Identification, synthesis and structural determination of some impurities of candesartan cilexetil

Article abstract of DOI:10.1135/cccc2008072

All principal candesartan cilexetil impurities and/or degradation products were synthesized and identified. The differentiation of N-l and N-2 ethylated candesartan cilexetil derivatives was performed using ID and 2D NMR experiments. The influence of the magnetic field on the resolution and sensitivity of NMR experiments is shown. Selective hydrolysis of these compounds then provided corresponding producrs of hydrolysis of the ethoxy group. Institute of Organic Chemistry and Biochemistry.

Full text of DOI:10.1135/cccc2008072

Some Impurities of Candesartan Cilexetil
347
IDENTIFICATION, SYNTHESIS AND STRUCTURAL DETERMINATION OF
SOME IMPURITIES OF CANDESARTAN CILEXETIL
b
Jaroslav HAVLÍČEKa1,*, Zuzana MANDELOVÁa2, Ruediger WEISEMANN ,
a4
a5
Ivo STŘELECa3, Lukáš PLAČEK and Stanislav RÁDL
a
Zentiva, U kabelovny 130, 102 37 Prague 10, Czech Republic;
e-mail: jaroslav.havlicek@zentiva.cz, zuzana.mandelova@zentiva.cz, ivo.strelec@zentiva.cz,
lukas.placek@zentiva.cz, stanislav.radl@zentiva.cz
Bruker BioSpin GmbH, Silberstreifen, 76287 Rheinstetten, Germany;
e-mail: ruediger.weisemann@bruker-biospin.de
1
2
3
4
5
b
Received April 17, 2008
Accepted July 3, 2008
Publish ed on lin e February 18, 2009
All prin cipal can desartan cilexetil im purities an d/or degradation products were syn th esized
an d iden tified. Th e differen tiation of N-1 an d N-2 eth ylated can desartan cilexetil derivatives
was perform ed usin g 1D an d 2D NMR experim en ts. Th e in fluen ce of th e m agn etic field on
th e resolution an d sen sitivity of NMR experim en ts is sh own . Selective h ydrolysis of th ese
com poun ds th en provided correspon din g producrs of h ydrolysis of th e eth oxy group.
Keyw ord s: Can desartan cilexetil; Im purity iden tification ; Im purity syn th esis; Tetrazoles;
Ben zim idazoles; NMR spectroscopy; 1H-15N HMBC; MS.
Can desartan cilexetil 1, a prodrug of can desartan 2, is a m em ber of a m od-
ern th erapeutic group of drugs kn own as an gioten sin II receptor an tago-
n ists (AIIRAs) used in th e treatm en t of h yperten sion (h igh blood pressure).
Th e drug is m arketed by AstraZen eca an d Takeda Ph arm aceuticals, un der
th e trade n am es Blopress, Atacan d, Am ias, an d Ratacan d1,2
.
Trityl can desartan cilexetil 3, a key in term ediate of th e syn th esis of can -
desartan cilexetil, can be prepared accordin g to th e paten t literature3 from
can desartan 2 in two steps via trityl can desartan 4. Detritylation of 3 un der
various con dition s h as been described in n um erous paten ts (Sch em e 1).
In spite of th e strict requirem en ts regardin g th e purity of drugs in devel-
oped coun tries, th e presen ce of both process-related im purities an d degra-
dation products can n ot be com pletely elim in ated. Th erefore, th e prin cipal
kn owledge n eeded for an y drug registration is th at of its im purities an d/or
degradation products. Th e specified im purities can be eith er iden tified or
un iden tified. Iden tified im purities sh ould be in cluded in th e specification
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362
© 2009 Institute of Organic Chemistry and Biochemistry
doi:10.1135/cccc2008072

348
Havlíček, Mandelová, Weisemann, Střelec, Plaček, Rádl:
wh en th ey are presen t at a level h igh er th an th e iden tification th resh old,
wh ich is usually 0.10%. Th ese im purities m ust n ot on ly be iden tified but
also in depen den tly syn th esized. In gen eral, th e im purities could be eith er
process-related or form ed by degradation of th e drug.
SCHEME 1
Syn th esis of can desartan cilexetil 1 from can desartan 2
Con siderin g th e structure of can desartan cilexetil 1, can desartan 2, can
be expected as th e m ajor degradation product. However, th is is true on ly
un der alkalin e con dition s; oth erwise usually th ree oth er degradation prod-
ucts are detected in h igh er am oun ts. LC-MS tech n ique iden tified th eir
probable structures 5–7 (Fig. 1). Com poun d 5 was described in paten t litera-
ture4 an d quite recen tly, both a paten t5 an d a paper6 listin g com poun ds 5–7
as im purities of can desartan cilexetil h ave appeared. However, n eith er
properties n or th e structure elucidation of th ese com poun ds h ave been
publish ed. Even th ough Rao et al.6 h ave described an isocratic HPLC for th e
separation of can desartan cilexetil an d its im purities, th e position s of th e
N-eth yl group in th e tetrazole rin g is n ot apparen t from th e paper. Th ere-
fore, we would like to report th e syn th esis an d structure determ in ation of
im purities 5–7 by several 1D an d 2D NMR experim en ts.
Literature search h as n ot revealed th e presen ce of com poun ds 8, 9 h avin g
com m on structural features with of 5 an d 6/7. Th erefore, we also decided
to syn th esize th ese com poun ds an d com pare th em to m in or im purities
form ed durin g th e can desartan cilexetil syn th esis.
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

