TIBS - January 1977
tamine synthetase. As a result, the cells will cease to produce the hut enzymes and glutamine synthetase, and begin to produce glutamate dehydrogenase. After several generations of growth in this medium, they will have acquired the enzymatic constitution appropriate for growth in excess ammonia. Acknowledgements The authors are recipients of Public Health Service research grants GM-07446 from the National Institute of General Medical Sciences and AM- 13894 from the National Institute of Arthritis and Metabolic Diseases, and grant GB-5322 from the National Science Foundation.
5 Ginsburg, A. and Stadtman, E. R. (1973) in The Enzymes of Glutamine Metabolism (Prusiner, S.
and Stadtman, E. R., eds), pp. 9-43, Academic Press, New York 6 Prival, M. J., Brenchley, J. E. and Magasanik, B. (1973) J. Biol. Chem. 248,4334-4344 7 Streicher, S. L., B:nder, R. A. and Magasanik, B. (1975) J. Bacferiol. 121, 32&331 8 Foor, F., Janssen, K. A. and Magasanik, B. (1975) Proc. Nat. Acad. Sci. U.S.A. 72, 4844.4848 9 DeLeo, A.B. and Magasanik, B. (1975) J. Bacteriol. 121,313-319
10 Streicher, S. L., DeLeo, A.B. and Magasanik, B. (1976) J. Bacterial. (in press) 11 Smith, G.R. and Magasanik, B. (1971) J. BioL Chem. 246,3330-3341
12 Hagen, D. C. and Magasanik, B. (1973) Proc. Nat. Acad. Sci. U.S.A. 70, SOS-812
13 Prival, M.J. and Magasanik, B. (1971) J. Biol. Chem. 246,62886296
14 Tyler, B., DeLeo, A. B. and Magasanik, B. (1974) Proc. Nat. Acad. Sci. U.S.A.
15 Friedrich, B. and Magasanik, B. (1976) Abstr.
1 Meers, J.L., Tempest, D. W. and Brown, C.M. (1970) J. Gen. Microbial. 64, 187-194 2 Tempest, D. W., Meers, J.L. and Brown, C.M. (1970) Biochem. J. 117,405-407 3 Wulff, K., Mecke, D. and Holzer, M. (1967) Biothem. Biophys. Res. Commun. 28, 740-745 4 Kingdon, M.S., Shapiro, B.M. and Stadtman, E.R. (1967) Proc. Nat. Acad. Sci. U.S.A. 58, 1703-1710
Annu. Meet. A.S.M.,
16 Streicher, S. L., Shanmugam, K.T., Ausubel, F., Morandi, C. and Goldberg, R.B. (1974) J. Bacferiol. 120, 815-821 17 Resnick, A. D. and Magasanik, B. (1976) J. Biol. Chem. 251,2722-2728
18 Brenchley, J. E., Prival, M. J. and Magasanik, B. (1973) J. Biol. Chem. 248,6122-6128
depletion of lactose, they rise again. This wave-like fluctuation of the levels during repeated cycles of growth shows the tendency for cyclic AMP to be low during periods of growth and high during periods of stasis or growth transition. The physiological significance of these alterations appears to be that the adaptation to a new growth situation always requires higher cyclic AMP levels than does the continued growth on that carbon source. It is likely that raised levels of cyclic AMP overcome the barrier to effective transcription of operons for catabolic enzymes when only low levels of inducer penetrate the cells. After induction of the transport system for a catabolite, effective transcription can persist with lowered concentrations of cyclic AMP. The following discussion will show that the presence of a sugar-transport system provides the key to catabolite-dependent lowering of cyclic AMP levels. Interaction of a sugar-transport system with adenylate cyclase creates a complex that can regulate the enzyme activity. Sugars inhibit adenylate cyclase activity
Regulation of Escherichia coli adenylate cyclase by phosphorylationdephosphorylation Alan Peterkofsky The mechanism of catabolite repression involves a sugar-dependent regulation of adenylate cyclase via phosphorylation-dephosphorylation. This is accomplished by an interaction of adenylate cyclase with the sugar transport system.
In Escherichia coli, the regulation of the rate of synthesis of induced enzymes required for the degradation of most carbon sources calls for not only the presence of inducer but also an optimal concentration of cyclic AMP [1,2]. A major advance in our understanding of the control of induced enzyme synthesis was the notion that catabolites exerted a negative control over cellular cyclic AMP levels [I,31 (the phenomenon of catabolite repression). Recent studies have mapped out the essential characteristics of this process. It has been shown that sugars inhibit adenylate cyclase if the appropriate transport system for that sugar is induced . A.P. is at the Laboratory of Biochemical Genetics, National Heart and Lung Institute, Bethesda, Maryland 20014, U.S.A.
