TIBS 16 - AUGUST 1991

biosynthesis of proinsulin, as most text- books assert that it does, what is it doing? It is now known that during the renaturation process, the yield of any refolded protein decreases with increasing temperature. For insulin, this decrease is accompanied by increases in products containing only either A or B chain. The role of the C peptide in this process could thus be to bring and keep the two chains together so as to favor the formation of the native, es- pecially the inter-chain disulfides.

Among the known proinsulin sequences from a number of species, the C peptide segment is the least con- servative and has a mutation rate approximately eight times that of the A and B chains 2°. This agrees with the suggestion that it is not providing essential structural information. How- ever, the C peptide may play important roles during the biosynthesis, process- ing and translocation across mem-

branes before the eventual secretion of the mature hormone z2,23.

References 1 Lehninger, A. L. (1982) Principles of

Biochemistry (3rd edn), pp. 202-203, Worth Publishers,

2 Sffyer, L. (1988) Biochemistry (3rd edn), p. 41, Freeman

3 Alberts, B. et al. (1989) Molecular Biolo~ of the Cell (2nd edn), p. 122, Garland Publishing

4 Damell, J., Lodish, H. and Baltimore, D. (1990) Molecular and Cell Biolo~/ (2nd edn), pp. 64-65, Scientific American Books

5 Matthews, C. K. and van Holde, K. E. (1990) Biochemisby, p. 194, Benjamin Commings

6 Du, Y. C., Zhang, Y. S., Lu, Z. X. and Tsou, C. L. (1961) ScL Sin. (Engl. edn) 10, 84-104

7 Du, Y. C., Jiang, R. Q. and Tsou, C. L. (1965) ScL Sin. (Engl. edn) 14, 229-236

8 Katsoyannis, P. G. et aL (1967) Biochemistry 6, 2642-2655

9 Chance, R. E. et aL (1981) in Peptides, Synthesis, Structure and Function (Rich, D. H. and Gross, E., eds), pp. 712-728, Proc. 7th American Peptide Symposium, Peirce Chemical Co.

10 Kauzmann, W. (1959) in Sulfur in Proteins

(Benesch R., ed.), pp. 90-108, Academic Press

11 Sela, M. and Lifson, M. (1959) Biochim. Biophys. Acta 36, 471-478

12 Qian, Y. Q. and Tsou, C. L. (1987) Biochem. Biophys. Res. Comm. 146, 437-442

13 Tang, J. G. and Tsou, C. L. (1990) Biochem. J. 268, 429-435

14 Anfinsen, C. B. and Haber, E. (1961) J. BioL Chem. 236, 1361-1363

I 5 Ahmed, A. K., Schaffer, S. W. and Wetiaufer, D. B. (1975) J. Biol. Chem. 250, 8477-8482

16 Givol, D., De Lorenzo, F., Goldberger, R. F. and Anflnsen, C. B. (1965) Proc. Natl Acad. ScL USA, 678-684

17 Varandani, P. T. and Nafz, M. A. (1970) Arch. Biochem. Biophys. 141, 533-537

18 Tang, J. G., Wang, C. C. and Tsou, C. L. (1988) Biochem. J. 255, 451-455

19 Brandenburg, D. et al. (1977) in Protein Crosslinking Part A (Friedman, M., ed.), pp. 261-282, Plenum

20 Wilson, A. C., Carlson, S. C. and White, T. J. (1977) Annu. Rev. Biochem. 46, 573-639

21 Wang, Z. X., Ju, M. and Tsou, C. L. (1987) J. Theor. BioL 124, 293-301

22 Steiner, D. F. and Clark, J. L. (1968) Proc. Natl Acad. Sci. USA 60, 622-629

23 Tager, H. S., Steiner, D. F. and Patxelt, C. (1981) Methods Cell BioL 23, 73-88

TEXTBOOK ERRORS

'HEXOKINASE AND GLUCOKINASE are two enzymes that catalyse the phos- phorylation of WD-glucose at the 6 pos- ition. Hexokinase will accept several other hexoses as substrates, but gluco- kinase, which predominates in liver, is highly specific for glucose. It also dif- fers from hexokinase in having a high Michaelis constant for glucose, about I0 mM, and in not being inhibited by the product of the reaction, glucose 6-phos- phate.' This composite statement is based on what appears in nearly all modern textbooks of biochemistry ~-9, and, indeed, in the opening paragraph of TIBS last year TM. While such state- ments are probably intended to provide a concise view of the physiological dif- ferences between the hexokinase isoen- zymes, from the enzymological point of view it would be hard to find a way o~f packing more errors into a few words.

