more than a single operon as was found for the Studies on the mode of regulation of enzyme synthesis in this system should be of considerable

Transcription

1 JOURNAL OF BACTERIOLOGY, May 1969, p Copyright 1969 American Society for Microbiology Vol. 98, No. 2 Printed in U.S.A. Biosynthesis of Branched-Chain Amino Acids in Yeast: Regulation of Synthesis of the Enzymes of Isoleucine and Valine Biosynthesis HOWARD BUSSEY AND H. E. UMBARGER Department of Biological Sciences, Purdue University, Lafayette, Indiana Received for publication 6 February 1969 Regulation of the levels of the five enzymes required for the biosynthesis of isoleucine and valine was studied in a Saccharomyces sp. When a mixture of isoleucine, valine, and leucine was added to the medium, the enzymes in the wild-type strain were repressed from about 30% (transaminase B) to about 907(7 (acetohydroxy acid synthetase) relative to the level in minimal medium-grown cells. Repression was also observed when threonine replaced isoleucine in the mixture but not when it replaced the other two amino acids. Significant derepression relative to the level in minimal-grown cells was not obtained by growing suitably blocked auxotrophs on medium containing limiting amounts of valine, isoleucine, or leucine. It has been known for some time (8, 17, 21) that Neurospora and yeast synthesize valine and isoleucine by pathways comparable to those in bacteria. The regulation of the synthesis of these enzymes has been well studied in Escherichia coli and Salmonella typhimurium (1, 3, 12, 13). However, very little is known about the process in eucaryotes. In Saccharomyces cerevisiae, it has been shown that control of enzyme activity at the level of endproduct inhibition is similar to that found in bacteria. Thus, threonine deaminase is inhibited by isoleucine and acetohydroxy acid synthetase is inhibited by valine (6, 9, 14). Little is known about the regulation of the synthesis of these enzymes. Magee and derobichon-szulmajster (Fed. Proc., p. 753, 1966) indicated that in yeast some repression could occur on complex media, but that the repression was apparently not multivalent, since repression did not occur upon the addition of the three branched-chain amino acids to the medium. Recently, the same authors (9) were unable to find any systematic changes in the level of threonine deaminase in the same organism. The gene-enzyme relationship for most of the enzymes in the isoleucine-valine pathway has been established, and the gene positions have been located on the chromosomes (7, 9, 20). These studies have shown that the genes are scattered on different chromosomes, ruling out the existence of an operon or a cluster of genes organized into 623 more than a single operon as was found for the corresponding genes in E. coli (13). Studies on the mode of regulation of enzyme synthesis in this system should be of considerable interest because of the unclustered arrangement of the genes, and because the yeast chromosomes are themselves more complex than those in bacteria and the expression of their genetic information may be controlled in a manner more nearly like that in mammalian systems. We used the Lindegren strain of Saccharomyces and several auxotrophic mutants from the Lindegren collection to study the general problem of regulation. The results obtained suggest that a noncoordinate, multivalent repression of the isoleucine-valine pathway does indeed operate in yeast, although the changes in enzyme levels are much smaller than those found in bacteria. MATERIALS AND METHODS Organism. The wild-type organism used was a haploid strain of Saccharomyces strain 60615, a mating type, from the Lindegren collection. Since this strain was derived from genetic crosses involving several species, a species designation is inappropriate. The mutants used, also haploid strains from the Lindegren collection, were: strain (is-3), lacking activity of the dihydroxy acid dehydrase and requiring isoleucine and valine; strain Q-869 (is-l), deficient in threonine deaminase and requiring isoleucine; and strain 89173, a leucine-requiring strain lacking both a-isopropylmalate isomerase and j3-isopropylmalate dehydrogenase owing to two separate genetic lesions, and also requiring histidine and tryptophan.