The membrane-bound complex between adenylate cylcase and the sugar transport system permits the sugar-dependent regulation of adenylate cyclase by way of a phosphorylation-dephosphorylation mechanism . When E. coli are presented with a growth medium containing a mixture of glucose and lactose, they grow sequentially on the glucose followed by the lactose (Fig. 1) (diauxic growth). An examination of the cellular cyclic AMP levels in such a situation showed that they rise abruptly during the transition period when cells are adapting from growth on glucose to growth on lactose. Just prior to the initiation of growth on lactose, the levels drop again and remain low throughout the period of lactose growth. When growth stops due to
While measurements of adenylate cyclase in broken cell preparations show no inhibition by glucose, intact or permeabilized cells are sensitive to glucose inhibition (Fig. 2) [6,7]. These observations have led to the idea that sugars do not inhibit adenylate cyclase directly but require some other factors, which have been suggested to be the sugar-transport systems [4,5]. When E. coli were grown on glucose, the inhibition of adenylate cyclase by glucose was relatively specific; most of the other stereoisomeric hexoses were inactive. Glucose 6-phosphate, which is the first product of glucose metabolism in E. coli, was also inactive, suggesting that the observed inhibition was due to free glucose itself. The absence of a requirement for glucose metabolism was also supported by the inhibitory effects of the glucose analogues 2-deoxyglucose and methyl-a-glucoside, neither of which support growth or are metabolized, although they are taken up by cells and phosphorylated. The first evidence that transport systems were involved in the regulation of adenylate cyclase activity came from the observation that the profile of sugars that inhibited adenylate cyclase varied with the growth conditions of the cells (Table I) . Irrespective of the carbon source used for growth, adenylate cyclase was always inhibited by glucose whose transport system is partially constitutive. However, when cells were grown on a variety of other carbon sources, like fructose or mannitol, the adenylate cyclase from such cells was
TIBS - January 1977
29 GLUCOSE ’
TABLE II Effect of glucose and phosphoenolpyruvate on adenylate cyclase activity in E. coli strains with pts mutations
Adenylate cyclase activity
a” (A) Wild-type 1260 Enzyme II glucose- 1362
(B) Wild-type HPrEnzyme I -
814 2055 52
71 138 26
1108 2305 2500
Cultures of the indicated strains were grown on nutrient broth medium. Adenylate cyclase activity (expressed as pmol cyclic AMP/mg per h) was measured  with the indicated additions (1 mM). Adapted from refs 5 and 9.
0.01 HOURS Fig. 1. The profile of extracellular and intracellular cyclic AMP accompanying diauxic growth on glucose and lactose. E. coli B were grown either on glucose (A) or glucose plus lactose (B). At the indicated times, the followh~g measurements were made: absorbance (0). &galactosiakse (m) [IO], extracellular cyclic AMP (A) _ . *[II], and intracellular cyclic AMP (0) [a. GLUCOSE
also inhibited by the grwoth compound. It was concluded that when metabolic pathways for utilization of sugars were induced, cells were then able to show inhibition of their adenylate cyclase by those sugars. Sugar transport systems for inhibition of adenylate cyclase Most sugars that can inhibit adenylate cyclase activity are transported into E. coli by way of the phosphoenolpyruvatedependent phosphotransferase system (pts) . Evidence has been presented that it is the phosphotransferase system which regulates adenylate cyclase . The pts system consists of the following reactions: Phosphoenolpyruvate
(PEP) + Enzyme 1-t + Enzyme I-P + pyruvate
Enzyme I-P + HPr+HPr-P + Enzyme I HPr-P+ Enzyme II-tEnzyrne II-P+HPr Enzyme II-P + sugar-rsugar-P + Enzyme II
(1) (2) (3) (4)
The two cytoplasmic components, Enzyme I and HPr are utilized for the transport of all pts sugars. The Enzyme II components are membrane bound, sugar specific and induced by growth on the specific sugars.
Test of the properties of adenylate cyclase in E. coli strains carrying mutations in the pts proteins have supplied conclusive evidence for the interaction of the pts system with adenylate cyclase (Table II). A strain carrying mutations in the glucose-specific Enzyme II expresses a normal level of adenylate cyclase (Table
Sugar in assay
Adenylate cyclase activity (%)
none glucose fructose m&nit01
100 5 102 120
none glucose fructose mannitol
100 11 13 165
none glucose fructose mannitol
100 27 104 33
E. coli B was grown on medium supplemented with the indicated sugars. Cells were assayed for adenylate cyclase  with the indicated sugars (3 mM) added to the assay. Data adapted from ref 4.
Fig. 2. The effect of glucose on adenylate cyclase in intact cells (a) French Press extracts (b) and toluenetreated cells (c). Adenylate cyclase was assayed [6,7j in the designatedpreparations with 1 mM glucose when indicated.
TABLE I Effect of carbon source on the profile of sugars that inhibit adenylate cyclase Sugar in growth medium
Fig. 3. Interaction of aaknylate cyclase with Enzyme I. The model suggests an opposing action of phosphoenolpyruvate (PEP) and glucose on the activity of aaknylate cyclase. The phosphoenolpyruvatedependent phosphorylation of Enzyme I (I), complexed to aaknylate cyclase (AC.) results in enzyme activation. Glucose transport through the phosphotransferase system results in akphosphorylation of Enzyme I, convertingaaknylate cyclase to the low activity state.
