Synthesis of 14C- and13C/2H-labeled SGLT inhibitors AVE2268 and AVE8887

Article abstract of DOI:10.1002/jlcr.1773

Isotopically labeled analogues of two structurally very similar SGLT inhibitors AVE2268 (1a) and AVE8887 (1b) have been synthesized by various routes. The radioactive labeled [¹⁴C]-AVE2268 was prepared in 5 steps including a Friedel-Crafts acylation as the key step for the ¹⁴C-label introduction. For [¹⁴C]-AVE8887 the same synthetic approach was not successful and therefore an alternative thiophene metalation/Weinreb amide sequence was developed. This pathway was also applied to obtain stable isotopically labeled analogs of both AVE2268 and AVE8887. Copyright r 2010 John Wiley & Sons, Ltd.

Full text of DOI:10.1002/jlcr.1773

Abstracts
Published online 13 May 2010 in Wiley Interscience
(www.interscience.wiley.com) DOI: 10.1002/jlcr.1773
10th International Symposium on the
Synthesis and Applications of Isotopes and
Isotopically Labelled Compounds—Synthesis
of Compounds Labelled with Long-Lived
Isotopes
Session 17, Thursday, June 18, 2009
SESSION CHAIRS: KARL CABLEa AND JOHN EASTERb
aGlaxoSmithKline, UK
bBristol-Myers Squibb, USA
Abstract: This session is a continuation of Session 1. A number of methods detailing the synthesis of tritiated compounds as well as
a discussion on the handling of tritium on a multi-Curie scale are presented. In addition, the descriptions of a number of
preparations of individual isotopically labelled compounds are detailed.
Keywords: tritium; [3H]-methyl nosylate; iridium catalysts; polyphenols; Rhodium black; Crabtree’s catalyst; Carbon-14; Deuterium;
PPAR; CXCR3; [3H]-methyl iodide; Tritec; Carbonylation; Delta-opioid receptor
THE SYNTHESIS OF TRITIUM-LABELED COMPOUNDS
CAROLEE F. LAVEY, DAVID HESK, SCOTT BORGES, DAVID KOHARSKI AND PAUL MCNAMARA
Schering-Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ, 07033, USA
Abstract: Tritium-labeled compounds are synthesized frequently in our group in order to assess metabolism. Tritium is the label of
choice because these compounds can be synthesized rapidly in most circumstances. Tritium gas and tritiated water are common
label sources, in addition to other commercially available tritiating reagents. This paper will focus on recent uses of sodium
borotritide, [3H]methyl nosylate, and iridium catalysts with tritium gas or tritiated water in our lab.
Keywords: tritium; sodium borotritide; [3H]methyl nosylate; iridium catalysts; tritium gas; tritiated water
Introduction: A number of tritium-labeled compounds are synthesized each year in our lab using a variety of approaches. Because of
the need for rapid turn-around, simple labeled starting materials are used. This paper will explore four methods that we commonly use in
order to label compounds with tritium. These methods include: 1) [3H]methyl nosylate, 2) the use of iridium catalysts for ortho-directed
exchange (specifically Ir(COD)(F6AcAc) with tritiated water [THO] and Ir(Ph3P)2(COD)BF4 with tritium gas [T2]), 3) sodium borotritide used
in reduction and dehalogenation reactions, and 4) the use of sodium borotritide/palladium acetate in dehalogenation reactions.
Results and Discussion: [3H]methyl nosylate
We have found [3H]methyl nosylate ([3H]methyl 4-nitrobenzenesulfonate) to be a very versatile and ‘user-friendly’ reagent for
oxygen, nitrogen, and sulfur methylation reactions. The compound is supplied at a specific activity of approximately 85 Ci/mmol and
is stored as a (1:3) ethyl acetate : hexanes solution at À801C. We have found that the material is reasonably stable when stored at
À801C. However, it can be easily purified using a small silica gel column and (1:1) hexane : dichloromethane as eluent, if necessary.
For example, the radiochemical purity (RCP) of a batch of [3H]methyl nosylate was 78% after 1.5 years of storage. Following
purification as described above, the RCP was 97%.
Two examples of the use of [3H]methyl nosylate to synthesize high specific activity tritium-labeled ligands are shown in Figure 1.
In the first example, [3H]methyl nosylate (50 mCi, 85 Ci/mmol) was reacted with approximately four equivalents of the phenol in
dimethylformamide (DMF) using sodium tert-pentoxide as base. The RCP of product in the reaction mixture was 66% after 3 h.
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.

Abstracts
Following purification, [3H]Sch A (22.1 mCi) at a specific activity of 85.2 Ci/mmol resulted. In the second example, [3H]methyl nosylate
(55.2 mCi, 85 Ci/mmol) was reacted with approximately four equivalents of the phenol in acetone/DMF using cesium carbonate as
base. The RCP of product in the reaction mixture was 85% after 2 h. Following purification, [3H]Sch B (46.8 mCi) at a specific activity
of 86.6 Ci/mmol resulted.
OH
OH
MeO
MeO
N
N
Cl
Cl
1. Sodium tert-pentoxide,
O
O
NMe2
DMF
NMe2
O
O
N
S
N
S
O
O
O
OMe
O
OMe
2. [3H]Methyl nosylate
OCT3
OH
[3H]Sch A
N
N
N
N
N
OCT3
OH
Cs2CO3, acetone, DMF
[3H]Methyl nosylate
N
N
O
O
S
S
N
[3H]Sch B
Figure 1. The use of [3H]methyl nosylate to synthesize high specific activity tritium-labeled compounds.
Iridium catalysts and THO or T2
Crabtree’s catalyst [Ir(Py)(PCy3)(COD)PF6] is frequently used in our lab for ortho-directed tritium-hydrogen exchange. There are
times, however, when either tritium incorporation is lower than we would like or the substrate is not soluble in dichloromethane,
the only solvent in which the catalyst is soluble. At these times, we must screen other iridium catalysts for their reactivity and
specificity.
The synthesis of Ir(COD)(F6AcAc) and an example of the use of this catalyst in our lab is shown in Figure 2. Although this catalyst is
now commercially available, at the time it was synthesized according to a literature procedure1 using bis(1,5-cyclooctadiene)diir-
idium(I) dichloride and hexafluoro 2,4-pentanedione. The catalyst can be used for ortho-directed labeling of aromatics functionalized
with nitrogen heterocycles using either THO or T2 in either DMF or dimethylacetamide (DMA)1,2. In the example shown, an oxazole
containing substrate (20 mg) was tritiated using Ir(COD)(F6AcAc) (10 mg) and THO (500 mCi, 50 Ci/mmol) in DMA, resulting in 32 mCi
of crude product 1 with a RCP of 27%. Although the tritium incorporation was modest, the results were reproducible so that an
adequate amount of the desired compound could be made.
O
O
CF3
Cl
Cl
O
O
F3C
CF3
Ir
Ir
Ir
ether, NaOH/H2O
CF3
OMe
N
OMe
N
N
CF3
N
CF3
Ir[COD(F6AcAc)]
THO, DMA, 106 oC
T
O
O
R'
R"
R'
R"
1
Figure 2. The synthesis of Ir(COD)(F6AcAc) and its use as a catalyst for tritiation.
When we needed to incorporate tritium into compound 2 (‘R’ = H, ‘acid sensitive group’ = CN, Figure 3), our first choice was
to use Ru(PPh3)3Cl2 and THO (500 mCi) in dioxane at 1201C, but there was very little tritium incorporation. When harsher
conditons were used (the reaction was run in DMSO at 1501C), the compound decomposed. Crabtree’s catalyst was tried next with
deuterium gas, with the results that when 1 eq, 0.25 eq, and 2 eq of catalyst were used, there was a small amount of
deuterium incorporation, no deuterium incorporation, and compound decomposition, respectively. Another labeling method
needed to be found.
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
PPh3, AgBF4, CH2Cl2
acid sensitive group
[Ir(COD)Cl]2
Ir(PPh3)2(COD)BF4
acid sensitive group
T
41%
T
This catalyst was used to label
59%
N
N
N
N
R
R
F
F
2
Figure 3. The synthesis of bis(triphenylphosphine)(cyclohexadiene)iridium (I) tetrafluoroborate.
The bis(triphenylphosphine)(cyclooctadiene)iridium(I) tetrafluoroborate catalyst was synthesized according to the literature
procedure3 as shown in Figure 3. This catalyst has been found useful in the incorporation of hydrogen isotopes alpha to
heteroatom-containing functional groups3. When the catalyst (1 eq) was used with substrate 2 (‘R’ = H) and deuterium gas, there was
better deuterium incorporation than when Crabtree’s catalyst was used. However, when the reaction was duplicated using tritium
gas, 40 mCi of crude product with a RCP of 8% resulted. Although the UV profile of the reaction mixture looked clean, there were
many radioactive impurities, suggesting that the conditions needed to be optimized further. When compound 2 (‘R’ = tert.-BOC) was
treated with catalyst (0.4 eq) and deuterium gas, there was a lot of deuterium incorporation. This reaction was repeated using tritium
gas (850 mCi) to give 155 mCi of crude product at a RCP of 54%. Tritium NMR showed that tritium distribution in the final compound
was as shown in Figure 3. This example illustrates that both varying the catalyst and the ratio of substrate to catalyst will affect the
results of the experiment, with the outcome dependent on the nature of the substrate.
Sodium Borotritide
Sodium borotritide has been used in our lab for ketone reductions and dehalogenation reactions. It has also been used
with palladium acetate for selective dehalogenation reactions in the presence of other reduction-sensitive functional groups.
Examples of each of these reactions will be discussed. In general, the sodium borotritide is supplied in a small amber ampoule. The
ampoule is cracked open and the reaction is run in the ampoule to minimize sodium borotritide loss and personnel/hood
contamination.
A typical ketone reduction reaction using sodium borotritide, for the eventual synthesis of compound 3, is shown in Figure 4.
Benzophenone was treated with sodium borotritide (890 mCi was accounted for in a vial containing ‘500 mCi;’ specific
activity = 50 Ci/mmol) to give 762 mCi of [3H]diphenylmethanol. To complete the synthesis of compound 3, [3H]chlorodiphenyl-
methane was synthesized by treating the alcohol with thionyl chloride, and the crude [3H]chloro compound (80% RCP) was used in
the Finkelstein reaction with a substituted piperazine to give 488 mCi of compound 3 with a RCP of 59%.
O
T
OH
T
Cl
[3H]NaBH4
EtOH/THF
SOCl2
toluene
"500 mCi" (890 mCi)
762 mCi
80% RCP
R
N
R
N
NH
N
T
K2CO3, KI, CH3CN, 85 oC
3, 488 mCi, 59% RCP
Figure 4. Typical ketone reduction using sodium borotritide.
The use of sodium borotritide in a dehalogenation reaction is shown in Figure 5. In this Suzuki reaction, a vinyl bromide was
treated with sodium borotritide (125 mCi, specific activity = 15 Ci/mmol) in the presence of tetrakis(triphenylphosphine)palladium(0).
There was a concern about potential side reactions with other functional groups on the molecule so the reaction did not go to
completion. However, 11.1 mCi of compound 4 was produced at a specific activity of 2.5 Ci/mmol.
R
R
[3H]NaBH4
Br
N
T
N
DMF, 85 oC
Pd(PPh3)4
R'
R'
4
Figure 5. The use of sodium borotritide in dehalogenation in tritiation.
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
Finally, we have used sodium borotritide with palladium acetate to carry out selective dehalogenation reactions. This reagent
combination is very useful because it can be used to dehalogenate aryl halides in the order of I4Br4Cl4F, double bonds and triple
bonds are not reduced, esters are not affected, and secondary hydrogen/tritium exchange does not occur4. In the reaction set-up, the
substrate is pre-treated with palladium acetate in methanol for 10 min. The resulting mixture is then added to the ampoule containing
the sodium borotritide. After one hour, the mixture is filtered and then immediately purified in order to prevent product decomposition.
An example of the use of sodium borotritide/palladium acetate is shown in Figure 6. A substrate containing an alkyne
and ester was treated with N-iodosuccinimide to give a mono- and di-iodinated product. The di-iodinated product was reacted
with the sodium borotritide (500 mCi)/palladium acetate mixture to give 21 mCi of compound 5 at a specific activity of 15.7 Ci/mmol.
R
R
R
O
O
O
NIS
+
I
O
O
O
HO
HO
HO
I
I
20%
50%
O
R
R
O
NaBT4, Pd(OAc)2
MeOH
I
T
O
O
HO
HO
I
T
500 mCi
5, 21 mCi
Figure 6. The use of sodium borotritide/palladium acetate in tritiation.
Conclusion: We use a variety of techniques in our lab to label compounds with tritium, and a few of these methods were
highlighted in this paper. To summarize, we consider [3H]methyl nosylate to be a very ‘user friendly’ reagent that can be
used to produce high specific activity ligands. The different iridium catalysts that are available complement each other in terms
of their reactivity and specificity. Sodium borotritide can be used to label compounds via reduction or dehalogenation reactions, or in
conjunction with palladium acetate for dehalogenation reactions on substrates containing other reduction-sensitive functional
groups.
