PEDS Advance Access published online on May 23, 2008
Protein Engineering Design and Selection, doi:10.1093/protein/gzn030
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Mutation analysis in UGT1A9 suggests a relationship between substrate and catalytic residues in UDP-glucuronosyltransferases
1Structural Biology and Biophysics, Institute of Biotechnology 2Drug Discovery and Development Technology Center (DDTC) and Division of Pharmaceutical Chemistry, Faculty of Pharmacy 3Neuroscience Center, University of Helsinki, Biocenter 3, PO Box 65, Viikinkaari 1, FIN-00014 Helsinki, Finland
4 To whom correspondence should be addressed. E-mail: adrian.goldman{at}helsinki.fi
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
UDP-glucuronosyltransferases (UGTs) catalyze the transfer of glucuronic acid from UDP-glucuronic acid to endo- and xenobiotics in our body. UGTs belong to the GT1 family of glycosyltransferases and many GT1s use a serine protease-like catalytic mechanism in which an Asp-His pair deprotonates a hydroxyl on the aglycone for nucleophilic attack on the sugar donor. The pair in human UGTs could be H37 and either D143 or D148 (UGT1A9 numbering). However, H37 is not totally conserved, being replaced by either Pro or Leu in UGT1A4 and UGT2B10. We therefore investigated the role of H37, D143 and D148 in UGT1A9 by site-directed mutagenesis, activity and kinetic measurements with several substrates. The results suggest that H37 is not critical in N-glucuronidation, but is so in O-glucuronidation. The Vmax of the H37A mutant was much less affected in N- than O-glucuronidation, while the reverse was true for the Asp mutations, particularly D143A. We suggest that this is due to the opposing properties of O- and N- nucleophiles. O-nucleophiles require the histidine to deprotonate them so that they become effective nucleophiles, while N-nucleophiles develop a formal positive charge during the reaction (RNH2+–GlcA), and thus require a negatively charged residue to stabilize the transition state.
Keywords: enzyme kinetics/general base/glycosyltransferase/UDP Glucuronosyltransferase/UGT
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
UDP-glucuronosyltransferases (UGTs, EC 2.4.1.1 [EC] 7) catalyze the transfer of glucuronic acid from UDP-glucuronic acid (UDPGA) to different xenobiotics and endogenous molecules, including drugs, phenolic compounds, environmental toxins, bile acids, bilirubin and steroid hormones (Radominska-Pandya et al., 1999
; Tukey and Strassburg, 2000; Fisher et al., 2001). This is an essential pathway for eliminating such compounds from the body. The human proteome contains 19 functional UGTs, all about 530 amino acids long. Based on sequence similarities and gene arrangement, they can be divided into two major families, UGT1 and UGT2 (Mackenzie et al., 2005). UGTs are endoplasmic reticulum (ER) membrane proteins and most of their mass is on the lumenal side of the ER membrane. They have a single trans-membrane helix close to the C-terminus and the last 20–25 residues are on the cytoplasmic side of the membrane (Radominska-Pandya et al., 1999). UGTs contain two nearly equal-sized domains: a rather variable N-terminal domain that binds the aglycone and a more conserved C-terminal domain that binds the co-substrate UDPGA (Mackenzie, 1990; Miley et al., 2007).
There is currently only one part-structure of UGTs, a segment of the C-terminal domain of UGT2B7 (Miley et al., 2007); the detailed structure of the N-terminal domain is unknown. UGTs belong to the GT1 family of glycosyltransferases (GT) (Campbell et al., 1997), and their fold is the ‘GT-B’ type (Bourne and Henrissat, 2001). ‘GT-B’ fold proteins, originally found in phage T4 β-glucosyltransferase (Vrielink et al., 1994), have two Rossmann-like domains separated by a linker region. The N-terminal domain binds the aglycone, the sugar acceptor, while the C-terminal domain binds the sugar donor, UDPGA in the case of the UGTs.
