Gram scale synthesis of the glucuronide metabolite of ABT-724

Article abstract of DOI:10.1021/jo0611972

A gram scale synthesis of the glucuronide metabolite of ABT-724 is reported. Glycosidic coupling between a trichloroacetimidate glucuronyl donor and a Cbz-protected hydroxypyridylpiperazine glycosyl acceptor is the key step in the synthesis, since attempts to directly glucuronicrossed d signate the aglycon, aglycon derivatives, and other truncated glycosyl acceptors were unsuccessful. The route was used to produce 2.1 g of metabolite in eight steps from 2-chloro-5-hydroxypyridine in 21% overall yield.

Full text of DOI:10.1021/jo0611972

Gram Scale Synthesis of the Glucuronide Metabolite of ABT-724
Kenneth M. Engstrom,*,† Jerome F. Daanen,‡ Seble Wagaw,† and Andrew O. Stewart‡
Global Pharmaceutical R&D, Process Research & DeVelopment, Abbott Laboratories, 1401 Sheridan
Road, North Chicago, Illinois 60064-4000, and Department of Neuroscience Research, Abbott
Laboratories, 100 Abbott Park Road, Abbott Park, Illinois 60064-6123
kenneth.engstrom@abbott.com
ReceiVed June 9, 2006
A gram scale synthesis of the glucuronide metabolite of ABT-724 is reported. Glycosidic coupling between
a trichloroacetimidate glucuronyl donor and a Cbz-protected hydroxypyridylpiperazine glycosyl acceptor
is the key step in the synthesis, since attempts to directly glucuronidate the aglycon, aglycon derivatives,
and other truncated glycosyl acceptors were unsuccessful. The route was used to produce 2.1 g of metabolite
in eight steps from 2-chloro-5-hydroxypyridine in 21% overall yield.
Introduction
Glucuronidation is a major pathway for drug metabolism.1
As a drug candidate advances through clinical development, a
synthesis of its glucuronide metabolite(s) often becomes neces-
sary to verify its structure, to provide an analytical standard for
use in quantification of metabolite levels in clinical samples,
and to provide material for further pharmacological evaluation.
While methods to synthesize simple glucuronides are relatively
well developed,2 the synthesis of structurally complex glucu-
ronides is not straightforward. The efficiency and scaleability
of such syntheses is often limited by low yielding or unselective
glycosidic couplings, complex protecting group strategies,
tedious isolations, or enzymatic reactions.
ABT-724 (1), a potent selective D4 dopamine receptor agonist,
has been identified as a promising treatment for erectile
dysfunction.3 ABT-724 is oxidatively metabolized to the hy-
droxy analogue 2, which is subsequently glucuronidated to give
3.4 As part of the pharmacological evaluation of ABT-724, a
synthesis of glucuronide 3 became necessary to confirm its
structure, to provide an analytical standard, and to supply gram
quantities of material for further pharmacological studies.
The synthesis of 3 presented significant challenges. In general,
the use of glucuronic acid derived glycosyl donors is often
inefficient due to the destabilizing effect the C-6 electron-
withdrawing group has on the glycosidic bond forming event.2
In addition, glycosyl acceptors such as 2 can be inefficient
coupling partners because of the presence of multiple basic
functional groups.5 Furthermore, we required a glycosidic
(3) (a) Brioni, J. D.; Moreland, R. B.; Cowart, M.; Hsieh, G. C.; Stewart,
A. O.; Hedlund, P.; Donnelly-Roberts, D. L.; Nakane, M.; Lynch, J. J., III;
Kolasa, T.; Polakowski, J. S.; Osinski, M. A.; Marsh, K.; Andersson, K.
E.; Sullivan, J. P. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 6758-6763. (b)
Cowart, M.; Latshaw, S. P.; Bhatia, P.; Daanen, J. F.; Rohde, J.; Nelson,
S. L.; Patel, M.; Kolasa, T.; Nakane, M.; Uchic, M. E.; Miller, L. N.;
Terranova, M. A.; Chang, R.; Donnelly-Roberts, D. L.; Namovic, M. T.;
Hollingsworth, P. R.; Martino, B. R.; Lynch, J. J., III; Sullivan, J. P.; Hsieh,
G. C.; Moreland, R. B.; Brioni, J. D.; Stewart, A. O. J. Med. Chem. 2004,
47, 3853-3864.
(4) Patel, M. V.; Kolasa, T.; Mortell, K.; Matulenko, M.; Hakeem, A.;
Rohde, J.; Nelson, S.; Cowart, M.; Nakane, M.; Miller, L.; Uchic, M.;
Terranova, M.; El-Kouhen, O. F.; Donnelly-Roberts, D. L.; Namovic, M.
T.; Hollingsworth, P.; Chang, R.; Martino, B.; McVey, J.; Marsh, K.; Martin,
R.; Darbyshire, J. F.; Gintant, G.; Hsieh, G.; Moreland, R. B.; Sullivan, J.;
Brioni, J. D.; Stewart, A. O. J. Med. Chem. Accepted for publication.
† Process Research & Development.
‡ Department of Neuroscience Research.