Acknowledgements: The authors would like to thank Drs. T. M. Chan and Mary Senior from the Schering-Plough Research
Institute Molecular Spectroscopy group for tritium NMR analysis of synthesized compounds. Thanks are also due to SPRI Chemical
Research for some starting materials and synthetic procedures.
References
[1] M. J. Hickey, J. R. Jones, L. P. Kingston, W. J. S. Lockley, A. N. Mather, B. M. McAuley, D. J. Wilkinson, Tet. Lett. 2003, 44,
3959–3961.
[2] B. McAuley, M. J. Hickey, L. P. Kingston, J. R. Jones, W. J. S. Lockley, A. N. Mather, E. Spink, S. P. Thompson, D. J. Wilkinson,
J. Label. Compd. Radiopharm. 2003, 46, 1191–1204.
[3] P. W. C. Cross, G. J. Ellames, J. S. Gibson, J. M. Herbert, W. J. Kerr, A. H. McNeill, T. W. Mathers, Tet. 2003, 59,
3349–3358.
[4] Y. S. Tang, W. Liu, A. Chaudhary, D. G. Melillo, D. C. Dean, in Synthesis and Applications of Isotopically Labelled Compounds,
Editors D. C. Dean, C. N. Filer, K. E. McCarthy, John Wiley & Sons, Ltd, England, 2004, pp. 71–74.
TOTAL SYNTHESIS OF 14C-LABELED PROCYANIDIN B2
FLORIAN VITON,a CYRILLE LANDREAU,b DAVID RUSTIDGE,b GILL LITTLE,b FABIEN ROBERT,a GARY WILLIAMSONa, AND DENIS
BARRONa
a
´
Nestle Research Center, PO Box 44, CH-1000 Lausanne 26, Switzerland
bSelcia Ltd, Radiochemistry, Fyfield Business & Research Park, Fyfield Road, Ongar, Essex, CM5 0GS, UK
Abstract: During the last decades, many in vitro and in vivo studies have shown the beneficial effects on health of procyanidins.
However, their absorption and metabolism is still not fully understood and some aspects are still controversial. In order to have a
clearer picture of the metabolism of procyanidins, the use of labelled compounds is essential. In this context, the enantioselective
synthesis of 14C-radiolabelled procyanidin B2 was developed in our laboratories. It was achieved in fourteen ‘hot’ steps, involving
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
as key steps the Sharpless dihydroxylation of an elaborated alkene, a stereoselective intramolecular cyclization to benzylated
(1)-catechin and the condensation of two (À)-epicatechin units. 11 mCi of protected procyanidin B2 were obtained from 524 mCi of
potassium [14C]cyanide.
Keywords: carbon-14C; radiolabeled synthesis; procyanidins; catechins; polyphenols
Introduction: Flavan-3-ols and their oligomeric procyanidin forms represent one of the major dietary families of polyphenols; fresh
fruits, tea, cocoa and dark chocolate are particularly rich in procyanidins.1 During the last few decades, many in vitro and in vivo
studies have shown the beneficial effects of procyanidins on health.2 However, their absorption and metabolism is still not fully
understood and some aspects are still controversial.3
Radiolabeled compounds have proven invaluable in metabolism studies for tracing parent compounds and their metabolites.
Therefore, to support our continuous efforts in studying polyphenols metabolism, we decided to undertake the preparation of
radiolabeled procyanidin B2 [14C]-1, one of the major procyanidin present in cocoa and chocolate.
Previous syntheses of procyanidins were based on the coupling of readily available flavan-3-ol monomers.4 However, this strategy
was not compatible with the introduction of a 14C-label in the target molecule.5 Proton/tritium exchange from a labile site of a
natural precursor was also discarded so as not to compromise the stability of the target molecule during biological assays.6
Our rational to delineate a synthetic pathway compatible with the constraints of a radiolabeled synthesis is described here. The
key milestones in the development of the first asymmetric total synthesis of procyanidin B2 ((À)-epicatechin-(4b-8)-(À)-epicatechin)
and its application to the preparation of a regioselectively radiolabeled 14C–analogue are also summarized.7
Results and discussion: Our retrosynthetic analysis of procyanidin B2 is shown in Scheme 1. In terms of a radiolabeling strategy,
preliminary work led us to consider position 2 of the upper C–ring moiety as the most feasible position to introduce the radioactive
label. Thus, the target 14C-labeled molecule was envisioned to arise from the heterologous coupling of radiolabeled/non-labeled
(À)-epicatechin functionalized units. This coupling step is known to require an excess of one of the two units, which precluded
labeling of both top and bottom moieties of procyanidin B2.
OH
OH
OBn
B
OBn
O
HO
A
C
O
BnO
OH
OH
O
OH
OH
HO
OH
OBn
2
1
OH
OH
BnO
OH
OBn
OBn
OH
OH
BnO
OBn
OBn
OBn
4
HO
3
OBn
OH
OBn
OBn
OBn
C-label location
OBn
H
Br
[
C]CuCN
O
5
6
Scheme 1. Retrosynthetic analysis of 14C-labeled procyanidin B2 [14C]-1.
Our work started with securing an efficient and reliable synthetic pathway, compatible with the constraints of a radioactive
synthesis. Scheme 2 presents the synthetic approach for the cinnamyl alcohol 4, starting from aryl bromide 6.8 The direct
condensation of 6 and CuCN provided in high yield nitrile 7 which was then elaborated into the cinnamyl alcohol intermediate 4 in
three steps.
OBn
OBn
OBn
OBn
a) CuCN (0.9 eq)
b) DIBAL
toluene
-78˚C, 3.5 h
DMF, reflux, 18h
(89%)
7
6
Br
(92%)
OBn
N
OBn
c) Ph3P=CHCO2Et
CH2Cl2, rt, 18h
OBn
OBn
(90%)
H
HO
d) DIBAL, toluene
-78˚C, 2h
(82%)
5
4
O
Scheme 2. Synthesis of cinnamyl alcohol intermediate 4.
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
The elaborated trans alkene 9 having the necessary C6-C3-C6 skeleton was then prepared from 4 (Scheme 3). In the synthetic
route we selected, this coupling step of C6-C3 and C6 units was generally limiting due to ‘over-alkylation’ of the C6 phenolic unit.
This potential issue of coupling alcohol 4 with 1,3–dibenzylphloroglucinol 89 was overcome by ensuring that the phloroglucinol was
permanently present in excess in the reaction medium. When using 3 equivalents of 8 and performing a slow addition of 4 the
condensation proceeded cleanly and with a satisfactory conversion to afford 9. This material was then converted into benzylated (1
)-catechin 10 using the sequence described by Wan et al.10 Thus, Sharpless asymmetric dihydroxylation of 9 led to diol 3, which was
cyclised to benzylated (1)–catechin 10 using triethyl orthoformate in the presence of PPTS followed by methanolysis.11 This
cyclization proceeded cleanly with inversion of the stereochemistry at carbon C-2, the free phenolic hydroxyl group reacting with
the intermediate cyclic orthoacetate.10
Scheme 3. Synthesis of benzylated (1)-catechin 10.
10 was then converted into the corresponding (À)-epicatechin derivative 2 as shown in Scheme 4.10,12 The chiral purity of 2 was
assessed to be492%.7
Having secured benzylated (À)-epicatechin 2 in 12 steps and 9.5% overall yield from aryl bromide 6 and with excellent chiral
enrichment, we turned our efforts towards the preparation of the dimer procyanidin B2 (Scheme 4). Treating 2 with DDQ in the
presence of 2-ethoxyethanol led to the activated monomer 11.13 Condensation of 11 with 4 equivalents of benzylated (À)-
epicatechin 12 (prepared from commercially available (À)-epicatechin) led to dimer 13.13,14 Finally, debenzylation of 13 into
procyanidin B2 1 was achieved by hydrogenation with palladium hydroxide.15
At this point, we felt sufficiently confident in our synthetic pathway to undertake the preparation of the 14C-analogue
labeled at position 2, without the need for further optimizations. [14C]CuCN was prepared from 524 mCi of [14C]KCN,16 and used as
the radioactive precursor in the radiolabeled synthesis of [14C]-1. In the synthesis of radioactive procyanidin B2, results
comparable to the synthesis of non-labeled procyanidin B2 1 were obtained: radiolabeled [14C]-2 was obtained in 12 ‘hot’ steps and
8.5% overall yield from [14C]KCN. Finally, the last steps of the radioactive synthesis (electrophilic activation and
coupling of (À)-epicatechin monomers) afforded 11.6 mCi (at 55.2 mCi/mmol) of [14C]-13. Small amounts of radiolabeled
procyanidin B2 [14C]-1 were obtained after hydrogenation and minor work-up in the form of diluted aqueous solutions, which
exhibited satisfactory purity and stability ([14C]-1 stable under N2 atmosphere at 41C in the dark for at least two days).
Thus, the radioactive material was kept in the protected form [14C]–13 for batch to batch deprotections prior to use in biological
assays.
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
Scheme 4. Synthesis of procyanidin B2 1.
Conclusion: In conclusion, we have developed the first asymmetric total synthesis of procyanidin B2 and applied this synthetic
route to the preparation of a regioselectively 14C-radiolabeled analogue.7 The 14C–labeled procyanidin B2 has been used to improve
our knowledge of procyanidins metabolism and these results will be published in due course.
Acknowledgements: We would like to thank the European Union 6th Framework project ‘FLAVO’ for partial support of this
research work.
References
[1] (a) P. M. Aron, J. A. Kennedy, Mol. Nutr. Food Res. 2008, 52, 79–104; b) C. Manach, A. Scalbert, C. Morand, C. Remesy, L. Jimenez,
Am. J. Clin. Nutr. 2004, 79, 727–747.
[2] (a) G. Williamson, C. Manach, Am. J. Clin. Nutr. 2005, 81, 243S–255S; b) C. Manach, G. Williamson, C. Morand, A. Scalbert,
C. Remesy, Am. J. Clin. Nutr. 2005, 81, 230S–242S.
[3] M. Hu, Mol. Pharmaceutics 2007, 4, 803–806.
[4] For reviews and recent examples, see: [7] and references therein (ref. [7]).
[5] For the preparation of isotopically labeled flavan-3-ols by a) biolabeling (14C), see: [7] and references therein (ref. [4]); by
b) total synthesis (13C), see: [7] and references therein (ref. [5]); by c) hemi-synthesis (2H), see: [7] and references therein
(ref. [6]).
[6] S. Deprez, S. Buffnoir, A. Scalbert, C. Rolando, Analusis 1997, 25, M43–M46.
[7] A full paper of this work has been published as: F. Viton, C. Landreau, D. Rustidge, F. Robert, G. Williamson, D. Barron, Eur. J. Org,
Chem. 2008, 6069–6078.
[8] Aryl bromide 6 was prepared by demethylation of commercially available 4-bromoveratrole with boron tribromide, followed by
benzylation.
[9] Adapted from: S. B. Wan, K. R. Landis-Piwowar, D. J. Kuhn, D. Chen, Q. P. Dou, T. H. Chan, Bioorg. Med. Chem. 2005, 13,
2177–2185.
[10] S. B. Wan, D. Chen, Q. P. Dou, T. H. Chan, Bioorg. Med. Chem. 2004, 12, 3521–3527.
[11] For selected examples of catechins asymmetric total syntheses, see: [7] and references therein (ref. [9]).
[12] W. Tueckmantel, A. P. Kozikowski, L. J. Romanczyk, J. Am. Chem. Soc. 1999, 121, 12073–12081.
[13] (a) A. Saito, N. Nakajima, A. Tanaka, M. Ubukata, Heterocycles 2003, 61, 287–298; b) A. Saito, N. Nakajima, A. Tanaka, M.
Ubukata, Tetrahedron 2002, 58, 7829–7837.
[14] A. Saito, Y. Mizushina, H. Ikawa, H. Yoshida, Y. Doi, A. Tanaka, N. Nakajima, Bioorg. Med. Chem. 2005, 13,
2759–2771.
[15] L. Romanczyk, P. K. Sharma, A. G. Kolchinski, H. A. Shea, Y. Gou (Mars Inc.), WO2007005248, 2007.
[16] Copper [14C]cyanide was prepared using a Selcia procedure in 90% yield.
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
TUNING UP RHODIUM BLACK
SØREN CHRISTIAN SCHOU
LEO Pharma, Medicinal Chemistry Research, Industriparken 55, DK-2750 Ballerup, Denmark
Abstract: A new catalytic system based on rhodium black using Crabtree’s catalyst as an additive for direct hydrogen isotope
exchange in aromatic compounds has been investigated. The level of deuterium incorporation can be improved from 16% to 93%.
The new catalyst mixture tolerates a variety of solvents. Other rhodium sources can be used, but the degree of crystallinity of the
rhodium (metal, black or on support) plays an important role. Rhodium sources with a low degree of crystallinity had the highest
catalytic activity.
Keywords: Rhodium black; Crabtree’s catalyst; deuterium; catalytic direct hydrogen isotope exchange.
