Proteomics and Models for Enzyme Cooperativity

Abstract

Cooperativity is a phenomenon of universal importance in biological systems and has almost as much variety as it has ubiquity. In virtually all ecosystems the cooperativity of groups, species, and individual organisms is evident. The cooperativity known as “mutualism” is one in which one species provides a haven or a metabolic advantage to another species. On a more microscopic level is the “metabolic cooperativity” in which an enzyme or substrate of one pathway can cooperate with another pathway by providing a component that can act as a substrate or a regulator of that pathway. Delving deeper into the molecular realm we find the type of cooperativity that will be the focus of this review: “allosteric cooperativity” (1Monod J. Wyman J. Changeux J.P. J. Mol. Cell Biol. 1965; 12: 88-118Google Scholar,2Koshland Jr., D.E. Nemethy G. Filmer D. Biochemistry. 1966; 5: 365-385Google Scholar). The term “allosteric” has been used to describe a ligand-enzyme interaction, which results in a measurable conformational change in proximal and distal regions of that protein. We will categorize as a subdivision of allosteric cooperativity the phenomenon of i3 cooperativity in which the three following conditions are met: (a) the binding of the ligand induces a conformational change in the protein; (b) the conformational changes are intramolecular in the subunits of a multisubunit enzyme; and (c) the sites are initially essentiallyidentical to each other. This type of cooperativity, which has since been shown in many enzymes, receptors, and ion channels, is of critical importance to both evolution and the field of proteomics because it can serve as a general model for the way in which the networks of interacting enzymes of metabolic pathways are regulated. Because these networks of interactions explain a lot of the factors that control these ensembles of networks, they play a major role in the differences in organisms and the understanding of proteomics. Cooperativity was originally found by C. Bohr in hemoglobin (3Bohr C. Hasselbach K.A. Krogh A. Skand. Arch. Physiol. 1904; 16: 401-412Google Scholar); he observed the sigmoid binding curve of O2 to hemoglobin, which he explained by saying that the binding of the first O2 molecule made it easier for the next O2 to bind and hence could be called “cooperative.” The α and β chains of hemoglobin satisfy the “essentially identical” criteria for i3 cooperativity because the binding affinities, sequence homology, and/or structural similarity of the O2binding sites reveal that they are nearly identical but not quite. After the importance of conformational changes was recognized, two different theories of the cooperative mechanism were postulated. One was the theory of Monod, Wyman, and Changeux (1Monod J. Wyman J. Changeux J.P. J. Mol. Cell Biol. 1965; 12: 88-118Google Scholar), herein referred to as the MWC model (and also mentioned as the “symmetry” model, “concerted” model, or “the two-state” model), and the other was the theory of Koshland, Nemethy, and Filmer (2Koshland Jr., D.E. Nemethy G. Filmer D. Biochemistry. 1966; 5: 365-385Google Scholar), herein referred to as the KNF model (and often mentioned as the “induced fit” model or the “sequential” model). The MWC model proposed that the subunits changed shape in a concerted manner to preserve the symmetry of the entire molecule as it was transformed from one conformation (T) to a second conformation (R) under the influence of ligand. The alternative KNF model postulated that each subunit changed shape as ligand bound, so that changes in one subunit led to distortions in the shape and/or interactions of other subunits of the protein. A mathematical examination of these theories showed that both gave sigmoid curves and could explain, within experimental error, how O2 bound to hemoglobin. Both theories were postulated to apply to many other enzymes that also gave sigmoid curves showing cooperativity. The KNF model, however, also predicted that in some cases the first ligand to bind could make it more difficult for subsequent ligands to bind. This was called “negative cooperativity” because there was (a) “cooperativity” between the subunits and (b) “negative” because binding of one ligand made the binding of subsequent ligands more difficult (4Conway A. Koshland Jr., D.E. Biochemistry. 