2 624 BUSSEY AND UMBARGER J. BACTiERIOL. Medium and growth of cells. Cells were grown on the medium of Halvorson (4), with 10 g of glucose per liter, and supplemented by the addition of L-amino acids or complex nutrients as indicated in the text. One-liter batch cultures were grown in 2-liter triplebaffled Erlenmeyer flasks (Bellco Glass, Inc., Vineland, N.J.) on a rotary shaker at 30 C. Chemostat cultures, also at 30 C, were grown in the apparatus described previously (1). Cell harvest and enzyme extraction. Cultures were harvested in the logarithmic phase of growth at cell densities of 0.8 to 1.0 mg (dry weight) of cells per ml, which gave Klett readings with a blue (no. 42) filter of 180 to 220. The cells were filtered through glassfiber filter paper (9-cm discs, 934AH; Reeve Angel, Clifton, N.J.). The filtered cells were washed well with 0.1 M potassium phosphate buffer (ph 7.5) and transferred to a Sorvall Micro-Omni-Mixer (Ivan Sorvall, Norwalk, Conn.) in which they were disrupted in 0.1 M potassium phosphate buffer (ph 7.5). For 0.8 to 1.0 g (dry weight) of cells, 4 to 5 g of glass beads (size ; 3M Co., St. Paul) and 5 ml of buffer were employed. During disruption (five 20-sec periods at top speed were sufficient), the stainlesssteel chamber was cooled by immersion in an ice bath. Between grinding periods, the chamber was allowed to cool in the ice bath. The cell-free crude extract was recovered by filtration through glass-fiber filter paper, and the extract was desalted by passage through a Sephadex G-25 column equilibrated with 0.1 M potassium phosphate buffer, ph 7.5, prior to the enzyme assays. Enzyme assays. Enzyme activities were determined TABLE 1. at 30 C immediately after preparation of the extracts. Threonine deaminase was assayed in a 1-ml reaction mixture containing (in micromoles): potassium phosphate buffer (ph 8.0), 100; pyridoxal phosphate, 1; L-threonine, 40; extract, 1 to 2 mg of protein. Routinely, the course of the reaction was followed by incubating replicate tubes for 1, 5, and 10 min. The reaction was stopped with 0.1 ml of 50% trichloroacetic acid. Keto acids were determined by the method of Friedemann and Haugen (2). Acetohydroxy acid synthetase was assayed as described by St0rmer and Umbarger (16) in 0.1 M potassium phosphate buffer (ph 7.5). Acetohydroxy acid isomeroreductase was determined by the assay of Szentirmai et al. (19), except that 5,umoles of a-acetohydroxybutyrate was substituted for a-acetolactate. Dihydroxy acid dehydrase was measured as described by Szentirmai et al. (19), except that 20 jumoles of dihydroxy-3-methylvalerate, was used in place of dihydroxyisovalerate. Transaminase B was assayed as described by Szentirmai and Umbarger (18). RESULTS Levels of isoleucine- and valine-forming enzymes in the wild-type strain grown in minimal medium and the effect of adding branched-chain amino acids to the minimal growth medium are shown in Table 1. Repression of the isoleucinevaline enzymes in this strain to levels below those found when the cells were grown on a minimal medium was obtained when all three branched- Effect of growth in the presence of the branched-chain amino acids on the levels of the isoleucine-valine-forming enzymes in yeast Supplement to minimal medium Specific activitya Threonine Acetibcrdroxy Isomero- cdihydrxy- Transaminase deaminase synthetase reductase dehydrase B None L-Leucine, 10-2 M L-IsoleuCine, 10-2 M, + L-valine, X 10-2 M L-ISOleUCinle, 10-2 M, + L-valine, X 10-2 M, + L-leucine, 10-2 M L-Threonine, 10-2 M, + L-valine, X 10-2 M, + L-leucine, 10-2 M L-Threonine, 10-2 M, + L-isoleuCine, M, + L-valine, 5 X 10-2 M, + L-leuCine, 10-2 M Yeast extract, 0.5%, + 0.5% tryptone a Expressed as micromoles per minute per milligram of protein. The uncertainties in the above values are estimated to be not more than :1:20%o in the range 0.3,moles per min per mg of protein to 0.02 ;moles per min per mg of protein, increasing to i50%0 at ;moles per min per mg of protein.

