Reversible inactivation of isocitrate dehydrogenase in Escherichia coli

600th MEETING, OXFORD

319

The Regulation of Microbial Enzymes by Covalent Modification Regulation in Metabolism Group Colloquium organized and edited by H. G. Nimmo (GIasgow)

Reversible inactivation of isocitrate dehydrogenase in Escherichia coli W. H. HOLMS and H. G. NIMMO Department of Biochemistry, University of Glasgow, Glasgow GI2 8QQ, Scotland, U.K. Escherichia coli can grow aerobically in defined media containing inorganic salts and any one of a wide range of carbon compounds. Whatever the nature of the carbon source. it is fed into the amphibolic pathways [glycolysis, pentose phosphate pathway, Krebs (tricarboxylic acid) cyclel, from which stem all the biosynthetic pathways required for making the components of new cell material. The operation of the amphibolic pathways also traps energy in the forms (ATP, NADPH, H+) required for biosynthesis and growth. These processes are common to growth on any carbon source, but biosynthesis also relies on anaplerotic sequences of metabolism which replenish the pools of some compounds (e.g. in the Krebs cycle) which are the starting points for biosynthesis (Kornberg. 1966). Different anaplerotic sequences are used according to the point at which any particular carbon source enters the amphibolic pathways. For example, growth on glucose (Fig. 1) depends on the

FDP

-

G6P

‘US’

‘PTS’-P

‘1 ‘I PEP

I

I

isoClT

OAA

I

I I

I

II

I SUC

MAL

A

’-\

b\

-. I

/

‘--AcCoA +----

AcP

*-----ACOH

Fig. I . Routes of metabolism used b.v Escherichia coli,for growth

on glucose or acetate Key: AcCoA, Acetyl-CoA; AcOH, acetate; AcP, acetyl phosphate; FDP, fructose 1,6-bisphosphate: G6P, glucose 6-phosphate: GLYOX, glyoxylate; isoCIT, isocitrate; MAL, malate: OAA, oxaloacetate; PEP, phosphoenolpyruvate; ‘PTS’, all the components of the glucose phosphotransferase system (‘PTS’P, phosphorylated); Py, pyruvate: SUC, succinate: +, routes used for growth on glucose: ---+, routes used for growth on acetate. VOl.

10

Consequences of acetate excretion during glucose catabolism For every mol of glucose used for aerobic growth. approx. 0.5 mol of acetate is excreted (Holms & Bennett, 1971). Glucose is taken into E. coli by a phosphotransferase system (Roseman, 1969) in which the translocation of glucose across the membrane is accompanied by its phosphorylation to glucose 6-phosphate. The phosphotransferase system is a complex chain of events in which the ultimate donor of the phosphate is phosphoenolpyruvate. For every mol of glucose brought in, lmol of phosphoenolpyruvate is consumed and lmol of pyruvate is liberated. Presumably the amount of pyruvate generated by the glucose phosphotransferase system is more than the needs of the organism, and half of the pyruvate is decarboxylated and excreted as acetate. Acetate accumulates until all the glucose is utilized, when the enzymes of the glyoxylate bypass are induced and acetate is used. When this occurs, the activity of ICDH falls by about 75% and. when all the acetate is consumed, the activity largely recovers (Holms & Bennett, 1971). The loss and recovery of ICDH activity does not occur after growth on other substrates (e.g. glycerol) which do not give excretion of acetate unless acetate is added to the culture at the end of the growth. Excretion of acetate during growth on glucose followed by loss and recovery of ICDH occurs in five strains of E. coli and one strain each of Salmonella t.vphimurium. Aerobacter aerogenes and Serratia marcescens. but not in E. coli K 12 or K 13. It would appear that modulation of ICDH in response to induction and operation of the glyoxylate bypass is common, but not universal, among the enteric bacteria (Bennett & Holms, 1975).

Mechanism orthe reversible inactivation oflCDH

GLYOX

-----------

carboxylation of phosphoenolpyruvate by the enzyme phosphoenolpyruvate carboxylase (EC 4.1.1.3 1) for the anaplerotic supply to the Krebs cycle (Kornberg, 1966). In contrast, growth on acetate requires the two enzymes of the glyoxylate bypass (Kornberg, 1966) to generate Krebs-cycle intermediates, and the decarboxylation of oxaloacetate to phosphoenolpyruvate to supply the intermediates that can be derived from glucose by glycolysis (Fig. I). During growth on acetate, isocitrate is a substrate for both isocitrate lyase (EC 4.1.3.1) in the anaplerotic pathway and for ICDH* (EC 1.1.1.42) in the Krebs cycle. Our interest in the possible competition between isocitrate lyase and ICDH was aroused when we rediscovered (Holms & Bennett, 1971) that E. coli growing aerobically on glucose excreted acetate (Roberts e f al., 1957).

The earlier work (Holms & Bennett. 197 1 : Bennett & Holms. 1975) showed that ICDH activity is lowered when E. coli. de-repressed for the enzymes of the glyoxylate bypass. metabolizes acetate. The activities of ICDH in crude extracts obtained before exposure to acetate and then throughout the cycle of deactivation and re-activation are apparently due to the same enzyme protein, as judged by substrate affinities and response to pH and heat, and this has now been confirmed by preliminary peptide mapping of the purified protein (W. H. Holms & H. G. Nimmo. unpublished work). The activities of extracts are not greatly enhanced or diminished by dialysis. which presumably eliminates small molecules as effectors of reversible inhibition of ICDH. The apparent concerted inhibition of ICDH by glyoxylate and oxalacetate is now known to be an artifact (H. G.

* Abbreviation: ICDH. isocitrate dehydrogenase.