TIBS - January 1977
ious pts components in the cell is shown in Fig. 4. It stresses the direct interaction of Enzyme I with adenylate cyclase, a complex which directly affects the enzyme activity (see Fig. 3). The nature of the multienzyme complex indicates that the inhibition of adenylate cyclase by glucose demands the function of all the pts components. This working model explains the interaction of the pts system with adenylate cyclase and demonstrates that transport of sugars into the cell is coupled to inhibition of adenylate cyclase. References 1 Pastan, I. and Perlman, R. (1970) Science 169, 339-344 2 Lis, J.T. and Schleif, R. (1973) J. Mol. Biol. 79, 149162 3 Makman, R.S. and Sutherland, E. W. (1965) J. Biol. Chem. 240, 1309-1314 4 Peterkofsky, A., Harwood, J.P. and Gazdar, C. (1975) J. Cvclic Nucleotide Res. 1. 1l-20 Peterkofsky, A. and Gazdar, C. (1975) Proc. Nat. Acad. Sci. U.S.A. 72,292&2924
Peterkofsky, A. and Gazdar, C. (1974) Proc. Nat.
Fig. 4. The interaction of the pts svstem with adenylate cyclase. The model suggests that membrane-bound adenylate cyclase functions as a multi-enzyme complex regulated by the pts system. It is hypothesized that the interaction of adenylate cyclase with the Enzyme I protein is direct. The inhibition of adenylate cyclase activity coupled to sugar transport is represented to be indirect and dependent on the presence of all the components of the transport system.
II, A) ; however, the adenylate cyclase can no longer be inhibited by glucose. It was therefore concluded that, while Enzyme II is not required for the activity of adenylate cyclase, it is necessary for the regulatory effect of glucose. Studies on E. coli strains carrying leaky mutations in the general cytoplasmic proteins, Enzyme I or HPr (Table II, B) produced results of a different nature . While a mutation in the HPr protein did not diminish the adenylate cyclase activity, a leaky mutation in the Enzyme I protein led to essentially complete loss in enzyme activity. The depressed levels of adenylate cyclase in the Enzyme I mutant were not the result of the absence of adenylate cyclase protein, but rather the ramification of a regulation defect; the addition of phosphoenolpyruvate fully restored the adenylate cyclase activity in the leaky Enzyme I mutant. It was concluded from these studies that Enzyme I plays an essential role in regulating adenylate cyclase activity .
low. The high activity state is favored by the presence of phosphoenolpyruvate and the absence of glucose, while the low activity state is favored by the presence of glucose and the absence of phosphoenolpyruvate. A comprehensive model indicating the location of adenylate cyclase and the var-
A phosphorylation-dephosphorylation model
It has been clearly established by experiments with the common bacterium Escherichia coli that it is possible for a cell to control its output of a given product in two ways: it may change the number of enzyme molecules available for some biochemical step in any sequential process
A current hypothesis to explain the interaction of ’the pts system with adenylate cyclase involves the idea that adenylate cyclase is normally complexed with Enzyme I (Fig. 3) . When Enzyme I exists in a phosphorylated form, adenylate cyclase can express a high level of activity; when Enzyme I is dephosphorylated, adenylate cyclase activity is
Acad. Sci. U.S.A. 71,2324-2328
Harwood, J. P. and Peterkofsky, A. (1975) J. Biol. Chem. 250,4656-4662
Roseman, S. (1972) in The Molecular Basis of Biological Transport (Woessner, Jr, J.F. and Huijing, F., eds), pp. 181-215, Academic Press, New York Harwood, J.P., Gazdar, C., Prasad, C., Peterkofsky, A., Curtis, S.J. and Epstein, W. (1976) J. Biol. Chem. 251,2462-2468 Experiments in Molecular Genetics (1972) (Miller, J.H., ed.), pp. 352-355, Cold Spring Harbor Laboratory, Cold Spring Harbor Peterkofsky, A. and Gazdar, C. (1971) Proc. Nat. Acad. Sci. U.S.A. 68,2794-2798
lomc control of biochemical reactions Pierre Douzou and Patrick Maurel Ionic strength jluctuations play an essential role in the efficiency of enzyme-catalyzed reactions occurring inpolyelectrolyte microenvironmentsprovidedby the ‘biological organization’ and by the polyelectrolyte nature of some proteins and nucleic acids. This ionic control might play a major physiological role in regulating systems involved in genetic translation as well as in the cellular metabolism ofplant tissues and higher animals.
P.D. and P.M. are at the French Institut National de la Sante et de la Recherche Medicale. P.D. is also working at the Foundation Edmond de Rothschild, in Paris.
[ 1,2], or it may change their rate of reaction by inhibition or by activation, the inhibitors and activators usually being small molecules, to which the enzymes automatically respond by structural modifications [3,4]. These might not represent the sole mechanisms for regulating enzyme reactions in eukaryotic and prokaryotic cells, in both of which physiologically important control might also be exerted by ionic exchanges at the macromolecular-compo-