3 VOL. 98, 1969 ISOLEUCINE-VALINE ENZYMES IN YEAST 625 chain amino acids were added, but not when isoleucine and valine only or leucine alone was added. Thus, repression of the isoleucine-valine enzymes is probably multivalent, that is, it requires isoleucine, valine, and leucine, although threonine can replace isoleucine. Experiments not reported in Table 1 have shown that threonine can replace only isoleucine, but not valine or leucine. This observation may indicate that the presence of threonine in the medium leads to a high isoleucine pool in the cell (presumably via a-ketobutyrate). Other combinations of exogenous amino acids tested caused no repression of these enzymes. Repression also occurred in a complex medium, e.g., Halvorson medium supplemented with tryptone and yeast extract (Table 1). This repression was of the same magnitude as that seen on addition of the branched-chain amino acid supplements. To study the kinetics of repression, some pilot experiments using the isomeroreductase, the most stable enzyme in the pathway, were performed. After the transfer of cells from a (repressing) medium supplemented with yeast extract and tryptone to minimal medium (Fig. 1), there was an approximately twofold elevation of the isomeroreductase level within the first generation after the medium shift to a value close to the steady state level of minimal medium-grown cells. On the reverse shift, from minimal to enriched medium, there was a repression which could be readily explained by cessation of synthesis, followed by dilution of the enzyme by the growth of the cells (Fig. 2).. 0D2 4O- 4%._ C.) UO.I transfer 0 I generation 0 I Time (hours) after transfer FiG. 1. Derepression of the acetohydroxy acid isomeroreductase after a transfer of cells from a (repressing) medium supplemented with yeast extract and tryptone (0.5% ofeach) to a minimal medium. The cells were transferred from the nutrient to the minimal medium at zero-lime (arrow). 0 0 ""0.02 -O a,, I I transfer I generation Time (hours) after transfer FIG. 2. Repression of the acetohydroxy acid isomeroreductase after a transfer of cells from a minimal to a (repressing) medium supplemented with yeast extract and tryptone. The cells were transferred from the minimal to the nutrient medium at zero-time (arrow). Absence of coordination of repression. Although all the enzymes of the pathway were repressed to some extent, the repression was apparently not tightly coordinate. Threonine deaminase and transaminase B were least subject to repression, the "repressed" (supplemented medium) levels always being less than the "unrepressed" (minimal medium) levels. This finding, nevertheless, is unlike that of Magee and derobichon-szulmajster (14), who were unable to find any systematic changes in the level of threonine deaminase in their strain. The acetohydroxy acid synthetase was more strongly repressed than any of the other enzymes tested, being as low as one-sixth the level in minimal medium-grown cells. The isomeroreductase and the dihydroxyacid dehydrase exhibited intermediate degrees of repression. Such a seemingly loose coordination of regulation of enzyme synthesis is clearly of interest in relation to the scattered genes of the pathway. It is perhaps worth noting that such an uncoordinated system would place some restraint on the constancy of structure of any multienzyme aggregate of the enzymes in this pathway, a possibility raised by Kakar and Wagner (7). Derepression of the isoleucine-valine-forming enzymes. To study further the regulation of enzyme synthesis in the pathway, several attempts were made to elevate the levels of the enzymes above those found when the cells were grown on a minimal medium. Despite a variety of approaches, no derepression was observed, although perhaps it is of value to list some of the approaches. Several isoleucine-valine and leucine auxotrophs were grown either in batch or chemostat culture with a required amino acid provided in 4