1968; 7: 4011-4023Google Scholar, 5Levitzki A. Koshland Jr., D.E. Proc. Natl. Acad. Sci. U. S. A. 1969; 62: 1121-1128Google Scholar). The MWC theory allowed no such alternative. Because only the KNF theory fit negatively cooperative enzymes, it is easy to select that model for such enzymes, but because both theories fit positively cooperative enzymes more sophisticated tools must be applied to such cases. However, positive cooperativity and negative cooperativity are easy to distinguish from each other and they are important to proteomics and evolution so we will address their significance first and the mechanism for achieving them next. To obtain an objective appraisal of the relative quantities of negatively and positively cooperative enzymes in nature, we first selected all publications that had cooperativity in their titles in the period 1980–1990. Tables 1–3 (see supplemental material) are a distillation from 7,316,007 documents in the Science Citation Index from the years 1980–1990 inclusive. Of these, 374 articles had “cooperativity” in the title. From there, articles focusing on enzymes consistent with i3 cooperativity were identified and are listed in Tables 1–3. Table 1 shows 29 of the 291 examples of protein cooperativity reported in the 1980–1990 period (Refs. 19Wilhelm F. Norman A.W. FEBS Lett. 1984; 170: 239-242Google Scholar, 20Damodoran S. Kinsella J.E. J. Biol. Chem. 1980; 255: 8503-8508Google Scholar, 21Geeves M.A. Halsall D.J. Biophys. J. 1987; 52: 215-220Google Scholar, 22Matsunoyagi A. Hatefi Y. J. Biol. Chem. 1990; 265: 82-88Google Scholar, 23Moore G.J. Scanlon M.N. Gen. Pharmacol. 1989; 20: 193-198Google Scholar, 24Andralojc P.J. Harris D.A. Biochim. Biophys. Acta. 1990; 1016: 55-62Google Scholar, 25Cox J.A. Stein E.A. Biochemistry. 1981; 20: 5430-5436Google Scholar, 26Togni A. Pastor S.D. J. Org. Chem. 1990; 55: 1649-1664Google Scholar, 27Iida S. Potter J.D. J. Biochem. (Tokyo). 1986; 99: 1765-1772Google Scholar, 28Coukell M.B. Differentiation. 1981; 20: 29-35Google Scholar, 29Benbaruch G. Schreiber G. Sokolovsky M. Mol. Pharmacol. 1982; 21: 287-293Google Scholar, 30Snow L.D. Doss R.C. Carraway K.L. Biochim. Biophys. Acta. 1980; 611: 333-341Google Scholar, 31Gordon M. Robertson W.R. Cell Biochem. Funct. 1987; 5: 255-261Google Scholar, 32Notides A.C. Lerner M. Hamilton D.E. Proc. Natl. Acad. Sci. U. S. A. 1981; 78: 4926-4930Google Scholar, 33Heyns W. Bossyns D. J. Steroid Biochem. Mol. Biol. 1983; 19: 1689-1694Google Scholar, 34Bach R. Gentry R. Nemerson Y. Biochemistry. 1986; 25: 4007-4020Google Scholar, 35Maitre M. Rumigny J.F. Mandel P. Neurochem. Res. 1983; 8: 113-120Google Scholar, 36Meglasson M.D. Schinco M. Matschinsky F.M. Diabetes. 1983; 32: 1146-1151Google Scholar, 37Bale J.R. Chock P.B. Huang C.Y. J. Biol. Chem. 1980; 255: 8424-8430Google Scholar, 38Cortese J.D. McIntrye J.O. Duncan T.M. Fleischer S. Biochemistry. 1989; 28: 3000-3008Google Scholar, 39Foreman J.E. Niehaus W.G. J. Biol. Chem. 1985; 260: 19-22Google Scholar, 40Bisswanger H. J. Biol. Chem. 1984; 259: 2457-2465Google Scholar, 41Haber R.S. Loeb J.N. J. Gen. Physiol. 1983; 81: 1-28Google Scholar, 42Kuo L.C. Lipscomb W.N. Kantrowitz E.R. Proc. Natl. Acad. Sci. U. S. A. 1982; 79: 2250-2254Google Scholar, 43Sweadner K.J. J. Biol. Chem. 1985; 260: 1508-1513Google Scholar, 44Royer W.E. Hendrickson W.A. Chiancone E. J. Biol. Chem. 1989; 264: 21052-21061Google Scholar, 45McGrew S.G. Wolleben C. Siegl P. Inui M. Fleischer S. Biochemistry. 