86 J. BOOTH, E. BOYLAND VOL. 12 {iCjSS) 



of complexes formed from polynitro compounds as described by Burton and Richards^'*. 



The fact that the ultraviolet absorption spectra of the molecular compounds in 

 solution described in the present paper were not very different from the sum of the 

 spectra of the components, indicates that the interaction between the two molecules is 

 of a rather indefinite nature and the molecules are possibly held together by hydrogen 

 bonds (in the case of the dibenzcarbazoles) or forces which are weak compared with those 

 existing between ions. Although molecular compounds have been isolated only for some 

 pairs of purines and polycyclic compounds, it seems probable that the increase in 

 solubility is due to similar complex formation in other cases. 



Leiter and Shear^^ showed that when xanthine, nucleic acid, hypoxanthine, 

 guanine and caffeine were added to 3 : 4-benzpyrene in lipoid solvents and injected into 

 mice, the production of tumours was retarded. Weil-Malherbe^ suggestes that the 

 affinity of polycyclic compounds for the purines in nucleo-proteins may cause inter- 

 ference with the normal nucleic acid metabolism. The protective action of the purines 

 in carcinogenesis may then be due to the fact that they compete with the nucleoprotein 

 for the carcinogenic polycyclic compounds. 



The aromatic compounds considered in this paper are carcinogenic but similar 

 reactions between purines and non-carcinogenic compounds also occur. The reaction of 

 aromatic carcinogens with the DNA of dividing cells might be sufficient to distort the 

 nucleoprotein and cause chromosome abnormalities or deficiencies, which are associated 

 with the induction of mutations and of cancer. Whether the carcinogenic compounds 

 produce more change of this kind than inactive analogues, and if such changes are 

 sufficient to account for the biological effects produced is not known. 



acknowledgements 



We are grateful to Messrs. F. Hoffmann-La Roche & Co Ltd. for the supply of 

 samples of dibenzcarbazoles and to Mr F. H. Oliver for the micro-analyses. This in- 

 vestigation has been supported by grants to the Royal Cancer Hospital and Chester 

 Beatty Research Institute from the British Empire Cancer Campaign, the Jane Coffin 

 Childs Memorial Fund for Medical Research, the Anna Fuller Fund and the National 

 Cancer Institute of the National Institutes of Health, U.S. Pubhc Health Service. 



SUMMARY 



1. The solubilities of three isomers of dibenzcarbazole and dibenzacridine (like those of other 

 aromatic carcinogens) in caffeine and tetramethyluric acid solutions are much greater than in water. 



2. The three dibenzcarbazoles examined and i :2-7:8-dibenzacridine are precipitated from 

 aqueous solutions of purines as molecular compounds formed between the aromatic substances and 

 the purines. 



3. A solution of sodium deoxyribonucleate also dissolves the dibenzcarbazoles and dibenz- 

 acridines. The apparent pA'^of 3:4-6:7-dibenzacridine is approximately 3.5 in 50% ethanol, caffeine 

 and tetramethyluric acid solutions and 6.8 in a solution of sodium deoxyribonucleate. 



RitSUMfi 



I. Les trois isomeres du dibenzcarbazole et de la dibenzacridine (de mcme que tous les canc^ri- 

 gcncs aromatiques), sont beaucoup plus solubles dans dcs solutions de caftnne et d'acidc t6tram6thyl- 

 urique que dans I'eau. 



References p. 87. 