Introduction: Isotopically labelled compounds are important in the screening of pharmaceutical candidates in drug development
programmes. Such compounds facilitate studies of pharmacokinetics and metabolism, as well as other important assays. Due to
an increased attention to speed for drug development and the quality of the drug candidate, isotopically labelled compounds
have to be prepared, preferably in a fast and efficient way. The synthesis of labelled compounds can often be laborious and
synthetically challenging. The use of catalysts for direct exchange of hydrogen isotopes can ameliorate some of these
difficulties associated with the preparation of isotopically labelled compounds because no synthesis of appropriate precursor is
required.
Recently, Alexakis et al.1 reported the use of heterogeneous systems in THF as suitable catalysts for promoting hydrogen isotope
exchange in pyridines and other nitrogen heteroaromatic compounds. These catalytic systems were based on group VIII metals such
as rhodium or ruthenium, used as rhodium black, rhodium on alumina or ruthenium black.
Crabtree’s catalyst, [Ir(COD)(PCy3)(Py)]PF6, is a homogeneous catalyst based on iridium, coordinated to pyridine and
tricyclohexyl phosphine (Figure 1). Crabtree’s catalyst is a well established catalyst used for hydrogen isotope exchange
reactions.2–9 It is able to promote deuterium or tritium incorporation into arenes at positions ortho to a directing group such as
carbonyl groups.[2–4],[6–9]
Figure 1. Crabtree’s catalyst.
Here we report improvements that can be obtained by adding Crabtree’s catalyst to rhodium black.
Results and discussion: As part of a medicinal chemistry programme, we were interested in labelling compound 110 with
deuterium. Three experiments were set up, one using Crabtree’s catalyst which was expected to promote H/D-exchange in the
ortho-position to the carbonyl group, a second experiment using rhodium black which was anticipated to effect H/D-exchange in
the ortho-position to the pyridine nitrogen and finally a third experiment where a mixture of the two catalysts was used to
potentially provide the dilabelled form of 1 (Figure 2). To our surprise there was little or no incorporation of deuterium using either
Crabtree’s catalyst alone or rhodium black (16% incorporation ortho to pyridine nitrogen). However, using a mixture of the two, an
incorporation level of 93% deuterium was observed in the position ortho to the nitrogen in the pyridine ring.
#
O
O
N
O
¤
H
H
N
N
N
O
Figure 2. The structure of 1. Marks indicate sites for anticipated deuteration using Crabtree’s catalyst (]) and Rhodium black (¤).
To further investigate this surprising observation, a set of experiments were undertaken to test the generality of the catalyst
mixture in comparison with the single catalysts. We investigated electron rich, electron poor and sterically hindered pyridines
(Table 1).
As shown in Table 1 (entries 2, 3, 4, 6, 7, 8 and 15), there is a clear trend towards greater incorporation of deuterium, although not
all compounds show increased incorporation of deuterium using the catalyst mixture.
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
Table 1. Deuteration of substituted pyridines, pyrimidine and benzene using Crabtree’s catalyst, rhodium black or the mixture.
Entry
Substrate
% incorporation of deuterium in the given position
Crabtree’s catalysta
Rhodium blackb
Mixturec
2
–
3
4
0
5
6
2
3
4
5
6
2
3
4
5
6
1
0
0
0
–
28
9
0
55
–
25
16
0
28
2
3
4
0
0
–
–
0
0
0
–
0
0
0
0
0
0
23
98
–
–
12
0
0
–
0
0
12
0
96
98
67
41
98
–
–
24
0
0
–
0
0
24
0
96
98
87
24
5
6
0
–
–
0
–
0
0
0
93
–
–
0
0
–
0
0
95
73
94
–
–
0
–
0
0
93
95
23
31
11
7
8
–
–
0
–
–
0
13
87
–
–
0
0
–
–
0
0
69
64
–
–
0
–
–
0
95
96
25
18
25
25
9
0
–
0
0
–
0
0
0
0
0
28
–
0
0
–
0
0
0
28
71
29
–
0
0
–
0
0
0
29
67
10d
e
]
e
]
e
]
e
]
11
–
0
0
0
0
–
–
0
0
0
80
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
Table 1. Continued
12
–
0
0
0
0
–
0
44
0
81
–
0
21
0
85
13
14
0
0
–
–
0
–
0
–
0
0
97
0
–
–
0
–
0
–
97
0
97
–
–
0
–
0
–
97
0
60f
15
0
0
0
0
0
25
0
0
0
25
36
0
0
0
36
aExperimental conditions: Substrate (0.016 mmol), Crabtree’s catalyst (0.008 mmol), dry DCM (3 mL), D2 (1342–1590 mbar), RT, 4 h.
bExperimental conditions: Substrate (0.016 mmol), Rh black (0.010 mmol), dry THF (3 mL), D2 (1342–1590 mbar), RT, 4 h.
cExperimental conditions: Substrate (0.016 mmol), Crabtree’s catalyst (0.008 mmol), Rh black (0.010 mmol), dry DCM:THF
(1:1)(3 mL), D2 (1342–1590 mbar), RT, 4 h.
dIncorporation of D in the Ph-ring in the 2- and 6- position (Crabtree’s catalyst: 88%; Rhodium black: 32%; Mixture: 75%).
eNo trace of starting material. Experiment repeated twice with the same outcome.
fExperiment repeated 3 times.
In the 2-substituted pyridines the most significant enhancement was observed using the mixture, except entry 1, where a lower
incorporation in the 6-position is observed. In 2-phenyl-pyridine (entry 10) incorporation of deuterium in the 2- and 6- position of
the phenyl group was also observed. The same deuteration pattern has earlier been reported using rhodium(III) hydride complexes
but with lower level of deuterium incorporation (9%) in the pyridine ring.11 Ellames et al.12 has investigated the use of iridium based
catalysts of the general form [Ir(PR3)2(COD)]1 for their capability of promoting H/D isotope exchange. Using PCy3 as ligand, which is
the closest catalyst related to Crabtree’s catalyst, they found no incorporation of deuterium into the phenyl group but a 50%
incorporation of deuterium in the C6-position of the pyridine ring. Surprisingly we found no incorporation of deuterium into the
pyridine ring using the iridium based Crabtree’s catalyst. No incorporation was observed in the substituent of 2-isopropyl-pyridine
(entry 11), but in this compound attempted exchange with rhodium black alone gave a reaction mixture with no trace of starting
material. The catalyst mixture, however, gave 80% deuterium incorporation. Whether deuterium was incorporated before or after
degradation of 2-isopropyl-pyridine using rhodium black was not determined.
Using electron rich pyridines (entries 12 and 13) the effect of adding Crabtree’s catalyst to rhodium black was less noticeable due
to the high level of deuterium incorporation using rhodium black alone.
Concerning the electron poor pyridines, no enhanced effect was observed by adding Crabtree’s catalyst to rhodium black. In
some cases (entries 2 and 3) high levels of deuterium incorporation resulted from using rhodium black alone; in other cases (entries
1 and 9) no significant enhancement was observed.
To investigate the scope, two aromatic non-pyridyl compounds, one benzene nucleus and one pyrimidine scaffold were
tested. In agreement with the results reported by Ellames et al.2 we found that aniline (entry 15) did not undergo exchange
using Crabtree’s catalyst. Using rhodium black we found a 51% incorporation of deuterium and using the mixture 72%
incorporation of deuterium was obtained. Using Crabtree’s catalyst or rhodium black, no incorporation of deuterium in 4-amino-
pyrimidine-5-carboxylic acid (entry 14) was obtained. On the other hand using the mixture an incorporation of 60% deuterium was
observed.
Sometimes, an additive effect was observed using the catalyst mixture, but less than an additive effect was also observed (entry 1).
The effect of the catalyst mixture cannot be predicted and must be determined by experiments. Most important however, is the
surprisingly positive outcome that sometimes can be found using the catalyst mixture, and substrates not able to undergo direct
hydrogen isotope exchange using Crabtree’s catalyst or rhodium black alone can be labelled using the catalyst mixture (entry 14).
The catalytic effect of rhodium black can be enhanced by adding Crabtree’s catalyst, while there is no enhanced catalytic effect of
Crabtree’s catalyst when rhodium black is added. In two cases, only the mixture was able to promote hydrogen isotope exchange
(entries 11 and 14), indicating that a new and more active catalytic system is at hand.
To investigate the mechanism further an experiment (mix and split) was undertaken where Crabtree’s catalyst and rhodium black
were stirred under D2-atmosphere for 4h without 1 added. The heterogeneous catalyst mixture was then centrifuged and divided
into solid and solution and each part was then added 1 and stirred under D2-atmosphere for 4 h. The solid was able to promote an
incorporation of 24% deuterium while no incorporation of deuterium was observed in the solution, supporting that the catalytic
exchange takes place exclusively on the rhodium surface and not in solution.
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
A hypothesis for the enhanced effect may involve a ligand transfer from Crabtree’s catalyst to rhodium. To investigate this
hypothesis, a set of experiments using 1 as a model compound were undertaken, where one or more of the components of
Crabtree’s catalyst were added to rhodium black (Scheme 1 and Table 2).
Scheme 1. Setup to investigate a possible ligand transfer.
Table 2. Deuteration of 1 using different catalyst components
Entry
Ir
Ir(COD)Cl
Rh(COD)Cl
PCy3
Py
%D
1
2
3
4
5
6
7
8
0.5
0.5
0.5
0.5
16
35
26
27
0
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
20
16a
13
11
17
34
26
27b
23b
42
35b
25c
0
0.5
0.5
9
0.6
0.6
0.6
0.6
10
11
12
13
14
15
16
17
18d
19d
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
1
10
0.5
1
0
Experimental conditions: 1 (0.016 mmol), Rh black (0.010 mmol), catalyst component/components, dry THF:DCM (1:1)(3 mL), D2
(1457–1639 mbar), RT, 4 h. Numbers in table are numbers of equivalents added.
aNo deuteration is also observed if 0.5 eq NH4PF6 is added.
bSame incorporation level observed if 0.5 eq COD is added.
cUnknown impurity formed.
dNo rhodium black added.
It was possible to improve the catalytic effect of rhodium black (Table 2) (from 16% to 42% deuterium incorporation), but not to
the same extent as adding Crabtree’s catalyst (93% incorporation). The most effective additive was pyridine (entry 15) however the
addition of larger amounts of pyridine did not promote further deuterium incorporation (entries 16 and 17). To investigate if the
enhanced effect of pyridine was an additive effect to rhodium black, pyridine was tested in absence of catalyst, and, as expected, no
incorporation could be detected (entries 18 and 19). These findings support the hypothesis that the enhanced catalytic exchange
reaction takes place at the rhodium surface.
Table 3 shows the effect of decreasing the amount of Crabtree’s catalyst added to rhodium black. Even a low amount (0.05 eq) of
Crabtree’s catalyst was able to enhance the catalytic effect of rhodium black noticeable, enhancing the incorporation level from 16
to 32% (entry 3).
Table 4 shows the catalytic effect of different rhodium sources. All the tested sources of rhodium were able to promote catalytic
hydrogen isotope exchange except non-amorphous rhodium, which was unable to promote exchange.
A batch variation of the different rhodium black sources was observed, noticeable when Crabtree’s catalyst was added. Rhodium
on different supports was quite active even without activation with Crabtree’s catalyst. An explanation for the high catalytic activity
could be that rhodium on support has a larger surface area than non-amorphous rhodium or even rhodium black. This is supported
by the powder X-ray diffractograms of the different sources (Figure 3). A sharp and distinguished peak indicates a high degree of
crystallinity whereas a blurred peak or non-existing peak indicates low or very low degree of crystallinity. The powder X-rays
diffractograms show a big difference between the different sources, and these findings correlate well with the amount of
incorporated deuterium. With respect to rhodium, sharp peaks are seen, indicating a high degree of crystallinity and no
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
Table 3. Incorporation of deuterium using different amounts of Crabtree’s catalyst
Entry
Rhodium black
Crabtree’s catalyst
%D
1
2
3
4
5
0.6
0.6
0.6
0.6
0.6
0.5
0.1
0.05
0.02
0
93
46
32
16
16
Experimental conditions: 1 (0.016 mmol), Rh black (0.010 mmol), Crabtree’s catalyst, dry THF:DCM (1:1)(3 mL), D2
(1463–1575 mbar), RT, 4 h. Numbers in table are numbers of equivalents added.
Table 4. Incorporation of deuterium observed using different rhodium sources
Entry
Rhodium source
%D
1 0.5 eq Crabtree’s catalyst, %D
1
2
3
4
5
6
7
8
Rh black (ABCR)
Rh black (Aldrich)
Rh (Acros)
16
16
0
57
91
0
Rh (Fluka)
0
0
5wt% Rh/act. aluminaa
5wt% Rh/aluminaa
5wt% Rh/carbona (Aldrich)
5wt% Rh/carbonb
75–92
71
88
8
90
93
85
16
Experimental conditions: 1 (0.016 mmol), catalyst/catalysts, dry THF:DCM (1:1)(3 mL), D2 (1550–1665 mbar), RT, 4 h.
aAdded amount correlates to 0.6 eq rhodium.
bAdded amount correlates to 0.6 eq rhodium, old batch with unknown origin.
incorporation of deuterium was observed. On the other hand looking at e.g. rhodium, 5 wt% on activated alumina, blurred or no
peaks are observed, indicating a very low degree of crystallinity and very high degree of incorporation (92%) was observed. The
powder X-ray can also clarify the difference in catalytic capacity of an old batch with unknown origin (entry 8) and a newly
purchased batch (entry 7) of 5% rhodium on carbon. The old batch (8% incorporation) has a higher degree of crystallinity compared
to the new one (88% incorporation).
