Adaptive (stationary-phase) mutagenesis occurs in the gram-positive bacterium Bacillus subtilis. Furthermore, taking advantage of B. subtilis as a paradigm for the study of prokaryotic differentiation and development, we have shown that this type of mutagenesis is subject to regulation involving at least two of the genes that are involved in the regulation of post-exponential phase prokaryotic differentiation, i.e., comA and comK. On the other hand, a functional RecA protein was not required for this type of mutagenesis. The results seem to suggest that a small subpopulation(s) of the culture is involved in adaptive mutagenesis and that this subpopulation(s) is hypermutable. The existence of such a hypermutable subpopulation(s) raises important considerations with respect to evolution, the development of specific mutations, the nature of bacterial populations, and the level of communication among bacteria in an ecological niche.For over a decade, there has been considerable interest in a phenomenon that has been called adaptive, or stationaryphase, mutagenesis. The result of the mechanism(s) responsible for this phenomenon is the production of mutations that arise in nondividing or stationary-phase bacteria when the cells are subjected to nonlethal selective pressure, such as nutrientlimited environments (6,11,15,32,61). While most of the research has involved Escherichia coli model systems, similar observations have been made in other prokaryotes (43) as well as in eukaryotic organisms (69).In the FЈ lac frameshift reversion assay system in E. coli, stationary-phase mutations that lead to the generation of Lac ϩ cells can be distinguished from normal growth-dependent spontaneous Lac ϩ mutations (21, 59, 63). Specifically, Lac ϩ mutations are generated in stationary-phase cells via a molecular mechanism that requires a functional homologous recombination system (11,21,36,37), FЈ transfer functions (20, 23), and a component(s) of the SOS system (50). Genetic evidence suggests that DNA polymerase III (18, 35) and DNA polymerase IV (51, 52) are responsible for the synthesis of errors that lead to these mutations. Furthermore, for the Lac ϩ mutations, different sequence spectra are generated for the stationaryphase mutations than for the types of mutations generated during growth.For instance, a majority of the Lac ϩ mutations that arise during stationary phase have a Ϫ1 deletion at mononucleotide repeats within the target gene. On the other hand, for the spontaneous mutations that arise during growth, various types of mutations occur in seemingly random locations (19, 62). These characteristics suggested that stationary-phase Lac ϩ reversions occur via a different molecular mechanism(s) than for those reversions of the same lac allele that are generated during growth. However, there is also evidence that demonstrates that the mutations generated by this lac system during stationary phase are the result of gene amplification followed by SOS-induced mutagenesis and selection (39).Although the very observations of ada...

Lysates of the virulent bacteriophage SPP1 were shown to be capable of mediating generalized transduction. Suppressible mutants of this bacteriophage (sus) were capable of transduction at a lower multiplicity of infection than virulent SPP1. Linkage analysis demonstrated that bacteriophage SPP1 transduced segments of the genome equal in size to that transferred by SP10. This bacteriophage should be useful in analyzing the regions of the genome where PBS1 appears to give anomalous results.

No abstract

A genetic library of Streptococcus mutans GS-5, constructed in an Escherichia coli plasmid vector, was screened for cells which could utilize sucrose as the sole carbon and energy source. The recombinant plasmid pFRU1, containing a 4.2-kilobase pair insert of S. mutans DNA, was shown to confer this phenotype. Further characterization of the gene product encoded by pFRUl revealed that the enzyme was a P-D-fructosidase with the highest specificity for the j(2-*6)-linked fructan polymer levan. The enzyme could also hydrolyze inulin [13(2-+1)-linked fructan], sucrose, and raffinose with 34, 21, and 12%, respectively, of the activity observed for levan. The gene (designated fruA) appeared to be expressed under its own control in E. coli, as judged by the lack of influence on gene product activity of induction or repression of the ,I-galactosidase promoter adjacent to the insertion site on the cloning vector. The protein was purified to homogeneity, as judged by silver staining of purified protein in denaturing and reducing conditions in polyacrylamide gels, from sonic lysate of E. coli, as well as from culture supernatants of S. mutans GS-5 grown in a chemostat at low dilution rate with fructose as the sole carbohydrate source. Both purified proteins had an apparent molecular mass of 140,000 daltons in sodium dodecyl sulfate-polyacrylamide gel electrophoresis, were immunologically related and comigrated in sodium dodecyl sulfate-polyacrylamide gel electrophoresis as determined by Western blotting with antisera raised against the cloned gene product, and were identical in all physical and biochemical properties tested. The pH optimum of the enzyme acting on fructan polymers was 5.5, with a significant amount of activity remaining at pH 4.0. The optimum pH for sucrose degradation was broader and lower, with a peak at approximately 4.5.Enzyme activity was inhibited almost completely by Hg2' and Ag2+, inhibited partially by Cu2+, not inhibited by fluoride ion or Tris, and slightly stimulated by Mn2' and Co2+. Fructan polymers were attacked exohydrolytically by the enzyme, fructose being the only product released. With sufficient time, both levan and inulin were degraded to completion, with no evidence of product inhibition.Several bacteria which normally inhabit the human mouth, including Streptococcus mutans (2,3,12,18,38), S. salivarius (3, 12, 13, 33), S. sanguis (24), and Actinomyces viscosus (3, 49), can synthesize polymers of D-fructose. The structures of the fructans produced by many of these organisms have been analyzed, and it appears as though S. salivarius (3,12,13,32) and the actinomycetes (3, 49) produce a levan-type fructan, consisting predominantly of ,B(2->6)-linked fructosyl units with a significant degree of branching in the 1 position (13, 32). S. mutans, on the other hand, produces an inulin-type fructan which is made up almost exclusively of 1(2-l) linkages with some branching in the 6 position (2, 3, 12, 38). In some cases, the physical properties of the fructan polymers have been examined, and gene...

