Lactim Tautomerism in Hydroxypurines

For this particular problem the theory drags somewhat behind experimental evidence, at least insofar as the question of the relative stability of the two forms is concerned. The experimental evidence coming mainly from infrared spectroscopy unambiguously shows that the three isomeric 2-, 6-, and 8-hydroxypurines (29-31) all exist essentially, both in the solid state and in solution, in the 0x0 form, as they all present the characteristic C=0 stretching vibration (near 1670 cih in the 2- and 6-hydroxypurines and near 1740 cm in the 8-hydroxy isomer) and show no band which could be attributed 

Attempts to account for this situation by the CNDO procedure surprisingly encounter difficulties. Fujita, Imamura, and Nagata while having apparently no trouble in finding that the usual amino and lactam forms are the most stable ones in adenine, thymine, cytosine, and uracil, obtain the surprising result that the lactim form of guanine (33) should be about 12 kcal/mole more stable than its 

134 jj. T. Miles aiul J. Frazier, Biochim. Biophys. Acta 79, 216 (1964). 

Unfortunately information about the C-0 and 0-H bond lengths in heterocyclic phenols is very scarce and does not permit a decision. 

Relative Stability (koal/mole) of the Lactim Form WITH Respect to the Lactam Form of Guanine 

If the study of the relative stability of the lactam and lactim forms of hydroxypurines thus escapes a rigorous quantitative treatment, the situation is more agreeable with respect to some other electronic properties of the two forms, which are much less sensitive to the details of the structure and, in particular, to the distance of the 

---

O-H bond. Among such properties a prominent one is the ultraviolet absorption spectrum and the theory may therefore be used for the examination of some of the spectroscopic shifts which accompany the lactam-lactim tautomerization. Much caution must, however, be exercised in this respect. Thus, in a recent paper Kwiatkowski135,137 performed Pariser-Parr-Pople-type calculations on the electronic structure of hydroxypurines, essentially to interpret their ultraviolet spectra. In these calculations he assumed that these compounds exist predominantly in their lactim form, and the results of his calculations, at least for 6- and 8-hydroxypurine, did not seem to contradict this assumption. It is only in the case of the 2-hydroxy isomer that a particularly striking disagreement between theory and experiment led him to admit that this last compound may exist in the lactam form. Calculations carried out for this form gave, in fact, a more satisfactory agreement with experiment.138 As we have seen, unambiguous infrared spectroscopy evidence clearly show s that all three isomers exist essentially in the lactam form. This shows that ultraviolet absorption may provide only very uncertain evidence about the lactam-lactim tautomerism in hydroxypurines and related compounds. 

Whatever be the difficulties in dealing satisfactorily with the problem of the lactam-lactim tautomerism in hydroxypurines, the predominance of the lactam tautomer granted, there remains the problem of the detailed structure of the most probable lactam form for each isomer. The problem is essentially that of the site of location of the imidazole proton. From that point of view forms 34-38 have to be considered for 2-hydroxypurine, forms 39—42 for 6-hydroxypurine (hypoxanthine), and forms 43-45 for 8-hydroxypurine. There are, in addition, some betaine tautomeric forms but these are probably of low stability and will not be considered further. Before describing the results of theoretical calculations, it may be useful to indicate that from the experimental point of view we may, in this respect, turn again for significant evidence to infrared spectroscopy... 

Because of the difficulties encountered in the study of the Iactam-lactim tautomerism in hydroxypurines, difficulties due to the delicate... 

There are various forms of tautomerism which operate in the different purine species. (1) Prototropy which involves attachment of the proton to any one of the four ring nitrogen atoms (Scheme 5). Corresponding CH tautomers, for example (52), seem to be of little significance. (2) Amine-imine tautomerism which operates in the aminopurines such as adenine (Scheme 6). (3) Lactam-lactim tautomerism as in the hydroxypurines such as hypoxanthine (Scheme 7) and the related thioxo-thiol tautomerism (53) and (54) in the biologically imporfant mercaptopurines (Scheme 8). The subject has recently been discussed in some detail <76AHC(Si)502>. 

---