ABORTIVE INFECTION BY BACTERIOPHAGE BF23 DUE TO THE COLICIN Ib FACTOR. I: GENETIC STUDIES OF NONRESTRICTED AND AMBER MUTANTS OF BACTERIOPHAGE BF23

Abstract

ACTERIOPHAGE BF23 was isolated by FREDERICQ, and has been used to detect bacterial strains that are resistant to colicins of group E (FREDERICQ 1949).NISIOKA and OZEKI (1968) noted that BF23 is related to T5 in its properties-both phages display virtually identical gross morphologies, both require calcium ions for multiplication, both have a minimal latent period of about 35 minutes, and both are restricted in their growth in cells of Escherichia coli carrying colicinogenic factor Ib (CoZIbf cells).Furthermore, these two phages undergo phenotypic mixing and, as we will report in a later publication, they readily recombine genetically with one another.Restriction of the growth of BF23 by colicinogenic factor Ib was first observed by STROBEL and NOMURA (1966).Their work was subsequently extended by NISIOKA and OZEKI (1968).Our present knowledge of this restriction may be summarized as follows.( 1) The efficiency of plating of BF23 on Cola+ cells (restrictive host) is less than when compared to that on CoZIb-cells (permissive host).(2) The adsorption of the phage and the injection of its DNA are both normal even when restrictive cells are used as the host.(3) In restrictive host cells the injected DNA is neither broken down into acid-soluble materials nor fragmented into small molecular species as judged by centrifugation in a sucrose density gradient.However, phage DNA is not synthesized.(4) Infection by the phage causes the degradation of host DNA in both the restrictive and permissive hosts.( 5 ) CoZIbf cells lyse 15 minutes after infection by wild-type BF23 whereas CoZIb-cells lyse at the usual time of about 45 minutes after infection.This unusual lysis of CoZIb+ cells is called "early abortive lysis" and requires the synthesis of phage-specific proteins other than lysozyme.(6) The rare plaques that do form on CoZIb+ cells infected with a preparation of wild-type BF23 are not due to the phenomenon of "host-controlled modification" as observed with phage A in certain E. coli strains (ARBER and LINN 1969) and with nonglucosylated T-even phages in cells lysogenic for P1 (REVEL 1967;REVEL and GEORGOPOULOS 1969).Rather, they result from a spontaneous mutation in the BF23 genome itself.These mutant phage, which can grow on ColIbf cells, are recessive to wild-type phage in determining early abortive lysis.