Some Impurities of Candesartan Cilexetil
349
FIG. 1
Structures of th e discussed im purities 5–9
RESULTS AND DISCUSSION
Identification of Impurities 5–9
Im purities 5–7 are presen t both in th e com m ercially available active ph ar-
m aceutical in gredien t of can desartan cilexetil an d in th e drugs on th e m ar-
ket. Th ese im purities are also form ed durin g th e stability tests un der all
tested con dition s. LC-MS provided [M + H]+ m asses of im purities 5–7 in TIC
scan m ode (m/z 583 for im purity 5 an d m/z 639 for im purities 6 an d 7).
Com poun ds 5–7 were syn th esized an d ch aracterized. Th e [M + H]+ values
were specified usin g h igh -resolution MS for elem en tal com position ;
m/z 583.2297 correspon din g to C31H31N6O6 (im purity 5). Sim ilarly, th e m/z
values of 639.2930 an d 639.2931 for im purities 6 an d 7, respectively, cor-
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

350
Havlíček, Mandelová, Weisemann, Střelec, Plaček, Rádl:
respon din g to C35H39N6O6, were obtain ed. An alogously, h igh -resolution MS
of th e poten tial im purities 8 an d 9 provided [M + H]+ m/z values of
611.26147 an d 611.26154 correspon din g to C33H35N6O6. All com poun ds
5–9 were detected by UPLC in th e crude reaction m ixtures of can desartan
cilexetil obtain ed from 3 by both acidic deprotection 3 an d deprotection un -
der n eutral con dition s7 (Table III).
Synthesis of Impurities 5–9
Com poun d 5 is form ed as th e m ajor side product in am oun ts up to 10%
durin g acidic detritylation of 3. Th e paten ted procedures for th e acidic
detritylation in clude, am on g oth er con dition s, h eatin g in aqueous h ydro-
ch loric acid3 or in eth an olic h ydrogen ch loride8. Prolon ged h eatin g un der
th e m en tion ed con dition s leads to com plex m ixtures con tain in g also prod-
ucts of solvolysis of th e relatively un stable cilexetil ester group. First, we
tried to obtain 5 by ch rom atograph ic purification of th ese m ixtures but we
failed to obtain pure sam ples. However, exten sive screen in g of th e con di-
tion s revealed th at acidic h ydrolysis of 3 in aqueous aceton e provided
relatively pure com poun d 5 an d trityl alcoh ol, an d subsequen t ch rom a-
tograph ic purification provided good yield of 5. Th e sam e treatm en t of
can desartan cilexetil 1 provided even better yield (88%) of 5 with out ch ro-
m atograph ic purification .
Com poun ds 6 an d 7, iden tified by LC-MS as eth yl derivatives of 1, were
prepared by sim ple eth ylation of 1 with iodoeth an e in aceton e in th e pres-
en ce of potassium carbon ate at am bien t tem perature. Th e form ed isom ers 6
SCHEME 2
Syn th esis of im purities 8 an d 9
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

Some Impurities of Candesartan Cilexetil
351
an d 7 were isolated by flash ch rom atograph y but th eir structure elucidation
was n ot trivial an d required th e use of advan ced NMR m eth ods described in
th e followin g paragraph .
For th e syn th esis of com poun ds 8 an d 9, acidic h ydrolysis of th e respec-
tive N-eth yl derivatives 6 an d 7 usin g th e con dition s developed for th e syn -
th esis of 5 was used (Sch em e 2).
Structure Elucidation of Impurities 5–9
For th e assign m en t of proton s an d carbon s, advan ced 2D NMR tech n iques
1
were used. Th e H-15N HMBC NMR spectra of im purities 6 an d 7 are sh own
1
in Figs 3 an d 4. Ch em ical sh ifts of H an d 13C of com poun ds 6 an d 7 are
given in Tables I an d II, respectively. Th e in fluen ce of th e m agn etic field on
th e resolution an d sen sitivity of th e NMR experim en ts is sh own in Fig. 2.
Th e differen tiation of im purities 6 an d 7 of can desartan cilexetil was per-
1
form ed usin g 2D H-15N HMBC experim en ts. In 1H-15N HMBC, th ree cross
peaks (35-46, 36-46, 37-46) for th e tetrazole rin g were foun d in im purity 7,
wh ile in th e case of im purity 6, th e cross peak 35-46 was n ot observed.
By an alogy with th e discussed structures of com poun ds 6 an d 7, th e NMR
spectra of com poun ds 5, 8 an d 9 were in terpreted (see Experim en tal).
FIG. 2
13C-APT NMR spectrum of im purity 6 at two m agn etic fields
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

