Isoleucine and Valine Metabolism of Escherichia coli

Transcription

1 JOURNAL OF BACTERIOLOGY, May 1968, P American Society for Microbiology Vol. 95, No. 5 Printed in U.S.A. Isoleucine and Valine Metabolism of Escherichia coli XVI. Pattern of Multivalent Repression in Strain K-12 SUSAN B. DWYER AND H. E. UMBARGER Department of Biological Sciences, Purdue University, Lafayette, Indiana Received for publication 24 February 1968 The levels of the five enzymes required for isoleucine and valine synthesis were examined under several growth conditions in strain K-12 of Escherichia coli and mutants derived from it. In strains with wild type repressibility, the same pattern of derepression was found on limiting isoleucine as is found to be constitutive in strain Tir-8, which has an altered isoleucine-activating enzyme. Homoserine dehydrogenase, which is essential for the biosynthesis of threonine and is normally derepressed on limiting isoleucine or threonine, is also derepressed in strain Tir-8. Threonine deaminase and homoserine dehydrogenase were partially repressed in strain Tir-8 by very high levels of isoleucine, but were not further derepressed over levels in minimal medium by limiting isoleucine. In the preceding paper (4), mutants of Escherichia coli strain K-12 that were resistant to thiaisoleucine were described. These organisms had altered isoleucyl soluble ribonucleic acid (srna) synthetases and had high levels of three of the five isoleucine and valine biosynthetic enzymes. In an effort to understand the significance of the physiological behavior of mutants bearing such altered synthetases, the levels of the isoleucine and valine biosynthetic enzymes have been examined under several growth conditions in the wild-type organism and in mutants derived from it. The results were compared with those obtained by using a thiaisoleucine-resistant mutant. Since enzymes in the pathway -to threonine had been shown (1) to be multivalently repressed by threonine and isoleucine, an enzyme in this pathway, homoserine dehydrogenase, was also examined. In addition, an attempt was made to demonstrate repression by thiaisoleucine in the wild-type strain K-12. The results of these studies are reported in this paper MATERIALS AND METHODS Organisms, media, and enzymatic assays. The wild-type strain K-12 of E. coli and several mutants derived from it were employed throughout. Strain Tir-8, a thiaisoleucine-resistant mutant, was described in the preceding paper (4). Strain 21, requiring threonine, leucine, and thiamine, was kindly provided by Robert Pritchard, University of Leicester. Strain AB2071 was kindly supplied by Frederick Neidhardt. This organism required isoleucine, valine, methionine, histidine, tryptophan, proline, arginine, and thiamine for growth. Strains 21 and AB2071 carry other non-nutritional genetic markers, but they were irrelevant for this study and were ignored. The minimal medium and procedures for growing the cells and for preparation of the extracts were those described previously (5). For growth of the auxotrophs, the minimal medium was supplemented with thiamine at 10-6 M and each of the required amino acids at a concentration of 4 X 10-4 M when an excess of the amino acid was desired. Growth factor-limited cultures were grown in a chemostat (see below). The medium in each case was supplemented with all components, except one, in excess. When either isoleucine or leucine was limiting, a concentration of 5 X 105 M was used; when threonine was limiting, a concentration of 7.5 X 10-5 M was used. Valine limitation was achieved by using 5.8 X 10-5 M glycyl- L-valine in place of L-valine. (These concentrations resulted in bacterial densities of 190 to 220,ug, dry weight, per ml.) Amino acid additions made to create repressing conditions are given below. The procedures for determining the activity of the isoleucine-valine biosynthetic enzymes were those described in the previous papers (4, 5). Homoserine dehydrogenase was determined by the method of Truffa-Bachi et al. (6). Reduced pyridine nucleotide formation was followed spectrophotometrically at 340 m, with a Gilford model 2000 recording spectrophotometer. The extracts contained an NADPH2 oxidase activity; however, at the concentrations of extract employed in the assay, the interference was not very significant. Nevertheless, the values given for homoserine dehydrogenase may be low by about 10%. Chemostat-grown cultures. The chemostat used consisted of a wide-mouth, 2-liter Erlenmeyer flask fitted with a no. 12 rubber stopper containing a cotton-plugged, 16-mm tube (for air escape), two 5-mm capillary tubes (for input of fresh medium and withdrawal of culture), and a -J-mm fritted glass sparger connected to an 8-mm tube. Medium was pumped into the culture by a model T-8 peristaltic pump (Sigmamotor Co., Mt. Vernon, N.Y.) through 1680

