Conferring Temperature-Conditional Allyl Alcohol Resistance

Selection for allyl alcohol resistance in respiratory incompetent yeast is a highly specific method for isolating functional mutations at ADHI, the gene coding for the cytoplasmic alcohol dehydrogenase, ADHI. Because of the nature of this selection scheme, the ADHI activity of such mutants is retained, but the kinetic characteristics of the enzymes are altered. The high specificity for targeting functional mutations at this locus suggested that selection for enzyme variants with more subtle phenotypic effects might be possible. Here, we describe functional ADHI mutants that are temperature-conditional in their allyl alcohol resistance. Haploid cells of one of these mutants grow well on plates at 10 mM allyl alcohol at 19", but not at 37", the restrictive temperature. A second mutant grows well at 10 mM at 37", but its growth is restricted at 19". What distinguishes these mutants from other temperature-sensitive mutants is that the temperature-conditional growth phenotypes described here must be due to interactions between allyl alcohol levels and ADHI functional properties and cannot be due to lability of the enzyme at the restrictive temperature. This system shows promise for the investigation of functional enzyme variants that differ by only one or two amino acid residues but have significant temperatureand substrate-conditional effects on growth phenotypes in both the haploids and the diploids. EVERAL laboratories have developed model sysS tems for investigating how the metabolic machinery of microorganisms may be altered by selection pressure for utilization of novel carbon sources (for reviews, see MORTLOCK 1984). The responses in such evolution "experiments" are remarkably diverse, and consequently, these systems have become useful for modeling the evolution of metabolic pathways and the adaptation of new enzyme functions from preexisting ones. Changes in an enzyme's substrate specificity are a common selective outcome, as in the "evolved" Bgalactosidase system of E. coli (HALL 1984) and the amidase system of Pseudomonas aeruginosa (CLARKE 1984). Another experimental system for studying evolution at the biochemical level is the alcohol dehydrogenase system of yeast. It differs from other selection schemes in being remarkably specific for the selection of functional mutations at a single enzyme locus, ADHI. In Saccharomyces cerevisiae, two genes, ADHI and ADH2, code for the two cytoplasmically expressed alcohol dehydrogenases, ADHI and ADHII (CIRIACY 1975; gene nomenclature revised by TAGUCHI, C RIACY and YOUNG 1984). ADHI is largely a constitutively expressed enzyme, whereas ADHII is repressed by glucose and in anaerobically grown or petite cells. This and the kinetics of the two cytoplasmic isozymes suggest that ADHII is involved in ethanol oxidation and ADHI is involved primarily in ethanol production Genetics 1 1 5 65-71 (January, 1987) during fermentation (WILLS 1976a). Yeast strains that cannot respire aerobically generate energy for growth by glycolysis alone. Because yeast strains having no cytoplasmic ADH activity cannot survive as petites, the presence of a functional cytoplasmic ADH ( i e . , ADHI) is apparently essential for survival under these conditions, since it is largely responsible for regenerating NAD+ in glycolysis (WILLS and PHELPS 1975; WILLS, KRATOFIL and MARTIN 1982). Taking advantage of this fact, WILLS developed a scheme for selecting large numbers of functional mutants at ADHI (WILLS and PHELPS 1975; WILLS 1976b). The basis for selection is the fact that allyl alcohol (2-propen-l-o1), when added to growth media, is oxidized by ADHI to acrolein (acrylaldehyde, 2-propenal), a toxic aldehyde that inactivates various proteins, including yeast ADH (RANDO 1974), binds to nucleic acids (IZARD and LIBERMAN 1978) and kills the cell. The only cells that survive are those that are able to minimize the toxic effects of acrolein. A large proportion (up to 40% in some experiments) of the allyl-alcohol-resistant mutants obtained in these experiments involve mutations in ADHl in which the activity of the enzyme is retained (WILLS 1976b; WILLS and JORNVALL 1979). This suggests that there are a large number of ways in which yeast can respond to this environmental challenge via alterations in this particular gene. The mechanism of resistance involves changes in 66 J. G. Hall and C. Wills the kinetic properties of the enzyme that, in turn, lead to increases in the ratio of NADH to NAD+ in the cytoplasm (WILLS 1976a; WILLS and PHELPS 1978). Since acrolein is apparently not metabolized further in the cell (WILLS and PHELPS 1978), this drives the reaction toward the harmless alcohol, decreasing acrolein levels. The specificity of this selection scheme for targeting functional mutations to this gene makes it extremely useful for investigating the structural and functional nature of single amino acid substitutions that leave the major function of the molecule intact but produce small changes in its kinetic parameters. Questions to be addressed include whether structural and functional constraints exist on the range of "adaptive" amino acid substitutions that confer allyl alcohol resistance. For example, do resistant substitutions occur in the neighborhood of the active site, or are they distributed throughout the molecule? Is resistance achieved by means of a limited number of kinetic changes, or can the same functional result be achieved in a variety of ways? Preliminary answers are now available for some of these questions [See WILLS (1 984) for review]. The five functional substitutions that have been localized are scattered throughout the ADHI subunit. The specificity of allyl alcohol selection also suggests potential for selecting ADHI mutants that are even more subtle in their phenotypic effects, i.e., in which allyl alcohol resistance is conditional upon some aspect of the physical or chemical cellular environment. Much comparative biochemical work does suggest, in fact, that the functions of enzymes can be changed via natural selection by the thermal, hydrostatic pressure, and intracellular solute environments of the organisms in which they are found (HOCHACHKA and SOMERO 1984). Here, we describe ADHI functional mutants that confer temperature-conditional allyl alcohol resistance. MATERIALS AND METHODS A spontaneous cytoplasmic petite was isolated from wildtype strain S288C (a mating type) and subcloned. Homogenates of cells from this clone (HP-73-4) grown in batch culture at 30", showed only ADHI activity on starch gels following electrophoresis and did not grow on complete medium (2% Bacto-peptone, 1% yeast extract, w/v from Difco) with 2% glycerol (v/v) as the carbon source. I t was later determined that this petite is, in fact, slightly leaky: it is able to grow very slowly with glycerol and produces low levels of ADHII, if cultures are aerated vigorously. T o select for allyl-alcohol-resistant mutants, cells of HP73-4 were inoculated into 100 ml of complete medium + 2% dextrose (YEPD) and were grown for 16 hr at 30". The cells were mutagenized with either 3% ethyl methanesulfonate (EMS; WILLS 1968) or 6.8% saccharin (Sigma, lot no. 98C-0276; c j , BATZINGER, OU and BUEDINC 1977) for 1 and 2 hr, respectively. The cells were plated at densities of 107-109 cells/plate on YEPD agar medium to which 2% (v/v) sterile ethanol and 5 mM allyl alcohol had been added after autoclaving. The plates were divided into two groups, sealed in plastic bags and incubated at 19" and 37". Resistant colonies appeared in about 2 weeks. Mutant strains were screened for ADHI electrophoretic variation. Cells were grown in 100 ml of YEPD broth at 30°, harvested by centrifugation and then lysed by vigorous vortexing for three l-min periods in 7-ml test tubes with one cell pellet volume each of 0.25-mm glass beads and homogenization buffer (1 00 mM potassium phosphate, pH 7.0, 5 p~ ZnClp and 0.1% (v/v) P-mercaptoethanol). Electrophoresis was conducted on supernatants of centrifuged extracts (1 0,000 X g for 2 min at 40") on horizontal starch gels (WILLS and PHELPS 1975). Resistant mutants with altered ADHI electrophoretic mobilities were then plated on YEPD + 2% ethanol (YEPDE) agar medium to which 5 or 10 mM allyl alcohol had been added. The plates were incubated at both 19" and 37" as a preliminary screen for temperature sensitivity in resistance. For genetic analysis, electrophoretic mutants were backcrossed to grande S288C or allyl-alcohol-resistant ADHI mutants. Clones of the tetrad segregants obtained for the sporulated diploids (MORTIMER and HAWTHORNE 1969) were screened both for segregation of electrophoretic mobility on starch gels and for segregation of allyl alcohol resistance by replica-plating onto YEPDE agar medium, containing 5 or 10 mM allyl alcohol and 1 ppm Antimycin A to convert the cells to phenotypic petites. Duplicate plates were incubated at 19" and 37" along with control plates from which allyl alcohol had been omitted. To measure the growth of haploid cells, three individual cells from each of four allyl-alcohol-resistant segregants were pulled by micromanipulation onto fresh agar slabs of the following media: (1) YEPD + 0, 1, 2.5, 5 or 10 mM allyl alcohol + 1 ppm Antimycin A; (2) YEPDE + 0, 5, 10, 15 or 20 mM allyl alcohol + 1 ppm Antimycin A; (3) YEPD + 0, 1, 2.5, 5 or 10 mM allyl alcohol + 0.4% freshly distilled sterile acetaldehyde + 1 ppm Antimycin A. Cells for each experiment were first grown in 5 ml of the appropriate liquid medium without allyl alcohol for 16 hr at 30". The agar slabs were transferred to plates of identically prepared media, which were then sealed with paraffin film and incubated at 19" or 3'7". Growth of resistant segregants was monitored by measuring clone diameter at least daily with a calibrated ocular micrometer. Growth experiments involving diploids were conducted in the same manner.