4 626 BUSSEY AND UMBARGER J. BACTERIOL. limiting amounts. Strain (is-3), an isoleucine and valine requirer lacking dihydroxyacid dehydrase, was grown in batch culture with isoleucine limiting and in a chemostat with limiting valine (Table 2). Under these conditions, where strong derepression would be found in E. coli (1), no derepression was observed. Strain Q-869 is-1, an isoleucine-requiring mutant lacking threonine deaminase, was grown in batch culture with an excess of isoleucine, valine, and leucine, and with isoleucine limiting. Although there was some repression with isoleucine, valine, and leucine present, the increase in specific activity on limiting isoleucine was very small (Table 2). When isoleucine was replaced exogenously by a-aminobutryrate as a source of a-ketobutyrate, a potential inducer of the pathway, some small elevation of the enzyme levels was observed, suggesting that inducers play no very active part in the regulation, at least under these conditions. Strain 8917 (le-l, le-2) is a leucine auxotroph, with lesions in both the a-isopropylmalate isomerase and the f3-isopropylmalate dehydrogenase, which showed a 10-fold derepression of the a- isopropylmalate synthetase when grown either batchwise or in a chemostat on limiting leucine (15). Under such conditions, no derepression of the isoleucine- and valine-forming enzymes was found; in fact, the threonine deaminase appeared lower in the chemostat (Table 2). Growth of wild-type cells in media containing a rich supply of nutrients with the absence of either isoleucine or valine has been shown to result in derepression of the isoleucine-valine enzymes in Bacillus subtilis (G. W. Hatfield, personal communication). However, this method caused no derepression in Saccharomyces (Table 3). DISCUSSION The apparent inability of yeast to be derepressed with respect to the isoleucine-valine enzymes, at least under the conditions examined here, could imply that, on minimal medium, these TABLE 2. Effect ofgrowth factor limitation on the isoleucine-valine-forming enzymes ofyeast Strain Specific activity' Culture conditions and supplement to minimal medium Threonine Acetohydroxy Isomero- Dihydroxy deaminase synthetase reductase acid Batch L-Isoleucine, 10- M + L-Valine, 103 M Batch L-ISoleUCine, 1o4 M (limiting) + L-Valine, 103 M Chemostat L-Isoleucine, 103 M + L-Valine, 10- (limiting) Q-869 Batch L-Leucine, 103 M + L-Valine, o10 M + L-Isoleucine, lo3 M Batch L-Leucine, 103 M + L-Valine, 103 M + L-Isoleucine, 104 M (limiting) 89173b Batch L-Leucine, 10-3 M Batch L-Leucine, 10r4 M (limiting) Chemostat O L-Leucine, 1r4 M (limiting) a Expressed as micromoles per minute per milligram of protein. b This strain also required L-tryptophan and L-histidine, which were added to the medium at concentrations of 103 M.

5 VOL. 98, 1969 ISOLEUCINE-VALINE ENZYMES IN YEAST 627 TABLE 3. Effect of amino acid assay media on the isoleucine-valine-forming enzymes of wild-type yeast Specific activitya Supplement to minimal Threo- Aceto- Dihymedium nine hydroxy Iso- droxy acid merore- acid demi nase nae synthe- tase ductase dehy- drase Isoleucine assay medium Valine assay medium a Expressed as micromoles per minute per milligram of protein. enzymes are being synthesized at their maximal rate. If one assumes that the corresponding enzymes in bacteria and yeast exhibit comparable catalytic capacity and that the lower activity represents low amount of enzyme, the regulatory mechanism in yeast may represent an evolutionary advance over that in bacteria. In the latter, the mere absence of the signal, "repress," leads in some enzyme systems to very high and physiologically detrimental levels of a single or even several enzymes. If yeast has preserved the capacity to repress enzymes when they are not needed but has developed a mechanism to limit derepression to minimal medium levels, the control mechanism is indeed a more sophisticated one. The multivalent repression of these enzymes in bacteria is known to involve at least the activation of valine and isoleucine by the corresponding aminoacyl transfer ribonucleic acid (trna) synthetases (11, 19). Although comparable findings have not been reported for any biosynthetic pathway in yeast, it