2 VOL. 95, 1968 REPRESSION OF ISOLEUCINE-VALINE ENZYMES 1681 gum rubber tubing (0.16-cm inside diameter) at a rate of one culture volume every 4 hr. The withdrawal tube was adjusted so that it maintained the culture volume constant at 1 liter. Fluid was withdrawn by means of the same peristaltic pump used to supply fresh medium, but the rubber tubing was 0.32-cm inside diameter, so that there was never any chance of the input exceeding the output owing to uneven pumping rates on the two sides of the peristaltic pump. The culture was aerated with air filtered through a cotton-filled drying tube. The cells remaining in the growth vessel were harvested after about three replacement volumes had been pumped through the apparatus. RESULTS Variation in enzyme levels in strains with normal regulation. Table 1 shows the levels of the five isoleucine and valine biosynthetic enzymes in extracts of cells grown under various conditions. The three strains had similar enzyme levels when grown in nutrient broth supplemented with 0.2% glucose. (Further supplementation with the salts of the minimal medium appeared to have little effect, the specific activities being intermediate between those of cells grown in nutrient broth and those of cells grown in minimal medium supplemented with excess isoleucine, valine, and leucine.) The effects of limiting valine and of limiting TABLE 1. Conditionsa isoleucine were conveniently observed by using extracts of cells of the mutant AB2071 grown in a chemostat with either valine or isoleucine limiting. When valine was limiting, the levels of the four enzymes of the pathway formed by this strain were derepressed over the levels observed in cells that had been grown batch-wise in nutrient plus glucose (Table 1). (The dehydrase is missing in this strain.) A different pattern of derepression was found in the cells grown in the chemostat with limiting isoleucine. Although threonine deaminase and transaminase B were more highly derepressed than they were in the valine-limited cells, the acetohydroxy acid synthetase and isomeroreductase were at essentially the same repressed level observed in nutrient-grown cells. Another way of creating an isoleucine deficiency is to incubate the wild-type strain K-12 in a minimal medium containing valine (2). Owing to the strong inhibition of the acetohydroxy acid synthetase in this strain, isoleucine formation is blocked and derepression of threonine deaminase, dihydroxy acid dehydrase, and transaminase B results (7). The results of such an experiment are shown in Table 1. One-half (500 ml) of a minimal medium culture was harvested and an extract was prepared. To the second half, 4 X 10-4 M L-valine was added, and the culture was incubated for an Levels of the isoleucine and valine biosynthetic enzymes in Escherichia coli strains under various conditions of growth Specific activityb Acetohydroxy Isomero- Threonine Dihydroxy acid Transaminase acid synthetase reductase deaminase dehydrase B Strain K-12 Nutrient + glucose Minimal + valine, leucine, isoleucine Minimal Minimal + valine, 3.5 hr Strain AB 2071 Nutrient + glucose Absent Isoleucine limiting (leucine, valine excess) Absent Valine limiting (isoleucine, leucine excess) Absent Strain 21 Nutrient + glucose Threonine limiting (leucine excess) Leucine limiting (threonine excess) Leucine limiting (threonine, isoleucine, valine excess) a Those cultures in which one amino acid was limiting were grown in a chemostat. Others were batch cultures. For other details, see Materials and Methods. b Expressed as micromoles of product per minute per milligram of protein.

