The gcr (Glycolysis Regulation) Mutation of Saccharomyces cerevisiue*

gcr is a mutation considerably decreasing the assayed amounts of most gIycoIysis enzymes in Smckuromyees cerevisiue (Clifton, D., Weinstock, S. B., and Fraenkel, D. G. (1978) Genetics 88, 1-11), We show here that although in the wild type strain the amounts of these enzymes do not greatly differ between cells from differ- ent media, in the gcr mutant strain most of the enzyme amounts are 5% or less, relative to wild type, from cells grown without sugars, but 20-50% from cells grown with sugars. Lower relative values were found for phosphoglycerate mutase and enolase. A corresponding al- teration in the mutant in the intensities of several major protein bands could even be seen in stained gels after electrophoresis of crude extracts; the profiles were otherwise normal. Results of titration of phosphoglyc- erate kinase with antibody accorded with activity, Transfer of ceils between the two types of media did not lead to a more rapid adjustment of enzyme amounts than expected from the steady state levels. gcr is not allelic to GPM (the gene for phosphoglycerate mutase) or to RNA1 (which affects transport of RNA from the nucleus). Translation of total RNA in a rabbit reticulo- cyte lysate gave a pattern of polypeptides similar to the in vivo one. Thus, gcr is likely to affect somehow mRNA synthesis or lifetime for a discrete number of proteins.

gcr is a mutation considerably decreasing the assayed amounts of most gIycoIysis enzymes in Smckuromyees cerevisiue (Clifton, D., Weinstock, S. B., and Fraenkel, D. G. (1978) Genetics 88, 1-11), We show here that although in the wild type strain the amounts of these enzymes do not greatly differ between cells from different media, in the gcr mutant strain most of the enzyme amounts are 5% or less, relative to wild type, from cells grown without sugars, but 20-50% from cells grown with sugars. Lower relative values were found for phosphoglycerate mutase and enolase. A corresponding alteration in the mutant in the intensities of several major protein bands could even be seen in stained gels after electrophoresis of crude extracts; the profiles were otherwise normal. Results of titration of phosphoglycerate kinase with antibody accorded with activity, Transfer of ceils between the two types of media did not lead to a more rapid adjustment of enzyme amounts than expected from the steady state levels. gcr is not allelic to GPM (the gene for phosphoglycerate mutase) or to RNA1 (which affects transport of RNA from the nucleus). Translation of total RNA in a rabbit reticulocyte lysate gave a pattern of polypeptides similar to the in vivo one. Thus, gcr is likely to affect somehow mRNA synthesis or lifetime for a discrete number of proteins.
Enzymes of the glycolytic pathway comprise a major fraction of yeast-soluble protein, and mutations are known for most of them (1). Such mutations affect single enzymes, with one exception, gcr ("glycolysis regulation"), which seems to decrease the level of most enzymes of the pathway (2). This paper is about the gcr mutation.
Enzyme Assays-Extracts were prepared as previously (2), using cells resuspended in extract buffer (50 mM KHzP04,2 mM Na-EDTA, and 2 mM 2-mercaptoethanol, pH 7.4), usually supplemented with 2 m~ phenylmethylsulfonyl fluoride. Enzyme assays were also as described previously (2) with the exception of aldolase (5). Protein was * This work was supported by Grant P C M -7 9 -1~2 from the National Science Foundation and Grant 2 ROI GM21098 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked ' ' a d u e r t~e~e n~" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. assayed by the biuret method (6) unless otherwise specified. ~~c~o p~o r e s~-P o l y a c r y l~i d e g e t e~e c t~p h o r e s~ was according to Laemmli (7) but with slabs (10 X 14 X 0.15 cm) @), 5% stacking, and 8-12% or 8-14% separating gels. Samples of extract were pretreated by 2-min boiling in treatment buffer (7). Electrophoresis was 16 h, 25 V. Protein staining was according to Fairbanks et al. (9). Fixing solution was 3.46% sulfosalicylic acid and 11.5% trichloroacetic acid.
I n Vitro Translation-Total RNA was extracted by the method of McAlister and Finkelstein (10)

