| dc.description.abstract | The harmonious coordination among the myriad of reactions operating simultaneously in a cell is achieved by conformational response of key enzymes to ligand binding. Most often, in a highly branched, interlinked metabolic pathway, the enzyme located at the branch points, or the enzyme catalyzing the first committed step in the sequence, is subject to regulatory restraints. The mechanisms operative in the maintenance of homeostasis of flavin coenzymes, which is under stringent control in cells, are to date ambiguous, despite a thorough understanding of the biogenesis, metabolism, and mechanisms of catalysis of these coenzymes. It was with a view to comprehend the regulation of the biosynthesis of the flavin coenzymes in plants, attempts were made to isolate flavokinase, the first enzyme in the pathway, in a homogeneous form and characterize its interactions with substrates and other effectors.
The mechanisms of the phosphotransferases, with special reference to the changes in the configuration of the phosphoryl group during its transfer from the donor to the acceptor molecule, are reviewed in the introductory chapter. The interactions of Pi at the genetic level, leading to the induction and regulation of alkaline phosphatase, are briefly summarized. The consequences of the changes in the conformation of ribulose 1,5-bisphosphate carboxylase, glutaminase, adenylate deaminase, luciferase, ATPase, nucleotide pyrophosphatase, and AMP nucleosidase upon their binding to Pi are also highlighted in this chapter.
The materials and methods employed in the study, viz., the preparation of affinity matrices, labelled substrates and products; fluorescence and visible difference spectroscopy; various types of paper and gel chromatography, electrophoreses; and assay of various enzymes, are described.
Flavokinase catalyzes the synthesis of FMN from riboflavin and ATP. The enzyme was isolated from mung bean (Vigna radiata) seeds by ammonium sulfate fractionation and affinity chromatography on Blue-Sepharose, riboflavin-Sepharose, and C-8 ATP agarose, in the presence of Pi, which was required for the binding of the enzyme to the affinity matrices (Fig. 14). The activity of the enzyme was enriched 107,000-fold with a recovery of 30 per cent (Table 1).
The enzyme was homogeneous as seen by PAGE under denaturing (Fig. 16 inset) and non-denaturing conditions (Fig. 15A). The identical molecular weight of 30,000 obtained both by gel filtration (Fig. 16) and SDS-PAGE (Fig. 16 inset) suggested that the enzyme was a monomeric protein.
The products of the reaction catalyzed by flavokinase were identified to be FMN and ADP by paper chromatography in several solvent systems (Tables 2 and 5). The stoichiometry of the reaction was established to be 1:1 (Table 4), by estimating the riboflavin consumed in the reaction and FMN formed at different intervals of time.
The reaction catalyzed by the enzyme was linear up to 15-20 min (Fig. 18) and with an enzyme concentration up to 6-7 g/ml (Fig. 19). The temperature (Fig. 21) and pH optimum (Fig. 20) for the enzyme were 55°C and 8.5-9.0, respectively. The enzyme required a divalent metal ion like Mg² for activity (Fig. 23).
A unique feature of the purification procedure was the requirement of Pi for the binding of the enzyme to the affinity matrices. The enzyme could be desorbed from the matrix by reducing the Pi concentration by washing the affinity matrices with water (Figs. 14-24).
The specificity of the interaction of Pi with flavokinase was established by estimating the capacity of other phosphate-containing compounds, or arsenate, to retain the enzyme bound in the presence of Pi on the affinity matrix when the concentration of Pi was lowered by washing the matrix with these compounds. The hypothesis was that if the other phosphate compounds were able to mimic the effects of Pi, they would retain the enzyme on the matrix in the absence of Pi. Only acetyl phosphate and arsenate were able to mimic the conformational changes brought about by Pi (Tables 5 and 6), by retaining the enzyme bound to the matrix in the presence of Pi and also were able to adsorb the enzyme to the matrix in the absence of Pi.
These reversible conformational changes of the enzyme (E) and the enzyme in the presence of Pi (Ep ) were examined in greater detail by monitoring the changes in the fluorescence spectra, quenching of fluorescence intensity, and visible absorption difference spectroscopy of the dye Cibacron Blue F5GA.
The fluorescence spectrum of the enzyme in the absence and presence of Pi and Cibacron Blue F5GA (in the presence and absence of Pi) were recorded (Figs. 26-28). There was a bathochromic shift of 5-20 nm, indicative of the changes produced in the tryptophanyl residues, which shifted to a relatively hydrophilic environment in the enzyme upon interaction with these ligands (Figs. 26-28).
These spectral shifts were accompanied in each case by quenching of the intrinsic fluorescence of the protein, suggesting that either the fluorescing residues interacted directly with the ligand, or the interaction of the protein with solvent molecules was altered due to the conformational change on ligand binding. The quenching was measured to calculate the binding constants of the dye Cibacron Blue F5GA in the presence and absence of Pi (Figs. 29 and 30). The value of 3.0 nM was obtained in the absence of Pi, which decreased to 0.5 nM in its presence. Accompanying this change, there was an increase in the number of sites from 2 to 9 in the presence of Pi (Fig. 50).
The conformational response of the enzyme to the dye was examined further by visible difference spectroscopy. The absorbance maximum of the dye was at 605-610 nm. In the absence of Pi, there was not much perturbation in the difference spectrum (Fig. 31), but in the presence of Pi, a distinct spectrum was obtained with absorption peaks at 585 and 630 nm (Fig. 31). This split peak is characteristic of ligand-protein interactions indicative of conformational changes in the enzyme, where some of the aromatic residues moved to a hydrophilic environment (610-585 nm) and others to a hydrophobic environment (610-630 nm).
The changes in the environment of the aromatic residues of flavokinase were examined by subjecting the dye-protein complex to varying degrees of hydrophobicity by the addition of anions according to the Hofmeister series (Fig. 32). The intensity of the distinct difference spectrum increased with increasing hydrophobicity of the environment. These results suggest that dye binding was facilitated by an increase in hydrophobicity around the protein. This experiment also indicates that in addition to the increasing hydrophobicity in the environment, potassium phosphate promotes specific interactions (Fig. 32).
It is a dream of every biochemist to relate an observed conformational change in the protein to the changes in the activity of the enzyme in question. This structure-functional relationship of flavokinase was probed by two methods: (i) the changes in the activity of the enzyme upon binding of ligands, and (ii) changes produced in the rates of inactivation of the E and Ep forms of the enzyme, between themselves and upon their interactions with various ligands (Figs. 53-41).
Increasing concentrations of Pi inhibited the activity of flavokinase. The inhibition produced by this ligand was subjected to iteration to obtain a Hill coefficient (n_H) value of 2.6 and a K_i value of 183 mM (Fig. 33A). The Hill coefficient indicated that more than one site may be present for Pi interaction with the enzyme, and also reiterated the conformational change produced by ligand-protein interaction.
The effect of varying substrate concentration on the activity of the enzyme was examined. The substrate saturation pattern of the enzyme was hyperbolic but gave n_H values of 1.0 and 2.0 in the Hill plots, which showed breaks in the saturation curve (Figs. 36 and 57). In the presence of Pi, the cooperative phase was lost; the saturation followed Michaelis-Menten behaviour and the Hill plots were linear with an n_H value of 1.0. These effects were specific to Pi, as other phosphate-containing compounds such as ADP, AMP, TTP, GMP, UTP, carbamyl phosphate, ribose 5-phosphate, ribose 1-phosphate, pyrophosphate, and cacodylate were unable to alter the substrate saturation pattern.
The ligand-protein interactions were next examined by measuring the changes produced in the pattern when the enzyme was heated at 50°C or subjected to proteolysis by chymotrypsin, in the absence and presence of Pi and other ligands, and by urea desensitization.
While the native enzyme was slowly inactivated at 50°C, Ep form was rapidly inactivated at this temperature, and the rate of inactivation was dependent upon the concentration of Pi (Figs. 38 and 39; Table 7).