3 1682 DWYER AND UMBARGER J. BACTERIOL. additional 3.5 hr before harvesting. During this time, the cell density increased by about 64%, while protein increased by about 30%, and the level of threonine deaminase increased 5.1-fold, the dehydrase 2.4-fold, and the transaminase 2- fold. The specific activities of the other two enzymes in the pathway, acetohydroxy acid synthetase and the isomeroreductase, were actually lower in the extract prepared from the valinetreated cells. When the increase in bacterial protein is considered, it can be calculated that, during the period of valine treatment, the loss in total isomeroreductase activity was only 10%, whereas the loss of the total acetohydroxy acid synthetase activity was about one-third. This observation again demonstrates the difference in stability of these two enzymes. Thus, there appears to be no difference between the effect of an isoleucine deficiency produced by valine inhibition of the parent and that produced by nutritional limitation of an auxotroph. In either case, there is derepression on limiting isoleucine of only the three enzymes that were found to be constitutively derepressed in the thiaisoleucine-resistant strains (4). In limiting the growth of E. coli strain 21 on leucine in a chemostat, all five enzymes were derepressed over the nutrient broth level, but the isomeroreductase was derepressed to only a limited extent (Table 1). When the same strain was grown on limiting threonine, the pattern of derepression typical of isoleucine limitation was observed. This pattern would be expected with threonine limitation, since threonine supplies four of the six carbons of isoleucine. Role of isoleucyl srna synthetase in the control of homoserine dehydrogenase. Isoleucine is involved with threonine in the multivalent regulation of the enzymes in the pathway to threonine, as well as with leucine and valine in multivalent regulation of the pathways to isoleucine and valine (1). Therefore, it was of interest to determine whether a mutant with a limited isoleucyl srna synthetase activity, and a reduced capacity to convert isoleucine to a form that participates in multivalent repression with leucine and valine, also had a reduced capacity to convert isoleucine to the form that participates in multivalent repression of the threonine-forming enzymes. To answer this question, homoserine dehydrogenase was measured in extracts of strains K-12 and Tir-8, a thiaisoleucine-resistant mutant. The enzyme in the wild-type strain was strongly repressed by growth on nutrient broth or on minimal medium supplemented with the three branched-chain amino acids (Table 2). (In these experiments, the isoleucine was added at a concentration 10-fold higher than that usually em- TABLE 2. Repression and derepression of threonine deaminase and homoserine dehydrogenase in Escherichia coli strains K-12 and Tir-8 Specific activityb Strain Conditionsa Threo- Homonine serine deami- dehydronase genase K-12 Nutrient + glucose Minimal + isoleucine, valine and leucine Minimal Minimal + valine, 3.5 hr Tir-8 Nutrient + glucose Minimal + isoleucine, valine and leucine Minimal Minimal + valine, 3.5 hr a Amino acid supplements, when used: L-isoleucine, 4 X 10-3 M; L-valine, 10-3 M, and L-leucine 4 X 10-4 M. b Expressed as micromoles per minute per milligram of protein. ployed.) Threonine deaminase was also measured in the same extracts, and its behavior in strain K-12 was as expected from the results shown in Table 1. In contrast, the levels of both enzymes in mutant cells grown in minimal medium were derepressed over those in wild-type cells grown in minimal medium. When the mutant cells were grown in nutrient broth plus glucose or in minimal medium supplemented with the branched-chain amino acids, some repression (below the level found when they were grown in minimal medium) of homoserine dehydrogenase and of threonine deaminase was observed. However, even when repressed, the level of neither enzyme was reduced to that of the wild type grown in minimal medium. That strain Tir-8 is nearly maximally derepressed is indicated by the observation that no further derepression of threonine deaminase was obtained when the cells were subjected to valine