Nature of the effector of catabolite repression of beta-galactosidase in Escherichia coli

Loomis, William F., Jr. (Massachusetts Institute of Technology, Cambridge, Mass.), and Boris Magasanik. Nature of the effector of catabolite repression of beta-galactosidase in Escherichia coli. J. Bacteriol. 92:170-177. 1966.-Many carbon sources were found to give rise to catabolite repression of beta-galactosidase in a mutant strain of Escherichia coli lacking hexose phosphate isomerase activity. Compounds containing glucose or galactose cannot be formed from several of these carbon sources in this mutant strain, and, therefore, appear not to be required for catabolite repression of beta-galactosidase. Glucose was observed to elicit catabolite repression of beta-galactosidase in another mutant strain under conditions in which the formation of compounds of the citric acid cycle is inhibited. If catabolite repression of the lac operon is mediated by a single compound, it appears that the compound is related to the pentoses and trioses of intermediary metabolism. The repression of beta-galactosidase by galactose in galactokinase negative strains was shown to be independent of the gene, CR, which determines catabolite sensitivity of the lac operon, and to be dependent on a functional i gene.     

Cyclic 3′,5′-Adenosine Monophosphate and N-Acetyl-glucosamine-6-Phosphate as Regulatory Signals in Catabolite Repression of the lac Operon in Escherichia coli

When an Escherichia coli mutant lacking the enzyme N-acetyl-glucosamine-6-phosphate (AcGN6P) deacetylase is grown in a succinate-mineral salts medium and exposed to an exogenous source of N-acetylglucosamine, approximately 20 to 30 pmoles of AcGN6P per mug of cell dry weight will accumulate in these cells. This accumulation occurs within 2 to 4 min after the addition of N-acetylglucosamine and is coincident with the production of a severe permanent catabolite repression of beta-galactosidase synthesis. This repression does not occur if adenosine 3',5'-cyclic phosphate (cyclic AMP) is added to the cells before AcGN6P accumulates. An immediate derepression occurs when cyclic AMP is added to cells that have already accumulated a large AcGN6P pool. These findings are consistent with the view that low-molecular-weight carbohydrate metabolites and cyclic AMP play key roles in the catabolite repression phenomenon, and that metabolites such as AcGN6P may participate in the represion mechanism by influencing either the formation or degradation of cyclic AMP in E. coli.     

Catabolite repression of the lac operon. Effect of mutations in the lac promoter

1. Several lac diploid strains of Escherichia coli were constructed and tested to discover whether mutations in the lac promoter alleviate catabolite repression. 2. In each of these diploids the chromosome carries one of the promoter mutations, L8, L29 or L1; so that the rate of synthesis of the enzymes of the lac operon is only 2-6% of the fully induced wild-type. Each diploid harbours the episome F'lacM15 that specifies the synthesis of thiogalactoside transacetylase under the control of intact regulator, promoter and operator regions, but has a deletion in the structural gene for beta-galactosidase. In each diploid more than 90% of the thiogalactoside transacetylase is synthesized from the episome, and 100% of the beta-galactosidase is synthesized from the chromosome, and comparison of the extent of catabolite repression that the two enzymes suffered indicated whether the chromosomal promoter mutation relieves catabolite repression. 3. In the strains in which the promoter carries either of the point mutations L8 or L29 the enzymes were equally repressed, suggesting that neither L8 nor L29 affects catabolite repression. 4. In a diploid strain harbouring the same episome but carrying deletion L1 on the chromosome, synthesis of beta-galactosidase suffered much less repression than that of thiogalactoside transacetylase. 5. In a diploid strain in which the chromosome carries L1 and also a second mutation that increases the rate of expression of lac to that permitted by L8 or L29, the synthesis of beta-galactosidase again suffered much less repression than the synthesis of thiogalactoside transacetylase. 6. The effect of L1 (which deletes the boundary between the i gene and the lac promoter) is ascribed to its bringing the expression of lac under the control of the promoter of the i gene. 7. Even in strains carrying L1, some catabolite repression persists; this is not due to a trans effect from the episome since it occurs equally in a haploid strain with L1.     

Glucose-lactose diauxie in Escherichia coli

Growth of Escherichia coli in medium containing glucose, at a concentration insufficient to support full growth, and containing lactose, is diauxic. A mutation in the gene, CR, which determines catabolite repression specific to the lac operon, was found to relieve glucose-lactose but not glucose-maltose diauxie. Furthermore, a high concentration of lactose was shown to overcome diauxie in a CR(+) strain. Studies on the induction of beta-galactosidase by lactose suggested that glucose inhibits induction by 10(-2)m lactose. Preinduction of the lac operon was found to overcome this effect. The ability of glucose to prevent expression of the lac operon by reducing the internal concentration of inducer as well as by catabolite repression is discussed.     

