Multiple strategies for the preparation of a sulfur-35 labeled NPC1L1 radioligand

Article abstract of DOI:10.1016/j.bmcl.2009.07.051

During our effort to design a receptor binding assay to aid in the elucidation of the molecular mechanism of ezetimibe, we prepared a sulfur-35 containing radioligand which exhibits improved potency over the glucuronide conjugate of ezetimibe in both nati

Full text of DOI:10.1016/j.bmcl.2009.07.051

Bioorganic & Medicinal Chemistry Letters 19 (2009) 5033–5036
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Multiple strategies for the preparation of a sulfur-35 labeled NPC1L1 radioligand
a
a
b
b
Joseph P. Simeone a, , Matthew P. Braun , Joseph F. Leone , Peter Lin , Robert J. DeVita ,
*
Margarita Garcia-Calvo c, Herb G. Bull c, JeanMarie Lisnock c, Dennis C. Dean d
a Department of Process Research, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065-0900, USA
b Department of Medicinal Chemistry, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065-0900, USA
c Department of Metabolic Disorders, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065-0900, USA
d Department of Drug Metabolism and Pharmacokinetics, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065-0900, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
During our effort to design a receptor binding assay to aid in the elucidation of the molecular mechanism
of ezetimibe, we prepared a sulfur-35 containing radioligand which exhibits improved potency over the
glucuronide conjugate of ezetimibe in both native enterocyte brush border membranes and membranes
from cells expressing recombinant NPC1L1. Herein, we describe the different synthetic strategies which
were used to obtain this compound as well as its effectiveness in the aforementioned assay.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 19 June 2009
Revised 6 July 2009
Accepted 8 July 2009
Available online 22 July 2009
Keywords:
Sulfur-35
NPC1L1
Radioligand
Synthesis
Ezetimibe is an effective therapeutic for the reduction of low
density lipoproteins and functions through the inhibition of cho-
lesterol absorption from the small intestine into the bloodstream.
The development of ezetimibe (1, Fig. 1) is uncommon in the mod-
ern drug discovery era in that its structure was optimized based on
in vivo efficacy and active metabolite identification in the absence
of an in vitro assay.1 Almost immediately after oral administration
in rodents, ezetimibe forms a glucuronide conjugate at the pheno-
lic hydroxyl group.2 This metabolite, ezetimibe glucuronide (2,
Fig. 1) has been shown to have comparable potency to ezetimibe
itself.3 Furthermore, glucuronide formation in humans leads to
enterohepatic circulation which localizes ezetimibe glucuronide
at the site of action, the small intestine.4
The mechanism by which ezetimibe regulates cholesterol flux
at the molecular level is becoming evident. Recent studies have re-
sulted in an increased understanding of intestinal cholesterol
absorption. The identification of Niemann-Pick C1 Like 1 Protein
(NPC1L1)5 was followed shortly thereafter by a study involving
NPC1L1-deficient mice which demonstrated that NPC1L1 is in-
volved in cholesterol uptake in enterocytes.6 Evidence that this
protein is the molecular target of ezetimibe was provided by the
development of a binding assay and the finding that tritiated eze-
timibe glucuronide binds to a single site in brush border mem-
branes (BBMs) and does not bind to BBMs prepared from NPC1L1
knockouts.7
Since these findings, further studies aimed at probing the mech-
anism by which NPC1L1 is involved in cholesterol transport have
been reported.8 These include experiments utilizing cell lines9
expressing recombinant and endogenous NPC1L1. Recently re-
ported data provides evidence that NPC1L1 possesses binding sites
for both ezetimibe and cholesterol and that ezetimibe binding may
force a conformational change in NPC1L1 which prevents its inter-
action with cholesterol.10 Furthermore, it has been shown that
cholesterol binding to NPC1L1 causes an association with the
clathrin/AP2 complex, resulting in internalization of both choles-
terol and NPC1L1. Ezetimibe acts by inhibiting this process.11
Competitive binding studies often require the synthesis of li-
gands containing a high specific activity radiolabel. This label must
be carefully chosen so as to ensure that the binding of the resulting
radioligand is both specific and of high affinity. In cases where
incorporation of tritium or iodine-125 does not meet these
requirements, sulfur-35 may be used as a surrogate.12 Sulfur-35
is most often introduced into a molecule in the form of a sulfon-
amide, via the treatment of an amine with methane[35S]sulfonyl
chloride.13 While this method is both general and efficient, it can
sometimes be problematic when the amine of interest suffers from
poor solubility in methylene chloride. These cases often give rise to
low conversions to [35S]sulfonamides and large amounts of meth-
ane[35S]sulfonic acid, arising from hydrolysis of the sulfene inter-
mediate.14 While the labeling reagent can sometimes be recycled,
it is a time consuming process often resulting in low recovery.