'Glucokinase', also known as hexo- kinase D (or type IV), is one of four hexo- ldnase isoenzymes (A to D, or types 1 to IV) that can be clearly distinguished in

A. Comish-Bowden and M. L. C~,rdenas are at the Centre de Biochimie et de Biologie Mol6culaire, Centre National de la Recherche Scientifique, BP 31, 13402 Ma~seille, France. © 1991, Elsevier Science Publishers, (UK) 0376-5067/91/$02.00

Rat liver contains four hexokinase isoenzymes, one of which: despite often being called 'glucokinase', is no more specific for glucose than the others. However, it does differ from them in displaying a sigmoid kinetic response to glucose, requiring much higher glucose concentrations for activity, and being insensitive to physiological concentrations of glucose 6-phosphate.

the rat liver ~'12. In most of the other 100 or so vertebrate species that have been studied, the adult liver likewise con- tains three or four isoenzymes :3.

The hexoldnase isoenzymes can catalyse the phosphorylation of both o¢- and ~D-glucose, although with different kinetic constants 14"~5. Generally, there is a somewhat higher limiting rate for phosphorylation of the [Yanomers o~ glucose and mannose, but much higher affinities for the a-anomers.

The idea that liver hexokinase D ('glu- cokinase') is more specific for glucose

than the other isoeazymes derives from the general agreement that it shows much more activity with glucose than with fructose at sugar concentrations of 100 mM, whereas no such striking differ- ence is apparent with the other isoen- zymes. However, for hexokinases A, B and c this concentration is saturating for both hexoses, whereas for hexoki- nase D it approaches saturation for glu- cose though it is well below saturation for fructose 16. Much earlier WalkeP 7, and later others ~-2°, pointed out that although the name "glucokinase' (and a

281

TIBS 16 - AUGUST 1991

different EC number) for this isoenzyme might be a useful guide to its physiologi- cal role it was unsatisfactory as the en- zyme was not wholly specific for glucose.

Assessing specificity by kinetic measurements at a single concen- tration, with substrates considered one at a time, was once quite standard. However, there is now widespread acceptance of Fersht's view that dis- crimination between substrates that are simultaneously present provides the only meaningful measure of speci- ficity 2=. Examination of the four hexoki- nase isoenzymes on ~ms nasis shows that they are similar in sugar specifici- ty, and that if any difference exists it is that hexokinase D is less effective at discriminating between glucose and fructose than some of the other isoen- zymes ~6. Thus the name 'glucokinase' is a misnomer for hexokinase D and it should be reserved for enzymes that are genuinely specific for glucose, such as those from moulds, bacteria and invertebrate animals.

Even Walker's suggestion ~7 that the name 'glucokinase' offers a 'useful guide to its physiological role' is arguable: although it is probably true that glu- cose is the only physiologically signifi- cant substrate for hexokinase o this is no less true for the other isoenzymes, as it has long been known that fructose metabolism h~ liver is initiated by phos- phorylation to fructose 1-phosphate, a reaction catalysed by fructokinase 22.

The Michaelis constant is a param- eter of the Michaelis-Menten equation; it has no meaning when applied to an enzyme that does not obey Michaelis-Menten kinetics 23. Kinetic data for hexokinase D published more than 20 years ago indicated easily observable deviations from Michaelis- Menten kinetics 24. Subsequent more detailed investigations 2~2~ showed that the plot of rate against glucose concen- tration is sigmoid, with a maximum slope at a concentration of about 2.5 raM, close enough to the 'normal' blood glucose concentration of 5 mM to suggest that the sigmoidicity has a physiologically important role. Yet none of the textbooks cited above men- tions the sigmoidicity, and those that include figures purporting to illustrate the kinetics of hexokinase and 'glucoki- nase' show cleanly hyperbolic curves for both isoenzymesT,9: these are little changed from ones published in text- !),~e, ks written before the si~moidicity was well-established 30.

It might be argued that,~statements 282

that 'glucokinase' has a Km of 10 mM are not intended to imply adherence to Michaelis-Menten kinetics, but only that the enzyme is half-saturated at a con- centration far higher than that needed for the other isoenzymes. This is cer- tainly correct, but it seems to be a rather indirect and confusing way of expressing it.

The statement that hexokinase D is not inhibited by glucose 6-phosphate is correct under physiological conditions, as substantial inhibition requires con- centrations above about 50 mM (Ref. 27). Nonetheless, from the point of view of enzymology the bare statement can be misleading, as it implies an anomaly that does not exist: the weak inhibition is quite consistent with the fact that not only glucose 6-phosphate, but also glu- cose, binds much more weakly to hexo- kinase D than to the other isoenzymes.

The statement with which we began contains or implies five errors, of which the most serious are the false statement of hexose specificity and the failure to recognize the sigmoid character of the response of hexokinase D to glucose, it is hard to understand the persistence with which these errors are copied from textbook to textbook, given that the facts are not controversial and that the main conclusion that their authors wish to draw from the discussion - that 'glucokinase' is better adapted than 'hexokinase' to respond to high blood- glucose concentrations - accords bet- ter with the actual kinetic behaviour of these enzymes than it does with the fic- titious behaviour that is described ~8.