Regulation of lac Operon Expression: Reappraisal of the Theory of Catabolite Repression

The physiological state of Escherichia coli with respect to (permanent) catabolite repression was assessed by measuring the steady-state level of beta-galactosidase in induced or in constitutive cells under a variety of growth conditions. Four results were obtained. (i) Catabolite repression had a major effect on fully induced or constitutive expression of the lac gene, and the magnitude of this effect was found to be dependent on the promoter structure; cells with a wild-type lac promoter showed an 18-fold variation in lac expression, and cells with the lacP37 (formerly lac-L37) promoter exhibited several hundred-fold variation. (ii) Exogenous adenosine cyclic 3',5'-monophosphoric acid (cAMP) could not abolish catabolite repression, even though several controls demonstrated that cAMP was entering the cells in significant amounts. (Rapid intracellular degradation of cAMP could not be ruled out.) (iii) Neither the growth rate nor the presence of biosynthetic products altered the degree of catabolite repression; all variation could be related to the catabolites present in the growth medium. (iv) Slowing by imposing an amino acid restriction decreased the differential rate of beta-galactosidase synthesis from the wild-type lac promoter when bacteria were cultured in either the absence or presence of cAMP; this decreased lac expression also occurred when the bacteria harbored the catabolite-insensitive lacP5 (formerly lacUV5) promoter mutation. These findings support the idea that (permanent) catabolite repression is set by the catabolites in the growth medium and may not be related to an imbalance between catabolism and anabolism.     

Polycistronic Effects of Catabolite Repression on the lac Operon

The catabolite repression caused by glucose and glucose-6-phosphate has been studied for both beta-galactosidase and thiogalactoside transacetylase, the products of the operator proximal and distal cistrons of the lac operon, respectively. We find that both cistrons are affected coordinately by this form of repression. We also find that a single alteration at the lac promoter region is sufficient to abolish sensitivity to repression of both cistrons. From this, we conclude that there is only one target site for catabolite repression in the lac operon.     

Control of mixed-substrate utilization in continuous cultures of Escherichia coli

The chemostat culture technique was used to study the control mechanisms which operate during utilization of mixtures of glucose and lactose and glucose and l-aspartic acid by populations of Escherichia coli B6. Constitutive mutants were rapidly selected during continuous culture on a mixture of glucose and lactose, and the beta-galactosidase level of the culture increased greatly. After mutant selection, the specific beta-galactosidase level of the culture was a decreasing function of growth rate. In cultures of both the inducible wild type and the constitutive mutant, glucose and lactose were simultaneously utilized at moderate growth rates, whereas only glucose was used in the inducible cultures at high growth rates. Catabolite repression was shown to be the primary mechanism of control of beta-galactosidase level and lactose utilization in continuous culture on mixed substrates. In batch culture, as in the chemostat, catabolite repression acting by itself on the lac enzymes was insufficient to prevent lactose utilization or cause diauxie. Interference with induction of the lac operon, as well as catabolite repression, was necessary to produce diauxic growth. Continuous cultures fed mixtures of glucose and l-aspartic acid utilized both substrates at moderate growth rates, even though the catabolic enzyme aspartase was linearly repressed with increasing growth rate. Although the repression of aspartase paralleled the catabolite repression of beta-galactosidase, l-aspartic acid could be utilized even at very low levels of the catabolic enzyme because of direct anabolic incorporation into protein.     

Regulation of the galactose-inducible lac operon and the histidine utilization operons in pts mutants of Klebsiella aerogenes.

Galactose appears to be the physiological inducer of the chromosomal lac operon in Klebsiella aerogenes. Both lactose and galactose are poor inducers in strains having a functional galactose catabolism (gal) operon, but both are excellent inducers in gal mutants. Thus the slow growth of K. aerogenes on lactose reflects the rapid degradation of the inducer. Several pts mutations were characterized and shown to affect both inducer exclusion and permanent catabolite repression. The beta-galactosidase of pts mutants cannot be induced at all by lactose, and pts mutants appear to have a permanent and constitutive inducer exclusion phenotype. In addition, pts mutants show a reduced rate of glucose metabolism, leading to slower growth on glucose and a reduced degree of glucose-mediated permanent catabolite repression. The crr-type pseudorevertants of pts mutations relieve the constitutive inducer exclusion for lac but do not restore the full level of glucose-mediated permanent catabolite repression and only slightly weaken the glucose-mediated inducer exclusion. Except for weakening the glucose-mediated permanent catabolite repression, pts and crr mutations have no effect on expression of the histidine utilization (hut) operons.     

Transient repression of catabolite-sensitive enzyme synthesis elicited by 2,4-dinitrophenol.

