Synthesis of [thiazolium-2,2′-14C2]-SAR97276A from [14C]-thiourea

Article abstract of DOI:10.1002/jlcr.1816

[thiazolium-2,2′-¹⁴C₂]-SAR97276A, a bis(thiazolium) antimalarial development candidate, was synthesized from [ ¹⁴C]-thiourea with an overall radiochemical yield of 15%. The synthetic route involves a modified procedure for the synthesis of [ ¹⁴C]-sulfurol, also a key intermediate in thiamine synthesis, which was developed due to unlabelled chemistry proving irreproducible with the radiolabelled substrate. Copyright

Full text of DOI:10.1002/jlcr.1816

Research Article
Received 7 April 2010,
Revised 10 June 2010,
Accepted 14 June 2010
Published online 16 September 2010 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.1816
Synthesis of [thiazolium-2,20-14C2]-SAR97276A
from [14C]-thiourea
John M. Herbert, Franck Le Strat, Delphine G. Oumeddour,
Ã
Stephen C. Passey, Keith Taylor, and David M. Whitehead
[thiazolium-2,20-14C2]-SAR97276A, a bis(thiazolium) antimalarial development candidate, was synthesized from [14C]-
thiourea with an overall radiochemical yield of 15%. The synthetic route involves a modified procedure for the synthesis of
[
14C]-sulfurol, also a key intermediate in thiamine synthesis, which was developed due to unlabelled chemistry proving
irreproducible with the radiolabelled substrate.
Keywords: antimalarial; carbon-14; thiazole; [14C]-sulfurol; [14C]-thiourea
low specific activity prior to being repeated at high specific
Introduction
activity. High specific activity materials were characterized by
comparative TLC (to known standards) and 1H-NMR. Merck silica
radio-TLC (RTLC) plates were analysed by electronic autoradio-
graphy using a Packard Instantimager. Activities were deter-
mined by liquid scintillation analysis using a Packard Tri-Carb
1900TR. All H-NMR data presented, apart from that of the final
product, are for low specific activity material and were recorded
on a Bruker DRX 500 NMR instrument.
Malaria threatens 41% of the world’s population, infects
between 350 and 500 million people each year, and is
responsible for the deaths of 1–3 million children each year.
Increasing resistance by strains of the malarial parasite
Plasmodium falciparum to the most commonly used drugs,
chloroquine and antifolate sulphadoxine/pyrimethamine, to-
gether with concerns over neuropathic side effects with
mefloquine and the reported first instances of resistance even
to artemisinin, urgently requires the development of inexpen-
sive antimalarials with a novel mode of action.1
1
a-Acetyl-a-chloro-c-butyrolactone
a-Acetyl-g-butyrolactone (5) (20.0 g, 0.156 mole) was stirred
while cooling to –101C (acetone/dry ice bath) in a 100 mL RB
flask, under a nitrogen atmosphere. Sulfuryl chloride (21.0 g,
0.156 mole) was added dropwise over 1 h while allowing the
mixture to warm to room temperature and the reaction mixture
was left to stir overnight at room temperature. After 24 h, the
reaction mixture was poured into ice (ꢀ100 g) and stirred with
ethyl acetate (100 mL) for 1 h. The phases were separated and
the aqueous layer was further extracted with ethyl acetate
(2 ꢁ 100 mL). The organic phases were combined, dried over
anhydrous magnesium sulfate and filtered before concentration
in vacuo and Kugelrohr distillation to afford a-chloro-a-acetyl-
g-butyrolactone, (23.8 g, 92%). dH (CDCl3) 2.48 (1H, m), 2.57
(3H, s), 3.21 (1H, m), 4.38 (1H, m), 4.45 (1H, m).
SAR97276A (1a) is a bis(thiazolium) development candidate
which inhibits phospholipid biosynthesis in P. falciparum, there-
by preventing the parasites’ growth. It is highly active against
both chloroquine resistant and chloroquine sensitive strains.2
In order to perform in vitro and in vivo metabolism and
pharmacokinetic studies, specifically labelled SAR97276A (1b)
was required. Owing to low dosage levels a high specific activity
was needed and therefore the route was designed to incorporate
two carbon-14 labels, with one in each thiazolium unit.
The reported route2 to 1a involves only one step in which two
molar equivalents of 4-methyl-5-thiazolylethanol (sulfurol) (2a)
are combined with one molar equivalent of 1,12-dibromo-
dodecane (3) to furnish the product as the dibromide salt
(Scheme 1). A synthesis of [14C]-sulfurol (2b) had already been
achieved by Williams and Ronzio3 in their synthesis of
[14C]-thiamine but, in developing the route to [14C2]-SAR97276A
(1b) based on this previous work, we found it necessary to
modify the synthesis of [14C]-4-methyl-5-thiazolylethanol (2b) to
improve yield and radioaccountability.
5-Acetoxy-3-chloro-2-pentanone (4)
a-Chloro-a-acetyl-g-butyrolactone (23.8 g, 0.146 mole) was mixed
with water (3 mL), acetic acid (18 mL) and concentrated
Department of Isotope Chemistry and Metabolite Synthesis, sanofi-aventis,
Alnwick Research Centre, Willowburn Avenue, Alnwick, Northumberland
NE66 2JH, UK
Experimental details
[14C]-Thiourea was purchased from GE Healthcare. Unlabelled 2-
Amino-4-methyl-5-thiazolylethyl acetate (6a) and other reagents
were supplied by Sigma-Aldrich. All steps were carried out at
*Correspondence to: Stephen C. Passey, Department of Isotope Chemistry and
Metabolite Synthesis, sanofi-aventis, Alnwick, Northumberland NE66 2JH, UK.
E-mail: stephen.passey@sanofi-aventis.com.
J. Label Compd. Radiopharm 2011, 54 89–92
Copyright r 2010 John Wiley & Sons, Ltd.

