A highly selective probe for UDP-glucuronosyltransferase 2B7 (UGT2B7) in human microsomes: isoform specificity, kinetic characterization, and applications

Abstract

3-Epideacetycinobufagin (EDCB) was found to be a highly isoform-specific probe substrate for 3-glucuronidation mediated by UDP-glucuronosyltransferase 2B7 (UGT2B7). The probe reaction was well-characterized, and our results strongly suggest that EDCB can be used specially to measure the catalytic activity of UGT2B7 in biological samples.

---

In humans, UDP-glucuronosyltransferases (UGTs), as an important superfamily of metabolic enzymes, are responsible for glucuronidation reactions involved in detoxification and elimination processes for drugs, dietary chemicals, environmental pollutants and endobiotics.¹ However, determination of catalytic activity for various UGTs is complicated, due to their overlapping substrate activities and broad substrate specificity.² Thus, it is imperative to find the highly selective probes for UGTs in order to distinguish a real enzymatic activity from different enzyme sources.

As an important isoform among the UGT superfamily, the enzyme UGT2B7 is involved in the glucuronidation of many clinically important drugs.³ As a result of either environmental or genetic influences, however, the inter-individual variations in the function of this enzyme, are likely to have an important impact on drug adverse reaction events, thereby affecting human health.⁴ Additionally, the UGT2B7 isoform is also commonly implicated in clinically relevant drug to drug interactions (DDIs) involving the UGTs.⁵ Therefore, the specific UGT2B7 bioactivity substrate characterized using a highly specific probe substrate. Generally, should possess a high level of selectivity, a strong sensitivity, proper kinetic parameters, and should constitute an accessible resource.⁶ Currently, AZT, morphine and codeine are all reported as potential UGT2B7 probes.⁷ However, some evidence describing the selectivity of these compounds as UGT2B7 substrates are far from convincing. For AZT, UGT2B family including UGT2B4, UGT2B7 and UGT2B17 isoforms are all responsible for its glucuronidation without significant differences (Fig. S1b, ESI†), demonstrating the unsatisfying level of selectivity associated with the use of AZT as a probe.⁸ In the case of morphine, in addition to the formation of two metabolites, morphine-6-glucuronidation reaction is catalyzed by UGT2B4 at 15%, by UGT2B7 at 79% and by UGT2B17 at 6%. This indicates that the selectivity of UGT2B7 for morphine is rather complex, making the use of morphine as a probe unsatisfactory (Fig. S1c, ESI†). Similarly, codeine is glucuronidated by both UGT2B4 and UGT2B7, and the catalytic activity of UGT2B4 for codeine was found to be even better than that of UGT2B7 (Fig. S1d, ESI†).

Bufadienolides are relatively abundant C24 steroids with unique skeleton characteristics. In our previous study,⁹ many novel bufadienolide analogues which are typically difficult to prepare by chemical synthesis, have been obtained using the microbial transformation approach, and used for investigating structure–activity relationships or new biological functions.¹⁰

Previous studies also reported that marinobufagin (a bufadienolide) can produce 3-glucuronide with HLMs.¹¹ In addition, our primary investigation also demonstrated that some bufadienolides can be metabolized by UGT2B7 enzyme in HLMs to produce 3-glucuronide. More importantly, the selectivity and reaction rate of UGT2B7 showed a significant relationship with the chemical structure of bufadienolides (Fig. S2–S4, ESI†). These findings prompted us to screen some bufadienolide derivatives in the hope of developing a highly selective probe substrate of human UGT2B7. In the present study, after screening a number of bufadienolides, the primary relationships between bufadienolides chemical structure and catalytic selectivity of UGT2B7 for 3-glucurondiation were established (Fig. S5, ESI†). Finally, 3-epideacetycinobufagin (EDCB) was identified as the most specific probe substrate for UGT2B7 (Fig. 1). Moreover, this biotransformation reaction obeys the Michaelis–Menten kinetics with a higher isoform-specificity and an lower K_m value, compared to those of AZT, morphine or codeine (Fig. S1, ESI†).

Fig. 1 Glucuronidation of ECDB at C-3α position by UGT2B7.