The catalytic mechanism of UGTs is still unclear. Ouzzine et al. (Ouzzine et al., 2000, 2003) suggested that a charge relay system between the ‘catalytic His’ and an unidentified Asp/Glu helps deprotonate the aglycone. They proposed that H371 (UGT1A6 numbering) might be the catalytic base. Several of the solved protein structures of GT1 family members, such as flavonoid glycosyltransferase (VvGT) and triterpene/flavonoid glycosyltransferase (UGT71G1) are in general agreement with this proposal, except that the ‘catalytic His’ is in the N-terminal domain (Shao et al., 2005; Offen et al., 2006). The structure of VvGT led Miley et al. (Miley et al., 2007) to suggest that H35 in UGT2B7 (H37 in UGT1A9) could play a catalytic role in UGTs. However, this histidine is not totally conserved among human UGTs; UGT1A4 has proline and UGT2B10 leucine instead. Mutating this histidine to proline essentially abolished 1-naphthol and 4-methylumbelliferone O-glucuronidation, but conferred lamotrigine and trifluoperazine N-glucuronidation in UGT1A3 (Kubota et al., 2007). However, Li et al. (Li et al., 2007a), in a recent extensive study, concluded that the corresponding histidine H38 in human UGT1A6 is rather involved in defining substrate specificity than playing a role of catalyst. In addition, in another member of the GT1 protein family, UDP-Glucose:Isoflavone 7-O-Glucosyltransferase from the roots of soybean seedlings, His-15 and Asp-125, which correspond to the catalytic residues of UGT71G1 and VvGT1, are unimportant for catalytic activity (Noguchi et al., 2007). Finally, a recent structure and mutagenesis study of the bifunctional plant enzyme UGT72B1 found that the corresponding histidine (H19) is required for O-glycosyltransferase, but not N-glycosyltransferase, activity. They proposed that the role of H19 in N-glycosylation is to direct and orientate nucleophile attack (Brazier-Hicks et al., 2007).
To examine the catalytic mechanism in the human UGTs, we have now studied the role of H37 of UGT1A9 in N- and O-glucuronidation reactions. We also investigated two conserved Asp that may be involved in the ‘charge relay’, D143 and D148. Our results show that the H37A mutant is active in N-, but not in O-glucuronidation; N-glucuronidation does not therefore require the catalytic dyad. This is consistent with the data on UGT1A4 and UGT2B10, because both tend to catalyze rather N- than O-glucuronidation reactions, unlike most other human UGTs (Sorich et al., 2006; Kaivosaari et al., 2007).
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Recombinant enzyme preparation
Recombinant control and mutant UGT1A9 were expressed as C-terminal His-tagged fusion proteins in baculovirus-infected insect cells and membranes were isolated and stored as previously described (Kurkela et al., 2003). Point mutations were done by PCR or by the QuikChange procedure (Stratagene, La Jolla, CA, USA), and the sequence of the entire DNA fragment that underwent PCR was checked by DNA sequencing. The relative expression level of each recombinant UGT, either mutant or control, was determined by immuno-detection on dot-blot using the monoclonal anti-His-tag IgG Tetra His (Qiagen, Hilden, Germany) as previously described (Kurkela et al., 2004a; Kurkela et al., 2007). All the activity measurements were normalized by their relative expression levels, as described elsewhere (Kurkela et al., 2004b) (Fig. 1). We determined protein concentration using the BCA (bicinchoninic acid) Protein assay kit (Pierce) with BSA as standard.
