Characterization of the human O-phosphoethanolamine phospholyase, an unconventional pyridoxal phosphate-dependent beta-lyase

Abstract

Abstract Subgroup-II aminotransferases (AT-II) are a structural subclass of PLP-dependent enzymes, specialized in transforming for compounds containing an amino group not adjacent to a carboxylate. As the name suggests, the vast majority of these enzymes are in fact transaminases, endowed with some structural peculiarities that help explain their specificity for terminal amines. In this thesis work, we analyzed the specificity of one of the few AT-II enzymes that is not a transaminase, namely O-phosphoethanolamine phospho-lyase lyase (PEA-PL), in terms of both reaction and substrate. This enzyme is quite peculiar for several reasons. First, it is the only known AT-II enzyme (together with its closely related paralog, O-phospho-5-hydroxy-L-lysine phospho-lyase) whose primary function is to catalyze an elimination reaction. Second, because its substrate is, quite unusually, an amine containing a phosphate group. Third, because the enzyme, despite being phylogenetically very close to pyruvate-dependent AT-II transaminases (in particular, to the promiscuous AGXT2) was reported to lack any transaminase activity. The obtained results show that this human PEA-PL is at least 500-fold more active as a lyase than as an aminotransferase and that the lyase reaction is very selective for O-phosphoethanolamine, strongly discriminating against closely related compounds. Dissecting the factors that contribute to such narrow substrate specificity we found that PEA-PL is in fact able to recognize only amines and not amino acids, while the phosphate was shown to be a key determinant of substrate specificity. The results of this first part of the work allowed us to conclude that both the substrate- and reaction specificity of PEA-PL are remarkably strict for a PLP-dependent enzyme. To rationalize these findings, and to better understand their mechanistic basis, we undertook a wide kinetic study encompassing (i) the relative ability of various anionic compounds in inhibiting the PEA-PL reaction; (ii) the pH dependence of the enzyme’s steady-state catalytic parameters and of the inhibitors effectiveness; (iii) a spectroscopic study of the interaction of the enzyme with substrates and inhibitors and (iv) a pre-steady state analysis of the PEA-PL reaction, to gain more information on the formation of catalytic intermediates and generally on the reaction mechanism of the enzyme. The results of these studies confirm that the active site of PEA phospho-lyase is optimized to bind dianionic groups and that this is a prime determinant of the enzyme specificity towards PEA. Single- and multiple-wavelength stopped-flow studies showed that upon reaction with PEA the main absorption band of PLP (lambdamax=412 nm) rapidly blue-shifts to ~400 nm. Further experiments suggested that the newly formed and rather stable 400-nm species most likely represents a Michaelis (non-covalent) complex of PEA with the enzyme. Accumulation of such an early intermediate during turnover is unusual for PLP-dependent enzymes and appears counterproductive for absolute catalytic performance, but it can contribute to optimize substrate specificity. PEA phospho-lyase may hence represent a case of selectivity-efficiency tradeoff. In turn, the strict specificity of the enzyme seems important to prevent inactivation by other amines (structurally resembling PEA) that occur in the brain. Having established the high substrate specificity of PEA-PL, we also decided to explore its practical usefulness, by designing an enzymatic assay for the selective detection of PEA in liquid samples. This assay would be most useful for the clinical laboratory: as PEA levels are increased in in some biological fluids in the course of metabolic, genetic and psychiatric diseases, a method for the rapid and quantitative detection of this analyte would be of relevance to the diagnosis and monitoring of such diseases. The fluorometric assay we begun to develop achieved the desired sensitivity, as it could satisfactorily quantitate micromolar amounts of PEA in a buffer-water system.