Figure 3. X-Ray powder diffractograms of the different rhodium sources. Numbers to the right are amount of deuterium incorporated (cf. Table 4).
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
To test the robustness of the catalytic mixture, experiments in different solvents were performed (Table 5). The catalytic system is
more or less independent of the solvent, and acceptable to high incorporation levels were observed in all solvents except diethyl
ether, which could be due to low solubility of 1.
Table 5. Incorporation of deuterium in different solvents.
Entry
Solvent
% incorporation of deuterium
1 0.5 eq Crabtree’s catalyst
No additive
1
2
3
4
5
Et2O
EtOAc
DCM
THF
0a
84
60
89
90–93c
THF:DCM (1:1)
16b
Experimental conditions: 1 (0.016 mmol), catalyst/catalysts, dry THF:DCM (1:1)(3 mL), D2 (1418–1666 mbar), RT, 4 h.
a1 is slightly soluble in Et2O
bExperiment performed in THF-d8:DCM-d2 show a similar incorporation level (12%).
cExperiment performed in THF-d8:DCM-d2 show a slightly decreased incorporation level (68%).
Consequently, the high solvent tolerance expands the scope and usefulness for the catalyst mixture, allowing a variety of
compounds to be labelled.
Conclusion: A new catalytic system based on rhodium black with Crabtree’s catalyst as an additive has been investigated. In
general, the mixture has an improved activity, compared to rhodium black alone. In two cases neither Crabtree’s catalyst nor
rhodium black was able to promote exchange, but using the mixture an incorporation of deuterium was observed. The new catalytic
system is active in a variety of solvents, extending its usefulness. This protocol will, of course, also be useful when tritiated
compounds are needed.
Experimental: General: All reactions were performed in a stainless steel manifold purchased from RC Tritec, Teufen, Switzerland.
˚
D2 (99.8 atom % D) was purchased from Isotec. Anhydrous solvents were dried over molecular sieves (4 A). All other solvents and
1
reagents were used as received, purchased from Sigma-Aldrich, ABCR, Fluka, Alfa Aesar and Acros. H and 13C NMR spectra were
obtained on a Bruker AV600 spectrometer with a 5 mm TCI-Cryoprobe or a Bruker DRX500 spectrometer with a 5 mm PAPPI-probe.
The deuterium content was measured on the basis of integration. Chemical shifts are reported in ppm with tetramethylsilane (TMS,
d = 0.00) as internal reference. The X-Ray Powder Diffractograms were obtained using on a X’pert PRO MPD from PANalytical with Cu
Ka radiation and operating at 45 kV and 40 mA. The samples were scanned from 3 to 70 degree 2 theta with a step size of 0.0020
degree 2 theta and at 121.18 s per step.
General procedure for deuterium exchange reaction: Catalyst was weighed into an 8 mL round bottom reaction flask
containing a new teflon coated stir bar (3 Â 10 mm), substrate (0.016 mmol) was dissolved in dry solvent (3 mL) and added. The
reaction mixture was frozen in liquid N2 and evacuated (below 3.5 Â 10À3 mbar), thawed and then stirred under D2-atmosphere
(1340–1666 mbar) for 4 h at RT. The reaction mixture was filtered through a syringe-filter (Whatman, 0.45 mm) and concentrated in
vacuo.
NMR-data for non-deuterated compounds are described in the literature.13
Acknowledgments: The author gratefully thanks Gunnar Grue-Sørensen (LEO Pharma) for fruitful discussions and invaluable
assistance and support, the Department of Spectroscopy and Physical Chemistry at LEO Pharma, especially Lene Hoffmeyer, Lone
Lerbech Dolleris and Zanni Winther for obtaining the X-Ray powder diffractograms, Grethe Aagaard for NMR assistance and Lone
Møss and Else Kristoffersen for obtaining high resolution NMR-spectra.
References
[1] E. Alexakis, J. R. Jones, W. J. S. Lockley, Tetrahedron Lett. 2006, 47, 5025–5028.
[2] G. J. Ellames, J. S. Gibson, J. M. Herbert, A. H. McNeill, Tetrahedron. 2001, 57, 9487–9497.
[3] J. S. Valsborg, L. Sørensen, C. Foged, J. Label. Compd. Radiopharm. 2001, 44, 209–214.
[4] D. Hesk, P. R. Das, B. Evans, J. Label. Compd. Radiopharm. 1995, 36, 497–502.
[5] A. Y. L. Shu, D. Saunders, S. H. Levinson, S. W. Landvatter, A Mahoney, S. G. Senderoff, J. F. Mack, J. R. Heys, J. Label. Compd.
Radiopharm. 1999, 42, 797–807.
[6] M. E. Powell, C. S. Elmore, P. N. Dorff, J. R. Heys, J. Label. Compd. Radiopharm. 2008, 50, 523–525.
[7] S. K. Johansen, L. Sørensen, L. Martiny, J. Label. Compd. Radiopharm. 2005, 48, 569–576.
[8] B. A. Czeskis, D. D. O’Bannon, W. J. Wheeler, D. K. Clodfelter, J. Label. Compd. Radiopharm. 2004, 48, 85–100.
[9] M. J. Hickey, J. R. Jones, L. P. Kingston, W. J. S. Lockley, A. N. Mather, D. J. Wilkinson, Tetrahedron Lett. 2004, 45, 8621–8623.
[10] J. Fensholdt, J. Thorhauge, B. Norremark, Patent 2005 WO 2005054179 A2.
[11] S. Chen, G. Song, X. Li, Tetrahedron Lett. 2008, 49, 6929–6932.
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
[12] G. J. Ellames, J. S. Gibson, J. M. Herbert, W. J. Kerr, A. H. McNeill, J. Label. Compd. Radiopharm. 2004, 47, 1–10.
[13] S. C. Schou, J. Label. Compd. Radiopharm. 2009, 52, 376–381.
SYNTHESIS OF 14C- AND 13C/2H-LABELED SGLT INHIBITORS AVE2268 AND AVE8887
VOLKER DERDAU, THORSTEN FEY, AND JENS ATZRODT
¨
Sanofi-Aventis Deutschland GmbH, Isotope Chemistry & Metabolite Synthesis, G876, 65926, Frankfurt/Hochst, Germany
Abstract: Isotopically labeled analogues of two structurally very similar SGLT inhibitors AVE2268 (1a) and AVE8887 (1b) have been
synthesized by various routes. The radioactive labeled [14C]-AVE2268 was prepared in 5 steps including a Friedel-Crafts acylation as
the key step for the 14C-label introduction. For [14C]-AVE8887 the same synthetic approach was not successful and therefore an
alternative thiophene metalation/Weinreb amide sequence was developed. This pathway was also applied to obtain stable
isotopically labeled analogs of both AVE2268 and AVE8887.
Keywords: SGLT; Lithium; Carbon-14; Deuterium
Introduction: Sodium glucose co-transporters (SGLT-2)1 are membrane proteins. They play an important role in maintaining glucose
equilibrium in the human body by re-absorption of glucose in the kidney. It is therefore expected that the inhibition of SGLT 2
transporters could decrease glucose blood levels by preventing re-absorption from the urine. The control of these transporters could
be an effective tool in the normalisation of high blood glucose levels that are associated with diseases such as Type 1 and Type 2
diabetes.2 Both radiolabeled AVE2268 and AVE8887 were required to measure the ADME (absorption, distribution, metabolism and
elimination) profiles of these new drug candidates in animals and humans. In addition stable-isotope labelled analogues of both
compounds were also required for use as an internal standard for bioanalytical assay validation. Due to the presence of sulphur in
the thiophene ring moieties of 1a and 1b at least 5 deuterium atoms were required in both cases.
HO
HO
HO
HO
O
O
HO
HO
O
O
*
*
OH
OH
F
F
F
S
S
O
O
1a
1b
AVE2268
AVE8887
Figure 1. Structures of AVE2268 and AVE8887 (14C-labelling position marked with Ã).
Results and Discussion: The synthesis of [14C]-AVE2268 (1a) was achieved following the simple 5 step synthesis path depicted in
scheme 1.3 Starting from commercially available 3-methoxythiophene (2) and 100 mCi [14CO]anisoyl chloride, [14C]-3a, purchased
from Amersham Biosciences, a completely regioselective Friedel-Crafts acylation followed by a selective methyl ether cleavage of 3-
methoxythiophene derivative 4a gave the hydroxyl-thiophene derivative [14C]-5a. Subsequent alkylation with acetobromo-alpha-
glucose under phase transfer conditions, carbonyl reduction with NaCNBH3/TMSCl4 and final basic acyl deprotection afforded the
desired compound [14C]-1a in a very good overall yield of 50% (50 mCi, 99.9% radiochemical purity, 57 mCi/mmol).
Scheme 1. Synthesis of [14C]-AVE2268 (1a).
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
For 14C-labelling of AVE8887 (1b) a very similar pathway was envisioned employing 4-trifluoromethoxy-[14CO]benzoyl chloride [14C]-
3b instead of [14C]-3a. Surprisingly we found in our cold elaboration experiments that more electron deficient benzoyl chlorides,
such as (3b), reacted much slower with thiophene (2) under Friedel-Crafts acylation conditions. Consequently the acid labile
thiophene (2) started to decompose resulting in low yields for this reaction step. Even under optimised conditions and a stepwise
addition (2–3 equiv.) of (2), the maximum yield achieved was only 32%. In addition, work-up and purification became difficult due to
the numerous side products formed. Therefore, we started to look for alternative strategies for the synthesis of (5b) in which greater
emphasise was taken in minimising decomposition by increased reactivity of the thiophene (2).
Stimulated by results reported by Miller and Yu5 we examined the regioselective lithiation of 3-methoxythiophene (2) in refluxing
diethylether over 5 min. However, when quenching the lithiated thiophene (2) either with anisoyl chloride (3a) or 4-trifluoromethoxy
benzoyl chloride 3b only traces of the desired products 4a and 4b respectively could be observed. Besides acyl chlorides the Weinreb
amide is well known to be reactive in organometallic reactions and would be readily accessible also in labeled form.6 After optimisation
of the reaction of 9b with the lithiated thiophene (2) we obtained the formation of 4b in 76% yield.
The 14C-synthesis of [14C]-1b was finally accomplished following the pathway shown in Scheme 2. Starting from 100 mCi
[14CO]trifluoromethoxy benzoic acid [14C]-10b, the corresponding Weinreb amide [14C]-9b was formed under Appel conditions.7
Addition of the lithiated thiophene to an ice-cold solution of the Weinreb amide [14C]-9b yielded [14C]-4b in 76%. Luckily the rest of
the synthesis showed no significant deviation from the synthesis of [14C]-1a reported above. The colorless compound was isolated
after 6 reaction steps with an overall yield of 45% (45 mCi, 99.5% radiochemical purity, 53.2 mCi/mmol).
Scheme 2. Synthesis of [14C]-AVE8887 (1b).
Synthesis of stable isotopically labeled AVE2268 (1a) and AVE8887 (1b): Stable isotopically labeled (1a) was synthesised
according to the path shown in Scheme 3 in 8 chemical steps with an overall yield of 25%. Starting from 4-hydroxy benzoyl
methylester (11) the hydroxy group was alkylated with deuterated methyliodide followed by basic ester hydrolysis to give the
corresponding acid [D3]-10a in 90% yield over two steps. The subsequent steps were described in the 14C-synthesis above, however
sodiumcyanoborodeuteride was employed in the carbonyl reduction to introduce two additional deuterium atoms. A co-injection of
AVE2268 and [D5]-AVE2268 revealed the expected mass difference but identical retention time in the LC-MS.
Figure 1. LC-MS-spectra of a 1:1 mixture of AVE2268 and [D5]-AVE2268 (M1Na).
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
Scheme 3. Synthesis of [D5]-AVE2268 (1a).
The synthesis of stable isotopically labeled AVE8887 (1b) was achieved applying exactly the same pathway starting from 13C-
labeled-trifluoromethoxy benzoic acid 10b.
Conclusions: We have described in detail the challenges of the synthesis of isotopically labeled compounds of SGLT inhibitors
(1a) and (1b). Although very similar in their chemical structure, different synthetic pathways had to be optimised.
References
[1] (a) F.-Q. Zhao, A. F. Keating Curr. Genomics 2007, 8, 113–128; b) S. I. Wood, P. Trayhurn British J. Nut. 2003, 89, 3–9; b) T. Asano,
M. Anai, H. Sakoda Drugs. Fut. 2004, 29, 461–466. c) A. L. Handlon Expert Opin. Ther. Patents 2005, 15, 1531–1540; d) M. Isaji Curr.
Opin. Invest. Drugs 2007, 8, 285–292.