J. M. Herbert et al.
Br
Br
S
OH
N+
3
N+
S
N
S
Br
HO
HO
Br
1a
2a
Scheme 1.
hydrochloric acid (0.9 mL) and heated to reflux for 18 h, then was extracted with ethyl acetate and the organic phases
cooled to room temperature. Acetic anhydride (19.8 g, combined, dried over anhydrous magnesium sulfate and
0.194 mole) was added, and the mixture was heated to 1201C filtered to give a solution of 2b (3462 MBq, 50% radiochemical
for a further 6 h, then cooled to room temperature and stirred yield). Carrying out the same procedure on the second batch of
for a further 72 h. The reaction mixture was filtered through a the mixture gave 3434 MBq, 51% radiochemical yield. dH (CDCl3)
plug of silica and washed through with ethyl acetate:hexane 2.30 (3H, s), 2.86 (2H, t), 3.55 (2H, m), 4.79 (1H, m), 8.76 (1H, s).
(1:1). The filtrate was concentrated in vacuo and purified by
column chromatography on silica (10% ethyl acetate in hexanes) [thiazolium-2, 20-14C2]-SAR97276A (1b)
to give 5-acetoxy-3-chloro-2-pentanone, (4) (20.2 g, 77%). dH
(CDCl3): 2.05 (3H, s), 2.11 (1H, m), 2.34 (4H, m), 4.19 (2H, m), 4.30
The combined batches of 2b, (6896 MBq, 3.16 mmole) and 1,12-
dibromododecane (638 mg, 1.94 mmole) were combined neat in
(1H, m).
a 50 mL round bottomed flask. The mixture was stirred and
heated to 1101C for 4.5 h. The crude reaction product was
washed once with diethyl ether to remove remaining starting
[thiazole-2-14C]-2-Amino-4-methyl-5-thiazolylethyl acetate (6b)
5-Acetoxy-3-chloro-2-pentanone, (4) (1.20 g, 6.72 mmole) was materials then purified by crystallization from a mixture of
added to a solution of [14C]-thiourea (7609 MBq at 2180 MBq/ acetone and iso-propanol to give [thiazolium-2,20-14C]-
mmol, 3.49 mmole) in ethanol (50 mL). The mixture was stirred SAR97276A, (1b) 2275 MBq, 33% radiochemical yield. The
under nitrogen and heated to reflux for 27 h, then cooled and radiochemical purity was assessed by HPLC to be 98%. dH
concentrated in vacuo. The residue was purified by column (CDCl3) 1.26-1.64 (16H, m), 1.77, (4H, m), 2.45 (6H, s) 3.01 (4H, t),
chromatography on silica gel (3–5% ethanol in dichloro- 3.63 (4H, m), 4.43 (4H, t), 5.20 (2H, s), 10.01 (2H, s); m/z (ESI-TOF)
methane). Pure fractions were combined to give 6b 457 ([M-H]1), 229 (M21, 100%). HRMS 457.2691 (calc. For
14
(6924 MBq, 91% radiochemical yield). This reaction was repeated C C2H41N2O2S2 457.2669).
with 7481 MBq of [14C]-thiourea to give 6733 MBq of 6, 90%
22
radiochemical yield. dH (CDCl3): 2.00 (3H, s), 2.11 (3H, s), 2.87 (2H,
t), 4.09 (2H, t), 9.25 (2H, s).
Results and discussion
All route development work was performed using low specific
activity material (44 MBq/mmol).
[thiazole-2-14C]-2-(4-Methylthiazol-5-yl)ethyl acetate (8b)
Unlabelled 5-acetoxy-3-chloro-2-pentanone (4) was first
[thiazole-2-14C]-2-Amino-4-methyl-5-thiazolylethyl acetate (6b)
synthesized from a-acetyl-g-butyrolactone (5). Chlorination of
5, followed by ring opening and acylation with acetic anhydride
in acetic acid gave the desired chloropentanone (4) in excellent
yield (Scheme 2).
Formation of the thiazole ring system and incorporation of
(6924 MBq, 2180 MBq/mmole, 3.18 mmole) was dissolved in
phosphinic acid (6 mL of 50% aqueous solution). This solution
was cooled to ꢂ51C under nitrogen and stirred while adding
sodium nitrite (228 mg, 3.3 mmole). Effervescence and a colour
change from bright yellow to orange were observed. After
the carbon-14 label was achieved using classical Hantzsch
45 min, no further change in the reaction mixture was observed
by RTLC (35:15:1, ethyl acetate:hexane:triethylamine, silica). The
chemistry. [14C]-Thiourea was treated with excess 4 in refluxing
ethanol for 24 h. The resultant [2-14C]-2-amino-4-methyl-5-
mixture was observed to contain 53% [thiazole-2-14C]-2-(4-
thiazolylethyl acetate (6b) was isolated by precipitation
methylthiazol-5-yl)ethyl acetate (8b) and 19% 2b by RTLC. The
from ether as the hydrochloride salt. The free base was
liberated by basification with sodium hydrogencarbonate
reaction mixture was kept at ꢂ51C, basified with 6 M aqueous
sodium hydroxide to pH 13, and then extracted with ethyl
acetate (4 ꢁ 100 mL) and the organic phases were combined,
dried over anhydrous magnesium sulfate and filtered to afford
(Scheme 2).
As with the Hantzsch synthesis of [2-14C]-2-amino-4-methyl-5-
thiazolylethyl acetate (6b), the diazotization/reduction of the
the mixture of products, 8b12b, which was carried through into
amino group and subsequent deprotection to give [2-14C]-4-
the next reaction. This reaction was repeated with 6733 MBq of
methyl-5-thiazolylethanol (2b) has previously been described by
6b, with similar results.
Williams and Ronzio3 in their synthesis of [14C]-thiamine. Here
we experienced a deviation from their results. Unlabelled trials
of the diazotization/reduction returned 2a in good yields after
[thiazole-2-14C]-4-Methyl-5-thiazolylethanol (2b)
The first batch of the mixture of 8b and 2b obtained above was
dissolved in concentrated hydrochloric acid (5 mL) and stirred at
room temperature under nitrogen. The deprotection was
followed by RTLC and reached completion after 1 h 30 min.
The reaction mixture was then cooled to ꢂ51C and basified
with 6 M sodium hydroxide solution to pH 14. The mixture
deprotection. Small quantities of the 2-chloro side-product 7
were also separated from the product by column chromato-
graphy (Scheme 3). However, repeated attempts to achieve
similar results with unlabelled material from the same batch
spiked with ꢀ1% radiolabelled substrate 6b (giving material
with a specific activity of 44 MBq/mmol) were unsatisfactory.
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2011, 54 89–92