1989; 28: 1686-1691Google Scholar, 46Konishi K. Fujioka M. J. Biol. Chem. 1988; 263: 13381-13385Google Scholar, 47Wolodko W.T. O'Connor M.D. Bridger W.A. Proc. Natl. Acad. Sci. U. S. A. 1981; 78: 2140-2144Google Scholar), Table 2 shows 27 of the 215 examples of negative cooperativity reported in 1980–1990 (Refs. 48Johal S. Partridge B.E. Chollet R. J. Biol. Chem. 1985; 260: 9894-9904Google Scholar, 49Takama M. Nosoh Y. FEBS Lett. 1981; 123: 100-102Google Scholar, 50Porteu F. Halbwachsmecarelli L. J. Biol. Chem. 1988; 263: 5091-5097Google Scholar, 51Teruel J.A. Tudela J. Carmona F.G. Fernandez J.C.G. Canovas F.G. J. Bioenerg. Biomembr. 1987; 19: 383-396Google Scholar, 52Palma M.S. Teno A.M. Rossi A. Phytochemistry. 1983; 22: 1899-1901Google Scholar, 53Benhayyim G. Ran U. Physiol. Plant. 1990; 80: 210-216Google Scholar, 54Delarco A. Burguillo F.J. Roig M.G. Usero J.L. Izquierdo C. Herraez M.A. Int. J. Biochem. 1982; 14: 127-140Google Scholar, 55Kitamura E. Kikkawa R. Fujiwara Y. Imai T. Shigeta Y. Biochim. Biophys. Acta. 1984; 800: 66-74Google Scholar, 56Clark S. Larkins R.G. Melick R.A. Biochem. Biophys. Res. Commun. 1980; 95: 1543-1549Google Scholar, 57Lohman T.M. Bujalowski W. Biochemistry. 1988; 27: 2260-2265Google Scholar, 58Usero J.L. Burguillo F.J. Izquierdo C. Delarco A. Herraez M.A. Thromb. Res. 1985; 39: 379-386Google Scholar, 59Marasco W.A. Feltner D.E. Ward P.A. J. Cell. Biochem. 1985; 27: 359-375Google Scholar, 60Kapoor M. Roland B. FEBS Lett. 1981; 123: 255-257Google Scholar, 61Hasinoff B.B. Davey J.P. Biochem. J. 1986; 253: 891-893Google Scholar, 62Scheek R.M. Kalkman M.L. Berden J.A. Slater E.C. Biochim. Biophys. Acta. 1980; 613: 275-291Google Scholar, 63Douzhenkova I.V. Asryants R.A. Nagradova N.K. Biochim. Biophys. Acta. 1988; 957: 60-70Google Scholar, 64Tait J.F. Weinman S.A. Bradshaw R.A. J. Biol. Chem. 1981; 256: 1086-1092Google Scholar, 65Arner P. Bolinder J. Ostman J. Acta Diabetol. Lat. 1983; 20: 197-203Google Scholar, 66Belknap W.R. Portis A.R. Biochemistry. 1986; 25: 1864-1869Google Scholar, 67Pry T.A. Hsu R.Y. Biochemistry. 1980; 19: 951-962Google Scholar, 68Sweadner K.J. J. Biol. Chem. 1985; 260: 1508-1513Google Scholar, 69Hanson K.R. Arch. Biochem. Biophys. 1981; 211: 564-574Google Scholar, 70Olefsky J.M. Green A. Ciaraldi T.P. Saekwo M. Rubenstein A.H. Tager H.S. Biochemistry. 1981; 20: 4488-4492Google Scholar, 71Scully M.F. Ellis V. Watahiki Y. Kakkar V.V. Arch. Biochem. Biophys. 1989; 268: 438-446Google Scholar, 72Kruger P. Gilge G. Cabuk Y. Wollmer A. Biol. Chem. Hoppe-Seyler. 1990; 371: 669-673Google Scholar, 73Degtyarev S.K. Beresten S.F. Denisov A.Y. Lavrik O.I. Kisselev L.L. FEBS Lett. 1982; 137: 95-99Google Scholar, 74Castuma C.E. Brenner R.R. Biochim. Biophys. Acta. 1983; 729: 9-16Google Scholar, 75Periyasamy S.M. Somani P. Eur. J. Pharmacol. 1985; 116: 17-24Google Scholar), and Table 3 shows 4 of the 61 examples of enzymes that show both negative and positive cooperativity in 1980–1990 (Refs. 76Saifullin S.R. Goldshtein B.N. Mol. Biol. 1986; 20: 187-194Google Scholar, 77Roustiau S. Batrel Y. Bernicardperon A. Legal Y. Biochem. Syst. Ecol. 1985; 13: 45-50Google Scholar, 78Neujahr H.Y. Biochemistry. 1988; 27: 3770-3775Google Scholar, 79Didonato A. Piccoli R. Dalessio G. Biochem. J. 1987; 241: 435-440Google Scholar). As can be seen in the list of enzymes given in Tables 1 and 2, the number of negatively cooperative enzymes is approximately the same as the number of positively cooperative enzymes suggesting that the sensitivity capabilities listed above have about equal evolutionary value with a slight evolutionary advantage to positive cooperativity. To check these results, in Tables 4–6 in the supplement are positive cooperativity and negative cooperativity in the years 1991–1993 (Refs. 80Gu J.L. Goldfine I.D. Forsayeth J.R. Demeyts P. Biochem. Biophys. Res. Commun. 1988; 150: 694-701Google Scholar, 81Ganela E.I. Franco R. Cascante M. Biosystems. 1989; 23: 7-14Google Scholar, 82Shilin V.I. Pechukin N.S. Brilkov A.V. Beliankina N.I. Terskov I.A. Dokl. Akad. Nauk SSSR. 1983; 271: 505-508Google Scholar, 83Foreman J.E. Niehaus W.G. J. Biol. Chem. 1985; 260: 19-22Google Scholar, 84Murray J.M. Weber A. Mol. Cell. Biochem. 1981; 35: 11-15Google Scholar, 85Bubis J. Saraswat L.D. Taylor S.S. Biochemistry. 