|
HPLC activity measurements
Scopoletin (7-hydroxy-6-methoxy-2H-1-benzopyran-2-one), 4-methylumbelliferone (7-hydroxy-4-methyl-2H-1-Benzopyran-2-one), 1-naphthol and 4-aminobiphenyl were purchased from Sigma (Sigma Chemical Co., St Louis, MO, USA), UDPGA triammonium salt from Fluka (Sigma Aldrich, Buchs, Switzerland), while entacapone (2-cyano-3-(5-dihydroxyamino-3, 4-dioxo-1-cyclohexa-1,5-dienyl)-N,N-diethyl-2-zpropenamide) was kindly provided by Orion Pharma (Espoo, Finland) and retigabine by Meda Pharma Gmbh & Co. KG (Radebuel, Germany). Single concentration activity measurements were made in 100 µl reactions, except for retigabine, which was done in 50 µl. Each reaction contained 50 mM phosphate buffer pH 7.4, 10 mM MgCl2, 5 mM saccharolactone and membranes containing 0.1–1.7 mg/ml of control or mutant UGT1A9. The mutants were 9H37A, 9H37D, 9D143A and 9D148A, where the prefix of ‘9’ indicates that this is an UGT1A9 mutant, as in our previous work (Patana et al., 2007). The concentration of aglycone was 10 µM for 1-naphthol, 100 µM for 4-MU and 500 µM for 4-aminobiphenyl, retigabine, entacapone and scopoletin. The reaction was started by adding 5 mM UDPGA and performed in the linear range at 37°C for 20–60 min depending on the substrate used. The reaction was stopped by adding 10 µl of 4 M ice-cold perchloric acid to the reaction mixture, except for 4-aminobiphenyl and retigabine. For these, we used 100 or 50 µl of ice-cold methanol instead because the N-glucuronides formed would have been hydrolyzed on addition of the acid. After this, the samples were centrifuged for 5–10 min at 16 000 g and the cleared supernatant used for HPLC analysis. All the samples were prepared in triplicates. An Agilent model 1100 (Agilent Technologies, Waldbronn, Germany) was used to analyze the glucuronidation products formed. However, retigabine glucuronides are quite labile and so we analyzed them with an Acquity Ultra Performance LC (Waters Corporation, MA, USA), because it is much faster.
For kinetic measurements, we used at least six different substrate concentrations and adjusted the enzyme concentration so that no more than 10% of substrate was consumed during the reaction. Twenty five to 2000 µM 4-aminobiphenyl, 10–500 µM retigabine, 10–2000 µM scopoletin and 2.5–1500 µM 4-MU with 5 mM of UDPGA were used for the aglycone kinetics and 25–5000 µM UDPGA with 2000 µM scopoletin for the UDPGA kinetics.
We obtained kinetic constants for scopoletin, 4-ABP and retigabine by non-linear curve fitting in GraphPad Prism using Michaelis–Menten kinetics. When 4-MU was used as aglycone, substrate concentrations above 250 µM clearly caused substrate inhibition for the control and the D143A mutant (Fig. 2). As we have shown before (Luukkanen et al., 2005), UGTs use a ternary complex bi bi mechanism following compulsory ordered kinetics (Fig. 3), and thus can be inhibited by the second substrate to bind (the aglycone). We, therefore, fit the data for 4-MU to Eq. (1):
|
| (1) |
).
|
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Initial studies and choice of mutants for full kinetic analysis
In this study, we first examined the effects of the four mutants, 9H37A, 9H37D, 9D143A and 9D148A, through single-concentration assays (Table I). Four substrates (entacapone, scopoletin, 4-MU and 1-naphthol) forming O-glucuronides and two substrates (4-aminobiphenyl and retigabine) forming N-glucuronides were selected for the activity studies (Fig. 4). The 9H37A mutant had very low activity (
0.3% of the control) with all four O-substrates, but had surprisingly high relative activity with 4-aminobiphenyl (4-ABP) and retigabine: around 30% of the control (Table I). Even the 9H37D mutant, which was essentially inactive with all other substrates, had 7.4% of the control 4-ABP activity (Table I). The effect of the 9D143A mutation was not as drastic as 9H37A, leading to 0.3–0.7% activity (Table I), while the 9D148A mutant showed the widest variation in residual activity, from 0.03% with entacapone to 2.53% with scopoletin.
|
|
These observations led us to more extensive kinetic analyses. Because of the higher overall activity, we chose 9H37A rather than 9H37D mutant for these studies. The kinetic analyses were done with two hydroxyl and two amide substrates.
Effect of mutation on activity towards UDPGA
The Km for UDPGA was investigated using scopoletin as aglycone substrate. Even though the Vmax was much lower, the mutants did not change the Km for UDPGA greatly (Table II). This is in contrast to the mutations in the C-terminal domain such as H369 and E377, where the UDPGA Km increased sharply (Patana et al., 2007). The Vmax was about 200-fold lower for 9H37A and 9D143A, but only 27-fold lower for 9D148A (Table II).
|
Effect of mutation on activity towards O-substrates
Next we investigated the effect of mutation on activity with scopoletin as substrate. All the activities were much lower than we observed for mutations in the C-terminal domain (Patana et al., 2007) (Tables I and II). Compared to control, there was a 11-fold increase in the Km for scopoletin in 9H37A, and 8-fold increase in 9D148A, but no significant change in 9D143A (Table II). The Vmax for these three mutants was low, but the most inactive mutant was 9H37A, while 9D148A had about 5% residual activity (Table II).
The results with 4-MU were generally similar to those with scopoletin. We could obtain kinetic parameters for the 9D143A and 9D148A mutants (Table II), but not for 9H37A because the activity was below our detection limit (Table I). The Km increased 46-fold for 9D148A but was as control for 9D143A. The Vmax of 9D143A was lower than that of 9D148A (Table II).
The kinetic analyses also reveal an interesting difference in sensitivity to substrate inhibition between the mutants (Table II, Fig. 2). UGT1A9 shows clear substrate inhibition with 4-MU as a substrate (Fig. 2A), so the data can not be fit with a simple Michaelis–Menten model. The data are best fit using Eq. (1), derived from the compulsory ordered bi bi mechanism (Luukkanen et al., 2005) (Fig. 3). In such a model, the aglycone acts as an inhibitor by binding to the enzyme-UDP product complex and thus depleting the active enzyme pool; this is non-competitive inhibition. We saw clear signs of this with control, as expected, but it was even more pronounced with the 9D143A mutant. In this mutant, the Vmax is only 0.4% of control Vmax and so Km is probably close to the value of Kd for substrate binding. The Ki for 4-MU decreased 3-fold, from 3.7 mM in the control to 1.2 mM in 9D143A (Table II, Fig. 2), with a much smaller change in Km. The 9D148A mutant did not show signs of substrate inhibition, presumably because the Km for 4-MU increased 46-fold to 967 µM (Table II), suggesting that 4-MU bound the enzyme poorly.
Effect of mutation on activity towards N-substrates
Using 4-ABP or retigabine as aglycone substrates gave the most interesting results. [Retigabine forms two N-glucuronides (Fig. 4) (Borlak et al., 2006), but the Kms are essentially the same (Table II) and so we summed these two together.) The H37A mutation lowered single-point activity only 3–4-fold for 4-ABP or retigabine, as opposed to over 290-fold for the other substrates (Table I, Fig. 4). This is clearly due to the changes in Vmax: 4–5-fold for 9H37A with N-aglycones and over 200 times lower for the O-aglycone scopoletin. Conversely, the Km value for 4-aminobiphenyl and both Km values for retigabine remained unchanged (Table II) even though the Km value for scopoletin increased 11-fold.
The 9D148A mutant had enough residual activity with 4-ABP for a full kinetic analysis, but the 9D143A mutant was virtually inactive. The Vmax for 9D148A decreased 50-fold compared to the control, i.e. much more than for 9H37A. In contrast, the mutation did not affect the Km value, unlike for O-substrates. In this respect, the 9D148A and 9H37A mutants behaved the same. The results thus suggest that a different catalytic mechanism may be used for N- than for O-glucuronidation (Fig. 5).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Recent evidence (Shao et al., 2005
; Offen et al., 2006; Bolam et al., 2007; Miley et al., 2007) indicates that some GT1 enzymes use a serine-protease-like mechanism with a His-Asp dyad to activate the aglycone. In UGT1A9, H37 would be the histidine that acts as a general base and accepts a proton from the aglycone hydroxyl nucleophile. In such a model, a conserved acidic residue, such as D143 or D148, would stabilize the protonated His through a ‘charge-relay-type’ mechanism (Fig. 5). We have therefore focused, in this study, on three key residues in the N-terminal domain that may be involved in the chemical catalytic mechanism: H37, D143 and D148. Two of these, D143 and D148 are totally conserved among human UGTs, but H37 is not; UGT1A4 has a proline and UGT2B10 a leucine in this important position (Fig. 6). How can H37 be a key catalytic residue if it is not always conserved?