DNA damage-inducible (din) (16,41). Regulation of the SOS system in E. coli is controlled by the products of the recA and lexA genes. The RecA protein has many functions in E. coli and is involved in the processes of recombination, DNA repair, and mutagenesis (31,41,45). The LexA protein is a repressor of as many as 20 unlinked, coordinately regulated loci which include the recA and lexA genes themselves (19,41). Following exposure of E. coli to agents that alter DNA structure or interfere with DNA replication (such as UV radiation, mitomycin, nalidixic acid, etc.), an inducing signal is generated. (4,7,17,34,35), and the UmuD protein (2, 39). As levels of LexA repressor decline, damage-inducible loci are derepressed, resulting in expression of'the physiological phenomena that compose the SOS response (16, 41).The SOS system of E. coli has served as a model for the study of similar inducible DNA repair systems in other gram-negative bacteria (14,36,37,43,44 and essentially error free (7a). This contrasts with W reactivation in E. coli, which is capable of repairing a variety of DNA lesions by an error-prone mechanism (32). Furthermore, while induction of all SOS phenomena in E. coli is dependent upon a functional RecA protein, filamentation in B. subtilis is a RecA-independent response (21). Finally, the SOB system in B. subtilis is developmentally regulated. As B. subtilis differentiates into the physiological state of natural competence (6), SOB phenomena are spontaneously induced in the absence of externally generated DNA damage (20,23,47,49,50).DNA damage-inducible loci in B. subtilis were first identified by using transposon-mediated gene fusions (20). Tn917-lacZ transposon insertions within din loci were isolated from a library of insertions by selecting those fusions that induced expression of the lacZ reporter gene after exposure to DNA-damaging agents (20). Fifteen independently isolated din gene transposon insertions were genetically mapped and localized to three loci (dinA, dinB, and dinC) on the B. subtilis chromosome (11). As mentioned above, induction of all three din loci was demonstrated to be dependent upon a functional RecA protein (20). In order to elucidate the mechanisms that regulate damage-inducible gene expression in B. subtilis, we have cloned and sequenced DNA fragments that contain the dinA, dinB, and dinC promoter regions. Described here is our initial characterization of these cloned din promoter regions and the identification of a putative SOB operator sequence.

DNA damage-inducible (din) operon fusions were, generated in Bacillus subtilis by transpositional mutagenesis. These YB886(din::Tn917-lacZ) fusion isolates produced increased fi-galactosidase when exposed to mitomycin C, UV radiation, or ethyl methanesulfonate, indicating that the lacZ structural gene had inserted into host transcriptional units that are induced by a variety of DNAdamaging agents. One of the fusion strains was DNA-repair deficient and phenotypically resembled a UV-sensitive mutant of B. subtilis. Induction of 18-galactosidase also occurred in the competent subpopulation of each of the din fusion strains, independent of exposure to DNA-damaging agents. Both the DNA-damage-inducible and competence-inducible components of ,B-galactosidase expression were abolished by the recE4 mutation, which inhibits SOS-like (SOB) induction but does not interfere with the development of the competent state. The results indicate that gene expression is stimulated at specific loci within the B. subtilis chromosome both by DNA-damaging agents and by the development of competence and that this response is under the control of the SOB regulatory system. Furthermore, they demonstrate that at the molecular level SOB induction and the development of competence are interrelated cellular events.