inhibition for 3 hr. (As will be shown below, strain Tir-8, like its parent, strain K-12, is sensitive to valine.) That the homoserine dehydrogenase activity was not derepressed upon exposure to valine is probably not significant, since exposure of the wild strain to valine for that period had very little effect on homoserine dehydrogenase. Under these conditions, the threonine deaminase of the wild strain was derepressed. Why the limitation of isoleucine caused by valine inhibition does not result in derepression of both enzymes is not clear. However, experience with isoleucine-requiring strains has indicated that appreciable derepression of homoserine dehydrog-

4 VOL. 95, 1968 REPRESSION OF ISOLEUCINE-VALINE ENZYMES 1683 enase can be more readily obtained by prolonged limitation of isoleucine, such as can be achieved in a chemostat. Effect of exogenous isoleucine on the growth of strain Tir-8. In view of the fact that some repression of threonine deaminase and homoserine dehydrogenase was obtained by growing strain Tir-8 in the presence of exogenous isoleucine (Table 2), it was of interest to determine whether exogenous isoleucine stimulated the growth of the organism. As shown in Fig. 1, 4 X 10-3M L-isoleucine did, in fact, increase the growth rate. When 4 X 1O-4 M L-isoleucine, a concentration sufficient for full growth of isoleucine auxotrophs, was used, only a very slight stimulation of growth rate was noted. Thus, with a high exogenous isoleucine concentration (which may lead to an increased internal pool of isoleucine), it is possible to compensate partially for the reduced affinity of the mutant isoleucyl srna synthetase. Effect of thiaisoleucine-resistance mutation on the sensitivity to valine. The derepression pattern found in strain Tir-8 is similar to that of a valineresistant mutant described by Ramakrishnan and Adelberg (3). In that mutant, the derepression was attributed to an Oc mutation affecting an operon consisting of the structural genes for the three enzymes affected. However, the growth of strain Tir-8 is quite sensitive to valine inhibition (Fig. 1B). Clearly, therefore, an elevated threonine deaminase is not a sufficient condition to lead w' 70 eo' w -J to resistance to valine. In strain Tir-8, however, there is presumably an additional phenotypic difference from the Oc mutant in that the isoleucyl srna synthetase of strain Tir-8 is abnormal. O 00X A B HOURS FIG. 1. Effect of L-isoleucinie and L-valine on the growth of Escherichia coli K-12 antd Tir-8. (A) Strain K-12; (B) strain Tir-8. Symbols: 0, minimal medium; 0, minimal medium with 5 X 10-5 M L-valine added at time indicated by arrow; a, minimal medium supplemented with 4 X 10O'M L-isoleucine. A Klett reading of 100 represenits 204 pig (dry weight) of cells per ml. DISCUSSION The finding that the limitation of isoleucine in E. coli strain K-12 leads to derepression of the same three of five isoleucine-valine enzymes that are derepressed in the thiaisoleucine-resistant mutants which have a reduced isoleucyl srna synthetase activity provides further evidence that the participation of isoleucine in the multivalent repression of those three enzymes is dependent upon isoleucine activation. That the three enzymes would respond as a unit is in accord with the evidence provided by the work of Ramakrishnan and Adelberg (3), that the structural genes for those three enzymes are under the control of a single operator region (i.e., they constitute an operon), whereas the other two structural genes are separately controlled. Since the mutation leading to thiaisoleucine resistance has not affected the level of the other two enzymes (acetohydroxy acid synthetase and acetohydroxy acid isomeroreductase), it is clear that multivalent repression of these enzymes does not involve isoleucine. Although both of these enzymes were derepressed by growth on limiting valine or on limiting leucine, the extent of derepression of the two enzymes was not the same (Table 1). Thus, isomeroreductase activity was eightfold higher on limiting valine than on limiting leucine, whereas synthetase activity was at least as high on limiting leucine as on limiting valine. That they do respond differently is in