(1) For a review of the importance of glucuronides in regulation of the
biological activity of pharmaceuticals, see: Mulder, G. J. J. Annu. ReV.
Pharmacol. Toxicol. 1992, 32, 25-49.
(2) For reviews on the chemical synthesis of glucuronides, see: (a)
Stachulski, A. V.; Jenkins, G. A. Nat. Prod. Rep. 1998, 173-186. (b)
Kaspersen, F. M.; van Boeckel, C. A. A. Xenobiotica 1987, 17, 1451-
1471.
10.1021/jo0611972 CCC: $33.50 © 2006 American Chemical Society
Published on Web 09/23/2006
8378
J. Org. Chem. 2006, 71, 8378-8383

Synthesis of the Glucuronide Metabolite of ABT-724
SCHEME 1
coupling that is selective for the â-anomer, a donor that is readily
available in gram quantities, and a practical method for the
isolation of the fully deprotected metabolite.
The most direct route to glucuronide 3 would use aglycon 2
as the glycosyl acceptor. However, glycosidic coupling between
2 and 4, 5, 6, or 7 failed under all conditions tested. The poor
reactivity of 2 as the glycosyl acceptor was attributed to the
presence of multiple basic functional groups and its poor
solubility in common organic solvents. Bis-borane complex 8
and zinc complex 9 were prepared to attenuate the basicity of
the acceptor, while tributylstannane 1012 was intended to
increase the solubility. Unfortunately, 8, 9, and 10 all failed to
couple with 4, 5, 6, or 7 under a variety of typical coupling
conditions.
Results and Discussion
1,2,3,4-Tetra-O-acetyl-â-D-glucuronic acid methyl ester (4)
was chosen as a convenient glucuronyl donor starting material
for several reasons. Glycosyl donors containing C-2 ester
protecting groups often result in â-selective glycosidic bond
formation.2 In addition, 4 is commercially available in significant
quantities6 and the 1-R-bromide (5), hemiacetal (6), and 1-R-
trichloroacetimidate (7) derivatives could all be prepared from
4 in one or two steps with literature procedures.7-9 Easy access
to these four distinct glucuronyl donors would allow a wide
range of coupling methods and conditions to be quickly
surveyed. Bromide 5 can be activated for glycosidic coupling
by using a variety of heavy metal salts, whereas acetate 4 and
trichloroacetimidate 7 can be activated with acid catalysts.10
Hemiacetal 6 can be used as a glucuronyl donor under either
acidic10 or Mitsonobu11 coupling conditions.
Truncated versions of aglycon 2 were then evaluated.
2-Chloro-5-hydroxypyridine 11 was coupled with 5 using Ag2-
CO3 to give 12 in approximately 54% yield (Scheme 1, eq 1).
However, Hartwig-Buchwald amination13 of 12 led exclusively
to elimination of the C-4 acetate group to give the R,â-
unsaturated ester14 despite repeated attempts with different
piperazines, bases, ligands, and catalyst loadings (Scheme 1,
eq 2).
Glycosidic coupling with piperazine-containing acceptors was
then evaluated. N-Protected hydroxypyridylpiperazines 13, 14,
and 15 were synthesized and tested in various coupling reactions
with donors 4-7. Glycosyl acceptors 13 and 14, containing the
Boc and benzyl protecting groups, respectively, failed to give
desired product in greater than 10% yield with any of the four
donors. Cbz-protected acceptor 15 likewise did not couple with
4, 5, and 6 in useful yield. However, reaction of 15 with 1 equiv
(5) For the synthesis of other structurally related glucuronides, see: (a)
Kim, S.; Wu, J. Y.; Zhang, Z.; Tang, W.; Doss, G. A.; Dean, B. J.; DiNinno,
F.; Hammond, M. L. Org. Lett. 2005, 3, 411-414. (b) Liu, C.-H.; Liu, H.;
Han, X.-Y.; Wu, B.; Zhong, B.-H.; Gong, Z.-H. Synth. Commun. 2005, 5,
701-710. (c) Stazi, F.; Palmisano, G.; Turconi, M.; Clini, S.; Santagostino,
M. J. Org. Chem. 2004, 4, 1097-1103. (d) Kawamura, K.; Horikiri, H.;
Hayakawa, J.; Seki, C.; Yoshizawa, K.; Umeuchi, H.; Nagase, H. Chem.
Pharm. Bull. 2004, 6, 670-671. (e) Kuo, F.; Gillespie, T. A.; Kulanthaivel,
P.; Lantz, R. J.; Ma, T. W.; Nelson, D. L.; Threlkeld, P. G.; Wheeler, W.
J.; Yi, P.; Zmijweski, M. Bioorg. Med. Chem. Lett. 2004, 13, 3481-3486.
(f) Wang, Y.; Yuan, H.; Wright, S. C.; Wang, H.; Larrick, J. W. Bioorg.
Med. Chem. 2003, 7, 1569-1576. (g) Hasuoka, A.; Nakayama, Y.; Adachi,
M.; Kamiguchi, H.; Kamiyama, K. Chem. Pharm. Bull. 2001, 12, 1604-
1608. (h) Rukhman, I.; Yudovich, L.; Nisnevich, G.; Gutman, A. L.
Tetrahedron 2001, 6, 1083-1092 and references therein. (i) Brown, R. T.;
Carter, N. E.; Mayalarp, S. P.; Scheinmann, F. Tetrahedron 2000, 56, 7591-
7594. (j) Suzuki, T.; Mabuchi, K.; Fukazawa, N. Bioorg. Med. Chem. Lett.
1995, 5, 659-662. (k) Dodge, J. A.; Lugar, C. W.; Cho, S.; Osborne, J. J.;
Philips, D. L.; Glasebrook, A. L.; Frolik, C. A. Bioorg. Med. Chem. Lett.
1997, 8, 9993-996.
(6) 50 g obtained from Toronto Research Chemicals, Inc.
(7) Bollenback, G. N.; Long, J. W.; Benjamin, D. G.; Lindquist, J. A. J.
Am. Chem. Soc. 1955, 77, 3310-3315.
(12) For use of tributyltin phenoxides as glycosyl acceptors, see: (a)
Clerici, F.; Gelmi, M. L.; Mottadelli, S. J. Chem. Soc., Perkin Trans. 1
1994, 985-988. (b) Hannessian, S.; Saavedra, O. M.; Xie, F.; Amboldi,
N.; Battistini, C. Bioorg. Med. Chem. Lett. 2000, 10, 439-442.
(13) (a) Yang, B. H.; Buchwald, S. L. J. Organomet. Chem. 1999, 576,
125-146. (b) Wolfe, J. P.; Wagaw, S.; Buchwald, S. L. J. Am. Chem. Soc.
1996, 118, 7215-7216.
(14) For an example of intentional elimination of the C-4 acetate group
with strong base, see: Adamczyk, M.; Chen, Y.-Y.; Fishpaugh, J. R. Org.
Prep. Proced. Int. 1992, 24, 546-548.
(8) Nudelman, A.; Herzig, J.; Gottlieb, H. E.; Keinan, E.; Sterling, J.
Carbohydr. Res. 1987, 162, 145-152.
(9) Jacquinet, J. C. Carbohydr. Res. 1990, 199, 153-181.
(10) Use of bromide 5, acetate 4, trichloroacetimidate 7, and hemiacetal
6 in glucuronidation has been extensively reviewed. For leading references,
see refs 2a and 2b.
(11) Badman, G. T.; Green, D. V. S.; Voyle, M. J. Organomet. Chem.
1990, 338, 117-121.
J. Org. Chem, Vol. 71, No. 22, 2006 8379

Engstrom et al.
SCHEME 2a
a Reagents and conditions: (a) Bu3SnOMe, THF, 60 °C, ref 8; (b) Cl3CCN, DBU, DCE, rt, ref 9; (c) MOMCl, K2CO3, DMF, 0 °C to rt, 72%; (d) 2.0
mol % of Pd(OAc)2, 4.4 mol % of Et3N, 2.4 mol % of rac-BINAP, 18, NaOtBu, PhMe, rt, 95%; (e) 2M HClaq,THF, 55 °C, 91%.
of trichloroacetimidate 7 promoted by BF3‚OEt2 in CH2Cl2 gave
the desired product 16 in approximately 50% yield. While
isolated in approximately 80% purity18 after silica gel chroma-
tography and was contaminated with a significant amount of
hydrolyzed donor (6) and several other unidentified impurities.
Fortunately, this mixture could be taken on to the next step
without further purification.
Removal of the Cbz group under hydrogenolysis conditions
gave amine 2019 in 83% isolated yield (Scheme 3). The product
was isolated by silica gel chromatography in 96% purity18 free
of 6 carried over from the previous reaction. Alkylation of amine
20 with 2-(chloromethyl)benzimidazole (21) in NMP with
triethylamine as base gave a 77% yield16 of protected glucu-
ronide 22. After aqueous workup to remove NMP, the product
was purified by silica gel chromatography and then recrystallized
from ethyl acetate to give 22 in 62% isolated yield from amine
20.
Removal of the acetate groups from 22 was accomplished
by using catalytic LiOH in methanol at -10 °C with warming
to 23 °C, which minimized formation of the R,â-unsaturated
ester side product 24 and gave methyl ester 23.20 Methyl ester
hydrolysis was then accomplished by addition of 1.9 equiv of
LiOH. Fortunately, R,â-unsaturated methyl ester 24 hydrolyzed
at a significantly slower rate than methyl ester 23. Because of
this rate difference, the formation of R,â-unsaturated acid 25
could be minimized by a timely quench of the reaction mixture
with 2 equiv of acetic acid. After addition of water, chloroform
extraction removed the small amount of remaining unhydrolyzed
methyl esters 23 and 24. The aqueous phase was then treated
reaction of 13 with 7 resulted in a complex mixture due to the
instability of the Boc group to BF3, reaction of 14 with 7 was
prevented by formation of an unreactive precipitate that is
presumably a complex formed by reaction of the tertiary
piperazine nitrogen with BF3. It appears that reaction between
15 and 7 succeeds because the Cbz group in 15 is stable under
the reaction conditions and attenuates the basicity of the
piperazine. Use of 15 as the glycosyl acceptor is therefore the
key to gaining synthetic access to glucuronide 3, the complete
synthesis of which is detailed below.