During the course of our effort to elucidate the mode of action
of ezetimibe we designed a sulfur-35 labeled radioligand (3,
* Corresponding author. Tel.: +1 732 594 4307; fax: +1 732 594 6921.
E-mail address: joe_simeone@merck.com (J.P. Simeone).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.07.051

5034
J. P. Simeone et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5033–5036
OH
OH
O
OH
OH
OH
OH
O
N
O
F
O
OH
N
F
O
F
1
2
F
OH
O
OH
OH
OH
3H
O
OH
OH
OH
O
OH
O
O
OH
N
O
3H
F
O
OH
N
F
O
NH35SO2CH3
F
4
3
Figure 1. Structure of ezetimibe and its analogs.
Fig. 1) which exhibits increased binding affinity over ezetimibe
glucuronide across species. However, the preparation of this com-
pound was complicated by its lack of solubility in chlorinated sol-
vents. To circumvent this problem, we have successfully employed
two novel synthetic methods for the synthesis of sulfur-35 con-
taining compounds. The first procedure involves a low mass palla-
dium-mediated Sonogashira-type coupling reaction of an aryl
iodide with a labeled propargylamine sulfonamide. Alternately, it
can be efficiently prepared via incubation of its phenol precursor
in the presence of dog liver microsomes.
Hydrolysis of the methyl ester resulted in the desired radioligand.
Unfortunately, this process proved to be unreliable, resulting in vari-
able radiochemical yields due to the limited solubility of 5 in haloge-
nated solvents. Several modifications of this route were attempted
including the use of different bases and various protected analogs of
5. However, these changes failed to result in a more reliablesynthesis.
In order to satisfy the need for additional quantities of this radioli-
gand, the design of an alternate route was necessary.
We envisioned that the molecule could be assembled via the
Sonogashira cross-coupling16 of labeled propargylamine sulfon-
amide and the appropriate iodoarene precursor (Scheme 2). Palla-
dium-mediated chemistry using sulfur-35 labeled substrates was
unknown at the time of this work.17 We were uncertain as to
whether the presence of palladium at concentrations many times
in excess of a sulfur containing compound would be detrimental be-
cause of the affinity of sulfur for palladium compounds.18 However,
we were pleased to find that compound 8 could be successfully
cross-coupled with iodoarene 915 by utilizing the copper-free condi-
tions developed by Leadbeater.19 Standard protocol using tetraki-
striphenylphosphine palladium (0) and copper(I) iodide resulted in
a muchlonger reaction time. Sulfur-35 propargylamine sulfonamide
8 could be prepared from propargylamine and methane[35S]sulfonyl
chloride. Removal of the methyl ester followed by HPLC purification
yielded the desired radioligand 3. This process was used to success-
fully prepare several batches and consistently resulted in an average
overall radiochemical yield of 20%. The specific activity ranged from
800 to 1100 Ci/mmol as determined by LC/MS.
Initial attempts to develop a binding assay to identify the target
of ezetimibe focused on the use of native rat intestinal brush bor-
der membranes. Four radiotracers were prepared to be used as li-
gands in the assay. These included tritiated and I-125 containing
ezetimibe and ezetimibe glucuronide. The ezetimibe analogs were
quickly shown to be unstable in aqueous and slightly basic buffers.
In addition, [125I]ezetimibe was a very poor radioligand resulting in
high background binding. Therefore, tritiated ezetimibe (4, Fig. 1)
was initially used to characterize the binding across species.7
When compound 3 was prepared as a precursor to a photoaffinity
ligand, it was found to have increased potency in this assay. Fur-
thermore, it was eventually shown to have a high affinity for re-
combinant NPC1L1 across species, including the mouse NPC1L1
receptor for which it is 27-fold more potent than tritiated ezetim-
ibe glucuronide (Table 1).