RESULTS
Growth-The gcr mutant strain DFY67 was obtained originally by selection for glucose resistance in a pyruvate kinase mutant (pyk) and subsequent outcrossing of the pyk mutation. Strain D N 6 7 was reported to grow very slowly on enriched medium with glucose but normally on enriched medium with pyruvate, and cells from the latter "permissive" medium had low assayed amounts of most glycolysis enzymes (2).
More data on growth are shown in Table I. In the usual enriched medium (which allows limited growth in the absence of additional carbon source), the mutant strain grew normally on gluconeogenic carbon sources, grew more slowly than normal on maltose, and barely grew at all on glucose or fructose. Glucose was inhibitory in an otherwise adequate medium (Le. one containing a gluconeogenic carbon source, such as lactate), while maltose was not inhibitory.
In the above characteristics, the gcr mutant strain did not differ markedly from some other known glycolysis mutants. The pattern of growth on minimal medium was somewhat different, however, being relatively less impaired on sugars but more impaired on lactate (Table I). Glycolysis enzyme levels were, therefore, assayed after growth in a variety of media.
Inducibility of Enzymes- Table 11, columns 1-4, shows data from four cultures: the wild type and mutant strains grown in enriched medium with lactate plus glycerol or lactate plus glycerol plus maltose. As is seen by normalizing to the values in the wild type strain from medium with lactate plus glycerol, the mutant had 1-10% levels of most glycolysis enzymes. {Similar results were reported for growth with pyruvate (Z).) In the cultures also containing maltose, on the other hand, although the wild type profile did not change much (compare columns 2 and 11, in the mutant the enzyme levels were about 5-fold higher (compare columns 4 and 3).

G~~~o~y s i s R e~~a t i o n M~t a t i o n (ger) of S. cerevisiae 13075
similar experiments were done with cells grown in other media. Table I1 shows that the average profile from three gluconeogenic media (column 7f and from nine media containing sugars (column 8) resembled the gluconeogenic and glycolytic profiies of columns 3 and 4, respectively. Again (columns 5 and 6), the wild type patterns differed little between the two types of media. Other wild type strains, including DFY56, the other parental strain used in construction of strain DF'Y67 (gcr) (2), had the usual wild type profile (data not shown). The general pattern of the gcr effect could also be seen in sodium dodecyl sulfate gels of total soluble protein, stained with Coomassie blue (Fig. 1). For the wild type strain in the two standard conditions, the protein patterns were similar, although there were some clear differences, one of which probably reflects the inducible enzyme, maltase, size 63,000 (12). By comparison with migration of known standards, some of which are yeast glycolysis enzymes, and the knowledge that a These minimal plates also contained lysine, leucine, and tryptophan, each 25 pg/ml. these enzymes comprise an appreciable fraction of yeast-soluble protein (1), some of the prominent bands of Fig. 1 may be tentatively assigned to particular glycolysis proteins or mixtures thereof. In the extracts of the gcr mutant strain grown on lactate plus glycerol, those prominent bands were considerably less intense, while in the extract from the mutant grown with maltose, they were stronger (see, for example, the band migrating like glyceraldehyde-3-P dehydrogenase (Gld)). Fig. 1 also shows that although one set of proteins was clearly affected by the gcr mutation, most other protein bands were of similar intensity in wild type and mutant strains. (Two enzymes known by assay to be unaffected are glucose-6-P dehydrogenase and isocitrate dehydrogenase (Table I).) Thus, it is possible that the gcr m u~t i o n affects only a small set of proteins.
Although the gel patterns make it unlikely that the enzymes are present in normal amount but have lower activity, one would also like to assess the amount of antigen. This was done for phosphoglycerate kinase (Fig. 2). Antibody titration showed that the amount of activity matched the amount of antigen in the three cases of wild type (high activity), mutant with uninduced low activity, and mutant with induced moderate activity.
The gcr mutation has not yet been mapped, but the earlier work (2), as well as experiments to be reported, has shown the phenotype of impaired growth on sugars to segregate 2 2 in crosses with wild type strains, as expected for a single gene mutation. To the degree it has been tested, the phenomenon of apparent partial inducibility of the affected enzymes by sugars is a consequence of the same mutation. For example, in three tetrads from the cross DFYl (GCR) X DFYll2 (gcr) giving 2 2 segregation of growth on glucose, the ~n d u c i b~~t y phenomenon was seen in all six glucose-negative segregants (gcr) but not in the others (GCR). Interestingly, however, induction was not observed with maltose itself in all cases, presumably because one of the parental strains (DFY 1) carries a MAL gene and the other (DFY112) does not. This result would be consistent with induction requiring metabolism of the sugar.
Aitered Synthesis o r A l t~r e~~e g r~d u t~o n of the Proteins?-The experiments above do not exclude the possibility that the gcr mutation affects protein turnover. Phenylmethylsuifonyl fluoride did not markedly affect the enzyme profile, but it was  Nonidentity with Other Mutants-We first considered the possibility that gcr might affect a single, known glycolytic enzyme, with the pleiotropic phenotype being an indirect consequence of that protein's alteration. The two most likely candidate enzymes might be phosphoglycerate mutase and enolase since their levels are most affected (Table 11) (3)). Nonetheless, to test whether gcr might be a special allele of gpm, a cross was performed between DFY67 (gcr) and strain DFY159 (gprn). The diploid grew normally on glucose and had normal glycolysis enzyme levels. If the two mutations were actually allelic, segregants would be 2:2 gpm/gcr. In fact, the usual tetrads contained one haploid with wild type enzyme profiie and three others, which by assay and backcrossing were found to be gpm, gcr, and the double mutant gpm gcr.
These results accord with gcr and gpm being in different genes.  usually included in the extraction buffer. Change of medium experiments was also done (Table 111). In the first one, growth was in gluconeogenic medium with lactate, and maltose was added for a 3-h incubation. During this time, amounts of the affected enzymes (in this experiment glyceraldehyde-3-P dehydrogenase and triose-P isomerase were assayed) increased, as expected, and the i.ncreases did not occur in the presence of cycloheximide. In the second experiment, cultures were transferred from a medium containing maltose and lactate to one without maltose. In this case, there was relatively little change in the enzyme profile over the 3 h. These and similar experiments did not reveal a rapid change in enzyme activities which might have indicated modification of preformed proteins; the data would best fit with the rate of synthesis of the affected  Another test for whether gcr might directly affect phosphoglycerate mutase or enolase was to examine the temperature lability of these enzymes in a revertant strain. The revertant chosen, DFY67R, contained approximately 50% of the wild type levels of phosphoglycerate mutase and enolase, and their labilities in crude extracts did not clearly differ from wild type, being for phosphoglycerate mutase, 50% loss of activity after 2 min at approximately 65 "C and for enolase, 50% loss a t approximately 50 "C. These results give no support to the  proteins are a, phosphorylase b, 92,500,  b, bovine serum albumin, 69,000, c, ovalbumin, 46 idea that gcr actually affects the structure of one of those two enzymes.
T h e r n a l mutation (4) is thought to affect RNA transport from the nucleus. The strain ts136 (mal ) did show some indication of being more temperature sensitive in growth with glucose than in gluconeogenic growth, but assay showed it to contain a normal profde of glycolytic enzymes, and a diploid between ts136 and DF112 (gcr) also gave a normal profde as well as growing well even at 37 "C on glucose minimal medium (data not shown). Hence, gcr and rnal are unlikely to be allelic.
mRNA-If the gcr mutation does not affect the structure of glycolysis enzymes, it is likely to affect their synthesis. T o test whether or not mRNA for these enzymes was present in normal amount, total RNA was obtained from the two mutants in the two standard growth conditions and used in an in vitro translation system from rabbit reticulocytes (Fig. 3). Total incorporation of radioactivity to hot trichloroacetic acidinsoluble material was similar in the four incubations (see legend), but the labeled polypeptide profdes differed. In the mutant, several prominent bands were missing or much less intense in the culture grown without maltose but more normal from the culture with maltose, while in the wild type strain, the two profdes were similar.
Thus, the in vitro polypeptide pattern was similar qualitatively to the in vivo pattern, and the deficiency of glycolytic enzymes is associated with a deficiency in their messages, at least as assessed in the heterologous translation system.

