Francis J. Ryan

Is DNA replication a necessary condition for spontaneous mutation

Previous communications from our laboratory have described a spontaneous mutat ion from a histidineless (h-) to a histidine independent (h +) condition in non-dividing Escherichia coli 15 during histidine starvation (RYAN 1959). Since then the possibility has been considered tha t errors may arise during DNA replication within these non-dividing cells. DNA is thought to be formed by replication. MESELSON and STAHL (1958) have used a density gradient centrifugation technique to show tha t bacterial DNA is bipart i te as was proposed by WATSON and C~ICK (1953) and tha t it reproduces in a semi-conservative fashion (DEL~ni~CK and STENT 1957), each original sub-unit being found associated, upon replication, with a newly-formed sub -unit. NAKADA and I~YAN (1961), who employed E. coli inhibited by chloramphenieol, showed tha t this is also the case in non-dividing bacterial cells while they engage in a residual DNA synthesis tha t lasts ~ few hours. Therefore an a t tempt was made to determine whether DNA replicates in non-dividing bacteria under conditions allowing mutation. A double auxotroph of E. coli 15 hl(histidineand leucine less) was starved for amino acids in minimal medium. The h + lor h l + mutants, pre-existing or occuring spontaneously during starvation, could not overgrow because they lacked leucine or histidine. In this system large enough numbers of bacteria (ca. 109 per ml) can be collected for the measurement of macromolecules. In the very early period of s tarvation for both histidine and leucine, there was up to a 35 per cent increase of DNA (Fig. 1). This is in good agreement with the finding of GOLDST~IN, GOLDSTEIN, BI~OWN and CHOU (1959) who observed a residual synthesis of DNA in an E. coli K-12 leucineless mutan t starved for leueine. But this residual synthesis ceased after 5 hours of starvation and the amount of DNA decreased gradually. After 100 hours of starvation, the amount of DNA was about the same as tha t at zero time. The number of mutants expected as a consequence of this residual synthesis of DNA was too small to be detected under the conditions of this experiment, on the assumption

Bacterial mutation and the synthesis of macromolecules

SummaryAn analysis was made of the effect of chloramphenicol on mutation in histidine-starvedE. coli from a histidine-dependent to a histidine-independent condition. Although mutations occurred in the absence of chloramphenicol, none could be found in its presence unless it was accompanied by a mixture of amino acids, excluding histidine. The role played by the amino acids was concluded to be twofold: 1.In the absence of chloramphenicol they increased the rate of mutation indirectly by accelerating the rate of gene replication.2.In the presence of chloramphenicol theyenabled gene replication; as a consequence, rare errors or mutations took place at a rate limited by the rate of turnover of the genetic material. The effect of chloramphenicol and amino acids on nucleic acid synthesis is such as to offer an account of this behaviour.

Is DNA replication a necessary condition for mutation

Since mutations occur in non-dividing bacteria (RYA~ 1959), the question arises whether the replication of genetic material is nonetheless required. With bacteriophage T2 it has been shown that mutations can be induced by 5-bromouracil during the DNA synthesis which occurs in the presence or absence of chloramphenicol (LITMA~ and PARDE],~ 1959, BRE~ER and SMIT~ 1959). I t has further been shown in Escherichiacoli that the frequency of mutants induced by caffeine is proportional to the amount of DNA synthesis that occurs when chloramphenicol is present (GLAss and NowcK 1959) and that mutations induced by ultraviolet light arise correlated With subsequent DNA synthesis (LIMB 1959). These findings are consistent with the working hypothesis of FRnnSE (1959) which proposes that some chemical mutagenesis involves the incorporation of abnormal bases followed by pairing errors during subsequent DNA replication. This hypothesis has been tested by the use of populations oi Salmonella typhimurium whose division has been Synchronized by filtration; mutations occur in accordance with DNA replication as the hypothesis predicts (RuD~Etr 1960). This Salmonella system involves mutation f rom tryptophane dependence (try D-I0, here called try-) to t ryptophane independence (try +) induced by 300 #g. 2-aminopurine per ml. I t has the advantage of a low background of spontaneous try + mutants. When trycells are exposed to 20 #g. of chloramphenicol per ml. in the presence or absence of 20 #g. of t ryptophane per ml., within 2 hours there is a net synthesis of 50 per cent more DNA as measured by the Dische reaction, while RNA determined by the orcinol reaction increases almost 3-fold. Thereafter the amounts of these compounds remain constant for as long as 5 hours. If 2-aminopurine is present at the onset of this synthesis, mutations are induced; on the other hand, if the cultures ar9 provided with 2-aminopurine after synthesis has ceased, no mutations aredetectable (Fig. 1). ,Under these conditions, therefore, the synthesis of nucleic acid is required for the induction of mutat ion while the synthesis of protein is not. In the presence of chloramphenicol, although protein synthesis is severely inhibited, a little DNA is replicated and RNA synthesis continues to a somewhat greater extent. When treated cells are plated on minimal medium, this RNA is expected to degrade, as has already been described by NWlD~Am)T and GRos (1957) and HoRowITz, LOMBARD and CHA~GAFF (1958), while the DNA and protein should remain almost constant in amount. In our experiments, when

Phenocopies induced by thymine in Escherichia coli.

SummaryA uracilless strain ofE. coli upon starvation for uracil adapts to synthesize this compound. These adaptations are of two sorts, heritable and non-heritable. The latter are induced by the presence of thymine although little or none of the uracil is synthesized by the demethylation of thymine. The non-heritable adaptations arise in a discontinuous fashion at a rate 10 times as high as the genetic reversions. The non-heritable uracil-independent cells are considered to be phenocopies because they mimic the phenotype of the genetic revertants.