Heteroduplex heterozygotes in bacteriophage T4 involving mutations of various dimensions.

Single-factor genetic crosses in bacteriophage T4 produce two distinct types of heterozygous progeny.' One kind probably corresponds to a terminal redundancy of the chromosome. The other kind, variously termed a heteroduplex heterozygote, an internal heterozygote, or a recombinational heterozygote, arises as an intermediate in recombination (Fig. 1). Internal heterozygotes are preferentially accumulated"' 2, 10 12 in the presence of FUDR, which inhibits DNA replication and therefore inhibits the conversion of internal heterozygotes into homozygous progeny. Furthermore, only certain types of mutational lesions are efficiently recovered from internal heterozygotes." 2 If a cross is performed between two different mutants, one of which (point mutation) contains a base pair substitution and the other an extended deletion, the heterozygous particles which emerge usually contain the point mutant. This result indicates that internal heterozygotes cannot easily accommodate mutations of extended dimensions, presumably because such heterozygotes would contain single-stranded loops. At present it is not clear whether deletion mutations fail to form internal heterozygotes initially, or whether the heterozygotes are formed but are later repaired by a mechanism leading to homozygosity.3 The sign mutants4 of bacteriophage T4 consist5 of additions and/or deletions of small numbers of base pairs; they are also called "reading-frame mutants" and "acridine-type mutants." An extensive fine-scale map6 of the rIl mutants in the B cistron strongly suggests that sign mutants possess extended dimensions (Fig. 2). An analysis of the ability of internal heterozygotes to encompass sign mutations should therefore yield information about the physical dimensions of the mutations, and also about putative single-stranded regions within the DNA molecule. Materials and Methods.-Strains of bacteriophage T4B and its rIl mutants,7 and of Escherichia coli, were obtained from the collection of Dr. Sydney Brenner and Mrs. Leslie Barnett in Cambridge. All incubations were at 37°. E. coli strain Bw plates r phages as large plaques, and r+ phages as small plaques. Strain BB does not distinguish r and r+ phages, and is used to grow stocks and for crosses. Strain OP33 is a K-12(X) derivative which transmits ril mutants at a very low frequency. Strain QA1 is a K-12(X) derivative which permits the growth only of certain rII mutants of the amber class, such as rX417. The various rII mutants employed have been extensively mapped,6 and a preliminary and condensed map appears in Figure 2. All of the mutants except r187 and r196b are capable of reversion. FUDR crosses were performed according to Shalitin and Stahl.2 BB cells in M9casamino acids medium were infected with 5 of each parental phage. At 2 min, FUDR was added to 4 X 10-5 M, and uracil to 2 X 10-4 M. The phage-cell complexes were aerated continuously thereafter. At 9 min, chloramphenicol was added to 125 jug/ml. At 90 min the complexes were chilled and centrifuged at low speed. They were resuspended in the same medium without chloramphenicol (but with FUDR and uracil) at about 107/ml and incubated without aeration for an additional 60 min. Lysis was completed with chloroform. Burst sizes ranged from 1 to 10, but were usually about 5. Enrichment plating for heterozygotes was also performed according to Shalitin and Stahl.2 OP33 cells were suspended in buffer containing 4 X 10-2 M KCN. The lysate from anFUDR cross was adsorbed to the cells at a total multiplicity of 0.01 or less. After 4 min of adsorption, during which