352
Havlíček, Mandelová, Weisemann, Střelec, Plaček, Rádl:
FIG. 3
1H-15N HMBC spectrum of im purity 6 in CDCl3, JN(XH) = 4 Hz
FIG. 4
1H-15N HMBC spectrum of im purity 7 in CDCl3, JN(XH) = 4 Hz
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

Some Impurities of Candesartan Cilexetil
353
TABLE I
1H, 13C an d 15N NMR assign m en ts for im purity 6 (500 MHz, CDCl3)
Position
δC
δH
δN
Multiplicity
JH,H
1
2
–
–
132.5
158.84
–
–
3
–
192.5
4
142.14
132.22
114.34
124.28
120.89
122.90
46.95
–
5
–
6
–
7
7.62
7.16
7.73
5.64
5.57
–
dd
t
7.9; 1.2
7.9
8
9
dd
d
7.9; 1.2
15.9
10
d
15.9
11
12
13
14
15
16
18
19
20
22
23
24
25
26
27
29
31
137.48
127.64
128.82
137.93
128.82
127.64
66.81
6.97
7.00
–
d
d
8.3
8.3
7.00
6.97
4.64
1.47
–
d
d
q
t
8.3
8.3
7.2
7.2
14.63
164.01
141.25
122.89
131.50
130.20
131.52
127.90
91.82
–
–
7.52
7.51
7.61
7.49
6.90
–
m
m
m
m
q
5.4
152.56
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

354
Havlíček, Mandelová, Weisemann, Střelec, Plaček, Rádl:
TABLE I
(Continued)
Position
δC
δH
δN
Multiplicity
JH,H
33
34
35
36
37
38
40
41
154.26
–
–
–
n ot foun d
n ot foun d
368.2
–
–
–
–
–
–
236.7
19.63
77.53
31.43a
1.56
4.61
1.98
1.44
1.73
1.34
1.51
1.25
1.73
1.34
1.98
1.44
3.47
0.84
d
5.5
m
m
m
m
m
m
m
m
m
m
m
q
42
43
44
45
23.60
25.14
23.60
31.40a
46
47
42.37
13.59
7.3
7.3
t
a
In terch an geable.
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

Some Impurities of Candesartan Cilexetil
355
TABLE II
1H, 13C an d 15N NMR assign m en ts for im purity 7 (500 MHz, CDCl3)
Position
δC
δH
δN
Multiplicity
JH,H
1
2
–
–
134.9
158.77
–
–
3
–
193.3
4
142.02
131.89
114.60
123.85
120.60
122.48
46.86
–
5
–
6
–
7
7.58
7.16
7.74
5.67
5.63
–
dd
dd
dd
d
7.9; 1.2
7.9
8
9
7.9; 1.2
15.9
10
d
15.9
11
12
13
14
15
16
18
19
20
22
23
24
25
26
27
29
31
116.33
126.52
129.30
135.83
129.30
126.52
66.65
7.00
7.05
–
d
d
8.2
8.2
7.05
7.00
4.67
1.49
–
d
d
q
t
8.2
8.2
7.1
7.0
14.56
163.95
141.50
126.30
130.20
127.40
129.78
130.53
91.71
–
–
7.83
7.44
7.48
m
dt
dt
m
q
1.5; 7.6
1.5; 7.6
6.94
–
5.4
152.49
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

356
Havlíček, Mandelová, Weisemann, Střelec, Plaček, Rádl:
TABLE II
(Continued)
Position
δC
δH
δN
Multiplicity
JH,H
33
34
35
36
37
38
40
41a
164.92
–
–
–
n ot foun d
377.2
–
–
–
–
288.9
–
–
302.6
19.51
77.42
31.31
31.30
23.49
23.46
25.05
1.50
d
5.5
m
m
m
m
m
m
m
m
m
m
m
q
1.91
1.43
1.72
1.27
1.46
1.22
1.72
1.27
1.91
1.43
4.40
1.30
42a
43
44a
45a
23.49
23.46
31.31
31.30
47.92
14.22
46
47
7.3
7.3
t
a
In terch an geable.
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

