¹H, ¹³C and ¹⁵N chemical shift assignments of the C-terminal domain of human UDP-Glucuronosyltransferase 2B7 (UGT2B7-C)

Abstract

The human UDP-glucuronosyltransferase (UGT) family of enzymes catalyze the covalent addition of glucuronic acid to a wide range of compounds, generally rendering them inactive. Although important for clearance of environmental toxins and metabolites, UGT activation can lead to inappropriate glucuronidation of therapeutics underlying drug resistance. Indeed, 50% of medications are glucuronidated. To better understand this mode of resistance, we studied the UGT2B7 enzyme associated with glucuronidation of cancer drugs such as Tamoxifen and Sorafenib. We report ¹H,¹³C and ¹⁵N backbone (> 90%) and side-chain assignments (~78% completeness according to CYANA) for the C-terminal domain of UGT2B7 (UGT2B7-C). Given the biomedical importance of this family of enzymes, our assignments will provide a key tool for improving understanding of the biochemical basis for substrate selectivity and other aspects of enzyme activity. This in turn will inform on drug design to overcome UGT-related drug resistance.

Biological Context.

UDP-glucuronosyltransferases (UGTs; EC 2.4.1.17) are phase II conjugative metabolism enzymes important for modifying a large number of endogenous and exogenous small lipophilic compounds, including carcinogens, steroid hormones, bilirubin, bile acids, fatty acids and therapeutic drugs (Allain et al. 2020a; Rowland et al. 2013; Tukey and Strassburg 2000). UGTs function by catalyzing the covalent addition of glucuronic acid from the cofactor uridine diphosphate glucuronic acid (UDPGA) to the targeted substrates. This typically deactivates drugs or metabolites and often increases solubility allowing for clearance via urine or bile. Approximately 50% of medications are inappropriately glucuronidated leading to resistance to these drugs; and this occurs in many clinical contexts including cancer (Allain et al. 2020b). A role for UGT enzymes in drug resistance is emerging (Allain et al. 2020a; Zahreddine et al. 2014). Recently, we showed that production of a subset of UGT1 enzymes was responsible for patients developing resistance to treatments in AML clinical trials and this impacted many FDA approved drugs (Assouline et al. 2015; Zahreddine et al. 2014; Zahreddine et al. 2019). Additionally, members of the UGT2B family have been implicated in cancer resistance (Allain et al. 2020b; Romero-Lorca et al. 2015; Sutiman et al. 2016).

There are 22 UGT enzymes identified to date. The majority of these enzymes can be divided into two major UGT family members based on sequence homology: UGT1A and UGT2B (Meech et al. 2019). Each family member glucuronidates selected metabolites or drugs, with some overlap in substrates between family members. Both families are comprised of two domains: the N-terminal and C-terminal domains. While the N-terminal domain is traditionally associated with substrate specificity and the C-terminal domain with UDPGA binding, homology modeling and biochemistry experiments suggest that these domains interact with each other leading to juxtaposition of substrate and donor at the interface between these domains (Dong et al. 2012; Laakkonen and Finel 2010). The C-terminal domain sequence in UGT2B enzymes differs substantially from that found in the UGT1A enzymes as do the substrates targeted by UGT2B and UGT1A families (Hu et al. 2019). This suggests that the C-domain plays roles not only in UDPGA binding but also in substrate selectivity and catalysis.

Any approach to target glucuronidation in patients in order to restore drug selectivity requires that only the specific family member is targeted; since, global targeting of UGTs impairs metabolite clearance and results in overt toxicity. With this in mind, we developed a strategy for identifying such selectivity by combining a fragment-based NMR approach with biochemical glucuronidation assays using the C-terminal of UGT1A as a proof-of-principle. This allowed identification of inhibitors selective for UGT1A4 and UGT1A1 (Osborne et al. 2019). We now continue these studies focussing on the UGT2B family member, UGT2B7, which glucuronidates ~ 35% of medications on the market (Williams et al. 2004), including many cancer therapeutics, including Sorafenib (Ye et al. 2014) and Tamoxifen (Romero-Lorca et al. 2015; Sutiman et al. 2016). The crystal structure for the C-terminus of UGT2B7 has been solved (Miley et al. 2007) but no NMR studies have been reported to date on any UGT2B family member. To study the molecular principles underlying drug selectivity in solution between UGT2B and UGT1A families, a critical first step is to obtain NMR assignments. Thus, we assigned the ¹H, ¹⁵N and ¹³C chemical shifts for the UGT2B7-C. These assignments will expedite identification of binding sites for substrates, donor, and compounds that inhibit UGT2B7 mediated glucuronidation.

Methods and experiments

Protein expression and purification.