Transient inhibition of catabolic enzyme synthesis in Escherichia coli occurred when a low concentration of 2,4-dinitrophenol (DNP) was simultaneously added with inducer. Using mutant strains defective for gamma-gene product or constitutive for lac enzymes, it was found that the inhibition is not due to the exclusion of inducer by uncoupling. The addition of cyclic adenosine 3',5'-monophosphate overcame repression. The components of the lac operon coordinately responded to DNP inhibition. From deoxyribonucleic acid-ribonucleic acid hybridization experiments, it was found that the inhibition of beta-galactosidase induction occurred at the level of messenger ribonucleic acid synthesis specific for the lac operon. It seems probable that DNP represses induction in a similar manner to that of transient repression observed upon the addition of glucose. Furthermore, it was found that transient repression disappeared if cells were preincubated with DNP before induction. This indicates that new contact of cells with DNP is obligatory for transient repression. From these results, it is suggested that the cell membrane may be responsible for regulation of catabolite-sensitive enzyme synthesis.     

Elucidation of enzyme control mechanisms using macroscopic measurements in a mixed substrate fermentation system

The objective of this work was to relate macroscopically measurable on-line fermentation parameters such as dissolved oxygen, off-gas oxygen and carbon dioxide, and cell mass, to the controlled production of key intracellular enzymes under carbon limited conditions. Both batch and perturbed batch aerobic fermentations were performed using two different strains of Escherichia coli, with glucose and lactose as the sole carbon sources. The two strains differed from each other only in the lac operon region of their genome. The parent strain, E. coli 3000, was inducible for the enzyme beta-galactosidase. The other strain, E. coli 3300, was a constitutive mutant in the production of beta-galactosidase. In all experiments, off-line assays of sugars and beta-galactosidase activity were performed. It was observed that there is a clear relationship between the macroscopic on-line measurements, dissolved oxygen tension, carbon dioxide evolution rate and oxygen uptake rate, and the microscopic control phenomena of catabolite repression, catabolite inhibition, and inducer repression.     

Transport regulation of recombinant gene expression in E. coli and B. subtilis

Expression kinetics of the lactose (lac) operon in Escherichia coli are reviewed for both wild-type and recombinant cell cultures under chemostatic conditions. A unified model which involves regulation of active inducer (lactose) transport, promoter-operator regulated expression of the lac operon, glucose-mediated inducer exclusion, and catabolite repression is summarized and supporting data is shown to verify its accuracy. The synthesis of alpha-amylase with a recombinant form of Bacillus subtilis is also reviewed to point out generic features in transport regulation, the lac operon model providing a point of departure. While there are many similarities in the influence of transport on both regulating models, there are also important differences. In a chemostat system, the synthesis of alpha-amylase is nongrowth associated, while beta-galactosidase is a growth-associated enzyme. Nevertheless, transport regulation is an important feature in both instances.     

Paradoxical effect of weak inducers on the lac operon of Escherichia coli

Previously, we reported the existence of a group of compounds whose function in the regulation of the lac operon was "paradoxical" in that they acted as either inducers or repressors depending on the circumstances. We now show that this group of compounds does not repress the lac operon by catabolite repression, transient repression, or by preventing the uptake of inducers. A model is presented which shows that "paradoxical" behavior is to be expected if a weak inducer is present at a concentration that is high relative to its binding affinity for the regulatory macromolecule. This model depends on the assumptions that the regulatory macromolecule is an allosteric protein which undergoes a transition between two conformational states and that the rate of enzyme synthesis depends on the fraction of protein molecules in each state. The previous observations on the responses of lac regulatory mutants to weak inducers have been extended to a series of such mutants. Weak inducers repress beta-galactosidase synthesis in several i(-) mutants. When this happens, enzyme synthesis can be reinduced by using a strong inducer such as isopropyl-beta-d-thiogalactoside. These compounds induce operator constitutives and the i(t) mutant more easily than they induce a wild-type strain.     

Relaxation of catabolite repression in streptomycin-dependent Escherichia coli

Acetohydroxy acid synthetase, which is sensitive to catabolite repression in wild-type Escherichia coli B, was relatively resistant to this control in a streptomycin-dependent mutant. The streptomycin-dependent mutant was found to be inducible for beta-galactosidase in the presence of glucose, although repression of beta-galactosidase by glucose occurred under experimental conditions where growth of the streptomycin-dependent mutant was limited. Additional glucose-sensitive enzymes of wild-type E. coli B (citrate synthase, fumarase, aconitase and isocitrate dehydrogenase) were found to be insensitive to the carbon source in streptomycin-dependent mutants: these enzymes were formed by streptomycin-dependent E. coli B in equivalent quantities when either glucose or glycerol was the carbon source. Two enzymes, glucokinase and glucose 6-phosphate dehydrogenase, that are glucose-insensitive in wild-type E. coli B were formed in equivalent quantity on glucose or glycerol in both streptomycin-sensitive and streptomycin-dependent E. coli B. The results indicate a general decrease or relaxation of catabolite repression in the streptomycin-dependent mutant. The yield of streptomycin-dependent cells from glucose was one-third less than that of the streptomycin-sensitive strain. We conclude that the decreased efficiency of glucose utilization in streptomycin-dependent E. coli B is responsible for the relaxation of catabolite repression in this mutant.     

