V. molp:cular mechanism of mutations 219 



strongest mutagen in this table is azaserine wliich inhibits purine syn- 

 thesis (see Handschumacher and Welch, 1960). But this chemical is also 

 a very labile alkylating agent; it may therefore owe its mutagenicity 

 to its alkylating ability rather than to its inhibitory action. Similarly 

 urethan is both an inhibitor of pyrimidine synthesis and a mild alkylat- 

 ing agent. But in this case Rogers (1957) has shown that in mice thymine 

 inhibits the production of chromosome breaks by urethan. All other 

 growth inhibitors are rather weak mutagens in bacteria and induce 

 chromosome breaks in higher organism. The most careful quantitative 

 experiments about the mutation induction have been done by Novick 

 and Szillard (see Novick, 1956). These authors have also shown that 

 even adenine increases the frequency of mutations while the normal 

 purine nucleosides have an anti-mutagenic effect. These observations 

 suggest that the frequency of mutations increases because the lack of 

 one base either causes a chromosome break, and thus a large alteration, 

 or it increases the frequency of pairing mistakes, through which one 

 normal nucleic acid base is incorporated in the place of another one. 

 However, it is just as likely that the inhibitor causes the increased 

 formation of another base analog which is the actual mutagen, perhaps 

 by being incorporated into the nucleic acids in place of the normal bases. 

 ]Many such natural base analogs have been observed in RNA [although 

 mostly for soluble RNA, e.g., Littlefield and Dunn (1958)] and con- 

 ditions of thymine starvation cause the accumulation of A"-methyl- 

 adenine in DNA (Dunn and Smith, 1955). In addition, a naturally 

 produced mutator factor has been genetically proved to exist in a certain 

 strain of Salmonella typhi murium (Miyake, 1959). 



B. IXCORPORATION OF BASE ANALOGS 



Apart from naturally occurring base analogs some artificial analogs 

 are incorporated into RNA and into DNA. Here we are only concerned 

 with those analogs that have shown a strong mutagenic effect. 5-Bromo- 

 uracil (BU), 5-chlorouracil, and 5-iodouracil can replace thymine in 

 DNA (bacteria: Weygand, Wacker, and Dellweg, 1952; Zamenhof and 

 Griboff, 1954; phages: Dunn and Smith, 1954; Litman and Pardee, 1960; 

 human cell lines: Szybalski and Djordjevic, 1959; Hakala, 1959; Eidin- 

 off et al, 1959), while 2-aminopurine (AP) (Wacker et al, 1960a, Gott- 

 schling and Freese, 1961; Rudner, 1961) is incorporated to such a small 

 extent that it has not been possible to determine chemically which base 

 it replaces. 2,6-Diaminopurine is also highly mutagenic (Freese, 1959a; 

 Benzer, 1961) but its effect has not been studied in great detail. These 

 base analogs are much less inhibitory for bacterial and phage growth 

 than other purine or pyrimidine analogs ; yet they are much more muta- 