320 Nimmo, unpublished work), and is in any event largely reversed by dialysis and is equally effective on extracts with high and low activities of ICDH. Enzyme degradation and resynthesis cannot be involved, as the full cycle of reversible inactivation can be shown in the presence of inhibitors such as chloramphenicol, provided that the glyoxylate bypass has previously been induced. By exclusion of all other processes, it was thought that only covalent modification of the enzyme protein or its association with another macromolecule could possibly be the mechanism of the cyclical changes in ICDH activity. Over the last three years evidence has accumulated to suggest that the mechanism is a reversible phosphorylation. Wang & Koshland (1978, 198 1) have demonstrated phosphorylation of proteins in Salmonella typhimurium. Phosphorylation of ICDH concomitant with inactivation of the enzyme has been shown both in K strains of E. coli (Garnak & Reeves, 1979) and ML308 (W. H. Holms & H. G. Nimmo, unpublished work). The physiological significance of the inactivation of ICDH is better established at the moment for E. coli ML308 than for K strains, in which the amplitude of the ICDH-activity changes over the full inactivationhe-activation cycle is small (Bennett & Holms, 1975). Effect of available carbon source on reversible inactivation of ICDH Inactivation of ICDH when E. coli ML308 adapts to use acetate as sole source of carbon and energy is relatively slow, but it very much more rapid in bacteria which already contain the enzymes of the bypass. Re-activation of ICDH occurs under three kinds of circumstances: (1) when acetate is exhausted or removed; (2) when acetate metabolism is prevented (e.g. by anaerobiosis); (3) when acetate metabolism by the glyxoylate bypass is made redundant by addition of an alternative carbon source. Of the compounds tested, pyruvate is the most efficient cause of re-activation, but dicarboxylic acids of the Krebs cycle and glucose will also lead to re-activation of ICDH. All of these compounds also eventually repress synthesis of the enzymes of the glyoxylate bypass, but over a much longer time scale (Bennett & Holms, 1975: W. H. Holms & P. M. Bennett, unpublished work). Function of reversible inactivation of ICDH When E. coli grows on acetate the glyoxylate bypass serves an anaplerotic role, and its first enzyme, isocitrate lyase. must compete with ICDH for the available supply of isocitrate (Fig. 2). However, ICDH has a greater affinity for isocitrate than does the lyase (see, e.g., Bautista et al., 1979). These considerations support the suggestion that the flux of acetate carbon through isocitrate lyase into the glyoxylate bypass is favoured by inactivation of its competitor (ICDH). The activity of the Krebs cycle is thus diminished but kept available for rapid re-activation whenever the quality of the carbon source improves (Holms & Bennett, 1971; Bennett & Holms, 1975). Frequently a switch to a new carbon source requires synthesis of enzymes specific for the new carbon source. The problem then arises that the enzymes specific for the new substrate must operate before the new substrate can be used to make the amino acids of which the new enzymes are composed. A switch to acetate must overcome this problem, but is even more difficult, because the first of the acetate-specific enzymes (isocitrate lyase) must compete for isocitrate with ICDH (Fig. 2), which has a

BIOCHEMICAL SOCIETY TRANSACTIONS Acetate

OAA

Krebs cycle

Glyoxylate

co, Amphibolic pathways

Fig. 2. Interdependence of isocitrate lyase synthesis and activity Key: ICDH, isocitrate dehydrogenase; ICL, isocitrate lyase; OAA. oxaloacetate. much higher affinity for the same substrate (Bautista et al., 1979). It seems to make teleological sense that E. coli ML308 has selected inactivation of ICDH by phosphorylation to enable adaptation to acetate to occur and to permit rapid re-activation in response to the supply of better carbon sources. However, this view may be too simple. K strains of E. coli apparently succeed without resource to ICDH inactivation (Bennett & Holms, 1975), even though they contain more ICDH when grown on acetate than on glucose (Bautista et al., 1979). Even in E. coli ML308, where the activity of ICDH is lower on acetate than on glucose, the absolute amount of ICDH protein is probably greater (A. Borthwick, W. H. Holms & H. G. Nimmo, unpublished work). Re-activation, also, is not totally straightforward. Re-activation of ICDH is rapid and independent of protein synthesis on addition of pyruvate. Re-activation in response to glucose (and other compounds) depends on protein synthesis (Bennett & Holms, 1975), probably because induction of glucose phosphotransferase (or other transport processes) is limiting (Clark & Holms, 1976). In conclusion, it would seem that reversible phosphorylation of ICDH is a means employed by many enteric bacteria to facilitate a shift to and from utilization of acetate as sole source of carbon. The full understanding of this shift and the possible involvement of other mechanisms await purification and characterization of ICDH kinase and phosphatase. Bautista, J., Satrustegui, J. & Machada, A. (1979) FEBS Left. 105. 333-336

Bennett, P. M. & Holms. W. H. (1975)J. Gen. Microbiol. 87,37-51 Clark, B. & Holms, W. H. (1976)J. Cen. Microbiol. 95, 191-210 Garnak, M. & Reeves, H. C. (1979) Science 203, 11 11-1 112 Holms, W. H. & Bennett, P. M. (1971)J. Gen. Microbiol. 6 5 , 5 7 4 8 Kornberg, H. L. (1966) Essays Biochem. 2, 1-31 Roberts, R. B., Abelson, P. H., Cowie, D. B., Bolton, E. T. & Bntten, R. J. (1957) Studies of Biosvnihesis in Escherichia coli. pp. 184-206, Carnegie Institution of Washington Publication 607, Washington D.C. Roseman. S. (1969) J. Gen. Physiol. 54, 138s-180s Wang, J. Y. J. & Koshland, D. E. (1978) J. Biol. Chem. 253, 7605-7608

Wang, J. Y. J. & Koshland, D. E. (1981) J. Biol. Chem. 256, 4640-4648

1982