will be of interest to see whether the isoleucyl-trna synthetase-deficient mutant recently isolated by Hartwell and McLaughlin (5) exhibits any alteration in the isoleucine-valine biosysnthetic enzymes. Even though large changes in the amounts of the enzymes involved in the biosynthesis of isoleucine and valine in yeast have not been found, the regulation of their pathway may be quite effective, since marked derepression does not seem to occur and, furthermore, metabolite flow is regulated as shown by others through end-product inhibition. ACKNOWLEDGMENTS We thank Gertrude Lindegren for supplying the yeast strains used in this study. This investigation was supported by Public Health Service grant GM from the National Institute of General Medical Science. LITERATURE CITED 1. Dwyer, S. B., and H. E. Umbarger Isoleucine and valine metabolism of Escherichia coli. XVI. pattern of multivalent repression in strain K-12. J. Bacteriol. 95: Friedemann, T. E., and G. E. Haugen Pyruvic Acid II. The determination of keto acids in blood and urine. J. Biol. Chem. 14: Freundlich, M., R. 0. Bums, and H. E. Umbarger Control of isoleucine, valine, and leucine biosynthesis. I. Multivalent repression. Proc. Nat. Acad. Sci. U.S. 48: Halvorson, H Studies on protein and nucleic acid turnover in growing cultures of yeast. Biochim. Biophys. Acta 27: Hartwell, L. H., and C. S. McLaughlin Mutants of yeast with temperature-sensitive isoleucyl-trna synthetases. Proc. Nat. Acad. Sci. U.S. 59: Holzer, H., C. Cennamo, and M. Boll Product activation of yeast threonine dehydratase by ammonia. Biochem. Biophys. Res. Commun. 14: Kakar, S. N., and R. P. Wagner Genetic and biochemical analysis of isoleucine-valine mutants of yeast. Genetics 49: Lewis, K. F., and S. Weinhouse Studies in valine biosynthesis. II. a-acetolactate formation in microorganisms. J. Amer. Chem. Soc. 80: Magee, P. T., and H. derobichon-szulmajster The regulation of isoleucine-valine biosynthesis in Saccharomyces cerevisiae. 2. Identification and characterisation of mutants lacking the acetohydroxyacid synthetase. Eur. J. Biochem. 3: Magee, P. T., and H. derobichon-szulmajster The regulation of isoleucine-valine biosynthesis in Saccharomyces cerevisiae. 3. Properties and regulation of the activity of acetohydroxyacid synthetase. Eur. J. Biochem. 3: Neidhardt, F. C Roles of amino acid activating enzymes in cellular physiology. Bacteriol. Rev. 30: Ramakrishnan, T., and E. A. Adelberg Regulatory mechanisms in the biosynthesis of isoleucine and valine. I. Genetic derepression of enzyme formation. J. Bacteriol. 87: Ramakrishnan, T., and E. A. Adelberg Regulatory mechanisms in the biosynthesis of isoleucine and valine. II. Identification of two operator genes. J. Bacteriol. 89: derobichon-szulmajster, H., and P. T. Magee The regulation of isoleucine-valine biosynthesis in Saccharomyces cerevisiae. I. Threonine deaminase. Eur. J. Biochem. 3: Satyanarayana, T., H. E. Umbarger, and G. Lindegren Biosynthesis of branched-chain amino acids in yeast: regulation of leucine biosynthesis in prototrophic and leucine auxotrophic strains. J. Bacteriol. 96: St0rmer, F. C., and H. E. Umbarger The requirement for flavine adenine dinucleotide in the formation of acetolactate by Salmonella typhimurium extracts. Biochem. Biophys. Res. Commun. 17: Strassman, M., J. B. Shatton, and S. Weinhouse Conversion of a-acetolactic acid to the valine precursor, a,#-dihydroxyisovaleric acid. J. Biol. Chem. 235: Szentrimai, A., and H. E. Umbarger Isoleucine and valine metabolism of Escherichia co I i. XIV. Effect of thiaisoleucine. J. Bacteriol. 95:

6 628 BUSSEY AND UMBARGER J. BACTERIOL. 19. Szentnrai, A., Szeatirmai, M., and Umbarger, H. E Isoleucine and valine metabolism of Escherlchla coll. XV. Biochemical properties of mutants resistant to thiaisoieucine. J. Bacteriol. 95: Von Borstel, R. C On yeast genetics. Microb. Genet. Bull. Suppl. K, no. 25. Oak Ridge National Laboratory Oak Ridge, Tenn. 21. Wagner, R. P., Radhakrishnan, A. N., and Snell, E. E The biosynthesis of isoleucine and valhie hn Neurospora crassa. Proc. Nat. Acad. Sci. U.S. 44: