GROWTH-RATE EFFECTS OF MUTATIONS CONFERRING STREPTOMYCIN-DEPENDENCE AND OF ANCILLARY MUTATIONS IN THE RPSL GENE OF ESCHERICHIA-COLI - IMPLICATIONS FOR THE CLUSTERING (HYPERMUTATION) HYPOTHESIS FOR SPONTANEOUS MUTATION

Colonies of newly arising streptomycin-dependent (Sm-D) mutants freque ntly contain a high proportion of cells with additional mutations (anc illary mutations) in the same gene (rpsL), The ancillary mutations app ear to have arisen at a rate greatly above expectation, To better esti mate this rate it is necessary to allow for any selective advantage co nferred by the ancillary mutations, We have previously measured their effect on growth rate of established Sm-D strains in the presence of s treptomycin. In the present work a pair of single and double mutant al leles (rpsL832 and rpsL852 respectively) has been employed together wi th the wildtype allele to model the situation soon after such mutation s first arise, i.e. when the cell still contains wild-type S12 protein (the rpsL gene product), When these alleles, under the control of an IPTG-inducible promoter and carried on a plasmid, were expressed in th e presence of a chromosomal wild-type allele, the double mutant allele permitted much faster cell growth than the single mutant allele, In t he presence of streptomycin, and with rpsL(+) on a plasmid, bacteria w ith a double mutant chromosomal gene grew faster than those with a sin gle mutant chromosomal gene. If these results can be extrapolated to a bacterial cell in which an Sm-D mutation has just occurred, the ancil lary mutation should be able to confer a selective advantage during a limited period when wild-type S12 protein is still present, both in th e absence and in the presence of streptomycin, We have used this infor mation, together with the previously measured selective advantage when the mutations are fully expressed, to estimate that the rate at which the ancillary mutation rpsL852 arises in a new clone of rpsL832 is of the order of 10(-4) per replication cycle, compared with an expected rate of similar to 10(-10) per base pair per replication cycle for a s pecific base pair substitution, It is not yet clear whether this hyper mutability reflects a stress response following the occurrence of the primary Sm-D mutation or whether temporal clustering of mutations is a feature of normal spontaneous mutation.