The full length UGT2B7 construct was purchased from Origene (cat SC119466). The C-terminal domain from human UGT2B7 (residues A285-D451) was inserted into a modified pET-28a vector between the BamHI and XhoI restriction sites. A TEV protease cleavage site was introduced downstream to the N-terminal His-tag for His-tag removal, introducing two extra residues (GS) at the N-terminus (169 residues total, 18,646 kDa). The plasmid was overexpressed in Escherichia coli BL21(DE3)-RIPL competent cells in minimum media at 37°C and induced when the OD₂₈₀ was 0.7 by addition of 1 mM isopropyl-β-D-thiogalactopyranoside (IPTG) (final concentration) at 20°C for 20 h. Cells were harvested by centrifugation and stored at −80°C until use. Uniformly ¹⁵N, ¹³C samples were obtained by growing the cells in M9 minimal media supplemented with 2 g/l of ¹⁵N labeled ammonium sulfate and 2g/l of ¹³C-labeled glucose as the sole sources of nitrogen and carbon respectively (Sigma-Aldrich). For purification, the frozen cells were resuspended in 8M urea supplemented with 50 mM Tris pH 7.5, 10 mM β-mercaptoethanol (βME), 10 mM imidazole, and lysed by sonication. The lysate was cleared by centrifugation (30 min, 20,000 rpm, 4°C) and purified over Ni-NTA beads (Qiagen) onto a gravity flow column. After the resin was extensively washed with the same buffer containing 30 mM imidazole, the protein was eluted in 6M urea, 50 mM Tris pH 7.5, 10 mM βME, 500 mM imidazole. The protein was refolded by dialysis with a gradient of decreasing concentration of urea (2M, 1M, 0.5M) in 20 mM Tris, pH 7.5 and 5 mM DTT. The UGT2B7-C concentration was kept lower than 1mg/ml. The protein was finally dialyzed against 50 mM sodium phosphate (pH 7.2), 300 mM NaCl, 1 mM DTT. The tag was further removed by the TEV protease. High level of purity (>95%) was achieved by gel filtration on Superdex 75pg column (GE-Biosciences) in 50 mM sodium phosphate (pH 7.2), 300 mM NaCl, 1 mM DTT.

NMR spectroscopy and data processing

NMR samples of UGT2B7-C (~ 300 μM) were prepared in 50 mM phosphate buffer (pH 7.2), 300 mM NaCl, 1 mM DTT, 0.02% NaN₃ in 93% H₂O/7 % D₂O and acquired at 23ºC on a Bruker AVANCE III HD 600 MHz spectrometer with a QCIP Z-axis gradient cryoprobe. Backbone assignments were obtained from the standard Bruker sequences for HNCO, HN(CA)CO, HNCA, HNCACB, CBCA(CO)NH, the latter two experiments were optimized for Cβ transfers. Side-chain assignments were facilitated by measurement of the following 3D experiments: HBHA(CO)NH, H(CCO)NH, CC(CO)NH and HCCH-TOCSY and (H)CCH-TOCSY experiments. All triple resonance experiments were acquired using non-uniform sampling schemes (typically between 15% and 20% sampling) using the default sampling schedules from TOPSPIN (Bruker). A 4D HCCH-TOCSY (with 4% NUS) experiment was also acquired, albeit with a low number of scans, to aid in the assignment of some side chains. All spectra were processed with NMRPipe (Delaglio et al. 1995) or SMILE (Ying et al. 2017) for NUS spectra and analyzed with NMRViewJ using the runabout feature (Johnson and Blevins 1994).

Extent of assignments and data deposition

The chemical shift assignments for human UGT2B7-C have been deposited in the Biological Magnetic Resonance Bank (BMRB) under the accession number 50789. The human UGT2B7-C protein comprises 169 residues. We assigned the majority of backbone residues (91% HN, 91% N, 93%Cα, 93%Cβ, 85% CO, 85% ¹Hα). Fig. 1 shows the assignments for the ¹H-¹⁵N HSQC of UGT2B7-C. In addition we have assigned a large number of ¹H and ¹³C side chain resonances (according to CYANA (Guntert 2004) the completeness of assignment was 78.5%). Fig. 2 shows the assignments for a selected number of methyl resonances, which are almost completely assigned for all non methionine methyls (93%). These assignments will be important in the future for accurately defining the interaction sites of UGT2B7-C with small molecules using the ¹³C-edited/¹³C-filtered transferred NOE experiments, which rely on the increased signal from the methyl groups (Abayev et al. 2018; Osborne et al. 2019). Secondary structure prediction from our backbone assignments using the TALOS-N webserver (Shen and Bax 2015) shows secondary structure in solution is in excellent agreement with the UGT2B7-C crystal structure (Fig. 3). We do note that we were unable to assign residues 309 to 314. This region forms part of a loop between structural elements β1 and α1 (Fig. 3a) in the UGT2B7-C crystal structure which has high B-factors. Indeed, coordinates for residues 311–314 are missing in one of the two models (pdb code, 2O6L). In all, our UGT2B7-C assignments will provide a powerful tool for studying this important enzyme and facilitate drug design efforts by identifying binding sites for inhibitors for selected UGTs.

Fig. 1.

¹H-¹⁵N HSQC spectrum of 300 μM UGT2B7-C in 50 mM sodium phosphate, 300 mM NaCl, 1 mM DTT (pH 7.2) at 23ºC. Assignment of the backbone amide resonances are shown for each residue: residues with an asterisk (*) denote folding in the ¹⁵N dimension. We note that residues K367 (10.96 ppm, ¹H and 118.87 ppm, ¹⁵N) and G341 (5.91 ppm, ¹H and 106.69 ppm, ¹⁵N) are out of this spectral window and not shown.

Fig. 2.

¹H-¹³C HSQC spectrum of ¹⁵N-¹³C labelled UGT2B7-C showing the regions containing peaks from methyl moieties. Assignments are indicated.

Fig. 3.

a) Secondary structure delineation based on the secondary structure elements in the UGT2B7-C crystal structure (2O6L.pdb). b) Secondary structure prediction of UGT2B7-C from TALOS-N (Shen and Bax 2013) using backbone ¹⁵N, ¹³C’, ¹³Cα, ¹³Cβ chemical shifts. The probability of α-helical (blue) or β-sheet (red) structure is represented by the height of the bars as predicted by the TALOS-N software.

Data availability

The assignments have been deposited to the BMRB under the accession code 50789