1988; 27: 1570-1576Google Scholar, 86Danieulius R.V. Mineyev A.P. Razjiavin A.P. FEBS Lett. 1989; 250: 183-186Google Scholar, 87Woolery G.L. Walters M.A. Suslick K.S. Powers L.S. Spiro T.G. J. Am. Chem. Soc. 1985; 107: 2370-2373Google Scholar, 88Delarco A. Burguillo F.J. Roig M.G. Usero J.L. Izquierdo C. Herraez M.A. Int. J. Biochem. 1982; 14: 127-140Google Scholar, 89Gordon M. Robertson W.R. Cell Biochem. Funct. 1987; 5: 255-261Google Scholar, 90Ricard J. Kellershohn N. Mulliert G. Biophys. J. 1989; 56: 477-487Google Scholar, 91Tung P.S. Skinner M.K. Fritz I.B. Ann. N. Y. Acad. Sci. 1984; 438: 435-446Google Scholar, 92Ehrlich R. Marin M. Gabarroarpa J. Rodier F. Schmitt B. Reiss C. C. R. Acad. Sci. Ser. III Sci. Vie. 1981; 292: 5-8Google Scholar, 93Jones M.N. Manley P. Holt A. Int. J. Biol. Macromol. 1984; 6: 65-68Google Scholar, 94Olavarria J.M. Cardenas M.L. Arch. Biol. Med. Exp. 1985; 18: 285-292Google Scholar, 95Winkle S.A. Krugh T.R. Nucleic Acids Res. 1981; 9: 3175-3186Google Scholar, 96Maderspach K. Fajszi C. Biochim. Biophys. Acta. 1982; 692: 469-478Google Scholar, 97Newton A.C. Koshland Jr., D.E. J. Biol. Chem. 1989; 264: 14909-14915Google Scholar, 98Cox J.A. Stein E.A. Biochemistry. 1981; 20: 5430-5436Google Scholar, 99Royer W.E. Hendrickson W.A. Chiancone E. J. Biol. Chem. 1989; 264: 21052-21061Google Scholar, 100Conti J. Halladay H.N. Petersheim M. Biochim. Biophys. Acta. 1987; 902: 53-64Google Scholar, 101Henis Y.I. Levitzki A. Eur. J. Biochem. 1980; 112: 59-73Google Scholar, 102Whitehead E.P. J. Theor. Biol. 1981; 93: 547-583Google Scholar, 103Cornishbowden A. Pollardknight D. Arch. Biol. Med. Exp. 1985; 18: 293-300Google Scholar, 104Lee S.L. Veis A. Int. J. Pept. Protein Res. 1980; 16: 231-240Google Scholar). Because the relative ratios of enzymes and other proteins with positive and negative cooperativity are roughly the same in the two arbitrarily selected time periods, we can assume that they are probably a pretty good indication of the relative ratios for all enzymes, i.e. about 50% positive and 50% negative. To some this may seem unusual because positive cooperativity was discussed first in relation to hemoglobin and to many people seems to be the reason that cooperativity exists. The large number of negatively cooperative enzymes shows that this type of cooperativity is not an aberration or a minor curiosity but rather that there must be good evolutionary reasons in metabolic systems for it as well as positive cooperativity. It is of interest, therefore, to consider why each should be selected over evolutionary time (6Koshland D.E. Curr. Opin. Struct. Biol. 1996; 6: 757-761Google Scholar). As seen in Fig. 1, positive cooperativity confers the metabolic advantage of amplifying the sensitivity of a signal, i.e. a small change in ligand L can have a far greater effect on the output response in a positively cooperative system than in a Michaelis-Menten system. For example, a 3-fold increase in the ligand (L = O2) concentration for hemoglobin changes the binding capacity 9-fold (from 10 to 90%), whereas in a Michaelis-Menten system, it requires an 81-fold change in ligand concentration to go from 10% binding capacity to 90% binding capacity. It is important, for example, that hemoglobin picks up the maximum amount of O2 in the lungs and unloads the maximum amount of O2 to the tissues, a biological phenomenon for which positive cooperativity is very important (7Perutz M.D. Nature. 1970; 228: 726-734Google Scholar, 8Ho C. Adv. Protein Chem. 1992; 43: 153-312Google Scholar). In fact, humans with hemoglobin mutations that lack the cooperativity are very sick people (9Perutz M. Lehmann H. Nature. 