A new route to the antitumour drug temozolomide, but not thiotemozolomide

Article abstract of DOI:10.1039/a606159a

Interaction of 5-aminoimidazole-4-carboxamide with alkyl isocyanates yields N-substituted 1-carbamoylimidazoles which can be cyclised to imidazo[5,1-d][1,2,3,5]tetrazin-4(3H)-ones, including temozolomide 3a, on nitrosation; a similar reaction with methyl isothiocyanate, followed by nitrosation, affords the nitrosomethylamino derivative 11 of a new ring-system, imidazo[l,5-b][1,2,4]thiadiazole.

Full text of DOI:10.1039/a606159a

View Article Online / Journal Homepage / Table of Contents for this issue
A new route to the antitumour drug temozolomide, but not thiotemozolomide
Yongfeng Wang,a,b Philip R. Lowe,b William T. Thomson,c Jonathan Clarka and Malcolm F. G. Stevens*a
a Cancer Research Laboratories, Department of Pharmaceutical Sciences, University of Nottingham, Nottingham, UK NG7 2RD
b Pharmaceutical Sciences Institute, Aston University, Birmingham, UK B4 7ET
c Aston Molecules Ltd., 10 Holt Court South, Aston Science Park, Birmingham, UK B7 4EJ
Interaction of 5-aminoimidazole-4-carboxamide with alkyl
isocyanates yields N-substituted 1-carbamoylimidazoles
which can be cyclised to imidazo[5,1-d][1,2,3,5]tetrazin-
4(3H)-ones, including temozolomide 3a, on nitrosation; a
similar reaction with methyl isothiocyanate, followed by
nitrosation, affords the nitrosomethylamino derivative 11 of
a new ring-system, imidazo[1,5-b][1,2,4]thiadiazole.
temperature range 5–25 °C gave a red solution from which
temozolomide 3a precipitated in 30–35% yield. A by-product,
2-azahypoxanthine (15%), is probably formed by competing
nitrosative cyclisation to the primary carboxamide group
followed by loss of methyl isocyanate. Compared with the ease
with which N-alkylanthranilamides cyclise to 3-alkyl-
1,2,3-benzotriazin-4(3H)-ones7 the yield of temozolomide is
disappointing, but accountable for by the weakly acidic nature
of the urea proton in the diazonium species 6a which suppresses
generation of the dipolar intermediate 2a, the presumed
precursor to temozolomide. Higher yields of the imidazo-
tetrazinones 3b–d were isolated (50–70%) by cyclisation of the
corresponding ureas 5b–d: also we have shown previously5
that, in the case of the urea 5e formed from ethyl isocyanato-
acetate, cyclisation gave the imidazotetrazine 3e (93%) con-
sistent with the hypothesis that an electron-withdrawing group
R facilitates ionisation of the urea proton, and hence cyclisation
(Scheme 2).
The antitumour drug temozolomide 3a has shown useful
clinical activity against brain tumours1 and metastatic melan-
oma,2 two malignancies hitherto considered poorly treatable.
The original synthesis of temozolomide3 started with
5-diazoimidazole-4-carboxamide 1 which was reacted with
methyl isocyanate following a route to azolotetrazinones
originally discovered by Ege.4 The (presumed) dipolar inter-
mediate 2a cyclised spontaneously to temozolomide in high
yield in a Me2SO–ethyl acetate mixed solvent system at 25 °C
(Scheme 1). Analogues of temozolomide 3b–d can be prepared
similarly. The diazoimidazole 1 does not react with methyl
isothiocyanate in Me2SO at 40 °C nor in the same solvent
containing the Lewis acids zinc iodide and boron trifluoride–
diethyl ether or the base triethylamine, so 4-thiotemozolomide
(3a; O = S) is not available by this route.
Interaction of the aminoimidazole carboxamide 4 with
methyl isothiocyanate gave a thiourea (90%) which on
nitrosation with excess nitrous acid at 25 °C gave a red product
(85%), originally considered to be thiotemozolomide on the
1
basis of its H and 13C NMR spectra which were comparable
Our earlier work to develop alternative syntheses of temozol-
omide had exploited imidazotetrazinones e.g. 3e,f prepared
from less volatile ‘masked’ methyl isocyanates, which can be
deprotected under conditions which retain the integrity of the
base-fragile ring-system.5 Here we report a new synthesis of
temozolomide starting with 5-aminoimidazole-4-carboxamide
hydrochloride 4, which adapts another route developed by Ege6
for the synthesis of pyrazolotetrazinones, so obviating the use of
the potentially unstable diazoimidazole 1. A parallel synthesis
designed to furnish 4-thiotemozolomide surprisingly took an
entirely different path and led to the construction of a new ring-
system, imidazo[1,5-b][1,2,4]thiadiazole.
Interaction of the free base of 4 with methyl, ethyl,
2-chloroethyl and benzyl isocyanates in Me2SO or acetonitrile
afforded the ureas 5a–d†‡ regioselectively (75–95%) despite
the availability of competing sites for carbamoylation. Nitro-
sative cyclisation of 5a in 2 m hydrochloric acid in the
with that of temozolomide. However, elemental and mass
spectral analysis confirmed a molecular formula C6H6N6O2S
(thiotemozolomide + 16 mass units). An X-ray structure
determination (Fig. 1) revealed that the product was 2-nitroso-
methylaminoimidazo[1,5-b][1,2,4]thiadiazole-4-carboxamide
11.
A logical route to this unexpected compound is outlined in
Scheme 3. Methylsulfanylcarbamoylation of the free base of
amine 4 evidently affords the thiourea 7†‡ substituted at the
exocyclic amino group. Nitrosation of the thiol tautomer 7A
would generate an S-nitroso intermediate 8 which might cyclise
to the imidazothiadiazole 10 directly (2HNO) or, probably,
undergo further oxidation in the excess nitrous acid to the
S-nitro intermediate 9 which would then cyclise (2HNO2) to
afford the bicycle 10. Further nitrosation of the sec-amine group
of 10 to afford the nitrosamine 11ठis to be expected, although
the exact timing of this step is not clear.
Of the five other imidazothiadiazole ring-systems reported in
the literature only one, imidazo[1,2-b][1,2,4]thiadiazole,8 con-
tains an imidazole bridgehead nitrogen atom attached to sulfur.
–
N
N
N
+
H2NOC
N
+
H2NOC
N
N
N
CONH2
+
N
R
N
N
N2
C
R
CONH2
CONH2
N
O
O
–
NH2
NH2
N
NHR
1
+
i
ii
2
HN
N
O
NH
N
NHR
a R = Me
b R = Et
c R = (CH2)2Cl
d R = Bn
e R = CH2CO2Et
6
Cl –
O
–H+
H2NOC
N
N
O
4
5
N
N
2
N
R
3a–e
f
R = CH2SiMe3
3
Scheme 2 Reagents and conditions: i, RNCO, Me2SO or MeCN, 0–25 °C,
NEt3; ii, NaNO2 in 2 m HCl, 0–25 °C
Scheme 1
Chem. Commun., 1997
363

