The effect of N-linked glycosylation on the substrate preferences of UDP glucuronosyltransferases

Use of a Baculovirus-Mammalian Cell Expression-System for Expression of Drug-Metabolizing Enzymes: Optimization of Infection With a Focus on Cytochrome P450 3A4

Heterologous expression systems are important for analyzing the effects of genetic factors including single nucleotide polymorphisms on the functions of drug-metabolizing enzymes. In this study, we focused on a baculovirus-mammalian cell (Bac-Mam) expression system as a safer and more efficient approach for this purpose. The baculovirus-insect cell expression system is widely utilized in large-scale protein expression. Baculovirus has been shown to also infect certain mammalian cells, although the virus only replicates in insect cells. With this knowledge, baculovirus is now being applied in a mammalian expression system called the Bac-Mam system wherein a gene-modified baculovirus is used whose promotor is replaced with one that can function in mammalian cells. We subcloned open-reading frames of cytochrome P450 3A4 (CYP3A4), UDP-glucuronosyltransferase (UGT) 1A1, and UGT2B7 into a transfer plasmid for the Bac-Mam system, and prepared recombinant Bac-Mam virus. The obtained virus was amplified in insect Sf9 cells and used to infect mammalian COS-1 cells. Expression of CYP3A4, UGT1A1, and UGT2B7 in COS-1 cell homogenates were confirmed by immunoblotting. Optimum infection conditions including the amount of Bac-Mam virus, culture days before collection, and concentration of sodium butyrate, an enhancer of viral-transduction were determined by monitoring CYP3A4 expression. Expressed CYP3A4 showed appropriate activity without supplying hemin/5-aminolevulinic acid or co-expressing with NADPH-cytochrome P450 reductase. Further, we compared gene transfer efficiency between the Bac-Mam system and an established method using recombinant plasmid and transfection reagent. Our results indicate that the Bac-Mam system can be applied to introduce drug-metabolizing enzyme genes into mammalian cells that are widely used in drug metabolism research. The expressed enzymes are expected to undergo appropriate post-translational modification as they are in mammalian bodies. The Bac-Mam system may thus accelerate pharmacogenetics and pharmacogenomics research.

Functional Interaction between Cytochrome P450 and UDP-Glucuronosyltransferase on the Endoplasmic Reticulum Membrane: One of Post-translational Factors Which Possibly Contributes to Their Inter-Individual Differences

Cytochrome P450 (P450) and uridine 5'-diphosphate (UDP)-glucuronosyltransferase (UGT) catalyze oxidation and glucuronidation in drug metabolism, respectively. It is believed that P450 and UGT work separately because they perform distinct reactions and exhibit opposite membrane topologies on the endoplasmic reticulum (ER). However, given that some chemicals are sequentially metabolized by P450 and UGT, it is reasonable to consider that the enzymes may interact and work cooperatively. Previous research by our team detected protein-protein interactions between P450 and UGT by analyzing solubilized rat liver microsomes with P450-immobilized affinity column chromatography. Although P450 and UGT have been known to form homo- and hetero-oligomers, this is the first report indicating a P450-UGT association. Based on our previous study, we focused on the P450-UGT interaction and reported lines of evidence that the P450-UGT association is a functional protein-protein interaction that can alter the enzymatic capabilities, including enhancement or suppression of the activities of P450 and UGT, helping UGT to acquire novel regioselectivity, and inhibiting substrate binding to P450. Biochemical and molecular bioscientific approaches suggested that P450 and UGT interact with each other at their internal hydrophobic domains in the ER membrane. Furthermore, several in vivo studies have reported the presence of a functional P450-UGT association under physiological conditions. The P450-UGT interaction is expected to function as a novel post-translational factor for inter-individual differences in the drug-metabolizing enzymes.

How glycosylation affects glycosylation: the role of N-glycans in glycosyltransferase activity

N-glycosylation is one of the most important posttranslational modifications of proteins. It plays important roles in the biogenesis and functions of proteins by influencing their folding, intracellular localization, stability and solubility. N-glycans are synthesized by glycosyltransferases, a complex group of ubiquitous enzymes that occur in most kingdoms of life. A growing body of evidence shows that N-glycans may influence processing and functions of glycosyltransferases, including their secretion, stability and substrate/acceptor affinity. Changes in these properties may have a profound impact on glycosyltransferase activity. Indeed, some glycosyltransferases have to be glycosylated themselves for full activity. N-glycans and glycosyltransferases play roles in the pathogenesis of many diseases (including cancers), so studies on glycosyltransferases may contribute to the development of new therapy methods and novel glycoengineered enzymes with improved properties. In this review, we focus on the role of N-glycosylation in the activity of glycosyltransferases and attempt to summarize all available data about this phenomenon.

The UGTome: The expanding diversity of UDP glycosyltransferases and its impact on small molecule metabolism

The UDP glycosyltransferase (UGT) superfamily of enzymes is responsible for the metabolism and clearance of thousands of lipophilic chemicals including drugs, toxins and endogenous signaling molecules. They provide a protective interface between the organism and its chemical-rich environment, as well as controlling critical signaling pathways to maintain healthy tissue function. UGTs are associated with drug responses and interactions, as well as a wide range of diseases including cancer. The human genome contains 22 UGT genes; however as befitting their exceptionally diverse substrate ranges and biological activities, the output of these UGT genes is functionally diversified by multiple processes including alternative splicing, post-translational modification, homo- and hetero-oligomerization, and interactions with other proteins. All UGT genes are subject to extensive alternative splicing generating variant/truncated UGT proteins with altered functions including the capacity to dominantly modulate/inhibit cognate full-length forms. Heterotypic oligomerization of different UGTs can alter kinetic properties relative to monotypic complexes, and potentially produce novel substrate specificities. Moreover, the recently profiled interactions of UGTs with non-UGT proteins may facilitate coordination between different metabolic processes, as well as providing opportunities for UGTs to engage in novel 'moonlighting' functions. Herein we provide a detailed and comprehensive review of all known modes of UGT functional diversification and propose a UGTome model to describe the resulting expansion of metabolic capacity and its potential to modulate drug/xenobiotic responses and cell behaviours in normal and disease contexts.

Characterization of human UGT2A3 expression using a prepared specific antibody against UGT2A3

UDP-Glucuronosyltransferase (UGT) 2A3 belongs to a UGT superfamily of phase II drug-metabolizing enzymes that catalyzes the glucuronidation of many endobiotics and xenobiotics. Previous studies have demonstrated that UGT2A3 is expressed in the human liver, small intestine, and kidney at the mRNA level; however, its protein expression has not been determined. Evaluation of the protein expression of UGT2A3 would be useful to determine its role at the tissue level. In this study, we prepared a specific antibody against human UGT2A3 and evaluated the relative expression of UGT2A3 in the human liver, small intestine, and kidney. Western blot analysis indicated that this antibody is specific to UGT2A3 because it did not cross-react with other human UGT isoforms or rodent UGTs. UGT2A3 expression in the human small intestine was higher than that in the liver and kidney. Via treatment with endoglycosidase, it was clearly demonstrated that UGT2A3 was N-glycosylated. UGT2A3 protein levels were significantly correlated with UGT2A3 mRNA levels in a panel of 28 human liver samples (r = 0.64, p < 0.001). In conclusion, we successfully prepared a specific antibody against UGT2A3. This antibody would be useful to evaluate the physiological, pharmacological, and toxicological roles of UGT2A3 in human tissues.

Introduction of an N-Glycosylation Site into UDP-Glucuronosyltransferase 2B3 Alters Its Sensitivity to Cytochrome P450 3A1-Dependent Modulation

Our previous studies have demonstrated functional protein-protein interactions between cytochrome P450 (CYP) 3A and UDP-glucuronosyltransferase (UGT). However, the role of carbohydrate chains of UGTs in the interaction with CYP is not well understood. To address this issue, we examined whether CYP3A1 modulates the function of UGT2B3 which lacks potential glycosylation sites. We also examined whether the introduction of N-glycosylation to UGT2B3 affects CYP3A-dependent modulation of UGT function. To introduce a potential glycosylation site into UGT2B3, Ser 316 of UGT2B3 was substituted with Asn by site-directed mutagenesis. A baculovirus-Sf-9 cell system for expressing CYP3A1 and UGT2B3/UGT2B3(S316N) was established using a Bac-to-Bac system. Glycosylation of UGT2B3(S316N) was demonstrated in this expression system. The microsomal activity of recombinant UGT was determined using 4-methylumbelliferone as a substrate. The effect of CYP3A1 co-expression on UGT function was examined by comparing the kinetic profiles between single (UGT alone) and double expression (UGT plus CYP) systems. The kinetics of the two expression systems fitted a Michaelis-Menten equation. When the 4-MU concentration was varied, co-expression of CYP3A1 lowered the V_max of UGT2B3-mediated conjugation. Conversely, for UGT2B3(S316N), the V_max in the dual expression system was higher than that in the single expression system. The data obtained demonstrate that the introduction of N-glycosylation to UGT2B3 alters its sensitivity to CYP3A1-dependent modulation while CYP3A1 enhanced UGT2B3(S316N) activity, and wild-type UGT2B3 was suppressed by CYP3A1. These data suggest that N-glycosylation of UGT is one of the determinants regulating the interaction between CYP3A and UGT.

Suppression of Cytochrome P450 3A4 Function by UDP-Glucuronosyltransferase 2B7 through a Protein-Protein Interaction: Cooperative Roles of the Cytosolic Carboxyl-Terminal Domain and the Luminal Anchoring Region

There is a large discrepancy between the interindividual difference in the hepatic expression level of cytochrome P450 3A4 (CYP3A4) and that of drug clearance mediated by this enzyme. However, the reason for this discrepancy remains largely unknown. Because CYP3A4 interacts with UDP-glucuronosyltransferase 2B7 (UGT2B7) to alter its function, the reverse regulation is expected to modulate CYP3A4-catalyzed activity. To address this issue, we investigated whether protein-protein interaction between CYP3A4 and UGT2B7 modulates CYP3A4 function. For this purpose, we coexpressed CYP3A4, NADPH-cytochrome P450 reductase, and UGT2B7 using a baculovirus-insect cell system. The activity of CYP3A4 was significantly suppressed by coexpressing UGT2B7, and this suppressive effect was lost when UGT2B7 was replaced with calnexin (CNX). These results strongly suggest that UGT2B7 negatively regulates CYP3A4 activity through a protein-protein interaction. To identify the UGT2B7 domain associated with CYP3A4 suppression we generated 12 mutants including chimeras with CNX. Mutations introduced into the UGT2B7 carboxyl-terminal transmembrane helix caused a loss of the suppressive effect on CYP3A4. Thus, this hydrophobic region is necessary for the suppression of CYP3A4 activity. Replacement of the hydrophilic end of UGT2B7 with that of CNX produced a similar suppressive effect as the native enzyme. The data using chimeric protein demonstrated that the internal membrane-anchoring region of UGT2B7 is also needed for the association with CYP3A4. These data suggest that 1) UGT2B7 suppresses CYP3A4 function, and 2) both hydrophobic domains located near the C terminus and within UGT2B7 are needed for interaction with CYP3A4.

In vitro–in vivo correlation for drugs and other compounds eliminated by glucuronidation in humans: Pitfalls and promises

Enzymes of the UDP-glucuronosyltransferase (UGT) superfamily are responsible for the metabolism of many drugs, environmental chemicals and endogenous compounds. Identification of the UGT(s) involved in the metabolism of a given compound ('reaction phenotyping') currently relies on multiple confirmatory approaches, which may be confounded by the dependence of UGT activity on enzyme source, incubation conditions, and the occurrence of atypical glucuronidation kinetics. However, the increasing availability of substrate and inhibitor 'probes' for the individual UGTs provides the prospect for reliable phenotyping of glucuronidation reactions using human liver microsomes or hepatocytes, thereby providing data directly relevant to drug metabolism in humans. While the feasibility of computational prediction of UGT substrate selectivity has been demonstrated, the development of easily interpretable and generalisable models requires further improvement in the datasets available for analysis. Quantitative prediction of the hepatic clearance of glucuronidated drugs and the magnitude of inhibitory interactions based on in vitro kinetic data is more problematic. Intrinsic clearance (CL(int)) values generated using human liver microsomes under-predict in vivo hepatic clearance, typically by an order of magnitude. In vivo clearances of glucuronidated drugs are also generally under-predicted by CL(int) values from human hepatocytes, but to a lesser extent than observed with the microsomal model. While it is anticipated that systematic analysis of the potential causes of under-prediction may provide more reliable in vitro-in vivo scaling strategies, mechanistic interpretation of in vitro-in vivo correlation more broadly awaits further advances in our understanding of the structural and cellular determinants of UGT activity.

The cynomolgus monkey (Macaca fascicularis) is the best animal model for the study of steroid glucuronidation

Intense research efforts performed during the past decade clearly established the major role of glucuronidation and uridine-diphospho-glucuronosyltransferase (UGT) enzymes for steroid metabolism in humans. However, a clear understanding of the physiological importance of this metabolic process requires in vivo studies. Numerous evidences ascertain that simians are the most appropriate animal models for such studies. Indeed human and monkey have a similar pattern of steroidogenesis, unlike common laboratory mammals such as rat or mouse. Furthermore, human and monkey are unique in having high levels of circulating androsterone glucuronide and androstane-3alpha-diol glucuronide (3alpha-Diol-G). In addition, characterization of eight monkey UGT proteins demonstrated the similarity of their conjugation activity toward steroid hormones. Like human ones, monkey enzymes are expressed in steroid target tissues, where they preferentially glucuronidate androgen and estrogen metabolites. In monkey tissues, immunohistochemical studies demonstrated that UGT2B proteins are expressed in a cell-type specific manner in ovary and kidney, where they control androgens and aldosterone inactivation. These results identify the cynomolgus monkey as an appropriate animal model for the determination of cellular localization of UGT enzymes in steroid target tissues and for the identification of endogenous or exogenous stimuli affecting steroid glucuronidation.

Structure and function of uridine diphosphate glucuronosyltransferases

1. The uridine diphosphate (UDP)-glucuronosyltransferases (UGT) are a family of enzymes that catalyse the covalent addition of glucuronic acid to a wide range of lipophilic chemicals. They play a major role in the detoxification of many exogenous and endogenous compounds by generating products that are more polar and, thus, more readily excreted in bile or urine. 2. Inherited deficiencies in UGT forms are deleterious, as exemplified by the debilitating effects of hyperbilirubinaemia and neurotoxicity in subjects with mutations in the enzyme that converts bilirubin to its more polar glucuronide. 3. The UGT protein can be conceptually divided into two domains with the amino-terminal half of the protein demonstrating greater sequence divergence between isoforms. This region apparently determines aglycone specificity. The aglycone binding site is presumed to be a 'loose' fit, as many structurally diverse substrates can be bound by the same UGT isoform. The carboxyl-terminal half, which is more conserved in sequence between different isoforms, is believed to contain a binding site for the cosubstrate UDP glucuronic acid (UDPGA). 4. Uridine diphosphate glucuronosyltransferase is localized to the endoplasmic reticulum (ER) and spans the membrane with a type I topology. The putative transmembrane domain is located near the carboxyl terminus of the protein such that only a small portion of the protein resides in the cytosol. This cytosolic tail is believed to contain an ER-targeting signal. The major portion of the protein is located in the ER lumen, including the proposed substrate-binding domains and the catalytic site. 5. The microsomal membrane impedes the access of UDPGA to the active site, resulting in latency of UGT activity in intact ER-derived microsomes. Active transport of UDPGA is believed to occur in hepatocytes, but the transport system has not been fully characterized. Uridine diphosphate glucuronosyltransferase activity is also highly lipid dependent and the enzyme may contain regions of membrane association in addition to the transmembrane domain.

Glycosylation-dependent activity of baculovirus-expressed human liver carboxylesterases: cDNA cloning and characterization of two highly similar enzyme forms

A cDNA, designated hCE, encoding the entire sequence of a carboxylesterase, was isolated from a human liver lambda gt11 library. The hCE-deduced protein sequence contained 568 amino acids, including an 18 amino acid signal peptide sequence, and had a calculated molecular mass of the mature protein of 60,609 Da. A second cDNA, designated hCEv, was isolated from the same lambda gt11 library and contained a 3-bp deletion resulting in the loss of the final amino acid in the signal peptide sequence (Ala-1) and a second 3-bp deletion leading to an in-frame loss of Gln345. Expression of mRNA corresponding to both hCE and hCEv was detected in eight adult human liver samples, with individual levels varying 5-fold (hCE) and 12-fold (hCEv). A single immunoreactive protein was detected in 13 adult human liver samples when probed with antibody directed against a rat carboxylesterase. Based on allele-specific oligonucleotide hybridizations, we believe that the hCE and hCEv cDNAs represent two distinct members of the carboxylesterase family. The carboxylesterase genes were localized to human chromosome 16 using a somatic cell hybrid mapping strategy. Baculovirus expression of hCE in Sf9 cells produced a protein with an estimated molecular mass of 59,000 Da. This enzyme was able to hydrolyze aromatic and aliphatic esters but possessed no catalytic activity toward amides or a fatty acyl CoA ester. Baculovirus-mediated expression of the hCEv cDNA yielded a second protein of 56,000 Da resulting from inefficient N-glycosylation of the hCEv protein. Although the substrate specificity for the hCEv protein was identical to that of expressed hCE for any given substrate, the specific activity for the hCE protein was always higher than that for the hCEv protein. Tunicamycin inhibition studies provided the first evidence that N-glycosylation of these luminal enzymes is essential for maximal catalytic activity.

Steroid UDP glucuronosyltransferases

The glucuronidation of steroids is a major process necessary for their elimination in the bile and urine. In general, steroid glucuronides are biologically less reactive than their parent steroids. However, in some cases often associated with disease and steroid therapy, more reactive or toxic glucuronides may be formed. The concentrations of specific steroid glucuronides in the blood may also indicate hormonal imbalances and may funnction as diagnostic markers of genetic defects in steroid synthesis and metabolism. In this review, the forms of UDP glucuronosyltransferase involved in steroid glucuronidation are described in terms of their specificities, functional domains and regulation. The available evidence suggests that steroid glucuronidation is mainly carried out by members of the UGT2B subfamily which are encoded by genes containing 6 exons. Members of this subfamily exhibit a regioselectively in their glucuronidation of steroids that is mediated by domains in the amino-terminal half on the protein encoded by exons 1 and 2. Although much of this review will describe studies in the rat, preliminary evidence indicates that a similar situation may exist in humans.