Gram scale synthesis of 3 began with the synthesis of
coupling partners 7 and 15 (Scheme 2). Trichloroacetimidate 7
was prepared from acetate 4 in two steps according to literature
procedures (Scheme 2).8,9 To prepare glycosyl acceptor 15, the
phenolic hydroxyl group of 2-chloro-5-hydroxypyridine (11)15
was first protected as a methoxymethyl ether (17) in 72% yield.
Hartwig-Buchwald amination of 17 with Cbz-protected pip-
erazine 18, which proceeded at room temperature with just a
2% catalyst loading, afforded pyridyl piperazine 19 in 95% yield.
Removal of the methoxymethyl ether with aqueous HCl gave
glycosyl acceptor 15 in 91% yield and 62% overall yield from
11.
Under optimized conditions, acceptor 15 was glycosylated
by using 2 equiv of glucuronyl donor 7 promoted by 4 equiv
of BF3‚OEt2 to give glucuronide 16 in 75% yield16 with the
required anomeric configuration (Scheme 3).17 The product was
with Dowex hydroxide resin to adsorb the carboxylate of 3.21
Filtration and washing of the resin with methanol and water
removed the lithium salts. The product was recovered from the
(15) For the synthesis of 2-chloro-5-hydroxypyridine, see: Lynch, J. K.;
Holladay, M. W.; Ryther, K. B.; Bai, H.; Hsiao, C.; Morton, H. E.; Dickman,
D. A.; Arnold, W.; King, S. A. Tetrahedron: Asymmetry 1998, 9, 2791-
2794.
(18) Purity was determined by HPLC weight percent assay versus a pure
(16) Yield was determined by HPLC weight percent assay using a pure
standard.
(17) The R-anomer was not isolated or identified.
standard.
(19) The relative stereochemistry of 20 was confirmed by a rOe
experiment.
8380 J. Org. Chem., Vol. 71, No. 22, 2006

Synthesis of the Glucuronide Metabolite of ABT-724
SCHEME 3a
a Reagents and conditions: (a) 2.0 equiv of 7, 1.0 equiv of 15, BF3‚OEt2, CH2Cl2, -60 °C to rt, 41 h, 75%; (b) 5% Pd/C, H2, MeOH, rt, 83%; (c) (1)
21, Et3N, NMP, 0 °C to rt, (2) EtOAc, rt, 62% from 20.
SCHEME 4a
a Reagents and conditions: (a) 0.1 equiv of LiOH, MeOH, -10 °C to rt; (b) 1.9 equiv of LiOH, MeOH, rt; (c) 2.0 equiv of AcOH, H2O, CHCl3 wash;
(d) Dowex hydroxide resin, filtration, wash, then AcOHaq, 87% from 22.
resin by treatment with aqueous acetic acid, which was then
removed by azeotropic distillation with 2-propanol and heptane.
This deprotection and isolation procedure gave glucuronide 3
as an amorphous solid22 in 87% overall yield. The current
synthesis was used to produce 2.1 g of glucuronide 3 in 21%
isolated yield from 2-chloro-5-hydroxypyridine 11 (8 chemical
steps, average yield of 82% per step).
Conclusions
In summary, although direct glucuronidation of the aglycon
2 and its derivatives was unsuccessful, synthetic access to 3
was realized by development of a three-step synthesis of
appropriately protected glycosyl acceptor 15 and its efficient
glycosidation, using the readily available glucuronyl donor 7.
Subsequent transformations installed the benzimidazole moiety
and a carefully controlled final deprotection and isolation
sequence allowed gram quantities of glucuronide 3 to be
produced for pharmacological evaluation.
(20) For another mild deprotection method that uses Na2CO3 in aqueous
methanol, see: Brown, R. T.; Scheinmann, F.; Stachulski, A. V. J. Chem.
Res. (S) 1997, 370-371.
(21) Use of hydroxide resin was critical for removal of lithium because
the carboxylate form of 3 would not adsorb to chloride resin. For a table of
selectivity coefficients of various anions on Dowex anion-exchange resins,
consult www.dowex.com.
(22) Attempts to crystallize 3 were unsuccessful. The final amorphous
solid was contaminated with 1-2% of R,â unstaurated acid 25 as measured
by HPLC at 260 nm.
Experimental Section
2-Chloro-5-methoxymethoxypyridine (17). To a 500 mL round-
bottom flask is charged 2-chloro-5-hydroxypyridine 11 (20.00 g,
154 mmol, 1.0 equiv) followed by DMF (200 mL). The solution is
J. Org. Chem, Vol. 71, No. 22, 2006 8381

Engstrom et al.
cooled to -10 °C and milled K2CO3 (42.68 g, 309 mmol, 2.0 equiv)
is added. Two minutes after addition of the K2CO3, chloromethyl
methyl ether (MOMCl, 11.7 mL, 154 mmol, 1.0 equiv) is charged.
After the reaction has been allowed to warm to ambient temperature
over 19 h, more MOMCl (5.9 mL, 77 mmol, 0.5 equiv) is added,
followed by another portion of MOMCl (4.7 mL, 62 mmol, 0.4
equiv) 1 h later. After an additional 2 h the reaction is complete as
determined by HPLC analysis (greater than 95% conversion of 11).