[2] H. Glombik, W. Frick, H. Heuer, W. Kramer, H. Brummerhop, O. Plettenburg, WO 2004007517.
[3] V. Derdau, L. Bierer, M. Kossenjans DE102004063099; V. Derdau, L. Bierer, M. Kossenjans USPTO 20080207882.
[4] Reaction conditions have been reported earlier: V. G. S. Box, P. Meleties Tetrahedron Lett. 1998, 39, 7059–7062.
[5] L. L. Miller, Y. Yu J. Org. Chem. 1995, 60, 6813–6819.
[6] (a) S. Nahm, S. M. Weinreb Tetrahedron Lett. 1981, 22, 3815–3818; b) S. Balasubramaniam, I. S. Aidhen Synthesis 2008, 23,
3707–3738.
[7] L. E. Barstow, V. J. Hruby J. Org. Chem. 1971, 36, 1305–1306.
STABLE ISOTOPE AND CARBON-14 SYNTHESES OF CP-945,598: A CANNABINOID CB1 RECEPTOR
ANTAGONIST
KEITH T. GARNES AND KLAAS SCHILDKNEGT
Pfizer Global Research & Development, Groton/New London Laboratories, Pfizer Inc, Groton, CT 06340, USA
Abstract: CP-945,598 is a cannabinoid CB1 receptor antagonist drug candidate for the treatment of obesity. Stable isotope
labeled CP-945,598 was required for use as a clinical assay internal standard, and carbon-14 labeled CP-945,598 was required to
perform detailed adsorption, distribution, metabolism and excretion (ADME) studies. A key consideration within the stable isotope
synthesis was the incorporation of sufficient H/C/N isotopes to increase the candidate’s molecular weight profile beyond that exhibited
by the non-labeled parent. The carbon-14 synthesis required unique attention to label position based on metabolism concerns.
Keywords: Stable Isotope; Carbon-14; Cannabinoid receptor
Introduction: CP-945,598, Figure 1, is a cannabinoid CB1 receptor antagonist under investigation for the treatment of obesity.1
Stable isotope labeled CP-945,598 was required for use as a clinical assay internal standard, and carbon-14 labeled CP-945,598 was
required to perform detailed adsorption, distribution, metabolism and excretion (ADME) studies. A key consideration within the stable
isotope synthesis was the incorporation of sufficient H/C/N isotopes to increase the candidate’s molecular weight profile beyond that
exhibited by the non-labeled parent. The carbon-14 synthesis required unique attention to label position based on metabolism concerns.
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
Cl
N
N
H
N
N
N
N
- Stable Isotope - MS internal standard
NH2
O
Cl
- Carbon-14 - Definitive biotransformation/ADME
CP-945,598
Figure 1. Structure of CP-945,598 and intended isotopic studies.
Results and Discussion: Central to preparing an appropriate CP-954,598 clinical assay internal standard was to incorporate a sufficient
number of stable isotope atoms within the target structure to increase its molecular weight beyond the nonlabeled two chlorine parent
molecular ion pattern. The established CP-945,598 synthesis involved a final coupling of a piperidine aminoamide to a chloropurine core–the
piperidine moiety presented the most straightforward target for stable isotope incorporation. To that end, commercially available piperidone 1
was treated with perdeuterated ethylamine and K13C15N in aqueous ethanol to afford the Strecker product 2 containing a total of seven stable
isotope atoms, in 81% yield, Scheme 1. Partial nitrile hydrolysis of 2 was affected with concentrated H2SO4 in CH2Cl2 to give stable isotope
labeled aminoamide 3. Debenzylation of 3 was accomplished by hyrogenolysis of the substrate in acidic methanol to yield piperidine
[2H,13C,15N]4 as an HCl salt. Finally, coupling of this piperidine to chloropurine 5 was performed in aqueous acetone with triethylamine to arrive
at [2H,13C,15N]CP-945,598. Of note, this coupling was completely regioselective with respect to the two available piperidine secondary amines.
2H
2H
2H 2H
2H 2H
2H
2H
H2NC2H2C2H3
H2, 53 psi
20%Pd(OH)2/C
2H
2H
O
K13C15
N
H2SO4
NH
13C15
NH
BnN
MeOH
CH2Cl2
EtOH, H2O
81%
15NH2
N
2.5 eq. HCl
13C
O
75% BnN
BnN
95%
1
2
3
Cl
N
N
Cl
N
N
Cl
N
N
2H
2H
2H 2H
Cl
H
2H
N
2H
. 2 HCl
N
N
2H
5
2H
N
2H 2H
NH
13C
Cl
TEA, Acetone/H2O
51%
15NH2
O
15NH2
13C
O
HN
[2H,13C,15N]4
[2H,13C,15N]CP-945,598
Scheme 1.
Importantly, mass spectral analysis of [2H,13C,15N]CP-945,598 showed it to be completely void of nonlabeled CP-945,598, Figure 2.
Figure 2. Mass spectral comparison of SIL CP-945,598 and unlabeled reference.
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
Carbon-14 labeled CP-945,598 was needed to perform ADME profiling studies. The compound’s central purine core was targeted
for radiolabeling because it was deemed metabolically stable and presented an attractive opportunity to utilize readily available
labeled cyanide as a carbon-14 source.
1-Chloro-2-iodobenzene (6) was treated with 150 mCi of K14CN and heated in refluxing acetone with a mixed palladium(0)/
copper(1) catalyst system to afford 110 mCi of 2-chlorobenzo[14C]nitrile (7), Scheme 2. It is important to note that this cyanation
method was completely selective for reaction at the aryliodide bond. In contrast, preliminary experiments involving metallation and
carbonation of 6 produced 8 contaminated with a minor quantity of 2-iodobenzoic acid–an impurity that would be difficult to
differentiate/purge from desired product. Nitrile 7 was hydrolyzed to carboxylic acid 8 in a biphasic refluxing solution of aqueous
sulfuric acid/dichloromethane. Base/acid workup of this reaction delivered 87 mCi of radiochemically pure 8. Conversion of 8 to its
corresponding acid chloride 9 was affected in refluxing thionyl chloride. The product was isolated from this reaction in crude form
by distillation of the solvent and then reacted directly with diaminpyrimidine 10 in dimethylacetamide to yield crude amide 11.
Chromatographic purification of the crude product delivered 53 mCi of high radiochemical purity 11 (61% yield from benzoic acid
8). There was no regioisomeric amide formed in this reaction.
Cl
Cl
Cl
I
14CN
14CO2H
150 mCi K14CN
aq. H2SO4
CH2Cl2,
79%
Pd(PPh3)4
CuI, CuCl
Acetone,
8
6
7
110 mCi
87 mCi
73%
Cl
Cl
N
N
N
N
N
Cl
O
N
Cl
NH3Cl
Cl
H
H
14COCl
HN
10
SOCl2
14C
DMAC
Cl
11
53 mCi
9
Scheme 2.
Cyclization of amide 11 to purine 12 was completed in a mixture of refluxing sulfuric acid and isopropanol, Scheme 3. The
product was easily isolated as a pure crystalline material from this reaction. An expected and unfortunate side reaction from this
process involved hydrolysis of the purine 6-chloro substituent. This halide was required as a leaving group for the final piperidine
addition to form [14C]CP-945,598. To reinstall the chloride, 36 mCi of hydroxypurine 12 was reacted with phosphorus oxychloride
and trietlylamine in refluxing toluene. Addition of isopropanol to this reaction precipitated crude product which was further purified
by flash column chromatography to afford 23 mCi of radiochemically pure chloropurine [14C]5. In similar fashion to the stable
isotope labeled synthesis of the target, [14C]5 was reacted with piperidine 4 in aqueous acetone and triethylamine to give crude
final product. This material was further purified by flash column chromatography to afford 19 mCi of high radiochemical purity
[
14C]CP-945,598. The specific activity of this material was measured at 55.1 mCi/mmol by mass spectrometry.
Cl
Cl
N
N
N
N
1. POCl3, TEA
toluen ,
e
OH
H2SO4
iPrOH,
68%
N
H
Cl
O
N
14C
N
HN
2. iPrOH
14C
Cl
64%
Cl
11
12
36 mCi
53 mCi
H
HN
Cl
Cl
N
N
N
N
N
H
N
Cl
NH2
N
14C
N
14C
N
4
O
N
N
NH2
TEA, acetone/H2O
81%
O
Cl
Cl
[
14C]5
23 mCi
[
14C]CP-945,598
19 mCi
Scheme 3.
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
Summary: [2H,13C,15N]CP-945,598 was prepared in 29% yield in 4 steps starting from deuterated ethylamine and K13C15N, and
14C]CP-945,598 was prepared in 12% yield in 7 steps starting from K14CN.
[
References
[1] (a) D. A. Griffith, A. M. Thomas, N. Nguyen, L. Hollywood. PCT Int. Appl. WO2004037823; (b) J. A. Ragan. PCT Int. Appl.
WO2006043175; (c) J. A. Ragan, D. E. Bourassa, J. Blunt, D. Breen, F. R. Busch, E. M. Cordi, D. B. Damon, N. Do, A. Engtrakul, D.
Lynch, R. E. McDermott, J. A. Mongillo, M. M. O’Sullivan, P. R. Rose, B. C. Vanderplas. Org. Process Res. Dev. 2009, 13, 186–197.
SYNTHESIS OF A DUAL CARBON-14 LABELED PEROXISOME PROLIFERATOR-ACTIVATED RECEPTOR a
AND g (PPARa/g) DUAL AGONIST FOR USE IN A HUMAN ADME STUDY
RICHARD C. BURRELL,a KAI CAO,b SAMUEL J. BONACORSI, JR.,b and BALU BALASUBRAMANIAN
Bristol-Myers Squibb Research and Development
aDepartment of Chemical Synthesis, 5 Research Parkway, Wallingford, CT 06492, USA
bDepartment of Chemical Synthesis, Route 206 and Province Line Road, Princeton, NJ 08540, USA
Abstract: A dual carbon-14 labeled peroxisome proliferator activated receptor a and g (PPAR a and g) dual agonist ([14C]-1) was
synthesized for use in a human ADME study. A total of 9.5 mCi of [14C]-1 was prepared with a specific activity of 10.18 mCi/mg and a
radiochemical purity of 99.85%. The title compound [14C]-1 was synthesized in 6 radioactive steps and a 23% radiochemical yield
from the dual carbon-14 labeled intermediate [14C]-6. The dual carbon-14 labeled intermediate [14C]-6 was prepared by mixing
equal millicurie amounts of mono labeled [14C]-6 Label A and mono labeled [14C]-6 Label B. Label A ([14C]-6 Label A) was
prepared in 4 radioactive steps in a 55% yield from carbon-14 labeled potassium cyanide. Label B ([14C]-6 Label B) was prepared in 1
radioactive step in a 96% yield from carbon-14 labeled 4-hydroxyacetophenone.
Key Words: Type 2 diabetes; peroxisome proliferator activated receptors; PPARa; PPARg; carbon-14 labeling
Introduction: Type 2 diabetes is a metabolic disease that currently affects about 180 million people worldwide and it is estimated to
rise to over 360 million by the year 2030.1 In addition to the characteristic combination of insulin resistance and insulin deficiency,
the type 2 diabetic often displays cardiovascular risk factors such as hypertriglyceridemia, low HDL-cholesterol and elevated LDL-
cholesterol levels. The peroxisome proliferator activated receptors (PPARa, PPARg and PPARd) are ligand-activated transcriptional
factors that belong to the nuclear hormone receptor superfamily which are essential in controlling glucose, lipid, and energy
homeostasis.2,3 The dual PPARa/g agonist appears well-suited as a treatment for type 2 diabetes because of the insulin-sensitizing/
glucose-controlling potential of the PPARg agonists, in combination with the positive lipid and cholesterol modulating activities of
the PPARa agonists.3 Compound 1 is a dual PPARa/g agonist that is being developed by Bristol-Myers Squibb. An important step in
the development of this compound was the synthesis of a carbon-14 labeled analog for use in a human absorption, distribution,
metabolism and excretion (ADME) study. This report describes an efficient synthesis of dual carbon-14 labeled 1.
O
N
CO2H
N
O
1
O
O
OCH3
Figure 1. Structure of the PPAR a/g dual agonist.
Results and discussion: In-vitro metabolism studies with unlabeled 1 revealed metabolites that arose from cleavage of the central
ether bond. This was not surprising since this metabolic pathway has been documented previously for a structurally similar PPAR
compound.4 The metabolic fate of the entire molecule needed to be elucidated by conducting a human ADME study with an analog
of 1 appropriately labeled with carbon-14. Since metabolic cleavage of the central ether bond would result in two major fragments, a
decision was made to prepare a dual carbon-14 labeled analog of 1 with labels on both sides of the central ether moiety. To minimize
the redundancy of two complete syntheses of carbon-14 labeled 1, a labeling strategy was developed in which an early intermediate
would be dual labeled. The dual carbon-14 labeled intermediate would then be carried forward to give dual carbon-14 labeled 1. This
labeling strategy would maximize the radiochemical yield and minimize the total number of radioactive steps.
