Intramolecular ionic interactions of lysine residues and a possible folding domain in fructose diphosphate aldolase.

1. Treatment with methyl acetimidate was used to probe the topography of the tetrameric fructose 1,6-diphosphate aldolase from ox liver. A single treatment with imido ester in the presence or absence of 20mM-fructose 1,6-diphosphate caused the number of amino groups in the enzyme to fall to approx. 30% of the starting number (assumed to be 30 per subunit). The catalytic activity of the aldolase modified in the presence of fructose 1,6-diphosphate was unaffected, whereas that of the enzyme modified in the absence of substrate fell by about 20%. 2. Use of methyl [1-14C]acetimidate and small-scale methods of protein chemistry showed that the amino group of lysine-27 (the numbering is that of the highly homologous rabbit muscle enzyme) is essentially unavailable for amidination in the native enzyme and is therefore predicted to be buried in a hydrophobic environment, probably in the form of an ion-pair with a negatively charged side-chain carboxyl group. All the other lysine residues that reacted poorly with methyl acetimidate in the native enzyme (a total of 7) were found to be within the primary structure bounded by lysine-107 and lysine-227. An important member of this group of lysine residues displaying aberrant reactivity is lysine-227, which is known to form an imine with the substrate as part of the catalytic mechanism of the enzyme. 3. The results of the amidination experiments can be correlated in an interesting way with previous studies of thiol-group modification in the aldolases. Taken together, and arguing in part by analogy with the results of identical experiments with glyceraldehyde 3-phosphate dehydrogenases where the three-dimensional structure is known [Lambert & Perham (1977) Biochem. 4. 161. 49-62], they suggest that the region of primary structure from residues 107-227 may form the whole or part of a three-dimensional structural feature, perhaps a folding domain. A three-dimensional structure deduced from X-ray-crystallographic analysis will be needed to interpret these findings more closely. 4. The amino groups of lysine residues are commonly thought to reside at the 'surface' of protein structures. The patterns of specific lysine residues in glyceraldehyde 3-phosphate dehydrogenases and in aldolases that have been found to react poorly with methyl acetimidate in the native enzymes can be attributed to intramolecular ionic interactions deep in hydrophobic pockets and at the protein 'surface'. Such ionic interactions may contribute significantly to the stability of a given protein.