|
The role of His37
Ouzzine et al. (Ouzzine et al., 2000; 2003) proposed that H37 in UGTs has a structural role, but our data clearly show that this is not the case. H37 is, in fact, the general base. The activity of the 9H37A and 9H37D mutants with the four substrates containing hydroxyl groups (Table I, Fig. 4) is very low, and the change in Km upon mutation is also consistent with the idea that His37 is in direct contact with the aglycone (Table II). The single-point assays of Miley et al. (Miley et al., 2007) on UGT2B7 mutants agree with our full kinetic analysis; and both agree with some of the GT1 family structures. Five of the 10 complete GT1 structures, UGT71G1 (Shao et al., 2005), VvGT (Offen et al., 2006), OleI, OleD (Bolam et al., 2007) and UGT85H2 (Li et al., 2007b) have the same catalytic mechanism with a His residue at the N-terminus acting as the proton-abstracting general base. It is, however, noticeable that in all these plant proteins mutating the catalytic histidine resulted in total loss of enzyme activity, unlike in human UGTs. In three structures, there is an Asp in the same position as the His (Mulichak et al., 2001; Mulichak et al., 2003; Mulichak et al., 2004).
The role of the aspartic acid residues
Based on modeling and mutational studies, Miley et al. (Miley et al., 2007) recently proposed that D148 is the general acid in human UGTs. In addition, Li et al. (Li et al., 2007a) investigated the role of this aspartate in various UGT1As by mutational and activity studies and concluded that it plays a major catalytic role in these enzymes. Our results, however, raise the possibility that D143 in UGT1A9 plays this role. The sequence alignment cannot reliably distinguish between these two residues; the high variability between GT1 sequences in the region around the two aspartates (Fig. 6) makes it hard to achieve a definitive match between the UGT sequences and the solved GT1 structures.
Our speculation is based on three lines of evidence. First, the D143A mutant had significantly lower values of Vmax than the D148A mutant (Table II). On the other hand, the D148A mutation tended to increase the Km values–for scopoletin 8-fold and for 4-MU 46-fold (Table II). It also did not show non-competitive inhibition by 4-MU (Table II), also implying that binding of 4-MU was weakened. Consistent with this, the equivalent mutant in UGT2B7, D151A (Fig. 6 in Miley et al., 2007), has over 20% residual activity with androsterone for a single-point assay. Such a high value is surprising if this residue is part of the charge relay. In the same study, the D151N mutant instead had no measured activity for androsterone, but had activity for hyodeoxycholic acid in single-point assays. We suggest that such variable levels of activity with different substrates reflect the changes in Km values, as we see in Table I, consistent with D151 being close to the active site, but not consistent with it being the general acid.
Second, the D143A mutation is the opposite of the H37A mutation; it is less active than H37A in N-aglycones but more active with O-aglycones. This is consistent with the expected chemistry; the histidine is not required to deprotonate an N-nucleophile, but the aspartate is required to stabilize the developing charge during attack by an N-nucleophile (Fig. 5; see below).