DISCUSSION
Lesion of the gcr Mutant-The gcr mutation reduces the amount of most glycolysis enzymes. The several types of experiments in this paper suggest that the enzymes themselves may be normal and that the lesion is likely to affect gene expression at the level of mRNA synthesis or stability. The effect may be on synthesis of a limited class of proteins, which includes most of the glycolysis enzymes. The exact lesion is not known.
It is tempting to speculate that gcr affects some regulatory component in glycolysis gene expression. However, it is uncertain even whether glycolysis enzyme levels are normally highly regulated. The work of Maitra and Lobo (18,19), who studied differential enzyme synthesis after glucose addition to gluconeogenically grown cells, suggested large induction factors depending on such metabolites as glucose-6-P, while the type of data given in Table I1 for the wild type strain does not show large differences in enzyme level in the steady state. Perhaps there is induction in all strains, but thresholds differ, so it is revealed by the gcr mutation in strains otherwise "constitutive." Growth of the gcr Mutant-The growth phenotype of the gcr mutant generally accords with its enzyme levels. In the usual enriched medium with sugars, the levels of phosphoglycerate mutase and enolase are low, and it may be that the growth impairment primarily reflects these blocks. In enriched medium with gluconeogenic carbon sources such as lactate, glycolysis is a biosynthetic pathway and certain end products are supplied by the medium, so perhaps the minimal levels of the several glycolysis enzymes are adequate. It is not clear, however, to what degree the phenotype of the gcr mutant will eventually prove to be explained in detail by the glycolysis enzyme pattern.