The enzymatic glucuronidation of 3-O-protected morphine - A new route to 7,8-dihydromorphine-6-glucuronide

Article abstract of DOI:10.1016/S0957-4166(99)00582-0

The selective enzymatic glucuronidation of the 6-position of morphine was probed by the glucuronidation of a series of 3-O-protected morphine derivatives. 3-O-Benzylmorphine 5 was converted to the corresponding 3-O-benzylmorphine-6-glucuronide 8 by human

Full text of DOI:10.1016/S0957-4166(99)00582-0

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 11 (2000) 413–416
The enzymatic glucuronidation of 3-O-protected morphine —
a new route to 7,8-dihydromorphine-6-glucuronide
Gareth N. Jenkins,a,∗ Andrew V. Stachulski,a Feodor Scheinmann a and Nicholas J. Turner b
aUltrafine (UFC Ltd), Synergy House, Manchester Science Park, Manchester M15 6SY, UK
bDepartment of Chemistry, University of Edinburgh, West Mains Road, Edinburgh EH9 3JJ, UK
Received 1 December 1999; accepted 13 December 1999
Abstract
The selective enzymatic glucuronidation of the 6-position of morphine was probed by the glucuronidation of
a series of 3-O-protected morphine derivatives. 3-O-Benzylmorphine 5 was converted to the corresponding 3-
O-benzylmorphine-6-glucuronide 8 by human liver microsomes. The identity of 8 was confirmed by independent
chemical synthesis. The glucuronide 8 was subsequently converted to 7,8-dihydromorphine-6-glucuronide. © 2000
Elsevier Science Ltd. All rights reserved.
We have been interested in morphine-6-glucuronide 1 (M-6-G) for many years and have previously
reported both chemical and enzymatic syntheses.1,2 Glucuronidation is carried out by the membrane
bound enzymes, UDP-glucuronosyl transferases (UGT), which are to be found in a number of tissues
but usually have a greater abundance in the liver. In common with the P450 enzymes responsible for
primary metabolism, there is a family of UGTs that has evolved to glucuronidate a variety of substrate
classes.3 These have been extensively characterised and a nomenclature system has been developed based
upon the DNA sequence of the genes coding for their production. In humans at least 20 different UGTs
have been identified and a handful of these have now been cloned and expressed. Of these, one isoform,
UGT2B7, has been shown to catalyse morphine glucuronidation at the 3- and 6-positions.4 The ratio of
3- to 6-glucuronide was 7:1 — similar to the reported ratio for human liver microsomes. This indicates
that it is unlikely that there is one UGT catalysing 6-glucuronidation and a separate UGT catalysing 3-
glucuronidation. In order to use a UGT enzyme to produce M-6-G 1 selectively, we needed some way to
prevent or inhibit this 3-glucuronidation.
∗
Corresponding author.
0957-4166/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00582-0
tetasy 3216

414
G. N. Jenkins et al. / Tetrahedron: Asymmetry 11 (2000) 413–416
We chose to use a protecting group on the phenolic 3-hydroxy group of morphine and investigate
whether this type of substrate would show increased 6-glucuronidation. A range of protecting groups
was chosen to include ester, ether and silyl protection (3-pivaloylmorphine 2, 3-cyclohexanecarbonyl-
morphine 3, 3-benzoylmorphine 4, 3-benzylmorphine 5 and 3-tert-butyldimethylsilylmorphine 6). As a
control, 3-methylmorphine 7 (codeine) was also used.
With these substrates prepared, we screened for glucuronidation activity using an assay developed by
Burchell et al.5 This is a radiolabel assay using C-14 radiolabelled UDP-glucuronic acid (UDP-GA). The
six substrates and a morphine control were all assayed in duplicate and the level of glucuronide formation
determined by the HPLC response. The assay was run with 200 nmol UDP-GA and, as the radiolabel
scintillation detector used shows only the unreacted UDP-GA and the formation of any glucuronide,
quantification of the conversion is possible. The results are reproduced in Table 1.
Table 1
Microsomes contain enzymes other than glucuronosyl transferases including a range of lipases. It
was not unexpected, therefore, to observe the formation of 3- and 6-morphine glucuronide from any
of the ester derivatives (2, 3 and 4) which were identified by comparison to the morphine control. The
TBS derivative 6 is of interest as it showed no glucuronidation and proved to be a weak inhibitor of
glucuronidation by human liver microsomes.
The most interesting result was the glucuronidation of 3-O-benzylmorphine 5 to give a glucuronidated