Some Impurities of Candesartan Cilexetil
357
Proposed Mode of Formation of Impurities 5–7
Form ation of can desartan cilexetil im purities 5–7 occurs durin g both th e
deprotection of trityl can desartan cilexetil 3 leadin g to can desartan cilexetil
1 an d th e purification of 1. Sin ce n o apparen t eth ylation agen t is presen t,
th e on ly reason able explan ation of th e form ation of th ese im purities is th at
th e eth oxy group of 1 serves as an alkylation species, an d alkylates, in an
in term olecular fash ion , position s N-1 an d N-2 of th e tetrazole rin g provid-
in g com poun ds 6 an d 7, respectively (Sch em e 3). To th e best of our kn owl-
edge, n o sim ilar in term olecular N-alkylation by an acyclic eth er h as been
reported.
SCHEME 3
Proposed m ech an ism of form ation of im purities 5–7
Wh ile un der acidic con dition s, com poun d 5 is form ed m ain ly by th e h y-
drolysis of th e eth oxy group of can desartan cilexetil or trityl can desartan
cilexetil (see Experim en tal, preparation of 5), th e h ydrolysis does n ot play
an im portan t role un der n eutral con dition s. Th is was proven by h eatin g
can desartan cilexetil solution s in various solven ts (see, e.g., Experim en tal,
Heatin g of can desartan cilexetil 1 in aqueous aceton e). Sim ilarly, form ation
of im purities 8 an d 9 from com poun ds 6 an d 7, respectively, can be ex-
plain ed.
CONCLUSIONS
Five im purities of can desartan cilexetil 5–9 were iden tified an d syn th esized.
Th e com plex structures of th e two N-eth yl substituted im purities were de-
term in ed by several 1D an d 2D NMR experim en ts. In 1H-15N HMBC, th ree
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

358
Havlíček, Mandelová, Weisemann, Střelec, Plaček, Rádl:
cross-peaks (35-46, 36-46, 37-46) for th e tetrazole rin g were foun d in im pu-
rity 7, wh ile in th e case of im purity 6, th e cross-peak 35-46 was n ot ob-
served. Th e advan tage of h igh er m agn etic field is n ot on ly in th e
im provem en t of sen sitivity (sh orter experim en tal tim e), but also in better
1
resolution of H an d 13C NMR spectra.
EXPERIMENTAL
Sam ples an d Reagen ts
Trityl can desartan cilexetil
3 was obtain ed from Zh ejian g Tian yu Ph arm aceutical
(h ttp://www.tian yuph arm a.com ). Oth er ch em icals used in th e syn th esis were purch ased
from Sigm a–Aldrich an d were used with out purification .
Syn th esis
Meltin g poin ts were m easured on a Kofler block an d are un corrected. Th e UV spectra were
recorded on a Hewlett–Packard 8452A spectroph otom eter (eth an ol) in th e ran ge 190–400 n m .
Flash ch rom atograph y was perform ed on silica gel Merck, particle size 0.04–0.063 m m . Th e
purity of th e substan ces prepared was evaluated by TLC on silica gel (FP KG F 254, Merck)
an d by Ultra Perform an ce Liquid Ch rom atograph y (UPLC) usin g a UPLC system Waters
Acquity with UV detection (colum n len gth 0.1 m , in tern al diam eter 2.1 m m , station ary
ph ase UPLC BEH-C8, tem perature 35 °C). Gradien t elution with m obile ph ase A (ph osph ate
buffer: 1.32 g (NH4)2HPO4 in 1000 m l of H2O, pH adjusted to 3.0 with 50% ph osph oric
acid), an d m obile ph ase B (aceton itrile) was used.
1-{[(Cyclohexyloxy)carbonyl]oxy}ethyl 2-Ethoxy-1-{[2′-(1H-tetrazol-5-yl)biphenyl-4-yl]m ethyl}-
1H-ben zim idazole-7-carboxylate (1)
A stirred m ixture of trityl can desartan cilexetil 3 (100 g, 117 m m ol), aceton itrile (1000 m l)
an d water (125 m l) was h eated at reflux for 10 h an d th e m ixture was stirred at am bien t
tem perature for addition al 2 h (UPLC results given in Table III, en try 1). Water (800 m l) was
th en added an d th e m ixture was stirred for addition al 10 h . Th e precipitate was filtered off,
m ixed with eth yl acetate (250 m l) an d stirred at 50 °C for 2 h . After coolin g, th e precipitate
was filtered off (Table III, en try 2) an d repeatedly crystallized from aqueous aceton e to pro-
vide 45.7 g (63.9%) of 1, m .p. 172–174 °C. Th e UPLC purity is given in Table III, en try 3.
UV (EtOH), λ (log ε): 214 (4.73), 254 (4.28), 304 (3.76). HRMS: m/z calculated for C35H39N6O6
[M + H]+ 639.2931, foun d 639.2930.
1-{[(Cycloh exyloxy)carbon yl]oxy}eth yl 3-{[2′-(1H-Tetrazol-5-yl)biph en yl-4-yl]m eth yl}-
2-oxo-2,3-dih ydro-1H-ben zim idazole-4-carboxylate (5)
Method A. A m ixture of 3 (2 g, 2.34 m m ol), aceton e (16 m l), water (4 m l) an d con cen -
trated h ydroch loric acid (0.2 m l) was h eated at reflux for 6 h . Th e m ixture was evaporated,
an d th e residue was purified by flash ch rom atograph y (h exan e–aceton e 8:2) to provide
0.95 g (69.7%) of 5, m .p. 246–248 °C (97.4% purity, RRT 0.66). 1H NMR (for n um berin g, see
Fig. 1): 1.22 m , 1 H (H-43); 1.31 m , 2 H (H-42, H-44); 1.35 d, 3 H, J = 5.5 (H-38); 1.39 m ,
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

