Mitochondrial Genetics. Vi the Petite Mutation in Saccharomyces Cerevzszae:

The survival of the pf factor and of DrugR mitochondrial genetic markers after exposure to ethidium bromide has been studied. A technique allowing the determination of Drug8 genetic markers among a great number of both grande and petite colonies has been developed. The results have been analyzed by the target theory. The survival of the p+ factor is always less than the survival of any DrugR genetic marker. The survivals of C R and ER are similar to each other, while that of OR is greater than that 3f the other two DrugR markers. All possible combinations of DrugR markers have been found among the Ppetite cells induced, while the only type found among the grande colonies is the preexisting one. The loss of the CR and E" genetic markers was found to be the most frequently concomitant, while the correlation between the loss of the OR marker and the other two DrugR markers is less strong. Similar results have been obtained after U.V. irradiation. Interpretations concerning the structure of the yeast mitochondrial genome are given and hypotheses on the mechanism of petite mutation discussed. WO classes of cytoplasmically inherited mitochondrial mutants are known to Toccur in baker's yeast: those leading to respiratory deficiency (pcytoplasmic petite mutants) and those conferring resistance to a variety of drugs (chloramphenicol, erythromycin, spiramycin, oligomycin, etc.). It has been established that mutants of the first class result from massive changes in the nucleotide sequence of the mit-DNA molecule. In some petite mutants no mit-DNA can be detected by extraction. The nature of the changes in mutants of the second class is less well known but thought to result from discrete point mutations. Recent reviews summarize the present knowledge in this field (WILKIE 1969; LINNANE and HASLAM 1970; PREER 1971; SAGER 1972; FAYE et al. 1973). A great number of agents are known to induce petite mutants, acriflavine being the first one studied (EPHRUSSI, HOTTINGUER and CHIMENES 1949). Genetic relations between the pmutation and the DrugR genes have already been considered (THOMAS and WILKIE 1968; LINNANE et al. 1968; GINGOLD et al. 1969; Genetics 76: 195-219 February, 1974. 196 J. DEUTSCH et al. SAUNDERS et al. 1971; NAGLEY and LINNANE 1972; COEN et al. 1969; BOLOTIN et al. 1971; MICHAELIS, PETROCHILO and SLONIMSKI 1973; AVNER et al. 1973; FAYE et al. 1973; WAKABAYASHI and GUNGE 1970; RANK 1970). The relations are not completely clear since petite mutants were first reported to obligatorily lose the DrugR markers (THOMAS and WILKIE 1968; LINNANE et al. 1968), while an independent reassortment between p+ and DrugR markers was later suggested ( GINGOLD et al. 1969). The present work was aimed at studying the interrelations between the p+ factor and the mitochondrial genes conferring resistance to chloramphenicol, erythromycin and oligomycin. The experiments have used the following experimental approach: DrugR pf cells are first treated with a mutagen (ethidium bromide or U.V.) in the strict absence of cell multiplication. They are then allowed to multiply in the absence of the mutagen to form individual clones. For each hereditary trait two types of clones are found: those still containing non-mutated cells (i.e. p+ clones and/or DrugR clones) and those composed exclusively of mutated cells (i.e. pclones and/or Drugo clones). Logically for two traits there are four types of clones possible, for three traits, eight possible types, etc. The interrelations between the genetic determinants are deduced from the frequency of different clonal types during a kinetic study of the action of the mutagens. The results allow new insights into the nature of the mitochondrial genome and the mechanism of petite mutation. MATERIALS A N D METHODS I. STRAINS The kinetic experiments presented here were carried out on the two following strains: IL8-8C IL828-4B p+ a+ CR,,, ER,,, OR, (Y his The loss of the OR, genetic marker was also tested in the strains: IL779-3C p+ U+ ER,,, OR, a his IL781-6C p+ a+ CR,,, OR, (Y his These strains were built by recombination in order to carry many different DrugR mitochondrial genes. Their construction and genetical properties are described elsewhere (COEN et al. 3969; COEN et al. 1974; AVNER et al. 1973). The symbol pis used to specify the cytoplasmic petite mutation, pf being the wild-type grande. After U.V. mutagenesis a few colonies appeared that could not grow on glycerol but did complement the pneutral tester strain used in the replica-cross technique (see also below RESULTS S 1.2) They were interpreted as nuclear petite mutants (CHEN, EPHRUSSI and HOTTINGUER 1950) and they were scored as p+. After ethidium induction, no nuclear petite mutants were detected. Following COEN et al. (1969) the following