YqjH and YqjW are Bacillus subtilis homologs of the UmuC/DinB or Y superfamily of DNA polymerases that are involved in SOS-induced mutagenesis in Escherichia coli. While the functions of YqjH and YqjW in B. subtilis are still unclear, the comparisons of protein structures demonstrate that YqjH has 36% identity to E. coli DNA polymerase IV (DinB protein), and YqjW has 26% identity to E. coli DNA polymerase V (UmuC protein). In this report, we demonstrate that both YqjH and the products of the yqjW operon are involved in UV-induced mutagenesis in this bacterium. Furthermore, resistance to UV-induced damage is significantly reduced in cells lacking a functional YqjH protein. Analysis of stationary-phase mutagenesis indicates that absences of YqjH, but not that of YqjW, decreases the ability of B. subtilis to generate revertants at the hisC952 allele via this system. These data suggest a role for YqjH in the generation of at least some types of stationary-phase-induced mutagenesis.In Escherichia coli, the dinB gene is required for bacteriophage untargeted mutagenesis (UTM), an error-prone pathway observed when undamaged DNA infects SOS-induced E. coli cells (4, 55). Overexpression of the dinB gene confers a mutator phenotype on the cells (22). However, mutations in the dinB gene only caused a modest UV sensitivity phenotype, indicating that this gene product might not play a major role in the tolerance of DNA lesions introduced by UV irradiation into E. coli (22). The genetic requirements for UTM include the recA, uvrA, uvrB, uvrC, and polA genes, as well as DNA polymerase III (DNA Pol III), in addition to dinB (22,26). However, when the dinB gene is overexpressed on a multicopy plasmid, these requirements for genes besides dinB for UTM are bypassed (22). In 1999, it was discovered that the purified DinB protein has a template-directed, DNA-dependent DNA polymerase activity and it was designated the fourth DNA polymerase in E. coli (DNA Pol IV) (51).The DNA damage-inducible UmuDЈ and UmuC proteins are required for another type of SOS mutagenesis in E. coli (40). UmuCD-dependent translesion DNA synthesis allows cells to replicate past DNA damage-induced lesions that would normally block the continuing polymerization by the major replication DNA polymerase (DNA Pol III) in E. coli. This translesion synthesis results in an increased mutation rate (21, 42). The translesion DNA synthesis process requires the products of the SOS-regulated recA gene and the umuDC operon, which was originally identified by screening for E. coli mutants that were not mutable by UV light and other agents (21, 42). The umuDC gene products are also known to be essential components of chromosomal UTM (9, 27), a transient increase in the mutation frequency of chromosomal genes following induction of the SOS response (9, 27, 30). In 1999, UmuC or UmuDЈ 2 C was discovered to be a template-directed, DNAdependent DNA polymerase that was designated the fifth DNA polymerase in E. coli (DNA Pol V) (34, 49).It has very recently become apparent that UmuC ...

Lysogenic strains of Bacillus subtilis 168 were reduced in their level of transformation as compared to non-lysogenic strains. The level of transformation decreased even further if the competent lysogenic cells were allowed to incubate in growth media prior to selection on minimal agar. This reduction in the frequency of transformation was attributable to the selective elimination of transformed lysogenic cells from the competent population. Concurrent with the decrease in the number of transformants from a lysogenic competent population was the release of bacteriophage by these cells. The lysogenic bacteria demonstrated this dramatic release of bacteriophage only if the cells were grown to competence. Both the selective elimination of transformed lysogens and the induction of prophage was prevented by the inhibition of protein synthesis. Additionally, competent lysogenic cells released significantly higher amounts of exogenous donor transforming deoxyribonucleic acid than did competent nonlysogenic cells or competent lysogenic cells incubated with erythromycin. These data establish that the induction of the prophage from the competent lysogenic cells was responsible for the selective elimination of the lysogenic transformants. A model is presented that accounts for the induction of the prophage from competent lysogenic bacteria via the induction of a repair system. It is postulated that a repair system is induced or derepressed by the accumulation of gaps in the chromosomes of competent bacteria. This hypothetical enzyme(s) is ultimately responsible for the induction of the prophage and the selective elimination of transformants. Lysogenic conversion can markedly influence the capacity of a wide range of bacterial species to undergo deoxyribonucleic acid (DNA) mediated transformation and transfection. For instance, Staphylococcus aureus can be made competent only in the presence of temperate bacteriophage 441 (29, 34). Similarly, Bacillus stearothermophilus requires the temperate bacteriophage TP-12 for the development of competence (N. E. Welker and M. E. Eager, Abstr. Annu. Meet. Amer. Soc. Microbiol. 1972, V98, p. 201). Apparently this virus enables the cell to produce the competence factor(s) required for the binding of DNA to the cytoplasmic membrane (38). On the other hand, in Bacillus subtilis (27, 44, 45, 47) and streptococci (25), lysogeny inhibits DNA-mediated transformation. Although transformation is reduced in competent lysogenic cultures of B. subtilis,