Some Impurities of Candesartan Cilexetil
359
2 H (H-41, H-45); 1.46 m , 1 H (H-43); 1.63 m , 2 H (H-42, H-44); 1.83 m , 2 H (H-41, H-45);
4.57 m , 1 H (H-40); 5.24 d, 1 H, J = 16.2 (H-10); 5.33 d, 1 H, J = 16.2 (H-10); 6.72 q, 1 H, J =
5.4 (H-29); 6.93 d, 1 H, J = 8.4 (H-12, H-16); 7.00 d, 2 H, J = 8.4 (H-13, H-15); 7.09 dd, 1 H,
J = 7.9 (H-8); 7.26 dd, 1 H, J = 7.9, 1.2 (H-7); 7.28 dd, 1 H, J = 7.9, 1.2 (H-9); 7.54 ddd, 1 H,
J = 15.0, 7.6, 1.5 (H-25); 7.49 dd, 1 H, J = 7.9, 1.5 (H-27); 7.63 m , 1 H (H-24); 7.64 m , 1 H
(H-26); 11.50 bs, 1 H (N-3). 13C NMR (for n um berin g, see Fig. 1): 18.98 (C-38); 22.86 (C-42
or C-44); 22.89 (C-42 or C-44); 24.51 (C-43); 30.70 (C-41 or C-45); 30.73 (C-41 or C-45);
44.50 (C-10); 76.74 (C-40); 91.83 (C-29); 113.00 (C-6); 113.18 (C-9); 120.70 (C-8); 122.24
(C-7); 123.42 (C-23); 126.42 (C-12, C-16); 127.66 (C-25); 128.28 (C-5); 128.93 (C-13, C-15);
130.00 (C-4); 130.47 (C-27); 130.55 (C-24); 130.91 (C-26); 136.22 (C-11); 138.02 (C-14);
140.90 (C-22); 151.84 (C-31); 154.83 (C-2); 154.83 (C-33); 163.40 (C-20). UV (EtOH), λ (log ε):
212 (4.74), 254 (4.27), 316 (3.73). HRMS: m/z calculated for C31H31N6O6 [M + H]+ 583.2305,
foun d 583.2297.
Method B. A m ixture of 1 (2 g, 3.28 m m ol), aceton e (16 m l), water (4 m l) an d con cen -
trated h ydroch loric acid (0.2 m l) was h eated at reflux for 6 h . Th e m ixture was evaporated,
an d th e residue was crystallized from aceton e to provide 1.68 g (87.9%) of 5, m .p. 248 °C
(99.8% purity, RRT 0.66).
TABLE III
Purity of can desartan cilexetil prepared by detritylation in aqueous aceton itrile7
UPLC in % (RRT)a
En try
1
3
5
6
7
8
9
(1.00)
(2.42)
(0.66)
(1.33)
(1.58)
(0.90)
(1.13)
1
2
3
93.91
96.18
99.87
4.11
3.50
0.07
0.14
0.09
0.02
0.42
0.06
0.01
1.14
0.17
0.03
0.02
0.06
–
–
–
–
a
Th e form ed trityl alcoh ol was n ot in tegrated.
1-{[(Cycloh exyloxy)carbon yl]oxy}eth yl 2-Eth oxy-1-{[2′-(1-eth yl-1H-tetrazol-5-yl)biph en yl-
4-yl]m eth yl}-1H-ben zim idazole-7-carboxylate (6) an d 1-{[(Cycloh exyloxy)carbon yl]oxy}-
eth yl 2-Eth oxy-1-{[2′-(2-eth yl-2H-tetrazol-5-yl)biph en yl-4-yl]m eth yl}-1H-ben zim idazole-
7-carboxylate (7)
Potassium carbon ate (2.5 g) was added to a solution of 1 (6.1 g, 10 m m ol) in aceton e (150 m l)
an d th e m ixture was stirred un der n itrogen at am bien t tem perature for 1 h . Iodoeth an e (1 m l)
was th en added, an d th e m ixture was stirred for addition al 24 h . Th e in soluble portion was
filtered off an d wash ed with aceton e (25 m l), th e filtrate was evaporated (8.2 g) an d sepa-
rated by flash ch rom atograph y (h exan e–aceton e 8:2) to provide 2.9 g of fraction I con tain -
in g 99.9% of com poun d 7 (RRT 1.58, m .p 109–112 °C), 0.8 g of com bin ed fraction II
con tain in g 54.3% of com poun d 6 (RRT 1.33) an d 45.6% of com poun d 7 (RRT 1.58), an d 2.5 g
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