symbols are used for mitochondrial antibiotic resistance markers: CR, CS for chloramphenicol resistance and sensitivity, ER, ES for erythromycin, OR, OS for oligomycin. Particular alleles are specified as such: CR,,,, OR, for example. The CR genetic marker (CRQ1,) used in the experiments presented here belongs to locus R,, the ER ones (ERsl4 and ER,,,) to locus R, the OR one (OR,) to locus 0, of the mitochondrial genetic map (AVNER et al. 1973) ; GRIVELL et al. 1973; NETTER et al. 1974). When the locus is not specified the Drug symbol is used, meaning either C, E or 0. The symbols CO, Eo, 00, Drug0 are used to specify cells which give a negative response on the replica-cross test and are interpreted as deletions for the genetic markers (see below RESULTS $ I and BOLOTIN et al. 1971). The symbol w p+ w f CR,,, ER,,& 0" CY his try THE PETITE MUTATION AND DRUG^ GENES 197 (U+ or U-) specifies the “mitochondrial sex”, i.e. a mitochondrial locus governing genetic recombination (see BOLOTIN et al. 1971; COEN et al. 1974). For testing the occurrence of pf Drug0 (see RESULTS s 111) many other strains were analyzed. These strains, carrying one or more mitochondrial genes, were isolated in our laboratory; their geentic properties are described in COEN et al. (1969; COEN et al. (1974); and AV“ et al. (1973). The tester strains (see Replica-cross technique) used mainly were strain 55R53C pf 0CsESOs a ura, and its derived neutral petite strain 55R5-3C/1035 pCOE’JO’J a ura. Occasionally, the strain D243-2B/R1 p+ w + CSESOS a ade, and its derived petite strain D243-2B/R1/6 were used as testers. No difference in the results was obseri ed on varying the tester strains. 11. MEDIA NO : Yeast Extract Difco I%, Bacto Peptone Difco I%, Sorensen phosphate buffer pH 6.2 0.05 M, glucose 2% as carbon source. N2 : Same as NO; but 0.1% glucose and 2% glycerol as carbon source. This medium is also called “differential medium” as it allows the discrimination between petite and grande colonies. N3 : Same as NO, but 2% glycerol as carbon source. N5 : Same as NO, but 2% galactose as carbon source. Antibiotics were added to N3 media at the following concentrations: Erythromycin 5 mg/ml, chloramphenicol 4 mg/ml, oligomycin 3 pg/ml. WO : Minimal medium: Yeast nitrogen base free of amino acids (DIFCO) added with 2% gluFor plating 3% DIFCO agar was added to all the media. 111. ETHIDIUM BROMIDE MUTAGENESIS The cells were cultured overnight in complete medium and harvested at the end of log phase. Either NO (glucose), N3 (glycerol), or N5 (galactose) media were used for culturing prior to mutagenesis without any marked effect on E.B.induction. The cells were washed, and resuspended in Sorensen phosphate buffer pH 6.5 0.1 M. The final titer of the suspension was 106 cells/ml. Cycloheximide was added at a final concentration of 2 pg/ml. Exposure to ethidium bromide was made in the dark, at 28”. Ethidium concentrations and exposure time are given in the text for each experiment. They varied from 1 to 5 pg/ml and from 10 min to 7 hours. After the treatment, the cells were washed twice, diluted and plated on NO and/or N2 media. Several times, two types of controls were made: zero time plating, and plating of the cells after shaking under the same conditions as treated samples in buffer plus cycloheximide, omitting ethidium bromide. No difference was ever found between the two controls. By comparing the number of cells and the number of colonies one finds that, on the average, two cells (the mother cell and its bud) form a colony. The plating efficiency does not vary during treatment. IV. FORMAL ANALYSIS OF THE KINETICS OF ETHIDIUM BROMIDE MUTAGENESIS We have analyzed the results of ethidium bromide mutagenesis by a formal treatment based on the target theory. The target theory (single-hit, multiple-target model) describes the inactivation of cells or viruses as follows (cf. DRAKE 1970). S = 1 (1 -&)la S is the frequency of surviving cells or viruses, h the number of hits and n the number of genetic targets. In our case S is the survival of a given genetic marker, as measured by the frequency of non-mutated colonies among the total population. Other mathematical treatments could have been applied (e.g., multiple hits, single-target model). However, as will be shown later, the one we have used has permitted us to single out a quantitative parameter, the relative target size, which is invariant of experimental conditions and more amenable, therefore, to interpretations than the survival curves alone. In our formal approach the number of hits h and the number of targets n as calculated do not necessarily represent the true number of physical hits or targets. We assume that among the total number of real hits only a number h will be effective. We also assume that among the total The media were prepared according to &EN et al. (1969).