accord with the suggestion of Ramakrishnan and Adelberg (3) that the operator (or whatever may be the controlling element responding to the repression signal) affecting the synthetase gene may be distinct from that affecting the isomeroreductase gene. However, it may be that the different response is due to a greater demand upon the limiting amino acid for the formation of one enzyme. Thus, if the isomeroreductase contains much more leucine in its protein than does the synthetase, less isomeroreductase derepression with leucine limiting would be expected (Table 2). On the other hand, the variation may be caused by differences in enzyme stability under different intracellular conditions. Experiments in progress may eventually eliminate one or both of these possibilities. The finding that the thiaisoleucine-resistance mutation also affected the level of homoserine dehydrogenase, an enzyme in the pathway to threonine, indicates that a derivative of isoleucine, and not isoleucine itself, participates in the multivalent repression of the threonine-forming enzymes. Presumably, the derivative is analogous in form as well as in function to that required for repression of threonine deaminase, if it is not

5 1684 DWYER AND UMBARGER J. BACTERIOL. identical. Presumably, too, threonine will also be found to require a functional threonyl srna synthetase for the formation of the active derivative. However, neither presumption is, as yet, supported by experimental observation. Whether thiaisoleucine itself can act as a repressor is not known. Attempts to demonstrate a specific repression have not been successful. The addition of thiaisoleucine to cells starved for isoleucine does, indeed, prevent the derepressed formation of threonine deaminase. Such a result would also be obtained if protein synthesis were inhibited or if false protein were made. A more critical test for false repression would be to determine whether thiaisoleucine could convert the derepression pattern in threonine-limited cells (derepressed formation of both homoserine dehydrogenase and threonine deaminase) to the pattern of threonine-limited cells in the presence of excess isoleucine (derepressed formation only of homoserine dehydrogenase). However, when various levels (between 5 X 10-5 and 5 X 10-4 M) of thiaisoleucine were added to a threoninelimited chemostat culture, there was either a continued derepression of both homoserine dehydrogenase and threonine deaminase (subinhibitory amount of thiaisoleucine), or there was cessation of derepression of both enzymes (a result that might have been due to false protein formation). At higher levels, the inhibition was so severe that washout of the culture resulted at the pumping rate employed. ACKNOWLEDGMENTS This investigation was supported by Public Health Service grant GM from the National Institute of General Medical Sciences. One of the authors (S.B.D.) was the recipient of U.S. Public Health Service Predoctoral Fellowship 7-F7-GM-21, 011. LITERATURE CITED 1. FREUNDLICH, M Multivalent repression in the biosynthesis of threonine in Salmonella typhimurium and Escherichia coli. Biochem. Biophys. Res Commun. 10: LEAvrrr, R. I., AND H. E. UMBARGER Isoleucine and valine metabolism in Escherichia coli. XI. Valine inhibition of the growth of Escherichia coli strain K-12. J. Bacteriol. 83: RAMAKRISHNAN, T., AND E. A. ADELBERG Regulatory mechanisms in the biosynthesis of isoleucine and valine. II. Identification of two operator genes. J. Bacteriol. 89: SZENTIRMAI, A., M. SZENTIRMAI, AND H. E. UMBARGER Isoleucine and valine metabolism of Escherichia coli. XV. Biochemical properties of mutants resistant to thiaisoleucine. J. Bacteriol. 95: SZENTIRMAI, A., AND H. E. UMBARGER Isoleucine and valine metabolism of Escherichia coli. XIV. Effect of thiaisoleucine. J. Bacteriol. 95: TRUFFA-BACHT, P., G. LEBRAs, AND G. N. COHEN The threonine-sensitive homoserine dehydrogenase and aspartokinase activities of Escherichia coli. III. Inactivation at ph 9. Biochim. Biophys. Acta 128: UMBARGER, H. E., AND M. FREUNDLICH Isoleucine and valine metabolism in Escherichia coli. XIII. The role of repression in the sensitivity of strain K-12 to valine. Biochem. Biophys. Res. Commun. 18: