Glucuronidation

Abstract

• Glucuronidation is a major detoxication process catalyzed by uridine diphosphate glucuronosyltransferases.

• The amount of enzyme can be modulated by numerous foreign compounds, such as common chemical inducers already implicated in the induction of other detoxication enzymes.

• Hormones such as thyroid hormones or growth hormone also are implicated in the control of glucuronidation.

• Because glucuronidation enzymes (isozymes) are anchored in the endoplasmic reticulum membrane, with their active site likely being located on the lumenal side of the membrane, the membrane environment of these enzymes was shown to modulate their functional state as evaluated by the conjugating activity per enzymatic molecular unit.

• In accord with a first, previously proposed model, it seems that this modulation can be attributed to different conformational states of the enzymes, depending on the physicochemical state of the membrane.

• In accord with a second model, the membrane may act as a barrier between the enzymes and the cosubstrate UDP-glucuronic acid, which is a polar and charged molecule synthesized in the cytosol. This would imply a transporting process for this molecule through the reticulum membrane, which has been characterized in vitro and could be of importance in vivo.

• Glucuronidation is under the control of a dual regulation, by means of a specific isozyme expression level and by the modulation of their functional state.

Introduction

Glucuronidation is a major detoxication process in all vertebrates (Dutton, 1980). Glucuronide formation is catalyzed by uridine diphosphate glucuronosyltransferases [EC 2.4.1.17] (UDPGTs), which convert a vast array of endogenous compounds such as bilirubin, steroid and thyroid hormones, retinoids or bile acids, as well as a remarkable number of different lipophilic xenobiotics, into more hydrophilic and thereby more readily excretable metabolites [for reviews, see Burchell and Coughtrie (1989), Tephly and Burchell (1990), Miners and Mackenzie (1991), Mulder (1992), and Bock (1994). But glucuronidation is not only a detoxication pathway, because it can be invoked in the metabolic activation of xenobiotics such as morphine or carcinogens [for reviews, see Kroemer and Klotz (1992) and Mulder (1992)]. Besides, some resulting products of glucuronidation are also secondarily implicated in the formation of protein adducts after treatment with nonsteroid anti-inflammatory drugs, as described by Smith et al. (1986).

All the UDPGT isoforms are members of the UDP-glycosyltransferase (UGT) superfamily of enzymes, which catalyze the addition of the glycosyl group of a nucleotide sugar to an aglycone. This superfamily is widely found in plants, animals and bacteria, indicating that these enzymes might have evolved from an ancestral gene before divergence of prokaryotes and eukaryotes, as is the case for the cytochrome P450 superfamily (Mackenzie et al., 1997). All these enzymes contain a characteristic “signature sequence” probably implicated in the binding of the UDP moiety of the nucleotide sugar (Mackenzie et al., 1997).

UDPGTs constitute the most studied superfamily of UGTs. These enzymes require uridine diphosphate glucuronic acid (UDPGA) as the donor substrate and are located principally in the endoplasmic reticulum of cells from a number of tissues and organs, though the greatest activity is usually found in the liver (Dutton, 1980); see also the aforecited reviews on glucuronidation. This superfamily is divided into two families named UGT1 and UGT2 that have less than 50% amino acid identity in common. UGT2 is divided again into two subfamilies, UGT2A and UGT2B, as described by Burchell et al. (1991) and Mackenzie et al. (1997). The structure of the UGT1 gene is highly interesting because it contains several promoters/first exons encoding isoform-specific sequences, followed by a set of common exons (2 through 5) encoding the C-terminal sequence that is identical in all UGT1 isoforms Emi et al. 1995, Ritter et al. 1992, Sato et al. 1990. This UGT1 gene complex, made up of various first exons undergoing differential splicing to the four common exons, necessarily shows a complex regulation of its expression. UGT1 isozymes are implicated in the glucuronidation of bilirubin and small planar phenol compounds such as 4-nitrophenol. Because bilirubin is an important substrate for glucuronidation, mutations in this UGT1 gene complex can lead to more or less severe unconjugated hyperbilirubinemias, owing to a partial-to-complete defect in bilirubin conjugation and excretion, such as in human Crigler-Najjar and Gilbert’s syndromes Arias et al. 1969, Koiwai et al. 1995 or in the Gunn rat (Iyanagi, 1991); for a review see Burchell and Coughtrie (1989). UGT2 isozymes are responsible for the glucuronidation of steroids, bile acids, retinoids and numerous xenobiotics [see the aforecited reviews and Mackenzie et al. (1992)].

UDPGT enzymes show a varying and often overlapping substrate specificity. The different isoforms are differentially regulated by foreign chemicals and hormones and differ in their developmental expression. Factors such as strain or ethnicity, age and diet were also shown to modify their expression (Miners and Mackenzie, 1991).

One of the most striking feature of UDPGTs is that those enzymes are anchored in the membrane by a transmembrane part of their C-terminal region, with the majority of the protein and probably the active site located on the lumenal side of the endoplasmic reticulum. Such a model of UDPGT topology was presented by Tephly and Burchell (1990). This topological location confers to UDPGTs a dependence on the membrane environment, which is displayed by the phenomenon of latency; that is, a constrained catalytic potential when UDPGTs are assayed in native microsomal preparations, as is the case for other membrane-bound enzymes such as glucose-6-phosphatase (Tephly and Burchell, 1990). This latency is released in vitro by the use of membrane perturbants such as detergents or sonication. The membrane dependence of UDPGTs is not yet fully understood and gave rise, more than 20 years ago, to two controversial hypotheses involving either a variation of the conformation of the enzyme as a function of its environment or a compartmentation between the enzyme and the hydrophilic UDPGA, the endoplasmic reticulum being considered a membrane barrier in the latter hypothesis Berry et al. 1975, Zakim and Dannenberg 1992.

Glucuronidation can then be considered to be under the control of a dual regulation. The first control, designated as conventional, is carried out on the expression levels of the isoenzymes and determines an amount of the different isoforms, whereas the second one is exerted, in a more general manner, on the modulation of the activity of the isoenzymes relatively to their “functional state.”