The synthesis of unlabeled 3 has been previously reported.15 The
radiolabeled analog of this compound was initially prepared using
standard chemistry (Scheme 1). In the presence of base, compound
While the Sonogashira chemistry proved to be a reliable route
for the preparation of the radioligand, we found that the final HPLC
purification was difficult because of the multicomponent reaction
5
15 was treated with a dichloromethane solution of methane[35S]sul-
fonyl chloride which was prepared from methane[35S]sulfonic acid.13
Table 1
NPC1L1 binding affinities across species expressed as IC50 (nM)
Compound
Human BBM
Rat BBM
Human NPC1L1
Rat NPC1L1
Mouse NPC1L1
3
4
4
150
4
127
4
200
4
173
44
1200
Mucosal intestinal membranes and membranes from human embryonic kidney (HEK) cells that have been transiently transfected with DNA constructs for expression of
human, rat, or mouse NPC1L1 were prepared as described earlier.7 Single-tube binding filtration assays were conducted in a final volume of 20
L as reported previously using
tritiated compound 4.7 Competition studies of 35[S]-compound 3 binding by unlabeled compound 3 and ezetimibe glucuronide (12 concentrations) were conducted by using
1–5 nM 35[S]-compound 3 and 5.7
g/assay for human BBM, 0.3 g for human NPC1L1, 1.5 g for rat BBM, 1.3 g for rat NPC1L1, and 15 g for mouse NPC1L1 in 30 mM
l
l
l
l
l
l
Hepes/117 mM NaCl/5.4 mM KCl, pH 7.4 containing 0.03% sodium taurocholate and 0.05% digitonin for maximal detection of binding sites in the samples. The values are the
average of at least five independent experiments run in triplicates.

J. P. Simeone et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5033–5036
5035
OH
OH
O
OH
OH
O
OH
OH
OH
OH
O
O
a
O
O
OCH3
OCH3
N
N
F
F
O
O
NH2
NH35SO2CH3
6
5
OH
O
OH
OH
OH
O
O
b, c
OH
N
F
O
NH35SO2CH3
3
Scheme 1. Reagents and conditions: (a) methane[35S]sulfonyl chloride, pyridine, CH2Cl2, room temperature, 15 min; (b) Et3N, CH3OH, H2O, room temperature, 1 h; (c) HPLC
purification.
OH
O
OH
OH
OH
O
NH2
NH35SO2Me
a
O
OCH3
N
H
H
F
O
7
8
9
I
OH
O
OH
OH
OH
O
O
b,c,d
OH
N
F
O
NH35SO2CH3
3
Scheme 2. Reagents and conditions: (a) methane[35S]sulfonyl chloride, pyridine, CH2Cl2, room temperature, 15 min; (b) Pd(PPh3)2Cl2, Et3N, DMF, 70 °C, 15 min; (c) Et3N,
CH3OH, H2O, room temperature, 1 h; (d) HPLC purification.
mixture. We therefore turned our attention to developing an alter-
nate strategy which would allow us to prepare multi-millicurie
amounts without the need for tedious purification. Installation of
the glucuronide conjugate onto a radiolabeled precursor was con-
sidered as a way to circumvent this problem.
In addition to chemical methods, phenolic glucuronides can be
prepared via enzymatic catalysis.20 This method is limited to mil-
ligram amounts of the phenol substrate and is often low yielding.
Because of this, enzymatic methods of preparation are most often
utilized in metabolite identification studies. We reasoned that a
sulfur-35 containing compound would be amenable to enzymatic
catalysis because of the extremely low mass present.
A suitable glucuronidation precursor (Scheme 3) was available
in the form of the diol protected propargyl amine derivative 10.21
This compound was converted to the [35S]sulfonamide 11. After
tert-butyldimethylsilyl (TBS) deprotection, phenol 12 was incu-
bated in the presence of dog liver microsomes with UDPGA as
the glucoronyl donor and saccharic acid lactone as a b-glucuroni-
dase inhibitor.
This process was repeated several times and each incubation
proceeded smoothly to give radiochemical yields as high as 67%.
It was, however, necessary to purify the phenol 12 by reverse
phase HPLC before subjecting it to the incubation conditions or
the yield was significantly lower. High specific activity tracers
(699 to 1100 Ci/mmol) could be prepared in this manner. It was
necessary to run multiple 2 mL incubations if the total radioactiv-
ity exceeded 5 mCi because repeated attempts to run 10 and
20 mCi incubations resulted in markedly slower reaction rates
and low turnover. To avoid this problem, incubations were run in
parallel and then combined for isolation and purification.
