Stable isotopic labelling of heterocyclic compounds

Article abstract of DOI:10.1002/jlcr.1235

Stable isotopically labelled (SIL) versions of Glimepiride 1 (10 steps, 11% overall yield), a blood glucose lowering drug, and Melagatran 13 (9 steps, 17% overall yield), an anticoagulant with similar uses to warfarin, were synthesized as internal standar

Full text of DOI:10.1002/jlcr.1235

Journal of Labelled Compounds and Radiopharmaceuticals
J Label Compd Radiopharm 2007; 50: 273–276.
Published online in Wiley InterScience
JLCR
(www.interscience.wiley.com). DOI: 10.1002/jlcr.1235
Short Research Article
Stable isotopic labelling of heterocyclic compoundsy
ANDREW J. BURTON* and ALAN H. WADSWORTH
GlaxoSmithKline, Gunnels Wood Road, Stevenage, Herts SG1 2NY, UK
Received 23 August 2006; Revised 18 December 2006; Accepted 18 December 2006
Abstract: Stable isotopically labelled (SIL) versions of Glimepiride 1 (10 steps, 11% overall yield), a blood glucose
lowering drug, and Melagatran 13 (9 steps, 17% overall yield), an anticoagulant with similar uses to warfarin,
were synthesized as internal standards for LCMS assays. Modifications of known routes (Weyer et al., US Patent
4379785, Hocchst, 1983; Antonsson et al., PCT International Application W09429336, 1994) to these compounds
are examples of the introduction of stable isotopes via heterocyclic intermediates. Copyright # 2007 John Wiley &
Sons, Ltd.
Keywords: Glimepiride; Melagatran; isotope labelling; pyrrolidinone; oxadiazolinone
Introduction
therefore the success of the synthesis hinged upon
construction of labelled pyrrolidinone 7. Fortunately, a
synthesis of the unlabelled pyrrolidinone had already
been developed by Rapoport.3 However, the synthesis
was low yielding and required the use of Raney nickel
for reduction of the cyanohydrin 4.
Marketed compounds Glimepiride, a blood glucose
lowering agent for use in the treatment of diabetes,
and Melagatran, an anticoagulant, were required in
stable isotopically labelled form by GSK as reference
standards for comparison work. Both syntheses
are variations of the published synthetic routes and
involve the stable labelling of heterocyclic precursors;
a pyrrolidinone and an oxadiazolinone in the case of
Glimepiride and Melagatran, respectively.
Thus, 2H5-ethyl iodide alkylation of ethyl acetoace-
tate under phase transfer conditions gave the mono-
alkylated b-ketoester in 76% yield. In contrast to
Rapoport’s reported conditions, no problem was
observed with over alkylation. The silyl cyanohydrin 4
was formed under standard conditions and reduced
with nickel hydride, generated in situ using Caddick’s
conditions.4 Here dibenzyl-dicarbonate was used
as the trap for the resulting primary amine. The
residual amine was protected with benzyl chorofor-
mate, giving an overall yield of 45% for the transforma-
tion. Catalytic hydrogenation of compound 5 and
spontaneous cyclisation, followed by acid induced
elimination gave the required labelled pyrrolidinone 7
(Scheme 2).
Results and discussion
2H5-Glimepiride
The Retrosynthesis of Glimepiride
1
is shown
(Scheme 1). In contrast to the published route,1 which
employs a coupling with phenylethyl isocyanate fol-
lowed by a chlorosulfonylation, commercially available
4-(2-aminoethyl)benzenesulfonamide 9 was selected as
a more suitable, alternative starting material
The activated mixed urea 8 could be formed on
treatment of 7 with 1,10-carbonyldiimidazole and could
then be reacted with the commercially available 4-(2-
aminoethyl) benzenesulfonamide 9 to give urea 10. The
formation of the sulfonyl urea was then readily
achieved by activation with methyl chloroformate to
give 11 and then reaction with trans-4-methycyclohex-
ylamine 12. Thus the overall synthesis of [M+5]
glimepiride had been completed in 10 steps, 11%
overall yield (Scheme 3).
Readily available 2H5-ethyl iodide seemed the most
appropriate isotopically labelled starting material and
*Correspondence to: Andrew J. Burton, GlaxoSmithKline, Gunnels
Wood Road, Stevenage, Herts SG1 2NY, UK.
E-mail: andrew.j.burton@gsk.com
yProceedings of the Ninth International Symposium on the Synthesis
and Applications of Isotopically Labelled Compounds, Edinburgh,
16–20 July 2006.
Copyright # 2007 John Wiley & Sons, Ltd.

274 A. J. BURTON AND A. H. WADSWORTH
I
Et
O
Me
O
O
EtO
Me
Et
H
H
O
S
O
S
N
N
N
NH2
H2N
O
N
O
O
O
Me
Me
O
O
N
H2N
H
Glimepiride
1
9
Scheme 1
1.2eq. Me3SiCN,
5mol% ZnI2,
CH2Cl2, 20˚C, 16h
1eq. EtI, K2CO3,
Et*
5 mol% BnEt3N+Cl-,
Et*
Me
NC
TMSO
20˚C, 41h
Me
CO2Et
CO2Et
CO2Et
95%
76%
O
Me
O
3
2
4
1. 2eq. NiCl2.6H2O, 2eq.
(CO2Bn)2O, 8eq. NaBH4, MeOH,
5-10˚C, 1.5h; 20˚C, 17h
45%
2. 4eq. ClCO2Bn, pyridine,
CH2Cl2, 20˚C
Et*
Et*
Me
Me
Et*
1. H2, 10% Pd-C,
EtOH
Me
c. H2SO4, 20ºC, 24h
95%
TMSO
(HO)
CO2Et
NHZ
TMSO
O
N
2. high vacuum,
20˚C
O
N
H
H
(~1/3 desilylated)
7
6
5
94%
* = position of 2H5 label
Scheme 2
13C1,15N3-Melagatran
of the amide and iodide. The final labelled atom
was introduced via reaction of the nitrile with 15N
labelled hydroxylamine. The resultant amidoxime
20 was then converted to the oxadiazolinone 21
via condensation with 1,10-carbonyldiimidazole.
Hoffmann rearrangement of 21 with Koser’s reagent
22 gave 13C1,15N3-benzylamine 23 as its tosylate salt
(Scheme 5).
The commercial synthesis of Melagatran2 13 involves
the coupling of dipeptide 14 with protected 4-amino-
methyl benzamidine 15 and glycolic acid 16. Previous
work published by GSK5 demonstrated that the desired
amidine could be obtained from mild hydrogenolysis of
oxadiazolinones, themselves easily synthesised and
compatible with the proposed labelling strategy
(Scheme 4)
Amidation of 4-iodophenyl acetic acid 17 with 15NH3
gave the corresponding primary amide 18. Next cyana-
tion using Buchwald’s conditions6 allowed introduction
of a doubly labelled nitrile group to give 13C1,15N2-
benzonitrile 19. In addition the coupling reaction
produced two major impurities: the dehalogenated
amide and a dimer formed from a Buchwald coupling
Peptide coupling of benzylamine 23 and N-protected
dipeptide 14 followed by Boc deprotection gave amine
24. Alkylation with the p-nosylate derivative of glycolic
acid lead to over-alkylation. Instead, reductive amida-
tion using benzyl glyoxylate yielded the direct precur-
sor to [M+4] Melagatran. Hydrogenolysis cleaved
both the benzyl protection and oxadiazolinone to
complete the synthesis in 9 steps and 17% overall yield
(Scheme 6).
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 273–276
DOI: 10.1002.jlcr