1968; 219: 902-909Google Scholar). Negative cooperativity, on the other hand, decreases the sensitivity to Michaelis-Menten, but in the process of doing that, it extends the range over which some response is generated. CTP acts as an allosteric inhibitor to inhibit carbamoyl-phosphate synthetase, which is at a branch point that leads to several other products besides CTP (10Gerhart J. Pardee A. J. Biol. Chem. 1962; 237: 891-897Google Scholar). Therefore, it is important that excess CTP not shut down the carbamoyl-phosphate synthetase enzyme completely because all of the other pathways would be inhibited as a consequence, so negative cooperativity is observed for CTP binding to carbamoyl-phosphate synthetase, which makes it very difficult to shut down the enzyme completely even with great excess of CTP. Although the present review does not prove that cooperativity is correlated with being a branch point enzyme, there seems to be a general trend indicating such a correlation. A second reason for the importance of i3 cooperativity may be its ability to shift the range of sensitivity without changing the amino acid sequences of the active site. In fact it is observed that there is a different midpoint of the affinity curve for O2between frogs and tadpoles (11Perutz M.F. Brunori M. Nature. 1982; 299: 421-426Google Scholar). One way that might have been tried to achieve such a change would be to alter amino acids where O2 binds to hemoglobin, but that binding site is exquisitely designed so that the ferrous ion of the heme is prevented from getting oxidized to ferric ion (7Perutz M.D. Nature. 1970; 228: 726-734Google Scholar, 8Ho C. Adv. Protein Chem. 1992; 43: 153-312Google Scholar, 9Perutz M. Lehmann H. Nature. 1968; 219: 902-909Google Scholar). The adjustments in side chain amino acids and their precise geometry have been selected over evolutionary time, and it is not clear that a new active site could be designed that was as effective at a different O2concentration range, so the same active site is used for both tadpole and frog but the intersubunit contacts, which are far from the active site, are mutated. How the subunit contacts changeS 0.5, the midpoint of the curve without changing the cooperativity (steepness) in the KNF model is shown in Fig.2, where a number of curves of the same steepness are seen to have different midpoints (2Koshland Jr., D.E. Nemethy G. Filmer D. Biochemistry. 1966; 5: 365-385Google Scholar) because they have different subunit interrelations. This property is used to optimize the O2 concentration range from aqueous ponds to dry land by changing subunit contacts without tampering with the active site (11Perutz M.F. Brunori M. Nature. 1982; 299: 421-426Google Scholar). This ability to change the midpoint of a saturation curve without changing its active site has undoubtedly been useful in changing the responsiveness range of the same enzyme in different tissues or in different species. At this point it may be useful to examine, first, whether it is appropriate to retain simplified models when x-ray crystallographic pictures exist, and second, which model, the concerted or the sequential, is more likely to be correct. First, the simplification with circles and squares for conformations is valuable even in the age of crystallography. By analogy, Schrodinger's equation is a far more accurate representation of the electrons and orbits of an atom than a simplistic symbol like H+ or Cu2+, but it will be a long time before a chemist will find the Schrodinger equation more useful in describing a simple reaction than using a simplification like Cu2+ + Fe2+ = Cu+ + Fe3+, so the simple models explain the phenomenon and the crystallography and complex math give the accurate details. Thus in the case of hemoglobin, Holt et al. (8Ho C. Adv. Protein Chem. 1992; 43: 153-312Google Scholar) using spin labels and NMR have shown clear examples of a stepwise change in structure that is in agreement with the KNF model but also shows a switch between two quaternary structures, which is in agreement with the MWC model. Perutz, an initial advocate of the MWC model for Hb, has recently said (12Perutz M.F. Annu. Rev. Biophys. Biomol. Struct. 1998; 27: 1-34Google Scholar) the changes induced by oxygen indicated that both the MWC model and the KNF model are partly right, and Holt and Ackers (13Holt J.M. Ackers G.K. FASEB J. 1995; 218: 210-218Google Scholar) have proposed a sequential model with a switch in quaternary structure at the 2O2 bound state. In the case of aspartyl transcarbamoylase, Stevens and Lipscomb (14Stevens R. Lipscomb W.N. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 5281-5295Google Scholar) showed a combination of sequential intrasubunit conformational changes on a broad background of a quaternary change. In the second place the KNF model can explain all cases,i.e. positive, negative and no cooperativity; where there is a ligand-induced conformational change, it is logical to deduce that KNF is the general model and see whether there are circumstances in which it would reduce to the MWC model. In fact, a calculation to clarify the relationship of the models to the protein forces in an enzyme was made (15Haber J. Koshland Jr., D.E. Proc. Natl. Acad. Sci. U. S. A. 1967; 58: 2087-2093Google Scholar) and shows that the general KNF model reduces to the MWC model or the simplified KNF model when appropriate subunit parameters are chosen (14Stevens R. Lipscomb W.N. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 5281-5295Google Scholar). Moreover, the KNF model can explain cases like aldolase and lactic dehydrogenase, where there are big changes in the conformation and no cooperativity by assumingK AA = K AB =K BB, i.e. no change in subunit interactions but still a change in the conformation of the individual subunits. Another class of enzymes that must fit the KNF model is those that show both positive (with one ligand) and negative cooperativity (with another ligand) or those that show both positive and negative cooperativity with a single ligand during the sequential binding of that ligand. Thus, a general way of looking at cooperativity is to say the ligand usually induces a sequential change such that the conformation change caused by the first ligand is transmitted to neighboring subunits in such a way that it may make subsequent ligands bind (a) more easily, (b) less easily, or (c) without any effect. In one extreme, the binding of the first molecule can cause all the subunits to change in a concerted way so that symmetry is preserved, and at another extreme they can change the conformation sequentially subunit by subunit as each ligand is bound with no net effect on neighboring subunits. The sequential binding of ligand may at some point cause the protein to shift from one quaternary structure to another (say T to R or A4 to B4). An elegant use of physical tools to distinguish between the models for positively cooperative enzymes has been developed by Ho and co-workers (17Kim H.W. Sliln T.J. Sun D. Ho N. Madrid M. Tan M.F. Zou M. Cottan P.F. Ho C. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 11547-11551Google Scholar). It is intriguing in this connection that it was recently found that an artificial point mutation changes pyruvate kinases from no cooperativity to positive cooperativity (16Ikeda Y. Tanaka T. Noguchi T. J. Biol. Chem. 1997; 272: 20495-20501Google Scholar). Similarly, the aspartyl receptor can be changed from positive to negative or to no cooperativity by changes in a single amino acid, Ser-68, at the subunit interface of that enzyme (18Kolodziej A.F. Tan T. Koshland Jr., D.E. Biochemistry. 1996; 35: 14782-14792Google Scholar). Thus, cooperativity in a multisubunit protein depends on a delicate balance between many forces in which it seems likely that a small change in ligand or an amino acid mutation can change the cooperativity and allosteric properties of an enzyme, so it becomes apparent how easy it would be in evolution to get the appropriate cooperativity. The initial pattern becomes solidified by selection to favor a cooperativity that benefits that organism, so allosteric cooperativity has been preserved and improved over evolutionary time as a mechanism that can increase sensitivity, expand the range, and adapt metabolism to a change in conditions such as the change in conditions from sea to land. Cooperativity is one of the established phenomena that makes living possible and will take its place as a key factor in the proteomics of the cell. Download .pdf (.21 MB) Help with pdf files