H2O (100 mL) is added followed by EtOAc (100 mL). This mixture
is transferred to a separatory funnel and diluted with more H2O
(400 mL) and EtOAc (400 mL). The layers are separated and the
organic layer is washed once with H2O (250 mL). The organic layer
is dried with Na2SO4, filtered, and concentrated, and the residue is
purified by silica gel chromatography (100% hexanes grading to
5% EtOAc in hexanes in 1% increments) to afford the product 17
(19.12 g, 71.5%) as a clear colorless liquid: 1H NMR (400 MHz,
CD3OD) δ 8.11 (d, J ) 3.0 Hz, 1H), 7.51 (dd, J ) 8.8, 3.2 Hz,
1H), 7.35 (d, J ) 8.8 Hz, 1H), 5.23 (s, 2H), 3.47 (s, 3H); 13C NMR
(100 MHz, CD3OD) δ 154.0, 143.7, 138.9, 128.0, 125.5, 95.8, 56.5;
FTIR (microscopic technique) 2960, 2910, 1570, 1460, 1375, 1255,
1230, 1205, 1155, 1115, 1085, 1015, 980, 925, 830 cm-1; EI HRMS
for C7H8ClNO2 (M + H+) calcd 173.0244, found 173.0246.
4-(5-Methoxymethoxypyridin-2-yl)piperazine-1-carboxylic Acid
Benzyl Ester (19). To an oven-dried 500 mL three-necked round-
bottom flask under an atmosphere of N2 is charged Pd(OAc)2 (590
mg, 2.63 mmol, 0.020 equiv), racemic BINAP (1.97 g, 3.16 mmol,
0.024 equiv), and toluene (50 mL). Triethylamine (0.81 mL, 5.79
mmol, 0.044 equiv) is added and the mixture is allowed to stir for
90 min before charging benzyl 1-piperazinecarboxylate 18 (25.4
mL, 132 mmol, 1.2 equiv), 17 (19.04 g, 110 mmol, 1.0 equiv),
and toluene (114 mL). After an additional 10 min, sodium tert-
butoxide (12.65 g, 132 mmol, 1.2 equiv) is added followed by
toluene (100 mL). The reaction exotherms slowly from 23 to 35
°C over approximately 30 min and is stirred for an additional 19 h
at ambient temperature until 17 is completely consumed as
determined by HPLC analysis. Aqueous NaCl (50%, 125 mL)
solution is added slowly. The mixture is transferred to a separatory
funnel and more toluene (100 mL) and 50% aqueous NaCl solution
(125 mL) are added. The layers are separated and the organic layer
is washed once with 20% aqueous NaCl solution (250 mL). The
organic layer is then dried with Na2SO4, filtered, and concentrated,
and the residue is purified by silica gel chromatography (100%
hexanes grading to 30% EtOAc in hexanes in 5% increments) to
afford the product 19 (28.70 g, 95.4%) as a crystalline off-white
solid: mp 59 °C; 1H NMR (400 MHz, CD3OD) δ 7.92 (d, J ) 3.0
Hz, 1H), 7.33 (m, 6H), 6.80 (d, J ) 9.2 Hz, 1H), 5.14 (s, 2H),
5.09 (s, 2H), 3.60 (br s, 4H), 3.45 (s, 3H), 3.40 (m, 4H); 13C NMR
(100 MHz, DMSO-d6) δ 154.3, 153.9, 145.5, 136.3, 136.0, 127.9,
127.4, 127.1, 126.9, 107.8, 94.9, 66.1, 55.4, 45.3, 43.1; FTIR
(microscopic technique) 2880, 2830, 1685, 1265, 1250, 1155, 1125,
980, 735, 695 cm-1; FAB HRMS for C19H23N3O4 (M + H+) calcd
357.1689, found 357.1676.
cake is washed once with heptane (50 mL). As determined by HPLC
assay, 1.00 g (7.0%) of 15 is lost to the combined filtrate and
washes. The wet cake is dried in a vacuum oven at 50 °C and 20
mmHg for 48 h to afford the product 15 (20.58 g, 97.0% purity,
91.2% purity adjusted yield) as an off-white crystalline solid: mp
1
92 °C; H NMR (400 MHz, DMSO-d6) δ 9.13 (s, 1H), 7.74 (d, J
) 3.0 Hz, 1H), 7.33 (m, 5H), 7.07 (dd, J ) 8.9, 3.0 Hz, 1H), 6.73
(d, J ) 8.9 Hz, 1H), 5.10 (s, 2H), 3.49 (br s, 4H), 3.29 (m, 4H);
13C NMR (100 MHz, DMSO-d6) δ 153.8, 152.5, 146.1, 136.3,
133.6, 127.9, 127.4, 127.1, 124.9, 108.3, 66.1, 45.9, 43.2; FTIR
(microscopic technique) 3350, 2850, 1675, 1490, 1440, 1365, 1270,
1240, 1130, 935, 760, 700 cm-1; FAB HRMS for C17H20N3O3 (M
+ H+) calcd 314.1505, found 314.1518.