STABLE ISOTOPIC LABELLING OF HETEROCYCLIC COMPOUNDS 275
O
S
NH2
Me
Et*
1.2eq.
Me
Et*
O
Me
Et*
O
1.2eq. CDI, toluene,
110˚C, 9h
H2N
9
NH2
MeOH, 20˚C
72%
O
S
N
O
N
O
O
quantitative
conversion
by NMR
N
O
N
H
O
N
N
7
H
10
8
ClCO2Me,
10 mol% DMAP,
85%
NEt3, CH2Cl2, 20˚C
H2N
12
Me
Et*
Me
Et*
O
CH3
H
H
O
S
O
H
N
N
N
OMe
dioxane, 100˚C
S
N
O
N
O
O
Me
O
O
64%
O
N
H
O
N
H
[M+5]-Glimepiride
11% overall yield
for 10 steps
11
2H5-1
2
* = position of H5 label
Scheme 3
NH
NH
Cbz
O
N
NH2
H
H
N
OH
HO
N
N
H2N
N
O
N
H
H
O
15
O
O
O
O
14
Melagatran
13
Ph
O
OH
O
16
Scheme 4
1.2eq. K
MeNHCH CH NHMe,
DMF, 90˚C
C N, CuI,
1. CDI, MeCN, 20˚C
2. 1.25eq. NH OAc,
K CO
15N
13C
I
I
O
O
O
H215
N
H215
N
70%
HO
94%
19
17
18
2eq. NH OH.HCl,
Pr NEt,
EtOH, 60˚C
~93% crude
OH
-
I
O
O
15N
13C
15N
13C
OH
SO3
15N
13C
O
O
OTs
1. CDI, DMF, 20˚C
2. DBU, 65˚C
15N
H
22
15N
H
O
15NH2
O
H315N+
MeCN, 80˚C
H125
N
H215
N
23
Scheme 5
Copyright # 2007 John Wiley & Sons, Ltd.
21
63%
89%
20
J Label Compd Radiopharm 2007; 50: 273–276
DOI: 10.1002.jlcr

276 A. J. BURTON AND A. H. WADSWORTH
O
O
15N
13C 15
15N
13C
O
O
15N
H
1. Me N(CH ) N=C=NEt .HCl,
DMAP, MeCN, 20˚C
N
H
+
O
H
15N
H215
N
N
N
H2N
.HCl
OH
O
N
70%
O
H
O
O
O
O
S OH
24
2. 4M HCl in dioxan, 20˚C
14
O
23
O=CH.CO CH Ph,
NaBH(OAc) , CHCl , 20˚C
multicomponent
mixture
(4-NO )-PhSO .OCH .CO CH Ph
84%
15NH
13C
O
15N
O
15NH2
H , Pd-C, EtOH, 20˚C
84%
13C 15
H
15N
N
H
HO
N
N
H
H
15N
O
N
O
O
O
N
H
O
[M+4]-Melagatran
13C1,15N3 -13
O
O
17% overall yield
for 9 steps
25
Scheme 6
REFERENCES
Conclusion
1. Weyer R, Hitzel V, Geisen K, Regitz G. US Patent
4379785 Hoechst, 1983.
Modification of known routes to Glimepiride and
Melagatran have been investigated and have allowed
the introduction of stable isotopes into compounds via
heterocyclic intermediates. These compounds have
been used successfully within GSK as reference
standards in therapeutic studies.
2. Antonsson KT, Bylund RE, Gustafsson ND, Nilson
NOI. PCT International Application WO9429336 Astra
AB, 1994.
3. Bishop JE, Nagy JO, O’Connell JF, Rapoport H.
J Am Chem Soc 1991; 113: 8024–8035.
4. Caddick S, Haynes AKdeK, Judd DB and Williams
MRV. Tetrahedron Lett 2000; 41: 3513–3516.
DOI:10.1016/S0040–4039(00)00410–X
5. Bolton RE, Coote SJ, Finch H, Lowdon A, Pegg N,
Vinader MV. Tetrahedron Lett 1995; 36: 4471–4474.
DOI:10.1016/0040-4039(95)00755-2
Acknowledgements
We are grateful for support and helpful suggestions
from other members of the GSK Isotope group in
Stevenage, UK.
6. Zanon J, Klapars A, Buchwald SL. J Am Chem Soc
2003; 125: 2890–2891. DOI: 10.1021/ja0299708
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 273–276
DOI: 10.1002.jlcr