Third, even though the GT1 sequences are not conserved, the structures are. We can therefore attempt to distinguish between the possibilities by placing D143 and D148 onto a GT1 scaffold, such as UGT72B1. The general acid, D117 in UGT72B1, occurs at the C-terminal end of strand β4 (Fig. 7). If D148 is the general acid and is placed at D117, as suggested (Li et al., 2007a; Miley et al., 2007), D143 would occur 
20 Å away on the outside of the protein at the N-terminal end of β4, aligned with UGT72B1 T112, UGT71G1 V116 and VvGT S114 (Fig. 6). The residues in this position are far from the active site and typically make a single contact to strand β1, so it is hard to see why mutation here should lead to the large changes in catalytic parameters observed for the D143A mutant. Conversely, if D143 is placed at D117 as the general acid, D148 occurs in the position of D122 in UGT72B1, at the beginning of helix D (Fig. 7). In UGT72B1, D122 hydrogen bonds to T91 in helix C and H199 in helix G (Fig. 8), and thus plays an important role in structural integrity in the closed form of the enzyme; helix C closes over the active site and makes contacts to the UDPG. The helix C region is the region of maximal difference between the open and closed forms of UGT71G1 (Shao et al., 2005) and VvGT (Offen et al., 2006) (data not shown), consistent with this region being involved in substrate binding—and thus consistent with D148A mutations affecting substrate binding.
|
|
Supporting such an assignment, the Vmax of the D143A mutant is very low, but the Km is like control (Table II). This seems reasonable because the charge transfer Asp does not make direct contact with substrate and so mutating it would not be expected to affect aglycone binding directly. Conversely, the aspartate could affect the orientation of the histidine and thus change the Km values as our 9D148A mutant does. This kind of role for aspartate has been reported, for example, in Ribonuclease A (Schultz et al., 1998
). On balance, we prefer a model where D143 is the general acid, rather than one where D148 is the general acid as proposed earlier (Li et al., 2007a; Miley et al., 2007). Further work will be needed to resolve this issue.
If H37 is the general base in UGT1A9, why is it not conserved in UGT1A4 and UGT2B10? We speculate that, in these UGTs, proximity and correct placement of the N-aglycone with respect to the UDPGA (Cannon et al., 1996) might be all that is required for glycosyl transfer. Because of the higher basicity of the amine group, the His general base may not be needed to activate the nucleophile (Fig. 5B). Furthermore, the binding of the substrate O-aglycone to a hydrophobic active site will tend to raise the pKa of the R-OH group, so that it protonates and becomes a worse nucleophile. The reverse is true for an N-aglycone, where the pKa will tend to be reduced and the RNH2+ will tend to deprotonate and become a better nucleophile.
To test this idea, we examined the effect of mutation on the activity with N-nucleophile 4-aminobiphenyl, known to be glucuronidated by UGT1A4 and UGT1A9 (Kaivosaari et al., 2002). As we hoped, the behavior of the 9H37A mutant did indeed depend on the type of the substrate used. The results were still confirmed with another N-nucleophile, retigabine. When O-glucuronide forming substrates were used, the activity was less than 1% compared to the control and the Km was significantly increased. Conversely, the activity in N-glucuronidation was around 30% of the control and the Km value remained unchanged (Tables I and II). It might be argued that, for UGT1A9, the overall N-glucuronidation activity in pmol/min/mg protein is low compared to the O-glucuronidation activity. However, in the 9H37A mutant even the specific activity with N-aglycones (as pmol/min/mg protein) was at least three times higher than with O-aglycones (Tables I and II). The effect of mutation at H37 is thus considerably different in O- than in N-glucuronidation.
The recent structure and mutagenesis study from Brazier-Hicks et al. (Brazier-Hicks et al., 2007) supports these ideas. They mutated H19, which corresponds to H37 in UGT1A9, to glutamine in the bifunctional O- and N-glucosyltransferase UGT72B1. The N-glycosyltransferase activity decreased only 2-fold whereas the kcat for O-glycosyltransferase activity decreased 300-fold. They proposed H19 directs and orientates nucleophile attack in N-glycosylation, while not necessarily abstracting a proton from acceptor (Brazier-Hicks et al., 2007). This viewpoint is also consistent with recent proposals that, in UGT1A3 and UGT1A6, the N-terminal histidine equivalent to H37 defines substrate specificity (Kubota et al., 2007; Li et al., 2007a). In these studies, the presence or absence of histidine switched catalysis between O- and N-glucuronidation.