J. M. Herbert et al.
AcOH, HCl
Ac2O, H2O
SO2Cl2
92%
O
O
O
O
O
O
Cl
77%
5
S
14C
NH2
14C
O
Cl
H2N
NH2
S
N
O
O
O
O
EtOH
91%
4
6b
Scheme 2.
NaNO2
NH2
Cl
phosphinic acid
S
N
S
N
S
N
+
O
c.HCl
O
HO
HO
6a
7
0 °C
2a
Scheme 3.
NH2
NaNO2
H3PO2
S
N
c. HCl
55%
S
N
S
N
S
N
+
HO
O
O
O
2a
HO
O
6a
2a
68%
8a
12%
Scheme 4.
Yields were low (16–30%), and radioaccountability was also with concentrated hydrochloric acid to give only the desired 2a
often below 95%, and therefore insufficient to permit transfer of (Scheme 4).
the method to high specific activity work. Also, observations
A trial with low specific activity starting material gave a
during the process did not match the literature: whereas radiochemical yield of 54% over two steps with 100% radio
Williams and Ronzio3 reported vigorous effervescence upon accountability enabling us to transfer to high specific activity
addition of phosphinic acid, a very slow reaction was observed work. Subsequent transfer to material with a specific activity of
in our case, with gas evolution reduced to the formation of a 2150 MBq/mmol gave a yield of 50–51% over two steps, again
thin film of bubbles on the surfaces in the flask. These results with quantitative recovery of all initial activity. Reproducibility
were not improved by using fresh acid, or by adding a higher was confirmed with a two further high specific activity syntheses
concentration of phosphinic acid (50% versus 30%) than that giving near-identical yields.
used in the original work. As the evolution of gas resulting from
The final step in the synthesis involved the reproduction of a
the reduction by phosphinic acid appeared slow, the reaction neat, melt reaction on a millimolar scale. Unlabelled trials on a
was attempted in neat 50% phosphinic acid without hydro- 1–5 g scale gave excellent yields of SAR97276A (1a) on heating
chloric acid (this would also have avoided formation of the an undiluted mixture of liquid 4-methyl-5-thiazolylethanol (2a)
chlorinated product). Unlabelled trials gave very good conver- and solid 1,12-dibromododecane (3) (melting point 38–421C). In
sions to 2a and 2-(4-methyl-thiazol-5-yl)ethyl acetate (8a) as a trials spiked with low levels of radioactivity and high specific
2:1 mixture. This step was followed by treatment of the mixture activity syntheses, the reaction was performed using 1-2 mmol
J. Label Compd. Radiopharm 2011, 54 89–92
Copyright r 2010 John Wiley & Sons, Ltd.
www.jlcr.org