While the average overall radiochemical yield for this process
was similar to the Sonogashira route (17%), the purification of
the final compound was easier due to the clean conversion of the
phenol to the glucuronide. Batches of the radioligand in excess of

5036
J. P. Simeone et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5033–5036
OTBS
OTBS
OTBS
OTBS
a
N
N
F
F
O
O
11
NH35SO2CH3
10
NH2
OH
OH
b,c
d,e
N
3
F
O
NH35SO2CH3
12
Scheme 3. Reagents and conditions: (a) methane[35S]sulfonyl chloride, pyridine, CH2Cl2, room temperature, 15 min; (b) 48% aq HF, CH3CN, rt, 6 h; (c) HPLC purification (d)
dog liver microsomes, bis-tris buffer pH 7.1, 0.1 M MgCl2, 20 mM UDPGA, 0.1 M saccharic acid lactone, alamethicin, CH3CN, water, 37 °C, 3 h; (e) HPLC purification.
7. Garcia-Calvo, M.; Lisnock, J.; Bull, H. G.; Hawes, B. E.; Burnett, D. A.; Braun, M.
10 mCi could be prepared and stored with minimal decomposition
provided methanol was used as the storage solvent.
P.; Crona, J. H.; Davis, H. R.; Dean, D. C.; Detmers, D. A.; Graziano, M. P.; Hughes,
M.; MacIntyre, D. E.; Ogawa, A.; O’Neill, K. A.; Iyer, S. P. N.; Shevell, D. E.; Smith,
In conclusion, we have discovered a radioligandwith high affinity
for the NPC1L1 receptor across multiple species. This compound can
be used as a diagnostic tool tofurther our understanding of intestinal
cholesterol transport. Its preparation required us to expand the
scope of sulfur-35 synthetic chemistry. In doing so, we have adapted
both chemical as well as enzymatic reactions for the synthesis of a
high specific activity radioligand. This work underscores the use of
sulfur-35 as a viable alternative for iodine-125 and tritium.
M. M.; Tang, Y. S.; Makerewicz, A. M.; Ujjainwalla, F.; Altmann, S. W.; Chapman,
K. T.; Thornberry, N. A. PNAS 2005, 102, 8132.
8. (a) Knopfel, M.; Davies, J. P.; Duong, P. T.; Kvaerno, L.; Carreira, E. M.; Phillips,
M. C.; Ioannou, Y. A.; Hauser, H. Biochim. Biophys. Acta 2007, 1771, 1140; (b)
Labonte, E. D.; Howles, P. N.; Granholm, N. A.; Rojas, J. C.; Davies, J. P.; Ioannou,
Y. A.; Hui, D. Y. Biochim. Biophys. Acta 2007, 1771, 1132.
9. (a) Davies, J. P.; Scott, C.; Oishi, K.; Liapis, A.; Ioannou, Y. A. J. Biol. Chem. 2005,
280, 12710; (b) Iyer, S. P.; Yao, X.; Crona, J. H.; Hoos, L. M.; Tetzloff, G.; Davis, H.
R., Jr.; Graziano, M. P.; Altmann, S. W. Biochim. Biophys. Acta 2005, 1722, 282; (c)
Sane, A. T.; Sinnett, D.; Delvin, E.; Bendayan, M.; Marcil, V.; Menard, D.;
Beaulieu, J. F.; Levy, E. J. Lipid Res. 2006, 47, 2112; (d) Alrefai, W.; Annaba, F.;
Sarwar, Z.; Dwivedi, A.; Saksena, S.; Singla, A.; Dudeja, P. K.; Gill, R. Am. J.
Physiol. 2007, 292, G369; (e) Field, F. J.; Watt, K.; Mathur, S. N. J. Lipid Res. 2007,
48, 1735; (f) Yamanashi, Y.; Takada, T.; Suzuki, H. J. Pharmacol. Exp. Ther. 2007,
320, 559; (g) Petersen, N. H.; Faergeman, N. J.; Yu, L.; Wustner, D. J. Lipid Res.
2008, 49, 2023.