360
Havlíček, Mandelová, Weisemann, Střelec, Plaček, Rádl:
of fraction III con tain in g 99.75% of com poun d 6 (RRT 1.33) an d 0.22% of com poun d 7
(RRT 1.58, m .p. 102–103 °C).
1-{[(Cyclohexyloxy)carbony]loxy}ethyl 2-ethoxy-1-{[2′-(1-ethyl-1H-tetrazol-5-yl)biphenyl-4-yl]methyl}-
1H-benzimidazole-7-carboxylate (6). UV, λ (log ε): 214 (4.65), 254 (4.25), 308 (3.68). HRMS:
m/z calculated for C35H39N6O6 [M + H]+ 639.2931, foun d 639.2930.
1-{[(Cyclohexyloxy)carbonyl]oxy}ethyl 2-ethoxy-1-{[2′-(2-ethyl-2H-tetrazol-5-yl)biphenyl-4-yl]methyl}-
1H-benzimidazole-7-carboxylate (7). UV, λ (log ε): 216 (4.68), 258 (4.15), 304 (3.67). HRMS:
m/z calculated for C35H39N6O6 [M + H]+ 639.2931, foun d 639.2931.
1-{[(Cycloh exyloxy)carbon yl]oxy}eth yl 3-{[2′-(1-Eth yl-1H-tetrazol-5-yl)biph en yl-4-yl]m eth yl}-
2-oxo-2,3-dih ydro-1H-ben zo[d]im idazole-4-carboxylate (8)
A m ixture of 6 (0.2 g, 0.31 m m ol), aceton e (2 m l), water (0.5 m l) an d con cen trated h ydro-
ch loric acid (0.02 m l) was stirred in a vial at 75 °C for 6 h . Th e m ixture was evaporated an d
th e residue was crystallized from a m ixture of dieth yl eth er–aceton e to provide 0.13 g
(68.8%) of 8, m .p. 189–192 °C (99.6% purity, RRT 0.90). 1H NMR (for n um berin g, see Fig. 1):
0.85 t, 3 H, J = 7.3 (H-47); 1.25 m , 1 H (H-43); 1.35 m , 2 H (H-42, H-44); 1.46 m , 2 H (H-41,
H-45); 1.50 d, 3 H, J = 5.4 (H-38); 1.53 m , 1 H (H-43); 1.74 m , 2 H (H-42, H-44); 1.93 m , 2 H
(H-41, H-45); 3.45 q, 2 H, J = 7.3, (H-46); 4.65 m , 1 H (H-40); 5.49 d, 1 H, J = 15.9 (H-10);
5.53 d, 1 H, J = 15.9 (H-10); 6.85 q, 1 H, J = 5.4 (H-29); 7.02 d, 2 H, J = 8.4 (H-13, H-15);
7.08 d, 1 H, J = 8.4 (H-12, H-16); 7.08 dd, 1 H, J = 7.9 (H-8); 7.31 dd, 1 H, J = 7.9, 1.1 (H-7);
7.45 dd, 1 H, J = 7.9, 1.1 (H-9); 7.49 ddd, 1 H, J = 15.0, 7.6, 1.5 (H-25); 7.50 m , 1 H (H-27);
7.62 ddd, 1 H, J = 15.0, 7.6, 1.5 (H-26); 7.83 dd, 1 H, J = 7.6, 1.5 (H-24); 10.27 bs, 1 H (N-3).
13C NMR (for n um berin g, see Fig. 1): 13.61 (C-47); 19.56 (C-38); 23.60 (C-42, C-44); 25.11
(C-43); 31.39 (C-41 or C-45); 31.44 (C-41 or C-45); 45.69 (C-10); 52.41 (C-46); 77.64 (C-40);
91.94 (C-29); 113.88 (C-9); 114.31 (C-6); 121.12 (C-8); 122.85 (C-23); 123.86 (C-7); 127.88
(C-25); 127.91 (C-12, C-16); 128.83 (C-13, C-15); 128.95 (C-5); 129.49 (C-4); 130.20 (C-27);
131.51 (C-26); 131.55 (C-24); 137.17 (C-11); 137.91 (C-14); 141.20 (C-22); 152.51 (C-31);
154.27 (C-33); 156.42 (C-2); 163.61 (C-20). UV, λ (log ε): 212 (4.70), 254 (4.28), 312 (3.70).
HRMS: m/z calculated for C33H35N6O6 [M + H]+ 611.26181, foun d 611.26147.
1-{[(Cycloh exyloxy)carbon yl]oxy}eth yl 3-{[2′-(2-Eth yl-1H-tetrazol-5-yl)biph en yl-4-yl]m eth yl}-
2-oxo-2,3-dih ydro-1H-ben zim idazole-4-carboxylate (9)
Usin g th e procedure described for th e syn th esis of 8, but startin g from N-eth yl derivative 7,
0.14 g (74.0%) of 9 was prepared, m .p. 173–176 °C (99.9% purity, RRT 1.13). 1H NMR (for
n um berin g, see Fig. 1): 1.24 m , 1 H (H-43); 1.32 m , 2 H (H-42, H-44); 1.34 t, 3 H, J = 7.4
(H-47); 1.44 d, 3 H, J = 5.4 (H-38); 1.46 m , 2 H (H-41, H-45); 1.53 m , 1 H (H-43); 1.73 m ,
2 H (H-42, H-44); 1.92 m , 2 H (H-41, H-45); 3.42 q, 2 H, J = 7.4 (H-46); 4.64 m , 1 H (H-40);
5.46 d, 1 H, J = 16.0 (H-10); 5.61 d, 1 H, J = 16.0 (H-10); 6.86 q, 1 H, J = 5.4 (H-29); 7.04 m ,
4 H (H-12, H-13, H-15, H-16); 7.05 m , 1 H (H-8); 7.29 dd, 1 H, J = 7.9, 1.0 (H-7); 7.36 dd,
1 H, J = 7.6, 1.5 (H-27); 7.41 dd, 1 H, J = 7.9, 1.0 (H-9); 7.45 ddd, 1 H, J = 15.6, 7.6, 1.6
(H-25); 7.48 ddd, 1 H, J = 15.0, 7.6, 1.6 (H-26); 7.82 dd, 1 H, J = 15.6, 7.6 (H-24); 10.24 bs,
1 H (N-3). 13C NMR (for n um berin g, see Fig. 1): 14.36 (C-47); 19.51 (C-38); 23.57 (C-42 or
C-44); 23.59 (C-42 or C-44); 25.12 (C-43); 31.39 (C-41 or C-45); 31.41 (C-41 or C-45); 45.71
(C-10); 48.06 (C-46); 77.63 (C-40); 91.93 (C-29); 113.51 (C-9); 114.71 (C-6); 120.89 (C-8);
123.59 (C-7); 126.38 (C-23); 126.52 (C-12); 126.72 (C-13, C-15); 127.46 (C-25); 128.86 (C-5);
129.35 (C-4); 129.41 (C-16); 129.85 (C-26); 130.32 (C-24); 130.62 (C-27); 135.56 (C-11);
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