J. M. Herbert et al.
Br
Br
H
14C
N+
S
H
14C
H
14C
S
OH
3
N+
S
N
Br
HO
HO
Br
1b
2b
Scheme 5.
of 2b. Mixing was ensured by utilizing as small a vessel as
possible with vigorous stirring.
Although the reasons for our failure to replicate the literature
procedure for the synthesis of [2-14C]-4-methyl-5-thiazolyl-
Low recoveries of activity were observed at higher tempera- ethanol (2b) are not fully clear, the apparent slowness of
tures (1251C), perhaps owing to volatility of the [2-14C]-4-methyl- the reduction step for the radiolabelled material appears to
5-thiazolylethanol (although the b.p. of sulfurol is estimated at be
a factor. That this radical reduction appears to be
around 2821C, it is a flavour/fragrance molecule with a distinctive affected negatively by radiolabelled substrate begs the
‘beefy’ aroma). Lowering the temperature slightly to 1101C gave question as to whether radiolysis may interfere with this
recoveries of495% of the starting activity.
process.
A conversion of 70% of [2-14C]-4-methyl-5-thiazolylethanol to
[14C2]-SAR97276A (1b) was observed by RTLC after overnight
stirring of a melt consisting of 2b and 1,12-dibromododecane
(3) at 1101C. Precipitation from ether and subsequent recrys-
tallization of the precipitate gave pure (498% RCP) [14C2]-
SAR97276A (1b) in 33% radiochemical yield (Scheme 5). This
means of purification proved to be the most convenient and
reliable, since attempts at purification of 2b by preparative HPLC
were less successful and, while the solid was comparatively
stable upon storage, solutions of 2b in lower alcohol solvents or
in mixtures containing these degraded significantly within 24 h.
Acknowledgements
The authors thank Dr L. Rivron and Mr D. Wilson for carrying out
specific activity measurements, and Dr S. J. Byard for recording
NMR spectra.
REFERENCES
[1] V. Sharma, Mini-Rev. Med. Chem. 2005, 5, 337.
´
[2] A. Hamze, E. Rubi, P. Arnal, M. Boisbrun, C. Carcel, X. Salom-Roig,
M. Maynadier, S. Wein, H. Vial, M. Calas, J. Med. Chem. 2005, 48,
3639.
[3] D. L. Williams, A. R. Ronzio, J. Am. Chem. Soc., 1952, 74,
2409.
Conclusion
A
reproducible synthetic route to [thiazolium-2,20-14C2]-
SAR97276A has been developed, with an overall radiochemical
yield of 15% from [14C]-thiourea.
www.jlcr.org
Copyright r 2010 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2011, 54 89–92