Incubation of EDCB with HLMs in the presence of UDPGA yielded a single glucuronide peak. The formation of this metabolite was dependent on time-, microsomes- and UDPGA. EDCB glucuronide, as a yellow powder in methanol, was prepared by biotransformation using HLMs. Its ESI-MS showed an [M − H]⁻ ion peak at m/z 574.2 in negative-ion mode with a characteristic m/z 176, indicating a molecular formula of C₃₀H₃₉O₁₁. Compared to 3-epi-acetycinobufagin (EDCB), six additional oxygen-bearing carbon signals were observed at δ 102.6, 74.9, 77.5, 73.2, 76.6 and 172.6, which indicated that EDCBG was a monoglucuronide metabolite. The carbon signal of C-3 shifted up to δ 80.2 (Δδ −10.4). In the HMBC spectrum, H-1′ (δ 4.46) had the key long-range correlations with C-3 (δ 80.2), which suggesting that the glucuronosyl substitution should be at the hydroxyl group of C-3 position. Therefore, based on the spectroscopic data including ¹H-, ¹³C-NMR, HMQC and HMBC, its structure was characterized as 3-epi-deacetycinobufagin-3-glucuronate (EDCBG) (Fig. S6 and Table S1, ESI†). Meanwhile, the preparation methods of EDCB and EDCBG were established and optimized. After comparing metabolism of EDCB in liver microsomes from various species (Fig. S7, ESI†), we found that DLM (Beagle dog liver microsomes) displayed the highest catalytic activity for EDCBG formation and could be used as a suitable model for large-scale preparation of EDCBG.

EDCB glucuronidation by recombinant UGT Supersomes™ was investigated using a panel of twelve recombinant UGT isozymes (1A1, 1A3, 1A4, 1A6, 1A7, 1A8, 1A9, 1A10, 2B4, 2B7, 2B15 and 2B17). Additionally, as shown in Fig. 2, UGT2B7 displayed an excellent ability to catalyze the formation of EDCBG at all three different substrate concentrations tested (5, 50 and 500 μM). Among other UGT isoforms, only UGT2B4 displayed a very limited activity towards EDCBG formation, which was approximately 70 times less than the activity of UGT2B7, even at the highest substrate concentration (500 μM). So far, this is the most UGT2B7-selective probe reaction to be reported.

Fig. 2 Isozyme specificity of EDCB 3-glucuronidation.

To confirm the key role of UGT2B7 in EDCB glucuronidation, chemical inhibition studies were performed. As shown in Fig. 3, the formation of EDCBG could be significantly inhibited by fluconazole (a selective and moderate UGT2B7 inhibitor), mefenamic acid (a broad UGT2B7 and UGT1A9 inhibitor) and diclofenac (a common inhibitor of UGT2B7, 1A1 and 1A9),¹² while the other inhibitors of UGT isoforms had minor effects on preventing formation of EDCBG. Additionally, the inhibitory effects of mefenamic acid and diclofenac on EDCB glucuronidation using HLMs and UGT2B7 were also evaluated. Similar IC₅₀ values were observed for mefenamic acid and diclofenac with HLMs and UGT2B7 (Fig. S8, ESI†). These results also demonstrated that EDCB glucuronidation was selectively catalyzed by human UGT2B7.

Fig. 3 Inhibition assays for ECDB glucuronidation using selective UGTs inhibitors or substrates (AZT, morphine, bilirubin) in HLMs.

To further characterize the isoform-specific biocatalysis by UGT2B7, a kinetic investigation was performed using HLMs and recombinant human UGT2B7 and UGT2B4. As evidenced by the Eadie–Hofstee plot, EDCB glucuronidation using both HLMs and UGT2B7 exhibited the Michaelis–Menten kinetic characteristics for the range of substrate concentrations tested in the kinetic analyses (Fig. 4). Furthermore, EDCB glucuronidation using HLMs from different sources displayed very similar K_m values as those obtained using UGT2B7 (Table 1), which suggesting that UGT2B7 was primarily responsible for the glucuronidation of EDCB.

Fig. 4 The enzyme kinetic analysis of EDCB glucuronidation as a potential UGT2B7 substrate probe using HLMs (A) and UGT2B7 isoform (B).

Table 1 Kinetic parameters of EDCB 3-glucuronidation determined in different enzyme resources^a

Enzyme resources	Vmax	Km (μM)	Vmax/Km
a Vmax values were expressed in nmol min^−1 mg^−1 protein. The substrate concentrations ranged from 0.5 to 400 μM. Each value is presented as the mean ± S.D and represents three determinations performed in duplicate.
HLMs	0.52 ± 0.01	48.69 ± 4.27	10.67
UGT2B7	0.15 ± 0.005	44.41 ± 6.15	3.38
UGT2B4	0.0019 ± 0.00018	182.1 ± 12.25	0.010

---

In the assay with recombinant human UGTs, although UGT2B4 was also involved in EDCB glucuronidation by HLMs, it exhibited very low glucuronosyltransferase activity and enzyme affinity. In contrast, UGT2B7 showed higher affinity and selective catalysis towards EDCB glucuronidation than UGT2B4 (Table 1). In addition, the clearance (V_max/K_m) of UGT2B7 in EDCB glucuronidation was more than 338-fold comparing with that of UGT2B4.