As all of the textbooks we cite reter to the bexokinases in the context of their physiological role they may, per- haps, be forgiven for omitting all men- tion of what enzymologists might regard as the most interesting aspect of hexokinase D, namely that its sigmoid response to glucose is a property of a monomeric enzyme with only a single binding site for glucose. Cooperative behaviour has been widely regarded as impossible for monomeric enzymes, and, as we have discussed elsewhere 3~, there are not many other examples.

We suggest an alternative textbook statement along the following lines: 'Mammals have several isoenzymes to catalyse the formation of glucose 6-phosphate from glucose. Hexokinase D (or type IV), the predominant isoen- zyme in the liver, is a low-affinity hexo- kinase and differs significantly from the others. Its levels vary markedly with dietary and hormonal status; it requires

much higher glucose concentrations (about 10 mM) for half-saturation, with a sigmoid dependence, and it is insensi- tive to physiological concentrations of glucose 6-phosphate. It is thus well adapted to respond to variations in blood-glucose levels. Despite its popu- lar but misleading name of "glucoki- nase", its sugar specificity is similar to that of the other isoenzymes.'

References 1 Lehninger, A. L. (1982) Principles of

Biochemistry, p. 431, Worth 2 Rawn, J. D. (1983) Biochemistry (1st edn),

p. 557, Harper & Row 3 Zubay, G. (1983) Biochemistry, p. 296, Addison-

Wesley 4 Bohinski, R. C. (1987) Modem Concepts in

Biochemistry (5th edn), pp. 502-503, Allyn & Bacon

5 Weil, J. H. (1987) Biochimie G~n6rale (5th edn), pp. 177-178, Masson

6 Stryer, L. (1988) Biochemistry (3rd ecln), p. 361, Freeman

7 Kuchel, P. W. and Ralston, G. B. (1988) in Schaum's Outline of Theory and Problems of Biochemistry, pp. 310-311, McGraw-Hill

8 Delaunay, J. (1988) Biochimie, p. 392, Hermann

9 Mathews, C. K. and van Holde, K. E. (1990) Biochemistry, pp. 438-439, Benjamin/Cummins

10 Watford, M. (1990) Trends Biochem. Sci. 15,1-2 11 Gonz&lez, C., Ureta, T., S,~lchez, R. and

Niemeyer, H. (1964) Biochem. Biophys. Res. Commun. 16, 347-352

12 Katzen, H. M., Soderman, D. D. and Nitowsky, H. M. (1965) Biochem. Biophys. Res. Commun. 19, 377-380

13 Ureta, T. (1982) Comp, Biochem. Physiol. 71B, 549-555

I4 Meglasson, M. D. and Matsch!nsky, F. M. (1983) J. Biol. Chem. 258, 6705--6708

15 Giroux, M. H., Sener, A, and Malatsse, W. J. (1985) Biochim. Biophys. Acta 829, 354-357

16 CSrdenas, M. L., Rabajille, E. and Niemeyer, H. (1984) Biochem. J. 222, 363-373

17 Walker, D. G. (1968) Essays Biochem. 2, 33-67 18 Niemeyer, H., Ureta, T. and Clark-Turri, L.

(1975) Mol. Cell. Biochem. 6, 109-125 19 Ureta, T., Radojkovi~:, J., Lagos, R., Guix6, V.

and NO~ez, L. (1979) Arch. BioL Meal. Exp. 12, 587-609

20 Pollard-Knight, D. and Comish-Bowde~, A. (1982) MoL Cell. Biochem. 44, 71-80

21 Fersht, A. (1985) Enzyme Structure and Mechanism (2nd edn), pp. 347-368, Freeman

22 Hers, H-G. (1955) J. Biol. Chem, 214, 373-381 23 Nomenclature Committee of IUB (1982)

Symbolism and Terminology in Enzyme Kinetics, Eur. ! Biochem. 128, 281-291

24 Parry, M. J. and Walker, D. G. (1967) Biochem. 3. 105, 473-482

25 Niemeyer, H., CSrdenas, M. L., Rabajille, E., Ureta, T., Clark-Turri, L. and Pefiaranda, J. (1975) Enzyme 20, 321-333

26 Storer, A. C. and Cornish-Bowden, A. ('1976) Biochem. J. 159, 7-14

27 Storer, A. C. and Comish-Bowden, A. (1977) Biochem. J. 165, 61-69

28 Tippett, P. S. and Neet, K. E. (1982) J. Biol. Chem. 257, 12839-12845

29 Cbrdenas, M. L., Rabajille, E. and Niemeyer, H. (1984) Eur. J. Biochem. 145,1~3-171

30 Lehninger, A. (1975) Biochemistry (2nd edn), pp. 423-424, Worth

31 Cornish-Bowden, A. and C~rdenas, M. L (1987) J. Theor. Biol. 124, 1-23