. phosphoric acid was added until the yH fell to 8. The addition of acetone (7.5 ml) initiated precipitation (completed at -10 "C overnight) of bis-(4-aini~ao-3-benz~~l-5-~izet/iylanzinonzetliyl-1,2,3-tviazolium) hydvogen p/zosphate (840/,), m.p. 180" from 150 parts of ethanol Found: C, 49.5; 13, 6.3; X, 26.4. (CllH15N5)2,-H,J?O, requires C, 49.6; H, 6.25; X,26.37',.The base was liberated by shaking the phosphate (1.25 g) with 2~- sodium hydroxide (2.5 nil) and chloroform (3 x 15 ml); 071; recovery; m.p. 93" (from benzene). The quanti- tative extraction sl~nvsthat no Iliniroth rearrangement to the acidic 4-benzylaniino isonier liad taken place during boiling with alkali. I)-Beizz~~l-l,G--dihycZro-l-i~zethyl-t)-a~~~p~ri~ze (3-(1 lb) Benzyl-6,7-dd1Jidv0-6-nzetliy1-v-tviazolo 4,5-dpyrimidine .-4-A1iiirio-3-benzyl-5-1iietli~lanii1ioniet~iyl-1,2,3-triazolium phosphate (0.266 g, equiv. to 0.001 11101 of base), formamidi- nium acetate (0.21 g, 2 equiv.), and sieve-dried butanol (3 nil) bsol;yere heated under reflux for 2 h. More forniamidinium acetate (0.21 g) was added and refluxing was continued for 2 h longer.The volatile portion was removed at 90 "C and 25 mmHg, and the residue quickly boiled with water (1.5 nil) ant1 refrigerated (the pH at this stage niust be main- tained above 4 to avoid loss as a soluble salt). Filtration gave 9-benzyl-1,6-di~iydro-l-nzetliyl-8-azapurine (80), m.p. IOG", from 23 parts of water. It was very soluble in cold benzene, but only slightly in boiling cyclohexane (Found : C, 63.2; H, 5.6; X, 30.5. C12H13X5requires C, 63.4; H, 5.8; S,30.80,;), illt 227 (other prominent peaks at m/e 198, 157, 108, and YI), T (CI),),SO 2.76 (5 H, Ph), 2.85 (1 H, 14-2); 4.69 (2 H, PliCH2), 5.34 (2 H, A-H,), and 7.08 (3 H, nie). Metliylntion of 9-Benzyl-l16-di/iydvo-8-azapurilze.-Iodo-methane (0.42 g, 3 equiv.) antl flanie-dried potassium carbonate (0.21 g, 3 equiv.) were added to a solution of 9-benzyl-l,6-dihydro-8-azapurine22 (0.213 g, 0.001 mol) in dried dinietliylformamide (2 nil).The suspension was stirred at 24 "C for 48 11, then filtered. The solid was washed with a little ethanol, then suspended in water (1 ml) and filtered off, yielding 9-benzyl-1,6-dihydro-x,y-dimethyl-8-azapuriniuni iodide (500,;), ni.p. 206" (effervesces), from 50 parts of ethanol or 8 parts of water; insoluble in acetone or ethyl acetate (Found: C, 42.3; H, 4.4; N, 19.0. Calc. for C13H161N5: C, 42.6; H, 4.4; K, 19.Oy'), T (CD,),SO 2.46 (1 H, H-z), 2.66 (5 H, rq, 4.47 (2 H, P~~cH,), 5.12 (2 H, 6-H,), antl 5.98 and 6.95 (eacli 3 H, 2 x Me). l-flIethy1-8-azapurine (6-;l.Iefliyl-~~-triazolo4,5-dpyrimi-4-A)nino-3-benzyl-5-(S-~~zet/i)Iltvi,fEuol.oaceta~~~ido~~zet~i.~il)-1,2,3-triazoZe ( 13e).-4-.Aniino-3-benzyl-5-trifluoroacet-dit2e) (with L).THACKI.;R)(3). --5-Aniino-l,4-dihydro- atnidomethyl- 1,2,3-triazole (0.30 g, 0.001 mol) dissolved in 4-imino- l-nietliylp~riniidine hytlrochloride 23 ( 16) (0.161 g, (2 nil) at 24 "C, flame-dried potassiuni 0.001 niol), propj'l nitrite (1.6 nil), and propanol (16 ml) dirnet1~)~lforruamide carbonate (0.21 g, 3 equii..), and iodomethane (0.28 g, 2"(124were stirred at for 4 h. 'The solution was concentrated equiv.) irere stirred for 1.5 11. Tlie volatile portion lvas then renioved at 110 "C and 25 mniHg. Water (4 nil) was added to the bsol;vell-cooletl residue. The precipitate, filtered off at once ant1 bsol;vashetl with much water and a 1itt.le 25;;) ethanol, gabsol;*e the 5-methyl derivative (740,/,), n1.p.142.5" (from about 4 parts of methanol) (Found: C, 49.9; H, in z:ncz~uand clilutecl with light petroleum (b.p. 40-60 "C). The deposited solid, recrystallizetl from propanol-light petroleum, gave 1,6-di/iydro-l-i~zet/i~~l-6-p~opoxy-8-azapurine ( 17) Iiydvoclilovide (6,$-tliliydro-6-niethyl- 7-propoxy-v-tri- azolo4,5-dpyviniinilze Iiydvorlilorzde) (600amp;), m.p. 137" (foams) (Found: C, 41.0; H, 6.0; C1, 15.55; N, 30.5. C,N,,Clrc',O requires C, 41.45; H, 6.1; C1, 15.3; N, 30.2o/b), pl:, 3.24 rk 0.03 (0.000 41~1, in water at 20 "C; analyt. A 250 ntii), j,,iLLx,261 nm (log E 3.94) (in propanol). Tliis liydrocliloride (0.2 g), silver carbonate (0.28 g), and inethanol (3 nil) were stirred overnight. The mixture was then filtered, the filtrate was taken to dryness, and the residue was repeatedly recrystallized from methanol-ether and dried at 140 "C and 0.01 mmHg to give l-methyl-8-az~ifiurim(40(,);)), m.p. about 235" (decomp.) (Found: C, 44.6; H, 4.1; X, 51.3. C,H,N, requires C, 44.45; H, 3.7; N, 51.8deg;(1),T(D,O) (a) peaks integrating to 85, 0.21, 0.88, and 5.57 (cf. 0.31, 0.82, and 5.40 for 8-methyI-8- azapurine 13), (h) peaks integrating to 15, 2.39 and 3.64 (cf. 2.56 and 3.57 for 8-methyl-8-azapurine 13) ; -r(D,O-DC1) 1.50, 3.39, and 6.43, A,,,,, 215 nm (log E 4.31) and 270 nm (3.80) (neutral species in H20 at pH 7.0) or 253 nni (log E 3.96) (hydrated cation at pH 1.0). I thank the staff of the Graduate Chemistry School in this University for accommodation and discussions. I thank Mrs. Clara J. Lin for many of the instrumental nieasure- nlents and Mr. A. M. Trotter for preparative assistance. This investigation was supported by the Xational Cancer Institute, Department of Health, Education, and Welfare, Washington, I. C. 7/1112 Received, 27th June 19771