On the mechanism of spontaneous reversion and genetic recombination in bacteriophage T4.

R E C E N T studies have distinguished two types of reversible mutations with respect to the type of change in the DNA molecule, in viruses and probably also in higher organisms (BRENNER, BARNETT, CRICK, and ORGEL 1961; MAGNI, personal communication). One type, involving substitution of one base for another and including transitions and possibly transversions, will be referred to here as “substitutions”; the other one, including deletions and/or additions of a few bases resulting in a shift of the reading frame, as “shift” mutations. A large part of the evidence concerning the two distinct mutation types has arisen from studies of the mutagenic effects of base analogues and acridines on the rll region of bacteriophage T4. It appears that the reversion of a given type of mutation is generally produced by a change of the same type that originally produced the mutation itself and this property has provided a means of distingishing between the two types. Thus a substitution may be corrected by another possibly inverse substitution (FREESE 1959) mostly in the same codon (HELINSKI and YANOFSKY 1962) and a shift by an opposite shift, which generally occurs at a site close to, but distinct from the site of the original mutation (CRICK, BARNETT, BRENNER, and WATTS-TOBIN 1961 ) . Most spontaneous mutants in T4, as a matter of fact, were shown to behave like acridine induced, namely shift, mutants (FREESE 1959; ORGEL and BRENNER 1961). Though the evidence concerning shift mutations is purely genetic (CRICK et al. 1961), a physical basis for their occurrence was provided by the structure of acridine-DNA complexes (LERMAN 1961, 1963, 1964). Since shift mutations are induced by acridines, LERMAN pointed out that a mutagenic effect of acridines has been reported only in recombining organisms, and he proposed a model which postulates that acridine molecules intercalated between bases in DNA would interfere with homologous pairing over short regions. According to this model, a crossover will be unequal if it occurs in a region in which base sequences of paired double helices are out of register as a result of structural changes produced by intercalated acridine. Local extension and untwisting of the paired double helices might also occur in the absence of acridine, and this might account for the majority of spontaneous mutations. An unequal crossing over at the molecular level may produce two opposite shifts in the daughter molecules (Figure 1 ), so accounting for both signs (additions and deletions) of shift mutants (thereafter called also sign mutants).