1-â-(4-(5-Hydroxypyridin-2-yl)-piperazine-1-carboxylic acid
benzyl ester)-2,3,4-tri-O-acetyl-D-glucuronic Acid Methyl Ester
(16). To a 200 mL flame-dried Schlenk flask under an atmosphere
of N2 is added trichloroacetimidate 7 (19.16 g, 40.0 mmol, 2.0
equiv) and 15 (6.27 g, 20.0 mmol, 1.0 equiv). Toluene (50 mL) is
added and the solvent is removed in vacuo to azeodry the solids.
After drying for 1 h under vacuum, CH2Cl2 (192 mL) is added and
the solution is cooled to -60 °C. BF3‚OEt2 (10.15 mL, 80.1 mmol,
4.0 equiv) is added and the reaction is allowed to warm to -20 °C
over 1 h. The reaction is then warmed from -20 to 23 °C over 16
h at approximately 3 deg per hour and then stirred for an additional
24 h at ambient temperature. When 99% of 15 has been consumed
as determined by HPLC assay, the reaction mixture is poured into
50% saturated aqueous NaHCO3 solution (200 mL) to quench the
remaining BF3‚OEt2. The mixture is transferred to a separatory
funnel and diluted with CH2Cl2 (100 mL), and the layers are
separated. The aqueous layer is washed once with CH2Cl2 (50 mL).
The combined organic layers are washed once with 50% saturated
NaCl solution (200 mL), the layers are separated, and the aqueous
layer is washed once with CH2Cl2 (50 mL). The combined organic
layers are dried with Na2SO4, filtered, concentrated, and redissolved
in EtOAc. As determined by HPLC assay, the EtOAc solution
contains 9.40 g (74.7%) of 16. The solution is concentrated and
the residue purified by silica gel chromatography (100% hexanes
grading to 55% EtOAc in hexanes in 5% increments) to afford the
product 16 (11.86 g, 79.3% purity, 74.7% purity adjusted yield) as
an off-white foam that is taken on to the next step without further
purification: 1H NMR (400 MHz, CD3OD) δ 7.92 (d, J ) 3.0 Hz,
1H), 7.34 (m, 6H), 6.79 (d, J ) 9.2 Hz, 1H), 5.42 (t, J ) 9.4 Hz,
1H), 5.24 (d, J ) 7.8 Hz, 1H), 5.18 (t, J ) 9.6 Hz, 1H), 5.14 (m,
3H), 4.41 (d, J ) 10.0 Hz, 1H), 3.71 (s, 3H), 3.60 (br s, 4H), 3.44
(m, 4H), 2.07 (s, 3H), 2.02 (s, 3H), 2.01 (s, 3H); 13C NMR (100
MHz, DMSO-d6) δ 168.8, 168.5, 168.4, 166.4, 155.0, 153.8, 145.0,
136.3, 136.2, 127.9, 127.5, 127.4, 127.1, 107.7, 98.4, 70.9, 70.7,
70.3, 68.8, 66.1, 52.5, 45.1, 43.0, 20.5, 20.4, 20.3; FTIR (micro-
scopic technique) 2950, 1755, 1700, 1485, 1430, 1370, 1225, 1125,
1075, 1045, 980 cm-1; FAB HRMS for C30H36N3O12 (M + H+)
calcd 630.2299, found 630.2316.
1-â-(4-(5-Hydroxypyridin-2-yl)piperazine)-2,3,4-tri-O-acetyl-
D-glucuronic Acid Methyl Ester (20). To a hydrogenation vessel
is added 16 (9.40 g, 11.86 g of 79.3% purity, 14.9 mmol) followed
by MeOH (94 mL). The reactor is purged with N2 and 5 wt % Pd
on carbon (940 mg) is added. The reactor is pressurized to 40 psi
with H2. After 8 h at ambient temperature and 99% conversion of
16 as determined by HPLC assay, the reaction is filtered and the
wet cake washed twice with MeOH (10 mL portions). The organic
solution is concentrated and the residue is purified by silica gel
chromatography (100% CH2Cl2 grading to 5% MeOH in CH2Cl2
in 1% increments followed by grading to 20% MeOH in CH2Cl2
in 5% increments) to afford the product 20 (6.45 g, 96.0% purity,
83.7% purity adjusted yield) as a yellow foam that is taken on to
the next step without further purification: 1H NMR (400 MHz,
CD3OD) δ 7.92 (d, J ) 3.0 Hz, 1H), 7.35 (dd, J ) 9.1, 3.0 Hz,
1H), 6.78 (d, J ) 9.2 Hz, 1H), 5.42 (t, J ) 9.4 Hz, 1H), 5.24 (d,
J ) 7.8 Hz, 1H), 5.19 (t, J ) 9.6, 1H), 5.15 (dd, J ) 9.5, 7.8 Hz,
1H), 4.42 (d, J ) 9.9 Hz, 1H), 3.72 (s, 3H), 3.44 (m, 4H), 3.96 (m,
4-(5-Hydroxypyridin-2-yl)piperazine-1-carboxylic Acid Ben-
zyl Ester (15). A 500 mL round-bottom flask is charged with 19
(25.00 g, 69.9 mmol) followed by THF (50 mL), aqueous 2 M
HCl (110 mL), and more THF (55 mL). The reaction is heated to
55 °C for 2 h and is complete when all of 19 is consumed as
determined by HPLC analysis. After cooling to room temperature,
the reaction mixture is poured into saturated aqueous NaHCO3
solution (300 mL) and stirred for 15 min. The mixture is transferred
to a separatory funnel, CH2Cl2 (300 mL) is added, and the layers
are separated. The aqueous layer is washed once with 100 mL of
CH2Cl2. The combined organics are dried with Na2SO4, filtered,
and concentrated. The residue is transferred to a 250 mL round-
bottom flask with EtOAc (100 mL) and concentrated to an oil.