Additionally, the significant differences between UGT2B7 and UGT2B4 for 3-glucuronidation, suggest the better reactivity and higher specificity of EDCB for UGT2B7 rather than UGT2B4 (Fig. S9†). These findings continue to suggest that EDCB could serve as an ideal selective probe for the accurate and specific determination of UGT2B7 activity in complex biological samples.

To evaluate the catalytic activity of UGT2B7 in HLMs, we measured UGT2B7 activity in a panel of HLMs from 14 different individuals using EDCB 3α-glucuronidation as the probe reaction (Fig. 5). Additionally, the expression levels of UGT2B7 in 14 individual HLMs were determined by Western blot analysis. The correlation study between EDCB 3α-glucuronidation rates and expression levels of UGT2B7 in these 14 individual HLMs, demonstrated a stronger correlation for EDCB than that for AZT. These results clearly indicate that EDCB has the higher selectivity than that of AZT, as the probe of UGT2B7. Additionally, as shown in Fig. S10,† EDCB could be used for measuring UGT2B7 activity in different human organ samples including liver, kidney and intestine from both males and females, which suggesting that DECB as a selective probe, has the wide application area.

Fig. 5 Western blots of recombinant UGT2B7 and human liver microsomes (A); EDCB 3-glucurondiation formation rate using 14 individual HLMs (B); correlation analyses (n = 14) were performed between the UGT2B7 protein and EDCB glucuronidation (C); correlation between the UGT2B7 protein and AZT glucuronidation (D).

Screening results of target compounds indicated that the configuration of 3-hydroxyl group in bufadienolides was closely related to UGT2B7 activity (Fig. S2–S4, ESI†). Therefore, the metabolic difference between 3-deacetycinobufagin (DCB, 2) and EDCB (6), as two 3-epi-isomers (Fig. S2†), mediated by UGT2B7 has been explored by docking into the active site within the homologous model of UGT2B7. As shown in Fig. S11,† the amino acids constituting the substrate binding pocket, which is adjacent to the uridine diphosphoglucuronic acid (UDPGA), include Ser34, His35, Ala152, Ile153, Phe154, Pro155, His 374, Gly375, Gly376, Phe396, Asp398, and Gln399. First, we analyzed the binding coordination of two isomers to propose a potential metabolic pathway. The distance between the end hydroxyl of UDPGA where the glucuronide and 3α-OH of EDCB (5.03 Å) were transferred, was shorter than that of DCB (6.93 Å) (Fig. S11†). Comparing the chemscore values, EDCB also exhibited a stronger binding affinity (−22.11) than DCB (−17.99) (Table S2 and Fig. S12, ESI†). Therefore, the UGT2B7 enzyme displayed more reactivity towards EDCB than DCB. The configuration of 3-OH was closely related with the catalytic activity of UGT2B7. Our findings presented herein will be an important supplement to further understand the function of UGT2B7 active site.

In summary, after a systemic screening of bufadienolides, EDCB was found to be a highly isoform-specific probe for UGT2B7, which is one of the most important metabolic enzymes in humans. EDCB 3-glucuronidation as a probe reaction showed more satisfactory selectivity and followed Michaelis–Menten kinetics in comparison with AZT or morphine. The isoform-specific probe reaction for UGT2B7 described herein can be used to characterize UGT2B7 activity specifically in different biological samples and to evaluate the inter-individual variations on UGT2B7-mediated glucuronidation, which can result from genetic or environmental influences.

Acknowledgements

We thank the National Natural Science Foundation of China (81473334, and 81274047), Program for Liaoning Excellent Talents (LR2014025), Liaoning BaiQianWan Talents Program, General Science Research Program of Liaoning Education Department (L2014352), Dalian Outstanding Youth Science and Technology Talent for financial support.

