Enzymes are large molecules. A number of arguments can be offered as to why this should be so, including the need for a large protein, consisting of a linear sequence of amino acids, to have a preferential folding into the biologically effective three-dimensional structure. There are also arguments that the dynamics of segment motion in a large protein can contribute to catalytic effectiveness. However, there is another argument for the large molecules characteristic of enzymes that is quite convincing-their size and structure let reactions occur inside a hydrophobic, nonaqueous, and nonpolar region of the protein while the outside of the protein has polar groups and is compatible with the aqueous solvents.Two interesting effects result from this. On the one hand, substrates that have hydrophobic groups will tend to bind in the nonpolar interior of the protein, where the catalytic groups can be located. On the other hand, water is in one respect an enemy of rate, even though the hydrophobic effect depends on water for its positive contribution to reaction rates. In a A C H T U N G T R E N N U N G reaction in which acid and base groups play a catalytic role, water that is hydrogen bonded to these groups must normally be removed before they can effectively act on the substrate. The energy cost of desolvation of these catalytic groups-and perhaps also desolvation of the substrates-can slow the reaction considerably. For this reason it has often been proposed that the interior of the protein, without any water molecules or highly polar groups in it, is a better place to perform a catalyzed reaction and that such a medium will increase the rates of the processes.
Water Exclusion in Polymeric Enzyme MimicsIn order to explore this, we have synthesized a number of enzyme mimics based on polymers. [1][2][3][4][5][6][7] Our most extensive studies have been directed to branched polyethylenimines, a group of commercial polymers whose catalytic properties have been previously studied by Klotz, [8,9] by Suh, [9,10] and by Kirby.[11] They explored hydrolysis and cleavage reactions, while we have been interested in synthetic processes. In our first studies, [1][2][3][4][5] we examined the amination of keto acids by pyridoxamine units attached to the polymers. Later, we examined the even more effective system in which the pyridoxamine units were reversibly bound to the hydrophobic core of such a polymer, in imitation of the binding of coenzymes by the natural enzymes.[6]As we have described, we synthesized our transaminase mimics by using a variety of polyethylenimines. In the first system, we used a rather large polyethylenimine with a number-average molecular weight of about 60 000 and with high polydispersity; the weight-average molecular weight was 750 000.[1] This commercial polymer has about 1400 nitrogens; about 25 % of them are primary amino groups, 50 % are secondary amines, and 25 % are tertiary amines. We alkylated about 10 % of the nitrogens with alkyl halides ranging from methyl iodide up to dodecyl iodide, and then ...