The intermediate [14C]-6 Label A was prepared in 4 radioactive steps from carbon-14 labeled potassium cyanide as shown in
Scheme 1.5–7 A total of 73 mCi of [14C]-6 Label A was prepared with a radiochemical purity of 99% and a specific activity of 83.9 mCi/
mg. The overall yield was 55%.
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
O
K14CN, KI
O
N
14CN
N
14C]-3
Cl
DMF, 85 °C
[
2
2-methoxyethanol
O
N
14C]-4
BH3 THF
78% for
three steps
O
14 C
KOH, water, 85 °C
OH
[
O
1. p-fluoroacetophenone
DMF, 43 °C
O
O
14C
14C
N
14C]-5
2. 1 M tBuOK in THF
71%
N
O
OH
[
[
14C]-6 Label A
RCP 99%
73 mCi
83.9 μCi/mg
Scheme 1. Synthesis of [14C]-6 Label A.
O
O
14 C
14 C
O
DMF, EtOH, Cs2CO3, 90 ˚C
O
N
O
HO
[
14C]-7
[
14C]-6 Label B
N
OMs
RCP 99%
8
96 mCi
SA 91.95 mCi/mg
96%
Scheme 2. Synthesis of [14C]-6 Label B.
The intermediate [14C]-6 Label B was prepared in 1 radioactive step from carbon-14 labeled 4-hydroxyacetophenone ([14C]-7) as
shown in Scheme 2.8–10 A total of 96 mCi of [14C]-6 Label B was prepared with a radiochemical purity of 99% and a specific activity
of 91.95 mCi/mg. The overall yield was 96%.
O
O
14C
N
O
[
14C]-6 Label A
14 C
O
6 steps
O
N
RCP 99%
N
CO2H
14C
70 mCi
SA 83.9 μCi/mg
O
O
[
14C]-1
O
14 C
O
RCP 99.85%
9.5 mCi
SA 10.18 μCi/mg
OCH3
N
O
[
14C]-6 Label B
RCP 99%
70 mCi
SA 91.95 μCi/mg
Scheme 3. Synthesis of dual carbon-14 labeled of [14C]-1.
Equal millicurie amounts of [14C]-6 Label A (70 mCi) and [14C]-6 Label B (70 mCi) were mixed to give dual carbon-14 labeled
material as shown in Scheme 3. This material ([14C]-6 Label A1[14C]-6 Label B) was converted to the desired product [14C]-1 in 6
radioactive steps.9,10 A total of 9.5 mCi of [14C]-1 was prepared with a radiochemical purity of 99.85% and a specific activity of
10.18 mCi/mg. The overall yield for the 6 steps was 23%. This material was used successfully in a clinical ADME study to elucidate the
human metabolism of [14C]-1.
Acknowledgements: The authors would like to thank the Bristol-Myers Squibb Process Research & Development Department, the
Medicinal Chemistry Department and the Biotransformation group.
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
References
[1] World Health Organization, Fact Sheet No. 312, November 2008. http://www.who.int/mediacentre/factsheets/fs312/en.
[2] Issemann, S. Green, Nature 1990, 347, 645–650.
[3] T. M. Willson, P. J. Brown, D. D. Sternbach,, B. R. Henke J. Med. Chem. 2000, 43, 527–550.
[4] (a) D. Zhang, L. Wang, G. Chandrasena, L. Ma, M. Zhu, H. Zhang, C. D. Davis, W. G. Humphreys, Drug Metab. Dispos. 2007, 35,
139–149; (b) D. Zhang, H. Zhang, N. Aranibar, R. Hanson, Y. Huang, P. T. Cheng, S. Wu, S. Bonacorsi, M. Zhu, A. Swaminathan, W.
G. Humphreys, Drug Metab. Dispos. 2006, 34, 267–80.
[5] P. T. W. Cheng, S. Chen, X. Qian, R. P. Deshpande, E. Tang, PCT Int. Appl. 2008, WO 2008006044 A2.
[6] H. O. Han, S. H. Kim, K.-H. Kim, G.-C. Hur, H. J. Yim, H.-K. Chung, S. H. Woo, K. D. Koo, C.-S. Lee, J. S. Koh, G. T. Kim, BioMed.
Chem. Lett. 2007, 17, 937–941.
[7] D. A. Brooks, G. J. Etgen, C. J. Rito, A. J. Shuker, S. J. Dominianni, A. M. Warshawsky, R. Ardecky, J. R. Paterniti, J. Tyhonas, D. S.
Karanewsky, R. F. Kauffman, C. L. Broderick, B. A. Oldham, C. Montrose-Rafizadeh, L. L. Winneroski, M. M. Faul, J. R. McCarthy, J.
Med. Chem. 2001, 44, 2061–2064.
[8] P. V. Devasthale, S. Chen, Y. Jeon, F. Qu, C. Shao, W. Wang, H. Zhang, D. Farrelly, R. Golla, G. Grover, T. Harrity, Z. Ma, L. Moore, J.
Ren, R. Seethala, L. Cheng, P. Sleph, W. Sun, A. Tieman, J. R. Wetterau, A. Doweyko, G. Chandrasena, S. Y. Chang, W. G.
Humphreys, V. G. Sasseville, S. A. Biller, D. E. Ryono, F. Selan, N. Hariharan, P. T. W. Cheng, J. Med. Chem. 2005, 48,
2248–2250.
[9] P. T. Cheng, P. Devasthale, Y. Jeon, S. Chen, H. Zhang US 2002, US 6414002 B1.
[10] P. T. W. Cheng, P. Devasthale, Y. T. Jeon, S. Chen, H. Zhang, PCT Int. Appl. 2001, WO 2001021602 A1.
SYNTHESIS OF [2H4]; [3H]; AND [14C] SCH 900875: A POTENT CXCR3 ANTAGONIST WITH POTENTIAL
THERAPEUTIC USE IN THE TREATMENT OF PSORIASIS
CHRIS V. GALLIFORD, KIMBERLY VORONIN, AMY SMITH, DAVID KOHARSKI, VAN TRUONG, SCOTT BORGES, CAROLEE
FLADER LAVEY, DAVID HESK, AND PAUL MCNAMARA
Schering-Plough Research Institute, 2015 Galloping Hill Road, K-15-A4545, Kenilworth, NJ 07033, USA
2
3
Abstract: H4, H and 14C–syntheses of the potent CXCR3 antagonist Sch 900875 were performed in order to aid the biological
evaluation of this compound.
Keywords: pyruvate; tritiation; CXCR3; deuterated; carbon–14
Introduction: Sch 900875 (1), (Figure 1) is a potent and selective CXCR3 antagonist with therapeutic potential in the treatment of
CXCR3 chemokine-mediated illnesses such as inflammatory disease (including psoriasis or inflammatory bowel disease); as well as
auto-immune diseases such as multiple sclerosis and rheumatoid arthritis.12H4, 3H and 14C-analogs were prepared in order to aid the
biological evaluation of this compound.
Figure 1. Structure of Sch 900875, 1.
Results and Discussion: Sch 900875 was synthesized in unlabelled form starting with a Fluoride–promoted nucleophilic aromatic
substitution of the commercially available pyrazine (2) with N-Boc–protected piperazine (3) (Scheme 1).2 After deprotection of the
Boc-group, reductive amination of the resultant piperazine with N-Boc-4-piperidinone yielded 6 in 73% yield. A second Boc-group
deprotection followed by amide bond formation with 4-benzoyl chloride (9) generated the advanced ester intermediate (10) in 82%
yield over three steps. The methyl ester was then elaborated via a three step sequence to the oxadiazole heterocycle furnishing 1 in
64% yield. Deuterated and tritiated intermediates were prepared and incorporated into the existing non-labeled synthesis depicted
in Scheme 1.
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
Scheme 1.
2H Chemistry: Synthesis of the 2H4-labeled analog 6 was accomplished using a modified procedure for the synthesis of N-benzyl-
4-hydroxypiperidine (12) reported by Grieco and co-workers (Scheme 2).3 Benzylamine trifluoroacetate salt, allyltrimethylsilane and
deuterated formaldehyde (employed as a 20% w/v aqueous solution), were heated to 401C and allowed to react overnight, yielding
57% of the desired N-benzyl hydroxypiperidine (12). 12 was then converted to the N-Boc-protected analog, before smoothly
2
undergoing TPAP-mediated oxidation to the desired labeled precursor H4-6 in 83% yield over two steps.4 The [2H4] synthesis was
completed according to Scheme 1, providing 503 mg of 2H4-1 at 97.8% purity. This procedure has been utilized for the synthesis of
several labeled-piperidinones in our laboratory, using an aqueous solution of labelled formaldehyde (Scheme 2).5
Scheme 2.
3H Chemistry: A tritium-label was readily introduced into the same synthetic route via 2,6-(3H)-4-chlorobenzoic acid by an amide
bond coupling reaction with 8. 48 mCi tritiated 4-chlorobenzoic acid 3H-9 was prepared by tritium exchange with THO using
3
Ru(acac)3 as the catalyst. H-10 was then converted to H-1 according to Scheme 1, yielding 31 mCi of 1 (Scheme 3).
3
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
Scheme 3.
14C Chemistry: The desired site of labeling was determined to be the aromatic pyrazine heterocycle 2. The synthesis of this
heterocycle has been published by Imai and coworkers, starting from methyl pyruvate and diaminomaleonitrile.6 Rather than carry
the labeled pyrazine through all of the synthetic steps outlined in Scheme 1, we envisaged coupling the whole of the right hand
heterocyclic system directly to the pyrazine core, generating 14C-10 directly and intersecting the existing route, with only a few
synthetic operations remaining (Figure 2).
Figure 2. Retrosynthetic analysis of C-14 labeled 10.
We were attracted to the idea of incorporating a 14C-labeled atom into the piperazine via a derivative of pyruvate. 14C-sodium
pyruvate (8) was selected as the labeling source, and incorporated into the heterocycle according to Scheme 4:
Scheme 4.
Treatment of sodium pyruvate 14C-(14) with concentrated HCl in methanol (7.5 N) generated methyl pyruvate 14C-(15) in situ.
After dilution with water, treatment with diaminomaleonitrile (16) led to the formation of heterocycle 14C-17. Refluxing aqueous HCl
then effected the hydrolysis and regioselective decarboxylation to yield 14C-18, which was then subjected to sequential treatment
with thionyl chloride and methanol, followed by phosphorus oxychloride giving 14C-2 in 68% yield at 98% radiochemical purity.
From 100 mCi of sodium pyruvate, 68 mCi of the desired chloropyrazine ester 14C-(2) was obtained. To utilize our modified
strategy required the synthesis of 13, which we accomplished in six steps starting from N-Boc-piperazine (3) (Scheme 5).
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
Scheme 5.
The key TBAF-promoted coupling of piperazine 13 with chloropyrazine 14C-2 proceeded smoothly to yield 18.6 mCi after flash
chromatography. 14C-6 was then converted to the target compound in identical fashion to Scheme 1 to yield 11.2 mCi at 98.6%
radiochemical purity.
Conclusions: The 3H2-Sch 900875-labelled fragment was prepared by ruthenium-catalyzed tritium-hydrogen exchange of
4-chlorobenzoic acid (9), which was readily incorporated into the existing synthesis. 2H4-Sch 900875 was prepared by introducing a
deuterium-labelled 4-piperidone (6) into the synthetic sequence. 14C-Sch 900875, with the label located in the pyrazine ring was
prepared by modifying the synthesis to introduce the 14C-label at the latest possible stage. 11.2 mCi was prepared starting from
1
100 mCi of sodium pyruvate. chemical purities of labelled compounds were assessed by HPLC, LC-MS, MS and H and 13C NMR.
Acknowledgements: The authors would like to thank the Schering-Plough Chemical Development group, for supplying synthetic
intermediates and Schering-Plough Research Institute Molecular Spectroscopy group for analytical support and structural analysis.
References
[1] (a) S. H. Kim, G. N. Anilkumar, M. K. C. Wong, Q. Zeng, S. B. Rosenblum, J. A. Kozlowski, Y. Shao, B. F. McGuinness, D. W. Hobbs
WO/2006/088837. (b) J. S. Fine, S. H. Kim, G. N. Anilkumar, M. K. C. Wong, Q. Zeng, S. B. Rosenblum, J. A. Kozlowski, Y. Shao, B. F.
McGuinness, D. W. Hobbs WO/2009/020534.
[2] 3 can be prepared via a four-step sequence starting with N-benzyl-protected alanine methyl ester. Homo-coupling with a 2-
amino propionic acid derivative leads to a diketopiperazine, which after reduction and protecting group manipulation gives 3.
For details see: G. N. Anilkumar, Q. Zeng, S. B. Rosenblum, J. A. Kozlowski, B. F. McGuinness, D. W. Hobbs WO/2006/088840.
[3] S. D. Larsen, P. A. Grieco, W. F. Fobare J. Am. Chem. Soc. 1986, 108, 12, 3512–3513.
[4] S. J. Ley, J. Norman, W. P. Griffith, S. P. Marsden Synthesis. 1994, 639.
[5] (a) D. Hesk, K. Voronin, P. McNamara, P. Royster, D. Koharski, S. Hendershot, S. Saluja, V. Truong, T. M. Chan J. Label. Compd.
Radiopharm. 2007, 50, 2, 131–137. (b) S. Ren, P. McNamara, P. Royster, J. Lee, S. Saluja, D. Koharski, S. Hendershot, V. Truong J.
Label. Compd. Radiopharm. 2007, 50, 7, 643–648.
[6] M. Mano, T. Seo, K. Imai Chem. Pharm. Bull. 1980, 28, 10, 3057–3063.
HANDLING OF TRITIUM REAGENTS IN MULTI-CURIE SCALE
¨
CHRISTIAN MEISTERHANS, STEFAN C. NUCKEL, AND ALBERT ZELLER
RC TRITEC AG, Speicherstrasse 60a, CH-9053 Teufen, Switzerland
Keywords: tritium; tritiated water; lithium tritide; tritiomethyl iodide
Abstract: Synthesis and applications of various tritium reagents (tritiated water, tritide reagents, tritiomethyl iodide) for the chemo-
and regioselective labelling are described and examples of products, which show high specific activities, are presented.
Introduction: Tritium-labelled compounds with high specific activities are applied in pharmaceutical research for studies of
metabolism (ADME), imaging technologies for the detection of tissue distribution (autoradiography), and experiments to elucidate
irreversible covalent binding to proteins.1 Metabolism studies in particular require chemo- and regioselectively labelled compounds.
Besides the standard reactions of halogen/tritium exchange and hydrogenation of unsaturated compounds, labelling with tritiated
water with high specific activity (SA), reduction of suitable precursor compounds with tritide reagents and the introduction of a
tritium-labelled methyl group have gained in popularity in the past decade.
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
Results and Discussion: Applications of tritiated water: Tritiated water (HTO, T2O) has been used for decades for H/T exchange.2
Sometimes the position of the exchange is predictable, but often additional exchange occurs, which is difficult to estimate. This
labelling method is of particular interest when no precursor is available for chemical transformations. Tritiated water is synthesized
in a well dried vessel using PtO2 and carrier-free tritium gas according to the literature.3 Pre-drying of commercially available PtO2,
which is not free of water, leads to an increased specific activity. In this context, aspects of radioprotection have to be considered
seriously as a tiny quantity of 7.5 ml (0.35 mmol) of T2O (assuming a specific activity of 57 Ci/mmol) corresponds to an activity of 20 Ci.
It is well known that tritiated water is rapidly incorporated through lungs and skin.4 Therefore, it is easy to understand that the
synthesis and all manipulations with tritiated water have to be carried out in a completely closed system. In case of release and
uptake of just a 1/100 of the above mentioned activity, an effective dose of 130 mSv would result. According to the Swiss
regulations, this is far above the limit for the effective dose of 20 mSv, the maximum annual exposure for an employee working with
radioactivity. Therefore, T2O is synthesized and handled using a tight tritium manifold system manufactured by RC TRITEC Ltd.,5,6,7
which allows manipulations without any release of T2O.
Typical approaches for H/T exchange using tritiated water vary pH or use different metal catalysts. Some of these methods were
deduced from acid and base promoted H/D exchange.8 The difficult radioprotection regulations allow the use of only small
quantities of tritiated water due to its high specific activity, whereas in an H/D exchange D2O can be utilized as neat solvent.
Figure 1. Acetyl chloride promoted H/T exchange.
Figure 1 shows an acid promoted H/T exchange at a steroid derivative yielding the tritium-labelled product with a specific activity
of 118 Ci/mmol due to additional exchange at the skeleton. We used tritiated water in twelvefold excess (19.4 Ci), acetyl chloride
with 18 equivalents, and produced a crude product with a radiochemical purity of480%.
Metal catalyzed H/D and H/T exchanges using RhCl3 Á 3 H2O and Ru acetylacetonate were found to promote ortho-exchange with
9,10
high regioselectivity.
The unspecific H/T exchange using PtO2 and tritium gas mostly gives products with low specific activities
(p1 Ci/mmol)11, and was successfully applied for H/T exchange of biopolymers in our laboratory.
An attractive method for the synthesis of compounds with high specific activities is the quench with tritiated water of carbanions
generated by deprotonation with strong lithium bases.
O
O
H
T
1. nBuLi, THF, -78 to -15˚C
2. T2O, -78˚C to rt
R1
N
R1
N
N
N
R2
S
R2
S
Figure 2. Deprotonation and quench with tritiated water gave a 2-thioxo-4-imidazolidinone derivative at 12 Ci/mmol.
The example in Figure 2 shows the deprotonation of a 2-thioxo-4-imidazolidinone derivative (21 mmol) with an equimolar quantity
of n-butyllithium and subsequent quench with 24 equivalents of tritiated water (28.5 Ci).
Figure 3. Specific activity of the product was 24 Ci/mmol.
Figure 3 illustrates the H/T exchange of a cyclopropyl derivative, when treated with a tenfold excess of lithium diisopropylamide
at À781C and quenched with 16 equivalents of tritiated water (18.8 Ci). The crude product had a radiochemical purity of 50%.
Applications of LiT: A milestone in regioselective tritium-labelling was the development of various lithium and other metal tritides,
all derived from lithium tritide. Than et al.12 used the observation to form finely powdered lithium hydride, obtained from stirring a
hexane solution of n-butyllithium and 1,2-bis(dimethylamino)ethane (TMEDA) under an atmosphere of hydrogen gas13 to synthesize
lithium tritide. With this method a specific activity of approximately the theoretically value (28.8 Ci/mmol) was achieved. LiT was
found to be the key compound for the syntheses of important reducing reagents: LiBT4, LiAlT4, LiBEt3T and Bu3SnT.3,14–16 These
compounds allow chemo- and regioselective labelling, making them versatile tools in tritium chemistry. However, the major
drawback of synthesizing tritides is the production of tritiated butane as an undesired by-product. The problem of gaseous waste
treatment prevents the wide spread use of these valuable reagents, which are known to give products with high specific activities.
The chemoselective reduction of an aldehyde in presence of an amide function was achieved with LiBT4 and the corresponding
primary alcohol had a specific activity of 21 Ci/mmol. The reduction of an ester function of sugar derivative with 40 mmol (4.0 Ci)
LiBT4 gave the alcohol with a specific activity of 55 Ci/mmol (Figure 4A). We encountered an enhanced specific activity of the
products when LiBT4 (estimated specific activity of approx. 100 Ci/mmol)17 was used compared to NaBT4 (estimated specific activity
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
40 Ci/mmol), which is synthesized by thermal H/T exchange with carrier-free tritium gas. In comparison with NaBT4, LiBT4 shows a
stronger reduction potential, allowing the reduction of ester and lactone functionalities. Furthermore, the solubility of LiBT4 in THF
and diethyl ether is higher. Since the Li cation is a stronger Lewis acid showing increased tendency to form complexes, a higher
diastereoselectivity is observed.17
Figure 4. A: Reduction of an ester with LiBT4 gave a primary alcohol with a specific activity of 56 Ci/mmol. B: In a one-step reaction, reduction of a carboxylic acid and
cleavage of the protected hydroxy groups was achieved, giving a primary alcohol at 55 Ci/mmol.
The powerful tritide reagents LiAlT4 and LiBEt3T (Supertritide), synthesized from LiT and AlCl3 or BEt3 respectively, can be used for
the reduction of various substrates and are well known to give products with high specific activities.3,14–16,18 In Figures 4B and 5, the
reduction of a carboxylic acid with LiAlT4 and the reduction of an ester with LiBEt3T are shown. Reactions were performed in typical
scales of 40 mmol of the tritide reagent, corresponding to an activity of 4.0 Ci. Depending on the substrate, the stoichiometry of the
tritide can vary as in the first example 6.7 equivalents of LiAlT4, in the second example 1.7 equivalents of Supertritide were used. In
both cases the theoretically expected values of the specific activities were almost achieved.
O
OH
T
S
S
R'
R'
LiBEt3T
O
T
N
N
R''
R''
Figure 5. Reduction of an ester with Supertritide giving a primary alcohol at 58 Ci/mmol.
Tributyltin tritide (nBu3SnT) is a versatile tool to accomplish dehalogenation of aliphatic compounds and deoxygenation of
alcohols.14 The synthesis of the reagent from LiT and nBu3SnCl with maximum specific activity of 28.8 Ci/mmol is facile and the
reagent can be readily purified by column chromatography prior to use.19 As nBu3SnT shows high chemoselectivity, it can be applied
even in presence of sensitive functional groups.3 Figure 6 illustrates the radical induced dechlorination using 3.28 Ci of tributyltin
tritide to give a ketone derivative at 35 Ci/mmol in a radiochemical yield of 20%.
R
R
nBu3SnT, AIBN
toluene, reflux
T
Cl
O
O
T
Cl
AIBN: α,α'-azoisobutyronitrile
Figure 6. Dechlorination with nBu3SnT yielded a ketone derivative at 35 Ci/mmol.
Applications of tritiomethyl iodide (CT3I): Different starting materials were reported for the syntheses of tritiomethanol with high
specific activity, which is converted to CT3I with hydriodic acid.20–23 Our approach was inspired by the synthesis of tritiomethanol
from carbon dioxide reduced with LiAlT4.22 Substitution of gaseous carbon dioxide by the non-volatile and commercially available
substrate diphenyl carbonate simplified the handling, and the final product CT3I shows specific activities in the range of 65 to 85 Ci/
mmol. With this reagent, methylation of alcohols, phenols, amines, thiols, etc. are accomplished in mmol scale. Due to its volatility
and toxicity, careful handling in a completely tight system has to be ensured. A selected example for an SN2 reaction with an alkynyl
derivative, which is generated in situ by deprotonation with butyllithium, is presented in Figure 7. The C-C coupling reaction was
performed in a 1.6 Ci (25 mmol) scale of CT3I and a 1.3-fold excess of acetylene derivative in a radiochemical yield of 32%.
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
Figure 7. Methylation of an alkynyl carbon gave a tritio-2-butin derivative at 65 Ci/mmol.
Conclusion: The need of tritium-labelled compounds with high specific activities in pharmaceutical research can be covered by
application of the reviewed tritium-labelled reagents. The use of tritiated water for H/T exchange should be considered if suitable
precursors are not available. While this method often gives rise to additional exchange on the skeleton, treatment of acidic C-H
positions with base may provide access to regioselective labelling. If regioselectivity is a prerequisite, the chemoselective reduction
of a precursor with a suitable tritide reagent is the method of choice. Methylated compounds can be obtained in a one-step reaction
using tritiomethyl iodide and the desmethyl precursor.
References
[1] D. Dean, Oral presentation at the 10th International Symposium of the IIS, Chicago 2009.
[2] J. L. Garnet, M. A. Long, Catalytic Exchange Methods of Hydrogen Isotope Labelling, in Isotopes in Physical and Biomedical
Sciences, Labelled Compounds (Part A), Editors E. Buncel, J. R. Jones, Elsevier, Amsterdam, 1987, pp. 86–121.
[3] M. Saljoughian, Synthesis 2002, 13, 1781–1801.
[4] E. A. Evans, Tritium and its compounds, Butterworths, London 1974.
¨
[5] E. Bannwart, A. Zeller, P. Strom, M. Skrinjar, in Synthesis and applications of isotopically labelled compounds, Vol. 7, Editors U.
Pleiss, R. Voges, Wiley, Chichester, 2001, pp. 664–666.
[6] K. Pfanner, A. Zeller, J. Label. Compd. Radiopharm. 1998, 41, 1033–1034.
¨
[7] P. Strom, J. Label. Compd. Radiopharm. 1998, 41, 1032.
[8] J. Atzrodt, V. Derdau, T. Fey, J. Zimmermann, Angew. Chem. Int. Ed. 2007, 46, 7744–7765.
[9] D. Hesk, J. R. Jones, W. J. S. Lockley, J. Label. Compd. Radiopharm. 1990, 28, 1427–1436.
[10] D. Hesk, J. R. Jones, W. J. S. Lockley, J. Label. Compd. Radiopharm. 1991, 30, 887–890.
[11] P. G. Williams, C. A. Lukey, M. A. Long, J. L. Garnet, J. Label. Compd. Radiopharm. 1990, 29, 175–192.
[12] C. Than, H. Morimoto, H. Andres, P. G. Williams, J. Org. Chem. 1995, 60, 7503–7507.
[13] R. Pi, T. Friedl, P. v. R. Schleyer, P. Klusener, L. Brandsma, J. Org. Chem. 1987, 52, 4299–4303.
[14] M. Saljoughian, P. G. Williams, Current Pharmaceutical Design. 2000, 6, 1029–1056.
[15] H. Andres in Synthesis and applications of isotopically labelled compounds, Vol. 7, (Eds: U. Pleiss, R. Voges), Wiley, Chichester,
2001, pp. 49–62.
[16] R. Voges, R. Heys, T. Moenius, Preparation of Compounds Labelled with Tritium and Carbon-14, Wiley, Chichester, 2009, pp. 146–177.
[17] C. Than, H. Morimoto, H. Andres, P. G. Williams, J. Org. Chem. 1996, 61, 8771–8774.