EtOAc (50 mL) is added to the residue and the resulting suspension
is stirred for 16 h, during which time crystals may form. Heptane
(150 mL) is added over 30 min to precipitate out the product. The
suspension is cooled to 0 °C for 1 h and then filtered. The wet
8382 J. Org. Chem., Vol. 71, No. 22, 2006

Synthesis of the Glucuronide Metabolite of ABT-724
4H), 2.07 (s, 3H), 2.02 (s, 3H), 2.01 (s, 3H); an rOe experiment
showed significant cross-peaks between the anomeric proton, H3,
and H5; 13C NMR (100 MHz, CD3OD) δ 170.7, 170.5, 170.3, 168.3,
157.4, 147.2, 137.9, 129.4, 109.3, 100.9, 73.1, 72.9, 72.4, 70.6,
53.3, 47.2, 45.9, 20.7, 20.7, 20.6; FTIR (microscopic technique)
2950, 2830, 1760, 1490, 1380, 1225, 1140 cm-1; FAB HRMS for
C22H30N3O10 (M + H+) calcd 496.1931, found 496.1917.
equiv) is added. The reaction is diluted with H2O (60 mL) and
CHCl3 (120 mL) and transferred to a separatory funnel. The 500
mL round-bottom flask is rinsed with additional H2O (60 mL) and
CHCl3 (120 mL), which are also transferred to the separatory funnel.
The layers are separated. The aqueous layer is washed with
chloroform (240 mL). As determined by HPLC assay, no product
3 is lost to the CHCl3 washes. The aqueous layer is transferred to
a 250 mL round-bottom flask. The separatory funnel is rinsed four
times with MeOH (10 mL portions) and the rinses transferred to
the 250 mL round-bottom flask. DOWEX 550A OH anion-
exchange resin (30.0 g) is added and the reaction is stirred at
ambient temperature for 18 h. The resin is collected by filtration
and washed sequentially with MeOH (100 mL) and H2O (100 mL).
As determined by HPLC assay, 0.017 g (0.7%) of 3 is lost to the
combined filtrate and washes. The resin is washed from the filter
into a 500 mL round-bottom flask twice with H2O (60 mL portions)
and twice with acetic acid (60 mL portions). The reaction is stirred
for 74 h at ambient temperature and then filtered. The resin is
washed twice with H2O (100 mL portions) and the combined filtrate
and washes contained 2.13 g (91.4%) of product 3. The acetic acid
and water are removed by azeotropic distillation with iPrOH and
heptane at 25 Torr and not more than 45 °C. The resulting solids
are dried for 14 d at 75 °C and 20 mmHg to give 3 (2.33 g, 86.8%
purity, 86.8% purity adjusted yield) as an amorphous tan powder.
The material is further purified by trituration with 30 mL of MeOH
1-â-Hydroxy-ABT-724-2,3,4-tri-O-acetyl-D-glucuronic Acid
Methyl Ester (22). A 100 mL round-bottom flask is charged with
20 (6.19 g, 6.45 g of 96.0% purity, 12.5 mmol, 1.0 equiv) followed
by NMP (31 mL). The solution is cooled to 0 °C and 2-(chloro-
methyl)benzimidazole 21 (2.08 g, 12.5 mmol, 1.0 equiv) is added.
The reaction is warmed to ambient temperature. After 1 h, the
reaction is cooled to 0 °C and triethylamine (1.74 mL, 12.5 mmol,
1.0 equiv) is added. After 30 min, the reaction is warmed to ambient
temperature. Ninety minutes after reaching ambient temperature
more triethylamine (0.35 mL, 2.5 mmol, 0.20 equiv) is added. The
reaction is stirred for an additional 15 h at ambient temperature.