Interestingly, isoenzymes where this histidine is not conserved appear to have ‘specialized’ in N-glucuronidation rather than O-glucuronidation reactions (Sorich et al., 2006; Kaivosaari et al., 2007). The differences between UGT1A4 and UGT2B10 at this position may then contribute to substrate-specificity differences between these two N-glucuronidating enzymes (Kaivosaari et al., 2007; unpublished results).
The unified view of the roles of His37, Asp143 and Asp148 in catalysis for UGTs goes beyond recent results (Kubota et al., 2007; Li et al., 2007a) that suggest that His37 is important for ‘substrate preference’ due to its role as a part of a charge relay mechanism. The explanation of the mechanistic differences between N- and O-glucuronidation raises the distinct and appealing possibility of devising highly specific mechanism-based inhibitors.
|
Funding |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
This work was supported by the Academy of Finland (A.G. project 1105157, M.F. projects 207538 and 210933) and the Sigrid Juselius Foundation (A.G. and M.F.). A.G. is a member of Biocentrum Helsinki, which partly supported this work.
|
Footnotes |
|---|
Edited by Roger Williams
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
We are grateful for the excellent technical assistance of Johanna Mosorin.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Bolam D.N., Roberts S., Proctor M.R., Turkenburg J.P., Dodson E.J., Martinez-Fleites C., Yang M., Davis B.G., Davies G.J., Gilbert H.J. Proc. Natl Acad. Sci. USA (2007) 104:5336–5341.
Borlak J., Gasparic A., Locher M., Schupke H., Hermann R. Metabolism (2006) 55:711–721.[CrossRef][Web of Science][Medline]
Bourne Y., Henrissat B. Curr. Opin. Struct. Biol (2001) 11:593–600.[CrossRef][Web of Science][Medline]
Brazier-Hicks M., Offen W.A., Gershater M.C., Revett T.J., Lim E.K., Bowles D.J., Davies G.J., Edwards R. Proc. Natl Acad. Sci. USA (2007) 104:20238–20243.
Campbell J.A., Davies G.J., Bulone V., Henrissat B. Biochem. J (1997) 326:929–939.[Web of Science][Medline]
Cannon W.R., Singleton S.F., Benkovic S.J. Nat. Struct. Biol (1996) 3:821–833.[CrossRef][Web of Science][Medline]
Fisher M.B., Paine M.F., Strelevitz T.J., Wrighton S.A. Drug Metab. Rev (2001) 33:273–297.[CrossRef][Web of Science][Medline]
Kaivosaari S., Salonen J.S., Taskinen J. Drug Metab. Dispos (2002) 30:295–300.
Kaivosaari S., Toivonen P., Hesse L.M., Koskinen M., Court M.H., Finel M. Mol. Pharmacol (2007) 72:761–768.
Kubota T., Lewis B.C., Elliot D.J., Mackenzie P.I., Miners J.O. Mol. Pharmacol (2007) 72:1054–1062.
Kurkela M., García-Horsman J.A., Luukkanen L., Mörsky S., Taskinen J., Baumann M., Kostiainen R., Hirvonen J., Finel M. J. Biol. Chem (2003) 278:3536–3544.
Kurkela M., Hirvonen J., Kostiainen R., Finel M. Biochem. Pharmacol (2004) a 68:2443–2450.[CrossRef][Web of Science][Medline]
Kurkela M., Morsky S., Hirvonen J., Kostiainen R., Finel M. Mol. Pharmacol (2004) b 65:826–831.
Kurkela M., Patana A.-S., Mackenzie P.I., Court M.H., Tate C.G., Hirvonen J., Goldman A., Finel M. Pharmacogenet. Genomics (2007) 17:115–126.[Web of Science][Medline]
Li D., Fournel-Gigleux S., Barre L., Mulliert G., Netter P., Magdalou J., Ouzzine M. J. Biol. Chem (2007) a 282:36514–36524.