Acknowledgments
We would like to thank Dr. David Schenk, Mr. Anson Chang,
Mr. Herb Jenkins, and Ms. Rosemary Marques of Merck Research
Laboratories for analytical support.
10. Weinglass, A. B.; Kohler, M. G.; Nketiah, E. O.; Liu, J.; Schmalhofer, W.; Thomas,
A.; Williams, B.; Beers, L.; Smith, L.; Hafey, M.; Bleasby, K.; Leone, J.; Tang, Y. S.;
Braun, M.; Ujjainwalla, F.; McCann, M. E.; Kacorowski, G. J.; Garcia, M. L. Mol.
Pharm. 2008, 73, 1072.
11. Ge, L.; Wang, J.; Qi, W.; Miao, H.-H.; Cao, J.; Qu, Y.-X.; Li, B.-L.; Song, B.-L. Cell
Metab. 2008, 7, 508.
Supplementary data
12. Dean, D. C.; Ellsworth, R. L.; Chen, R.; Staskiewicz, S. J.; Melillo, D. G. In Synthesis
and Applications of Isotopically Labelled Compounds 1994; Allen, J., Ed.; John
Wiley and Sons Ltd, 1994; p 795.
13. Dean, D. C.; Nargund, R. P.; Pong, S.-S.; Chaung, L.-Y. P.; Griffin, P.; Melillo, D. G.;
Ellsworth, R. L.; Van Der Ploeg, L. H. T.; Patchett, A. A.; Smith, R. G. J. Med. Chem.
1996, 39, 1767.
Supplementary data (detailed experimental procedures for the
preparation of compounds 3, 6, 8, 11, and 12, radiochromatograms,
and mass spectra) associated with this article can be found, in the
online version, at doi:10.1016/j.bmcl.2009.07.051.
14. Truce, W. E.; Campbell, R. W. J. Am. Chem. Soc. 1966, 88, 3599.
15. Goulet, M. T.; Ujjainwalla, F.; Von Langen, D. PCT Patent Publication No.
WO2005/062824, July 14, 2005.
References
16. Chinchilla, R.; Najera, C. Chem. Rev. 2007, 107, 874.
1. Clader, J. W. J. Med. Chem. 2004, 47, 1.
17. Concurrent experiments in these laboratories resulted in the successful
utilization of sulfur-35 labeled sulfonamide derivatives in Stille and
Buchwald–Hartwig cross-coupling reactions. See: Wallace, M. A.; Raab, C.;
Dean, D.; Melillo, D. J. Label. Compd. Radiopharm. 2007, 50, 347.
18. Voskresenskaya, T. P.; Chinakov, V. D.; Nekipelov, V. M.; Mashkina, A. V. Catal.
Lett. 1986, 32, 359.
19. Leadbeater, N. E.; Tominack, B. J. Tetrahedron Lett. 2003, 44, 8653.
20. Brunelle, F. M.; Verbeeck, R. K. Biochem. Pharmacol. 1993, 46, 1953.
21. Goulet, M. T.; Ujjainwalla, F.; Ogawa, A.; Von Langen, D. US Publication No.
2005/0267049 A1, Dec 1.
2. Van Heek, M.; Farley, C.; Compton, D. S.; Hoos, L.; Alton, K. B.; Sybertz, E. J.;
Davis, H. R. Br. J. Pharmacol. 2000, 129, 1748.
3. Hawes, B. E.; O’Neill, K. A.; Yao, X.; Crona, J. H.; Davis, H. R., Jr.; Graziano, M. P.;
Altman, S. W. Mol. Pharmacol. 2007, 71, 19.
4. Patrick, J. E.; Kosoglou, T.; Stauber, K. L.; Alton, K. B.; Maxwell, S. E.; Zhu, Y.;
Statkevich, P.; Iannucci, R.; Chowdhury, S.; Affrime, M.; Cayen, M. N. Drug.
Metab. Dispos. 2002, 30, 430.
5. Davies, J. P.; Levy, B.; Ioannou, Y. A. Genomics 2000, 65, 137.
6. Altmann, S. W.; Davis, H. R., Jr.; Zhu, L. J.; Yao, X.; Hoos, L. M.; Tetzloff, G.; Iyer,
S. P.; Maguire, M.; Golovko, A.; Zeng, M.; Wang, L.; Murgolo, N.; Graziano, M. P.
Science 2004, 303, 1201.