The interaction of the substrates with the two different enzyme conformers (E and Ep ) was examined. ATP·Mg² afforded little protection to E, but Ep was protected against heat inactivation at 50°C for at least 20 min (Figs. 44 and 45). Riboflavin protected both the E and Ep forms against heat inactivation (Figs. 44 and 45). ATP, like Pi, rendered the enzyme more labile to heat inactivation but had no effect on the Ep form of the enzyme (Figs. 44 and 45). These results were interpreted to indicate that ATP and Pi might be inducing similar types of conformational changes in the enzyme (Figs. 44-46).
This contention was examined in detail by studying the heat inactivation patterns in the presence of varying concentrations of ATP (Fig. 46). Upon plotting k and n_H values obtained versus log[ATP], the values were 24 and 2.1, respectively (Fig. 46), which indicated that ATP, like Pi, was probably interacting at more than one site on the enzyme.
The determination of the conformational response of a protein to the binding of a ligand is incomplete unless examined using more than one probe. It was therefore imperative to check these ligand interactions with the enzyme by yet another probe, namely preferential proteolysis. Plant flavokinase was subjected to chymotrypsin proteolysis in the presence and absence of various ligands—Pi, ATP, riboflavin, ATP·Mg² , and MgSO . The native enzyme, E, was very rapidly inactivated by chymotrypsin, but the presence of Pi, ATP, or other ligands protected the enzyme, indicating again that the enzyme was drawn into a different conformation (Figs. 41 and 43) upon ligand binding.
Conformational change induced in plant flavokinase, probed by fluorescence (Figs. 26, 27, and 42), heat (Figs. 38, 39, 44-46), and proteolysis (Figs. 41 and 43), may vouch for the fact that both ATP and Pi may be drawing the enzyme into similar conformations. These versatile conformational changes, coupled with the observation of cooperativity in substrate interactions, suggest that the enzyme could be a regulatory protein.
The existence of cooperative interactions in a protein is best demonstrated by desensitization. Chaotropic agents have been extensively used to desensitize enzymes to allosteric interactions. Urea denaturation-renaturation desensitized plant flavokinase to cooperative interactions (Figs. 47 and 47A) with its substrates but failed to distinguish between the various conformers of the enzyme in the presence of different ligands.
One of the prerequisites for a mnemonic enzyme catalyzing a bi-bi reaction is that it should catalyze an ordered sequential reaction. It was therefore necessary to study the kinetic mechanism of the reaction catalyzed by plant flavokinase. This was carried out by initial velocity and isotope exchange studies.
The Lineweaver-Burk plots for the substrate saturation at varying, but fixed, concentrations of the other substrate in the range of 10-80 M showed an intersecting pattern of lines (Figs. 51 and 52), diagnostic of a sequential mechanism for the reaction catalyzed by plant flavokinase.
The order of substrate addition was next ascertained by the isotope trap method. The burst in the radioactivity incorporated into FMN was observed only when [³H]-riboflavin was incubated with the enzyme and the ‘chase’ solution containing excess substrates and reaction components was added, and progress of the reaction stopped after one catalytic cycle. This indicates that riboflavin could be the leading substrate in the catalytic reaction (Table 8).
The order in the release of products from the central product complex was then examined by the binary complex detection method. In this, the transfer of label from the product to the substrate is observed only when the free enzyme is able to complex with the incubated pair of substrate and product. Increasing amounts of label were transferred only from FMN to riboflavin with time of incubation when [¹ C]-FMN was incubated with the enzyme and cold riboflavin. This experiment indicates that FMN may be the last product to leave the enzyme surface (Table 9).
These observations, along with the initial velocity and isotope exchange studies, permit the postulation of a probable kinetic mechanism for the reaction catalyzed by plant flavokinase:
Riboflavin + ATP [Riboflavin·ATP·E] FMN + ADP
The results hitherto summarized, coupled with the observations of a burst in the time course of the reaction (Fig. 53) and the ordered sequential kinetic mechanism for the catalytic reaction followed by plant flavokinase, as well as the conformational flexibility of the enzyme in the presence of ligands, monitored by a wide variety of methods, significantly suggest that the monomeric flavokinase, isolated for the first time to homogeneity, could be a mnemonic enzyme, channeling the flux of flavin nucleotides and thus contributing to the overall homeostasis of flavin coenzymes observed in a normal cell. | |