Some Impurities of Candesartan Cilexetil
361
139.97 (C-14); 141.61 (C-22); 151.10 (C-2); 152.53 (C-31); 163.67 (C-20); 165.15 (C-33). UV,
λ (log ε): 214 (4.74), 256 (4.20), 306 (3.62). HRMS: m/z calculated for C33H35N6O6 [M + H]+
611.26181, foun d 611.26154.
Heatin g of Can desartan Cilexetil 1 in Aqueous Aceton e
Candesartan cilexetil 1 (1.0 g) was heated at reflux in a mixture of acetone (8 ml) and water (2 ml)
for 24 h . Th e reaction m ixture was an alyzed by UPLC an d th e results are given in Table IV.
TABLE IV
Heatin g of can desartan cilexetil 1 in aqueous aceton e
UPLC in % (RRT)
Tim e
h
1
5
6
7
8
9
(1.00)
(0.66)
(1.33)
(1.58)
(0.90)
(1.13)
0
1
99.80
98.77
98.68
89.30
0.02
0.55
0.60
4.91
0.03
0.16
0.18
1.31
0.05
0.52
0.54
4.00
–
–
–
–
4
–
–
24
0.12
0.36
MS Experim en ts
Im purities 5–7 were iden tified by LC-MS usin g a Perkin –Elm er 200 series HPLC system cou-
pled to a Sciex API 3000 m ass spectrom eter with positive atm osph eric pressure ion ization
(TurboIon spray) an d by h igh -resolution MS tech n ique usin g an LTQ Orbitrap h ybrid m ass
spectrom eter with direct in jection in to APCI source in th e positive m ode. HPLC system
(C8 sym m etry 5 µm , 4.6 × 250 m m colum n , 25 °C) in gradien t elution (m obile ph ase A:
10 m M am m on ium form ate in water, pH adjusted to 3.5 with form ic acid, m obile ph ase B:
aceton itrile, 0.8 m l/m in ) was used.
NMR Experim en ts
NMR experim en ts were carried out on a Bruker Avan ce 500 (Bruker Biospin Gm bH) at
500.13 MHz (1H), 125.77 MHz (13C) an d 50.70 MHz (15N). Referen ce for 1H δCDCl3 7.26 ppm ,
for 13C δCDCl3 77.0 ppm an d for 15N δNH4+ 0 ppm . Couplin g con stan ts (J) are given in Hz. To
im prove spectral resolution an d sen sitivity, a Bruker Avan ce-III 800 MHz spectrom eter oper-
atin g at frequen cies of 800.13 MHz (1H), 201.19 MHz (13C) an d 81.11 MHz (15N) was used
for selected experim en ts. All experim en ts were perform ed in CDCl3 at 298 K. COSY, HSQC,
1H-13C HMBC an d 1H-15N HMBC spectra were recorded usin g pulse program s from th e
Bruker NMR stan dard library. At 500 MHz, stan dard 5 m m TXO (triple-n ucleus X-observe)
an d TBI (triple-broadban d in verse) probeh eads equipped with z-gradien t coils were em -
ployed for all m easurem en ts. At 800 MHz, a Bruker 5 m m TCI (triple C-optim ized in verse)
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362