Li L., Modolo L.V., Escamilla-Trevino L.L., Achnine L., Dixon R.A., Wang X. J. Mol. Biol (2007) b 370:951–963.[CrossRef][Web of Science][Medline]
Luukkanen L., Taskinen J., Kurkela M., Kostiainen R., Hirvonen J., Finel M. Drug Metab. Dispos (2005) 33:1017–1026.
Mackenzie P.I. J. Biol. Chem (1990) 265:3432–3435.
Mackenzie P.I., Walter Bock K., Burchell B., Guillemette C., Ikushiro S., Iyanagi T., Miners J.O., Owens I.S., Nebert D.W. Pharmacogenet. Genomics (2005) 15:677–685.[Web of Science][Medline]
Miley M.J., Zielinska A.K., Keenan J.E., Bratton S.M., Radominska-Pandya A., Redinbo M.R. J. Mol. Biol (2007) 369:517–522.
Mulichak A.M., Losey H.C., Walsh C.T., Garavito R.M. Structure (2001) 9:547–557.[Medline]
Mulichak A.M., Losey H.C., Lu W., Wawrzak Z., Walsh C.T., Garavito R.M. Proc. Natl Acad. Sci. USA (2003) 100:9238–9243.
Mulichak A.M., Lu W., Losey H.C., Walsh C.T., Garavito R.M. Biochemistry (2004) 43:5170–5180.[CrossRef][Web of Science][Medline]
Noguchi A., Saito A., Homma Y., Nakao M., Sasaki N., Nishino T., Takahashi S., Nakayama T. J. Biol. Chem (2007) 282:23581–23590.
Offen W., Martinez-Fleites C., Yang M., Kiat-Lim E., Davis B.G., Tarling C.A., Ford C.M., Bowles D.J., Davies G.J. EMBO J (2006) 25:1396–1405.[CrossRef][Web of Science][Medline]
Ouzzine M., Antonio L., Burchell B., Netter P., Fournel-Gigleux S., Magdalou J. Mol. Pharmacol (2000) 58:1609–1615.[Web of Science][Medline]
Ouzzine M., Barre L., Netter P., Magdalou J., Fournel-Gigleux S. Drug Metab. Rev (2003) 35:287–303.[CrossRef][Web of Science][Medline]
Patana A.-S., Kurkela M., Goldman A., Finel M. Mol. Pharmacol (2007) 72:604–611.
Radominska-Pandya A., Czernik P.J., Little J.M., Battaglia E., Mackenzie P.I. Drug Metab. Rev (1999) 31:817–899.[CrossRef][Web of Science][Medline]
Schultz L.W., Quirk D.J., Raines R.T. Biochemistry (1998) 37:8886–8898.[CrossRef][Web of Science][Medline]
Shao H., He X., Achnine L., Blount J.W., Dixon R.A., Wang X. Medicago truncatula. Plant Cell (2005) 17:3141–3154.
Sorich M.J., McKinnon R.A., Miners J.O., Smith P.A. J. Chem. Inf. Model (2006) 46:2692–2697.[CrossRef][Web of Science][Medline]
Tukey R.H., Strassburg C.P. Annu. Rev. Pharmacol. Toxicol (2000) 40:581–616.[CrossRef][Web of Science][Medline]
Vrielink A., Ruger W., Driessen H.P., Freemont P.S. EMBO J (1994) 13:3413–3422.[Web of Science][Medline]
Received April 30, 2008; revised April 30, 2008; accepted April 30, 2008.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  O. Kerdpin, P. I. Mackenzie, K. Bowalgaha, M. Finel, and J. O. Miners Influence of N-Terminal Domain Histidine and Proline Residues on the Substrate Selectivities of Human UDP-Glucuronosyltransferase 1A1, 1A6, 1A9, 2B7, and 2B10 Drug Metab. Dispos., September 1, 2009; 37(9): 1948 - 1955. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||