G. N. Jenkins et al. / Tetrahedron: Asymmetry 11 (2000) 413–416
415
species 8 with a retention time of 10.75 min. This compares to 9 min for the morphine glucuronides and
9.75 min for the codeine (3-O-methylmorphine) glucuronide 9 (see Fig. 1).
Fig. 1.
The assay was repeated with unlabelled UDP-GA and the supernatants analysed by LC-MS to
confirm the identity of the glucuronide 8 formed with 3-O-benzylmorphine 5. The retention time of
3-O-benzylmorphine 5 was 11.74 min and the LC of the production of the glucuronidation of 3-O-
benzylmorphine showed a small peak at 11.28 min corresponding to the glucuronide 8 peak we had
observed with the radiolabel study. The MS corresponding to this peak at 11.28 min showed a molecular
ion of 552.18 (EI+), consistent with the structure of 3-O-benzyl-6-morphine glucuronide 8.
The structure of the enzymatic product was corroborated by chemical synthesis and conversion
to a known compound. Thus, reaction of 3-O-benzylmorphine 5 with the tri-isobutyryl imidate 106
[BF3·Et2O, (CH2Cl)2] afforded glucuronate ester 11 in 48% yield after chromatography; use of the
stronger Lewis acid Me3SiO3SCF3 gave concomitant debenzylation.
This provides another example1 of the value of 10 in the glucuronidation of morphine derivatives,
despite erroneous reports to the contrary.7 Hydrolysis of 11 using NaOH in aq. PriOH followed by
neutralisation, then purification by reverse-phase silica chromatography, afforded the free glucuronide
8 (66%) which proved identical (MS, HPLC) with the enzymatic product.

416
G. N. Jenkins et al. / Tetrahedron: Asymmetry 11 (2000) 413–416
Finally, hydrogenolysis of 8 (10% Pd–C, H2O–MeOH–AcOH) gave both debenzylation and reduction
of the 7,8-double bond, leading to 7,8-dihydromorphine-6-glucuronide 12. We have reported previously8
that this derivative is a more potent analgesic than either morphine or morphine-6-glucuronide.
Selected spectroscopic data: compound 11: δH (300 MHz, CDCl3), inter alia, 2.80 (3H, s), 3.79 (3H,
s), 4.20 (1H, d), 4.45 (1H, m), 5.00 (2H, m), 5.15–5.50 (6H, m), 5.85 ( 1H, br d), 6.52 and 6.73 (2H, 2 d);
m/z (E.I.) 775 (M+, 10%). Compound 8: δH [300 MHz, (CD3)2SO+D2O], inter alia, 2.85 (3H, s), 4.05
(1H, m), 4.45 (1H, m), 4.56 (1H, d), 5.05 (2H, s), 5.16 (1H, narrow d), 5.25 and 5.75 (2H, 2 d), 6.60 and
6.78 (2H, 2 d), and 7.20–7.40 (5 H, m); m/z (E.S.+ve mode) 574 (MNa+, 20%) and 552 (MH+, 100%).
Acknowledgements
We are very grateful to Prof. Brian Burchell for his advice and allowing us to carry out the radiolabel
experiment in his laboratories. This work was partially funded by the Teaching Company Scheme.
References
1. Brown, R. T.; Carter, N. E.; Lumbard, K. W.; Scheinmann, F. Tetrahedron Lett. 1995, 36, 8661; PCT 1992, GB 92/01449;
EP 0597915; US 5977326.
2. Brown, R. T.; Carter, N. E.; Scheinmann, F.; Turner, N. J. Tetrahedron Lett. 1995, 36, 1117; EP 0733120.
3. Burchell, B. Rev. Biochem. Toxic. 1981, 3, 1; Irshaid, Y. M.; Tephly, T. R. Mol. Pharmacol. 1987, 31, 27.
4. Coffman, B. L.; Rios, G. R.; King, C. D.; Tephly, T. R. Drug Metab. and Dispos. 1997, 25, 1.
5. Ethell, B. T.; Anderson, G. D.; Beaumont, K.; Rance, D. J.; Burchell, B. Anal. Biochem. 1998, 255, 142.
6. Scheinmann, F.; Lumbard, K. W.; Brown, R. T.; Mayalarp, S. P. International Patent, 1993, WO 93/3051; Brown, R. T.;
Mayalarp, S. P.; McGown, A. T.; Hadfield, J. A. J. Chem. Res. 1993, 496.
7. Lacy, C.; Sainsbury, M. Tetrahedron Lett. 1995, 36, 3949; Berrang, B.; Brine, G. A.; Carroll, F. I. Synthesis 1997, 1165.
8. Joel, S. P.; Scheinmann, F.; Stachulski, A. V. PCT 1998, GB 98/01071.