When 95% of 21 has been consumed as determined by HPLC assay,
the reaction is diluted with H2O (120 mL), transferred to a
separatory funnel, and diluted with EtOAc (120 mL). The layers
are separated and the organic layer is washed first with 10%
saturated aqueous NaCl solution (100 mL) and then 5% saturated
aqueous NaCl solution (100 mL). As determined by HPLC assay,
0.28 g (3.6%) of 15 is lost to the combined aqueous washes. The
organic layer is dried with Na2SO4 and filtered. As determined by
HPLC assay, the EtOAc solution contains 6.04 g (77.3%) of product
22. The solution is concentrated and the residue purified by silica
gel chromatography (100% EtOAc to 5% MeOH in EtOAc in 1%
increments) to afford the product 22 (6.15 g, 93.4% purity, 73.5%
purity adjusted yield) as an off-white solid. To the solids is added
EtOAc (48 mL). After 6 h the suspension is filtered, washed once
with 25% EtOAc in hexanes (48 mL), and dried at 50 °C and 20
mmHg for 3 d to give 22 (4.93 g, 100% purity, 63.0% isolated
1
to give material of 95.0% purity: mp 190 °C dec; H NMR (400
MHz, pyridine-d5) δ 8.50 (d, J ) 2.9 Hz, 1H), 7.85 (m, 2H), 7.60
(dd, J ) 9.1, 3.0 Hz, 1H), 7.33 (m, 2H), 6.58 (d, J ) 9.2 Hz, 1H),
5.58 (d, J ) 7.3 Hz, 1H), 4.78 (d, J ) 9.7 Hz, 1H), 4.71 (t, J ) 9.0
Hz, 1H), 4.44 (t, J ) 9.7 Hz, 1H), 4.40 (d, J ) 9.3 Hz, 1H), 3.97
(s, 2H), 3.45 (m, 4H), 2.60 (m, 4H); 13C NMR (100 MHz, CD3-
CO2D) δ 174.4, 152.2, 146.8, 145.4, 135.4, 132.7, 130.4, 125.3,
115.5, 112.2, 101.2, 76.3, 75.1, 73.5, 72.2, 52.6, 52.5, 45.3; FTIR
(microscopic technique) 3300, 2850, 1605, 1495, 1460, 1410, 1245,
1070 cm-1; FAB HRMS for C23H28N5O7 (M + H+) calcd 486.1989,
1
yield from 20) as a semicrystalline white powder: mp 93 °C; H
found 486.2007; [R]23 -45.5 (c 2.099, DMF).
NMR (400 MHz, CD3OD) δ 7.90 (d, J ) 2.9 Hz, 1H), 7.53 (br s,
2H), 7.33 (dd, J ) 9.1, 3.0 Hz, 1H), 7.21 (m, 2H), 6.77 (d, J ) 9.2
Hz, 1H), 5.41 (t, J ) 9.4 Hz, 1H), 5.22 (d, J ) 7.8 Hz, 1H), 5.18
(t, J ) 9.7 Hz, 1H), 5.14 (dd, J ) 9.3, 7.8 Hz, 1H), 4.41 (d, J )
9.9 Hz, 1H), 3.84 (s, 2H), 3.71 (s, 3H), 3.49 (m, 4H), 2.65 (m,
4H), 2.07 (s, 3H), 2.02 (s, 3H), 2.01 (s, 3H); 13C NMR (100 MHz,
CD3OD) δ 168.8, 168.5, 168.3, 166.4, 155.3, 150.9, 144.8, 142.4,
136.3, 133.9, 127.4, 121.4, 120.5, 118.0, 110.8, 107.5, 98.4, 70.9,
70.7, 70.3, 68.8, 55.7, 52.5, 52.5, 45.2, 20.4, 20.4, 20.3; FTIR
(microscopic technique) 3400, 2950, 2830, 1755, 1490, 1455, 1410,
1385, 1230, 1095, 1055 cm-1; FAB HRMS for C30H36N5O10 (M
D
Acknowledgment. We would like to thank Dr. John
Darbyshire of Abbott Drug Metabolism for synthesis of initial
milligram quantities of glucuronide 3 through enzymatic means,
as well as Stephen Latshaw for the first isolation of glucuronide
3 by preparative HPLC chromatography. We would also like
to thank Brian Kotecki of the Abbott Gas Reactions Lab for
performing all reactions involving hydrogen gas, Dr. Stanley
Agre of the Abbott Structural Chemistry Department for
obtaining IR data, Tanveer Ahmed and Dr. Paul West of the
Abbott Structural Chemistry Department for obtaining high-
resolution MS data, Dr. Steven Hollis of the Abbott Structural
+ H+) calcd 626.2462, found 626.2469; [R]23 -16.7 (c 2.235,
D
MeOH).
1-â-Hydroxy-ABT-724-D-glucuronic Acid (3). A 500 mL
round-bottom flask is charged with 22 (3.00 g, 4.80 mmol, 1.0
equiv) followed by MeOH (120 mL). The solution is cooled to
-10 °C, lithium hydroxide monohydrate (0.0201 g, 0.480 mmol,
0.10 equiv) is added, and the reaction is gradually warmed to
ambient temperature over 1 h. After 4 h at ambient temperature
and when greater than 99% of 22 has been consumed as determined
by HPLC assay, additional lithium hydroxide monohydrate (0.382
g, 9.12 mmol, 1.9 equiv) is added at ambient temperature. After
19 h and when greater than 99% of 23 has been consumed as
determined by HPLC assay, acetic acid (0.27 mL, 9.6 mmol, 2.0
1
Chemistry Department for obtaining H NMR data, and Dr.
Marius Naris of Abbott Process Analytical Chemistry for
obtaining melting point data.
1
Supporting Information Available: Copies of H NMR and
13C NMR spectra for compounds described in the Experimental
Section. This material is available free of charge via the Internet at
http://pubs.acs.org.
JO0611972
J. Org. Chem, Vol. 71, No. 22, 2006 8383