Notes and references

1. (a) K. M. Knights and J. O. Miners, Drug Metab. Rev., 2010, 42, 63; (b) P. G. Wells, P. I. Mackenzie, J. R. Chowdhury, C. Guillemette, P. A. Gregory, Y. Ishii, A. J. Hansen, F. K. Kessler, P. M. Kim, N. R. Chowdhury and J. K. Ritter, Drug Metab. Dispos., 2004, 32, 281.
2. (a) J. O. Miners, P. I. Mackenzie and K. M. Knights, Drug Metab. Rev., 2010, 42, 196; (b) B. Burchell, D. J. Lockley, A. Staines, Y. Uesawa and M. W. Coughtrie, Methods Enzymol., 2005, 400, 46.
3. (a) F. Innocenti, L. Iyer, J. Ramirez, M. D. Green and M. J. Ratain, Drug Metab. Dispos., 2001, 29, 686; (b) O. Barbier, D. Turgeon, C. Girard, M. D. Green, T. R. Tephly, D. W. Hum and A. Belanger, Drug Metab. Dispos., 2000, 28, 497; (c) H. Wang, L. Yuan and S. Zeng, Biochem. Pharmacol., 2011, 82, 1757.
4. (a) J. Y. Chung, J. Y. Cho, K. S. Yu, J. R. Kim, K. S. Lim, D. R. Sohn, S. G. Shin and I. J. Jang, Clin. Pharmacol. Ther., 2008, 83, 595; (b) M. H. Court, S. Krishnaswamy, Q. Hao, S. X. Duan, C. J. Patten and D. J. Greenblatt, Drug Metab. Dispos., 2003, 31, 1125; (c) C. R. Bhasker, W. McKinnon, A. Stone, A. C. Lo, T. Kubota, T. Ishizaki and J. O. Miners, Pharmacogenetics, 2000, 10, 679.
5. (a) K. M. Knights, K. Bowalgaha and J. O. Miners, Drug Metab. Dispos., 2010, 38, 1011; (b) K. M. Knights, A. Rowland and J. O. Miners, Br. J. Clin. Pharmacol., 2013, 76, 587.
6. (a) M. H. Court, Methods Enzymol., 2005, 400, 104; (b) G. B. Ge, J. Ning, L. H. Hu, Z. R. Dai, J. Hou, Y. F. Cao, Z. W. Yu, C. Z. Ai, J. K. Gu, X. C. Ma and L. Yang, Chem. Commun., 2013, 49, 9779.
7. (a) M. H. Court, S. Krishnaswamy, Q. Hao, S. X. Duan, C. J. Patten, L. L. Von Moltke and D. J. Greenblatt, Drug Metab. Dispos., 2003, 31, 1125; (b) J. J. Engtrakul, R. S. Foti, T. J. Strelevitz and M. B. Fisher, Drug Metab. Dispos., 2005, 33, 1621; (c) J. C. Precht, W. Schroth, K. Klein, H. Brauch, E. Krynetskiy, M. Schwab and T. E. Mürdter, Drug Metab. Dispos., 2013, 10, 1124.
8. M. H. Court, S. Krishnaswamy, Q. Hao, S. X. Duan, C. J. Patten, L. L. Von Moltke and D. J. Greenblatt, Drug Metab. Dispos., 2003, 31, 1125.
9. (a) X. C. Ma, X. L. Xin, K. X. Liu, J. Han and D. A. Guo, J. Nat. Prod., 2008, 71, 1268; (b) X. L. Xin, L. B. Zhan, F. Y. Li, X. C. Ma, K. X. Liu, J. Han and D. A. Guo, J. Asian Nat. Prod. Res., 2009, 11, 7; (c) X. C. Ma, J. Cui, J. Zheng and D. A. Guo, J. Mol. Catal. B: Enzym., 2007, 48, 42.
10. (a) M. Ye, G. Qu, H. Guo and D. Guo, J. Steroid Biochem. Mol. Biol., 2004, 91, 87; (b) X. C. Ma, J. Ning, G. B. Ge, S. C. Liang, X. L. Wang, B. J. Zhang and L. Yang, Drug Metab. Dispos., 2011, 39, 675.
11. Y. Miyashiro, T. Nishio and K. Shimada, J. Chromatogr. Sci., 2008, 46, 534.
12. (a) U. Verawan, K. W. Leanne, I. M. Peter, J. E. David, J. A. Williams and J. O. Miners, Br. J. Clin. Pharmacol., 2006, 61, 427; (b) M. T. Donato, S. Montero, J. V. Castell, M. J. Lechón and A. Lahoz, Anal. Bioanal. Chem., 2010, 396, 2251; (c) P. Gaganis, J. O. Miners and K. M. Knights, Biochem. Pharmacol., 2007, 73, 1683.

---

Footnotes

† Electronic supplementary information (ESI) available: Materials and methods, enzyme sources, screening for the highly selective probe, preparing and identifying EDCB and its metabolite, inhibition assays, correlation studies, docking simulation, and comparing the selectivity of EDCB with AZT and morphine. See DOI: 10.1039/c4ra09819f

‡ These authors contributed equally to this work.

---