362
Havlíček, Mandelová, Weisemann, Střelec, Plaček, Rádl:
cryoprobe equipped with a z-gradien t was used to record gradien t-en h an ced h eteron uclear
experim en ts, i.e. 1H-13C HSQC, 1H-13C HMBC an d 1H-15N HMBC, at h igh sen sitivity. For th e
1H-13C HSQC, a dataset was acquired with 8 scan s for each t1 in crem en t at a resolution of
2048 an d 256 poin ts in th e F2 an d F1 dim en sion s, respectively. Th e tim e dom ain data were
zero-filled to 2048 an d 1024 data poin ts in F2 an d F1 dim en sion s, an d m ultiplied with a si-
n usoidal squared sin e-bell win dow fun ction in both dim en sion s prior to Fourier tran sform .
Th e gradien t-selected 1H-13C HMBC an d 1H-15N HMBC data sets were recorded with 4 K an d
512 poin ts in th e F2 an d F1 dim en sion s, respectively. Th e m agn etization tran sfer in th e
1H-13C HMBC experim en t was optim ized for a th ree-bon d couplin g con stan t 3J(C,H) of
8 Hz. Th e correspon din g 1H-15N HMBC was optim ized for a lon g-ran ge couplin g con stan t of
6 Hz. 1 K in crem en ts of 2 K data poin ts were recorded with 128 scan s for each in crem en t.
Th e data were subsequen tly processed em ployin g zero-fillin g to 2 K an d 1 K data poin ts in
th e F1 an d F2 dim en sion s, usin g a sin usoidal squared sin e-bell win dow fun ction for
apodization prior to Fourier tran sform in both dim en sion s.
REFERENCES
1. Gohlke P., Jürgensen T., van Kügelgen S., Unger T.: Drugs Today 1999, 35, 105.
2. Hansson L.: Drugs Today 1999, 35, 117.
3. Naka T., Nishikawa K., Kato T. (Takeda Chemical Industries, Ltd.): U.S. 5196444; Chem.
Abstr. 1992, 116, 128924.
4. Etinger M. Yu., Niddam-Hildesheim V., Fedotev B., Dolitzky B.-Z. (TEVA Pharmarma-
ceutical Industries, Ltd.): WO 2005/111021; Chem. Abstr. 2005, 143, 4483142.
5. Kurgan Z., Pesachovich M. (TEVA Pharmarmaceutical Industries, Ltd.): WO 2006/122254;
Chem. Abstr. 2007, 145, 445638.
6. Rao D. V. S., Radhakrishnanand P., Suryanarayana M. V., Himabindu V.:
Chromatographia 2007, 66, 499.
7. Rádl S., Stach J. (Zentiva a.s.): WO 2007/048361; Chem. Abstr. 2007, 146, 462268.
8. Shida Y., Kokichi Y. (Takeda Chemical Industries, Ltd.): U.S. 5578733; Chem. Abstr.
1995, 123, 340152.
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 2, pp. 347–362