Transient Repression of the lac Operon

Severe transient repression of constitutive or induced beta-galactosidase synthesis occurs upon the addition of glucose to cells of Escherichia coli growing on glycerol, succinic acid, or lactic acid. Only mutants particularily well adapted to growth on glucose exhibit this phenomenon when transferred to a glucose-containing medium. No change in ribonucleic acid (RNA) metabolism was observed during transient repression. We could show that transient repression is pleiotropic, affecting all products of the lac operon. It occurs in a mutant insensitive to catabolite repression. It is established much more rapidly than catabolite repression, and is elicited by glucose analogues that are phosphorylated but not further catabolized by the cell. Thus, transient repression is not a consequence of the exclusion of inducer from the cell, does not require catabolism of the added compound, and does not involve a gross change in RNA metabolism. We conclude that transient repression is distinct from catabolite repression.     

Physiological studies of beta-galactosidase induction in Kluyveromyces lactis

We examined the kinetics of beta-galactosidase (EC 3.2.1.23) induction in the yeast Kluyveromyces lactis. Enzyme activity began to increase 10 to 15 min, about 1/10 of a cell generation, after the addition of inducer and continued to increase linearly for from 7 to 9 cell generations before reaching a maximum, some 125- to 150-fold above the basal level of uninduced cells. Thereafter, as long as logarithmic growth was maintained, enzyme levels remained high, but enzyme levels dropped to a value only 5- to 10-fold above the basal level if cells entered stationary phase. Enzyme induction required the constant presence of inducer, since removal of inducer caused a reduction in enzyme level. Three nongratuitous inducers of beta-galactosidase activity, lactose, galactose, and lactobionic acid, were identified. Several inducers of the lac operon of Escherichia coli, including methyl-, isopropyl- and phenyl-1-thio-beta-d-galactoside, and thioallolactose did not induce beta-galactosidase in K. lactis even though they entered the cell. The maximum rate of enzyme induction was only achieved with lactose concentrations of greater than 1 to 2 mM. The initial differential rate of beta-galactosidase appearance after induction was reduced in medium containing glucose, indicating transient carbon catabolite repression. However, glucose did not exclude lactose from K. lactis, it did not cause permanent carbon catabolite repression of beta-galactosidase synthesis, and it did not prevent lactose utilization. These three results are in direct contrast to those observed for lactose utilization in E. coli. Furthermore, these results, along with our observation that K. lactis grew slightly faster on lactose than on glucose, indicate that this organism has evolved an efficient system for utilizing lactose.     

Catabolite repression of tryptophanase in Escherichia coli

Catabolite repression of tryptophanase was studied in detail under various conditions in several strains of Escherichia coli and was compared with catabolite repression of beta-glactosidase. Induction of tryptophanase and beta-galactosidase in cultures grown with various carbon sources including succinate, glycerol, pyruvate, glucose, gluconate, and arabinose is affected differently by the various carbon sources. The extent of induction does not seem to be related to the growth rate of the culture permitted by the carbon source during the course of the experiment. In cultures grown with glycerol as carbon source, preinduced for beta-galactosidase or tryptophanase and made permeable by ethylenediaminetetraacetic acid (EDTA) treatment, catabolite repression of tryptophanase was not affected markedly by the addition of cAMP (3',5'-cyclic adenosine monophosphate). Catabolite repression by glucose was only partially relieved by the addition of cAMP. In contrast, under the same conditions, cAMP completely relieved catabolite repression of beta-galactosidase by either pyruvate or glucose. Under conditions of limited oxygen, induction of tryptophanase is sensitive to catabolite repression; under the same conditions, beta-galactosidase induction is not sensitive to catabolite repression. Induction of tryptophanase in cells grown with succinate as carbon source is sensitive to catabolite repression by glycerol and pyruvate as well as by glucose. Studies with a glycerol kinaseless mutant indicate that glycerol must be metabolized before it can cause catabolite repression. The EDTA treatment used to make the cells permeable to cAMP was found to affect subsequent growth and induction of either beta-galactosidase or tryptophanase much more adversely in E. coli strain BB than in E. coli strain K-12. Inducation of tryptophanase was reduced by the EDTA treatment significantly more than induction of beta-galactosidase in both strains. Addition of 2.5 x 10(-3)m cAMP appeared partially to reverse the inhibitory effect of the EDTA treatment on enzyme induction but did not restore normal growth.