Guillermo Mulliert

Phosphorylation and sulfation of oligosaccharide substrates critically influence the activity of human β1,4-galactosyltransferase 7 (GalT-I) and β1,3-glucuronosyltransferase I (GlcAT-I) involved in the biosynthesis of the glycosaminoglycan-protein linkage region of proteoglycans

We determined whether the two major structural modifications, i.e. phosphorylation and sulfation of the glycosaminoglycan-protein linkage region (GlcAβ1–3Galβ1–3Galβ1–4Xylβ1), govern the specificity of the glycosyltransferases responsible for the biosynthesis of the tetrasaccharide primer. We analyzed the influence of C-2 phosphorylation of Xyl residue on human β1,4-galactosyltransferase 7 (GalT-I), which catalyzes the transfer of Gal onto Xyl, and we evaluated the consequences of C-4/C-6 sulfation of Galβ1–3Gal (Gal2-Gal1) on the activity and specificity of β1,3-glucuronosyltransferase I (GlcAT-I) responsible for the completion of the glycosaminoglycan primer sequence. For this purpose, a series of phosphorylated xylosides and sulfated C-4 and C-6 analogs of Galβ1–3Gal was synthesized and tested as potential substrates for the recombinant enzymes. Our results revealed that the phosphorylation of Xyl on the C-2 position prevents GalT-I activity, suggesting that this modification may occur once Gal is attached to the Xyl residue of the nascent oligosaccharide linkage. On the other hand, we showed that sulfation on C-6 position of Gal1 of the Galβ1–3Gal analog markedly enhanced GlcAT-I catalytic efficiency and we demonstrated the importance of Trp243 and Lys317 residues of Gal1 binding site for enzyme activity. In contrast, we found that GlcAT-I was unable to use digalactosides as acceptor substrates when Gal1 was sulfated on C-4 position or when Gal2 was sulfated on both C-4 and C-6 positions. Altogether, we demonstrated that oligosaccharide modifications of the linkage region control the specificity of the glycosyltransferases, a process that may regulate maturation and processing of glycosaminoglycan chains.

Identification of aspartic acid and histidine residues mediating the reaction mechanism and the substrate specificity of the human UDP-glucuronosyltransferases 1A

The human UDP-glucuronosyltransferase UGT1A6 is the primary phenol-metabolizing UDP-glucuronosyltransferase isoform. It catalyzes the nucleophilic attack of phenolic xenobiotics on UDP-glucuronic acid, leading to the formation of water-soluble glucuronides. The catalytic mechanism proposed for this reaction is an acid-base mechanism that involves an aspartic/glutamic acid and/or histidine residue. Here, we investigated the role of 14 highly conserved aspartic/glutamic acid residues over the entire sequence of human UGT1A6 by site-directed mutagenesis. We showed that except for aspartic residues Asp-150 and Asp-488, the substitution of carboxylic residues by alanine led to active mutants but with decreased enzyme activity and lower affinity for acceptor and/or donor substrate. Further analysis including mutation of the corresponding residue in other UGT1A isoforms suggests that Asp-150 plays a major catalytic role. In this report we also identified a single active site residue important for glucuronidation of phenols and carboxylic acid substrates by UGT1A enzyme family. Replacing Pro-40 of UGT1A4 by histidine expanded the glucuronidation activity of the enzyme to phenolic and carboxylic compounds, therefore, leading to UGT1A3-type isoform in terms of substrate specificity. Conversely, when His-40 residue of UGT1A3 was replaced with proline, the substrate specificity shifted toward that of UGT1A4 with loss of glucuronidation of phenolic substrates. Furthermore, mutation of His-39 residue of UGT1A1 (His-40 in UGT1A4) to proline led to loss of glucuronidation of phenols but not of estrogens. This study provides a step forward to better understand the glucuronidation mechanism and substrate recognition, which is invaluable for a better prediction of drug metabolism and toxicity in human.

Sphingobium sp. SYK-6 LigG involved in lignin degradation is structurally and biochemically related to the glutathione transferase omega class

SpLigG and SpLigG bind by X‐ray crystallography (View interaction).

Identification of Key Functional Residues in the Active Site of Human β1,4-Galactosyltransferase 7 A MAJOR ENZYME IN THE GLYCOSAMINOGLYCAN SYNTHESIS PATHWAY

Glycosaminoglycans (GAGs) play a central role in many pathophysiological events, and exogenous xyloside substrates of β1,4-galactosyltransferase 7 (β4GalT7), a major enzyme of GAG biosynthesis, have interesting biomedical applications. To predict functional peptide regions important for substrate binding and activity of human β4GalT7, we conducted a phylogenetic analysis of the β1,4-galactosyltransferase family and generated a molecular model using the x-ray structure of Drosophila β4GalT7-UDP as template. Two evolutionary conserved motifs, (163)DVD(165) and (221)FWGWGREDDE(230), are central in the organization of the enzyme active site. This model was challenged by systematic engineering of point mutations, combined with in vitro and ex vivo functional assays. Investigation of the kinetic properties of purified recombinant wild-type β4GalT7 and selected mutants identified Trp(224) as a key residue governing both donor and acceptor substrate binding. Our results also suggested the involvement of the canonical carboxylate residue Asp(228) acting as general base in the reaction catalyzed by human β4GalT7. Importantly, ex vivo functional tests demonstrated that regulation of GAG synthesis is highly responsive to modification of these key active site amino acids. Interestingly, engineering mutants at position 224 allowed us to modify the affinity and to modulate the specificity of human β4GalT7 toward UDP-sugars and xyloside acceptors. Furthermore, the W224H mutant was able to sustain decorin GAG chain substitution but not GAG synthesis from exogenously added xyloside. Altogether, this study provides novel insight into human β4GalT7 active site functional domains, allowing manipulation of this enzyme critical for the regulation of GAG synthesis. A better understanding of the mechanism underlying GAG assembly paves the way toward GAG-based therapeutics.Glycosaminoglycans (GAGs) play a central role in many pathophysiological events, and exogenous xyloside substrates of β1,4-galactosyltransferase 7 (β4GalT7), a major enzyme of GAG biosynthesis, have interesting biomedical applications. To predict functional peptide regions important for substrate binding and activity of human β4GalT7, we conducted a phylogenetic analysis of the β1,4-galactosyltransferase family and generated a molecular model using the x-ray structure of Drosophila β4GalT7-UDP as template. Two evolutionary conserved motifs, 163DVD165 and 221FWGWGREDDE230, are central in the organization of the enzyme active site. This model was challenged by systematic engineering of point mutations, combined with in vitro and ex vivo functional assays. Investigation of the kinetic properties of purified recombinant wild-type β4GalT7 and selected mutants identified Trp224 as a key residue governing both donor and acceptor substrate binding. Our results also suggested the involvement of the canonical carboxylate residue Asp228 acting as general base in the reaction catalyzed by human β4GalT7. Importantly, ex vivo functional tests demonstrated that regulation of GAG synthesis is highly responsive to modification of these key active site amino acids. Interestingly, engineering mutants at position 224 allowed us to modify the affinity and to modulate the specificity of human β4GalT7 toward UDP-sugars and xyloside acceptors. Furthermore, the W224H mutant was able to sustain decorin GAG chain substitution but not GAG synthesis from exogenously added xyloside. Altogether, this study provides novel insight into human β4GalT7 active site functional domains, allowing manipulation of this enzyme critical for the regulation of GAG synthesis. A better understanding of the mechanism underlying GAG assembly paves the way toward GAG-based therapeutics.

Molecular characterization of β1,4-galactosyltransferase 7 genetic mutations linked to the progeroid form of Ehlers–Danlos syndrome (EDS)

β1,4‐Galactosyltransferase 7 (β4GalT7) is a key enzyme initiating glycosaminoglycan (GAG) synthesis. Based on in vitro and ex vivo kinetics studies and structure‐based modelling, we molecularly characterized β4GalT7 mutants linked to the progeroid form of Ehlers–Danlos syndrome (EDS), a severe connective tissue disorder. Our results revealed that loss of activity upon L206P substitution due to altered protein folding is the primary cause for the GAG synthesis defect in patients carrying the compound A186D and L206P mutations. We showed that R270C substitution strongly reduced β4GalT7 affinity towards xyloside acceptor, thus affecting GAG chains formation. This study establishes the molecular basis for β4GalT7 defects associated with altered GAG synthesis in EDS.

Phylogenetic and mutational analyses reveal key residues for UDP‐glucuronic acid binding and activity of β1,3‐glucuronosyltransferase I (GlcAT‐I)

The β1,3‐glucuronosyltransferases are responsible for the completion of the protein–glycosaminoglycan linkage region of proteoglycans and of the HNK1 epitope of glycoproteins and glycolipids by transferring glucuronic acid from UDP‐α‐D‐glucuronic acid (UDP‐GlcA) onto a terminal galactose residue. Here, we develop phylogenetic and mutational approaches to identify critical residues involved in UDP‐GlcA binding and enzyme activity of the human β1,3‐glucuronosyltransferase I (GlcAT‐I), which plays a key role in glycosaminoglycan biosynthesis. Phylogeny analysis identified 119 related β1,3‐glucuronosyltransferase sequences in vertebrates, invertebrates, and plants that contain eight conserved peptide motifs with 15 highly conserved amino acids. Sequence homology and structural information suggest that Y84, D113, R156, R161, and R310 residues belong to the UDP‐GlcA binding site. The importance of these residues is assessed by site‐directed mutagenesis, UDP affinity and kinetic analyses. Our data show that uridine binding is primarily governed by stacking interactions with the phenyl group of Y84 and also involves interactions with aspartate 113. Furthermore, we found that R156 is critical for enzyme activity but not for UDP binding, whereas R310 appears less important with regard to both activity and UDP interactions. These results clearly discriminate the function of these two active site residues that were predicted to interact with the pyrophosphate group of UDP‐GlcA. Finally, mutation of R161 severely compromises GlcAT‐I activity, emphasizing the major contribution of this invariant residue. Altogether, this phylogenetic approach sustained by biochemical analyses affords new insight into the organization of the β1,3‐glucuronosyltransferase family and distinguishes the respective importance of conserved residues in UDP‐GlcA binding and activity of GlcAT‐I.

Spatial order as a source of kinetic cooperativity in organized bound enzyme systems.

When enzyme molecules are distributed within a negatively charged matrix, the kinetics of the conversion of a negatively charged substrate into a product depends on the organization of fixed charges and bound enzyme molecules. Organization is taken to mean the existence of macroscopic heterogeneity in the distribution of fixed charge density, or of bound enzyme density, or of both. The degree of organization is quantitatively expressed by the monovariate moments of charge and enzyme distributions as well as by the bivariate moments of these two distributions. The overall reaction rate of the bound enzyme system may be expressed in terms of the monovariate moments of the charge density and of the bivariate moments of charge and enzyme densities. The monovariate moments of enzyme density do not affect the reaction rate. With respect to the situation where the fixed charges and enzyme molecules are randomly distributed in the matrix, the molecular organization, as expressed by these two types of moments, generates an increase or decrease of the overall reaction rate as well as a cooperativity of the kinetic response of the system. Thus both the alteration of the rate and the modulation of cooperativity are the consequence of a spatial organization of charges with respect to the enzyme molecules. The rate equations have been derived for different types of organization of fixed charges and enzyme molecules, namely, clustered charges and homogeneously distributed enzyme molecules, clustered enzyme molecules and homogeneously distributed charges, clusters of charges and clusters of enzymes that partly overlap, and clusters of enzymes and clusters of charges that are exactly superimposed. Computer simulations of these equations show how spatial molecular organization may modulate the overall reaction rate.

Dynamics of an open metabolic cycle at the surface of a charged membrane. I.: Multiple steady states and oscillatory behavior generated by electric repulsion effects

Abstract If an open metabolic cycle, made up with two antagonistic enzyme reactions, takes place at the surface of a charged membrane, the electric partitioning of the two charged reaction intermediates may generate the existence of three steady states. Two of these steady states are stable and another one is unstable. This means that the steady-state concentration of the reaction intermediate, S1, may display a hysteresis loop when plotted versus the input rate of matter in the system. Similarly there exists a hysteresis loop of the electric potential at the surface of the membrane when the input rate of matter is varied. This type of dynamic behavior implies that the open multi-enzyme system is able to sense not only the intensity of a chemical signal (the input rate) but also the direction of variation of this signal, that is, whether this signal is being increased, or decreased. Another remarkable property of electric partitioning of charged reaction intermediates is that when increasing the substrate concentration, the local pH in the membrane rises. If the reaction rate of the bound enzyme decreases under the these conditions, the overall system may display sustained oscillations and limit cycle of the two reaction intermediates, as well as of the electric partition coefficient. The existence of sustained oscillations at the surface of the membrane may be viewed as a device which increases the efficiency of the metabolic cycle. These remarkable nonlinear properties of the open enzyme network are not properties of the enzyme network but of the partitioning of the reaction intermediates in the network.

Dynamics of an open metabolic cycle at the surface of a charged membrane. I.: a simple general model

Abstract The dynamics of an open metabolic cycle, composed of two antagonistic enzyme reactions taking place at the surface of a charged membrane, has been investigated. The corresponding general dynamic equations may be written in terms of individual enzyme reaction rates, electric partition and sensitivity coefficients. These sensitivity coefficients express how a local concentration of a charged reaction intermediate in the matrix varies as a response to a variation of a bulk ligand concentration. These coefficients may be expressed in terms of the electric partition coefficient and must follow summation relationships, exactly as the parameters of the metabolic control theory do. The dynamic behavior of the bound multi-enzyme system close to a steady state is defined by the trace and the determinant of the Jacobian matrix of the corresponding variational system. In these expressions, the part played by the electric partition coefficient has been determined regardless the expression of the equations of the individual reaction rates. The electric partitioning of the charged reaction intermediates may alter so profoundly the expression of the trace and of the determinant of this Jacobian matrix that the local stability of the dynamic system may be qualitatively changed. Electric partitioning of the charged reaction intermediates of an open metabolic cycle at the surface of a charged membrane may thus generate dynamic properties that are novel relative to the properties of the metabolic network. The nature of the dynamic properties that may emerge from this electric partitioning is discussed in the companion paper.

Dynamics aspects of long distance functional interactions between membrane-bound enzymes.

The aim of this mini review is to study how an organized charged milieu, such as a membrane, may alter functional long-distance interactions between bound enzymes. Two questions are more specifically considered. The first is to know whether the overall response of a bound enzyme is dependent upon the degree of spatial order of fixed charges and enzymes molecules. The second is to determine whether electric interaction between the fixed charges of the matrix and the charged substrate may generate hysteresis loop of substrate concentration as well as oscillations of this concentration at the surface of the membranes. These effects that have been shown to occur at the surface of membranes, are not the result of intrinsic properties of enzymes. They appear as the consequence of the interplay between functional long-distance interactions between bound enzyme systems and electric repulsion effects of mobile ions. They may be viewed as supramolecular devices that allow storing information from the external milieu.