[18] H. Andres, H. Morimoto, P. G. Williams, J. Chem. Soc., Chem. Commun. 1990, 627–628.
[19] D. K. Jaiswal, H. Andres, H. Morimoto, P. G. Williams, J. Chem. Soc., Chem. Commun. 1993, 907–909.
[20] M. Saljoughian, H. Morimoto, P. G. Williams, J. Chem. Soc. Perkin Trans 1 1990, 1803–1808.
[21] D. G. Ott, V. N. Kerr, T. H. Whaley, T. Benziger, R. K. Rohwer, J. Label. Compounds Radiopharm. 1974, 10, 315–324.
[22] S. Lee, H. Morimoto, P. G. Williams, J. Label. Compounds Radiopharm. 1997, 39, 461–470.
[23] R. Bolton, J. Label. Compounds Radiopharm. 2001, 44, 701–736.
SYNTHESIS OF A C-14 LABELED DELTA-OPIOID RECEPTOR AGONIST BY CARBONYLATION
CHARLES S. ELMORE, WILLIAM E. FRIETZE, DONALD W. ANDISIK, GLEN E. ERNST, J. RICHARD HEYS, AND CATHY L. DANTZMAN
CNS Chemistry, AstraZeneca Pharmaceuticals LP, 1800 Concord Pike, Wilmington, DE 19350, USA
Abstract: Opioid drugs have long been known to bind to specific receptors in both the brain and periphery. First identified in the
1970s, these receptors are classified as mu, kappa and delta.1 While each of these receptors appear to be involved in regulation of
pain pathways and of mood, the delta opioid receptor produces a unique pharmacology. AstraZeneca has designed a number of
agonists for the receptor, and in order to better understand this receptor’s pharmacology, a C14-labeled compound was prepared.
The structure of this compound was ideal for a late stage carbonylation using C-14 carbon monoxide to give the C-14 labeled
compound an 18% overall radiochemical yield.
Key words: C-14 carbonylation; Delta-Opioid receptor agonist
Introduction: There are 3 distinct opioid receptor subtypes : mu, kappa, and delta. While these G-protein coupled receptors have a
relatively high sequence homology, they differ considerably in terms of their pharmacology and distribution both in brain as well as
in the periphery.2
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
Mu receptors underlie the analgesic, sedative, addictive, gastrointestinal, as well as respiratory depressant effects of the prototype
agonist, morphine and related drugs. Both enkephalin and endorphin peptides serve as endogenous ligands for mu receptors.
Kappa receptors, on the other hand, subserve a different type of analgesia than mu receptors, and agonists additionally are known
to produce diuresis, negatively regulate mood and produce hallucinogenic effects.3 Agonists of delta opioid receptors may confer
still a different type of analgesia than mu or kappa agonists, but without the same deleterious side effects associated with the other
opioid receptors. Enkephalins are the endogenous ligands for delta receptors, and agonists could therefore be referred to as
‘enkephalinergics’. Delta agonists have been demonstrated to produce antidepressant and anxiolytic effects in a variety of animal
models, suggesting further therapeutic indications.4
O
O
Et2N
Et2N
R
OR
H
N
N
H
N
N
H
R= H
BW373U86
R=Me SNC 80
O
Me2N
O
N
N
Et2N
R
N
N
R
S
1
Figure 1. Several non-peptidic, small molecule delta-opioid selective agonists.
Several small molecules have been reported to have high affinity and good selectivity for the delta-opioid receptor (Figure 1).
BW373U86 was the first reported non-peptidic delta-opioid selective receptor agonist;5 it displays high potency at the delta-opioid
receptor (1 nM) with450 selectivity over the mu-opioid receptor. Methylation of the BW373U86 and resolution of the enantiomers
afforded SNC 80 which displayed improved selectivity against mu-opioid binding (4300 fold) while also maintaining potency
(0.5 nM).6 Further modification of BW373U86 demonstrated that the piperazine methyls were not required for activity7 and in fact an
achiral molecule–substituting an olefin for the benzylic amine - could have high potency and good selectivity.8 In an effort to find a
compound with better drug-like properties, dimethylamide 1 was prepared, and in order to more fully understand its properties, it
was required labeled with C-14.
Results: The most obvious site for C-14 labeling of compound 1 was in the amide carbonyl (Scheme 1). We felt that incorporation
of the label using either late-stage carbonylation or carbonylation9 of an appropriate precursor would provide the target compound
in one to two steps. The requisite starting iodide 2 could be prepared in a straightforward method similar to that previously
reported.8
O
I
I
Me2N
O
N
N
N
N
N
I
N
N
N
Boc
ClH2C
N
S
S
S
14C]-1
2
[
Scheme 1. Retrosynthetic analysis of 1.
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

Abstracts
Isonipecotic acid was protected as the t-butylcarbamate and a standard peptide coupling was used to prepare the Weinreb amide from the
acid (Scheme 2). The amide was then coupled with the Grignard reagent prepared from 2-bromopyridine using the methodology of Knochel
to give the ketone in 63% yield.10 A similar coupling of a Grignard reagent prepared from 1,4-diiodobenzene afforded the tertiary alcohol
which was readily eliminated to give the olefin with concomitant removal of the t-butylcarbamate in 90% yield over the two step sequence.
Coupling of the secondary amine with the corresponding chloride afforded iodide 2 in moderate yield after silica gel chromatography.
Both carbonylation with C-14 carbon monoxide and carboxylation with C-14 carbon dioxide were considered for the label
incorporation. Carboxylations are in general higher yielding and less complex to run; however, carbonylation can provide amide 1 directly
while carboxylation would necessitate a further step to form the peptide.11 Additionally, the acidity of the thiazole might complicate the
halogen metal exchange reaction. Therefore, carbonylation was the first approach investigated for the synthesis of [14C]-1.
One equivilent of [14C]formic acid was dehydrated with sulfuric acid and the resulting 14CO was reacted with two equivilents of
iodide 2, 0.4 eq of Pd(Dppf)Cl2 and excess dimethylamine in a 1:1 mixture of THF:DMF mixture at 801C overnight (Scheme 3). After
workup, HPLC assay showed a radiochemical purity of 90% and a radiochemical yield of 47%. Purification by preparative HPLC gave
1.5 mCi (18% radiochemical yield) at a radiochemical purity of 97.9% and specific activity of 52 mCi/mmol.
OMe
N
O
OH
O
1. (Boc)2O
Br
N
iPrMgCl, THF
63%
2. EDCi, HOBt
NEt3 HN OMe
N
H
N
Boc
85%
I
I
O
H2SO4
N
I
N
OH
iPrMgCl, THF
90% over
two steps
N
N
Boc
Boc
I
ClH2C
N
I
N
S
N
K2CO3, H2O,
MeCN
N
N
N
55%
H
S
2
Scheme 2. Preparation of iodide 2.
H14CO2Na
H2SO4
100 ˚C
O
14C
I
14CO
Me2N
N
N
Pd(dppf)CH2Cl2,
Me2NH, THF, DMF
100 ˚C
N
N
N
N
47% (18% after purification by HPLC)
S
S
2
[
14C]-1
Scheme 3. Synthesis of [14C]-1 via carbonylation.
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 368–397

Abstracts
Conclusions: Delta-opioid agonist [14C]-1 was prepared in one radiochemical step from iodide 2. Close eluting radiochemical
impurities necessitated the discarding of several product containing fractions to give an overall radiochemical yield of only 18%.
Experimental: General: Pd(Dppf)Cl2 Á CH2Cl2, 2 M Me2NH in THF and DMF were obtained from Sigma-Aldrich Chemical Company
1
and were used directly in reaction. Sodium [14C]formate was obtained from American Radiolabeled Chemical Company. H NMR
spectra were recorded on an Avance 500 and were reference to residual solvent peak (7.26 for CDCl3).
N,N-dimethyl-4-((pyridin-2-yl)(1-((thiazol-4-yl)methyl)piperidin-4-ylidene)methyl)benzamide ([14C]-1) A three necked flask contain-
ing 13 mg (8.5 mCi, 0.16 mmol) of sodium [14C]formate was fitted with a rubber septum, a vacuum adapter, and a 901 bent adapter.
A flask containing 114 mg (0.14 mmol) of Pd(Dppf)Cl2 Á CH2Cl2, 156 mg (0.33 mmol) of iodide 2, 4 mL of DMF and 4 mL of 2 M Me2NH
in THF was attached to the other end of the bent adapter (N2 was bubbled through the solution for 2 min). The flask containing the
THF/DMF solution was cooled in liquid nitrogen and the apparatus evacuated. The valve leading to the vacuum was closed and
20 mL of concentrated sulfuric acid was injected through the septum. The sulfuric acid solution was warmed to 1101C for 1 h. and
the reaction flask was heated and stirred at 801C for 12 h. The DMF/THF mixture was concentrated to dryness and was then taken up
in MeCN to give 4.5 mCi. HPLC analysis (5 to 20% MeCN-0.1% TFA on a Phenomenex Luna C-18(2) column over 20 min) indicated a
radiochemical purity of 90%. Preparative HPLC (5 to 15% MeCN-0.1% TFA on 21.2 Â 250 mm Phenomenex Luna C-18(2) column over
45 min, 15 mL/min) afford 1.5 mCi of [14C]-1 as a solution in EtOH with a radiochemical purity of 97.9% (5 to 20% MeCN-0.1% TFA on
Phenomenex Luna C-18(2) over 20 min followed by wash with 100% MeCN) and a specific activity of 52 mCi/mmol. 1H NMR
(500 MHz, CDCl3) ꢀ ppm 2.7 (m, 4 H), 2.98 (br. s., 3 H), 3.10 (br. s., 3 H), 3.5 (m, 4 H), 4.41 (br. s., 2 H), 7.20 (m, 3 H), 7.39 (m, 3 H), 7.74
(br. s., 1 H), 7.86 (t, J = 7.5 Hz, 1 H), 8.72 (br. s., 1 H), 8.82 (br. s., 1 H). LC/MS (M1H): 421 (100%), 419 (19%), 422 (25%).
References
[1] M. Waldhoer, S. E. Bartlett, J. L. Whistler. Annu. Rev. Biochem. 2004, 73, 953–990.
[2] A. Mansour, R. C. Thompson, H. Akil, S. J. Watson. Journal of Chemical Neuroanatomy, 1993, 6, 351–362.
[3] B. L. Roth, K. Baner, R. Westkaemper, D. Siebert, K. C. Rice, S. Steinberg, P. Ernsberger, R. B. Rothman. Proc. Natl. Acad. Sci. U.S.A.,
2002, 99, 11934–11939.
[4] M. M. Torregrossa, E. M. Jutkiewicz, H. I. Mosberg, G. Balboni, S. J. Watson, J. H. Woods. Brain Research, 2006, 1069, 172-182;
E. M. Jutkiewicz. Molecular Interventions, 2006, 6, 162–169.
[5] K. Chang, G. Rigdon, J. Howard, R. McNutt. J. Pharma. Exp. Thera., 1993, 267, 852–857.
[6] S. N. Calderon, R. B. Rothman, F. Porreca, J. L. Flippen-Anderson, R. W. McNutt, H. Xu, L. E. Smith, E. J. Bilsky, P. Davis, K. C. Rice.
J. Med. Chem., 1994, 37, 2125–2128; S. N. Calderon, K. C. Rice, R. B. Rothman, F. Porreca, J. L. Flippen-Anderson, H. Kayakiri, H.
Xu, K. Becketts, L. E. Smith, E. J. Bilsky, P. Davis, R. Horvath. J. Med. Chem. 1997, 40, 695–704.
[7] N. Plobeck, D. Delorme, Z.-Y. Wei, H. Yang, F. Zhou, P. Schwarz, L. Gawell, H. Gagnon, B. Pelcman, R. Schmidt, S. Y. Yue, C.
Walpole, W. Brown, E. Zhou, M. Labarre, K. Payza, S. St-Onge, A. Kamassah, P.-E. Morin, D. Projean, J. Ducharme, E. Roberts. J.
Med. Chem. 2000, 43, 3878–3894.
[8] Z.-Y. Wei, W. Brown, B. Takasaki, N. Plobeck, D. Delorme, F. Zhou, H. Yang, P. Jones, L. Gawell, H. Gagnon, R. Schmidt, S.-Y. Yue,
C. Walpole, K. Payza, S. St-Onge, M. Labarre, C. Godbout, A. Jakob, J. Butterworth, A. Kamassah, P.-E. Morin, D. Projean, J.
Ducharme, E. Roberts. J. Med. Chem. 2000, 43, 3895–3905.
[9] C. S. Elmore, D. C. Dean, D. G. Melillo. J. Label. Compd. Radiopharm. 2000, 43, 1135–1144.
[10] P. Knochel, W. Dohle, N. Gommermann, F. F. Kneisel, F. Kopp, T. Korn, I. Sapountzis, V. A. Vu. Angew. Chem. Int. Ed. Engl. 2003,
42, 4302–4320.
[11] C. S. Elmore, D. C. Dean, R. J. DeVita, D. G. Melillo. J. Label. Compd. Radiopharm. 2003, 46, 993–1000.
J. Label Compd. Radiopharm 2010, 53 368–397
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org