View Article Online
5-Amino-1-(N-ethylcarbamoyl)imidazole-4-carboxamide
5b:
mp
150–152 °C (decomp.); nmax (KBr) 3476, 3360, 3276, 1718, 1656, 1532,
1294, 847 cm21; dH [(CD3)2SO] 8.51 (m, 1 H, NH), 7.64 (s, 1 H, H-2), 6.92
(br s, 1 H, NH), 6.81 (br s, 1 H, NH), 6.37 (br s, 2 H, NH2), 3.22 (m, 2 H,
CH2), 1.11 (t, 3 H, CH3); dC [(CD3)2SO] 167.2, 150.2, 150.9, 144.4, 127.1,
112.1, 35.8, 15.3; m/z 198 (M+ + 1); 5-Amino-1-[N-(2-chloroethyl)carba-
moyl]imidazole-4-carboxamide 5c: mp 102–105 °C (decomp.); nmax (KBr)
3432, 3364, 3257, 1720, 1651, 1550, 1502, 1325 cm21; dH [(CD3)2SO] 8.79
(m, 1 H, NH), 7.68 (s, 1 H, H-2), 6.92 (br s, 1 H, NH), 6.84 (br s, 1 H, NH),
6.40 (br s, 2 H, NH2), 3.78 (t, 2 H, CH2CH2Cl), 3.59 (q, 2 H, CH2CH2Cl);dC
[(CD3)2SO] 170.3, 154.4, 147.6, 130.1, 115.3, 46.9, 45.9; m/z 231/233
(M+); 5-Amino-1-(N-benzylcarbamoyl)imidazole-4-carboxamide 5d: mp
163–168 °C (decomp.); nmax (KBr) 3323, 3202, 1735, 1637, 1532, 1492,
1316, 1250 cm21; dH [(CD3)2SO] 9.09 (t, 1 H, NH), 7.74 (s, 1 H, H-2), 7.34
(m, 5 H, Ph), 6.92 (br s, 1 H, NH), 6.84 (br s, 1 H, NH), 6.42 (br s, 2 H, NH2),
4.45 (d, 2 H, CH2); dC [(CD3)2SO] 168.2, 152.3, 145.5, 140.2, 130.3, 129.2,
129.0, 128.0, 113.1, 45.2; N-(4-Carbamoylimidazol-5-yl)-NA-methylthio-
urea 7; mp 200–205 °C (decomp.); nmax (KBr) 3374, 3205, 2914, 1663,
1574, 1528, 1481, 1327 cm21; dH [(CD3)2SO] 12.63 (br s, 1 H, NH), 10.26
(br s, 1 H, NH), 9.89 (br s, 1 H, NH), 7.82 (s, 1 H, H-2), 7.43 (br s, 2 H,
NH2), 3.08 (d, 3 H, CH3); dC [(CD3)2SO] 183.8, 167.6, 149.5, 139.1, 110.9,
37.2; m/z 199 (M+); 2-Nitrosomethylaminoimidazo[1,5-b][1,2,4]thiadia-
zole-4-carboxamide 11; mp 145–150 °C (decomp.); nmax (KBr) 3481, 3423,
3365, 3139, 3126, 3039, 1675, 1625 1507, 1125 cm21; dH [(CD3)2SO] 8.08
(s, 1 H, H-6), 7.30 (br s, 1 H, NH), 7.16 (br s, 1 H, NH), 4.50 (s, 3 H, CH3);
Fig. 1 ORTEP view of the structure of 2-nitrosomethylaminoimidazo-
[1,5-b][1,2,4]thiadiazole-4-carboxamide 11. Displacement elipsoids are
shown at the 50% probability level.
CONH2
CONH2
NHC(S)NHMe
NHMe
N
MeNCS
4
N
N
SH
NH
NH
7
7′
NO+
–H+
dC [(CD3)2SO] 163.9, 154.8, 145.1, 126.4, 119.9, 31.6; m/z 227 (M+
+
CONH2
CONH2
1).
§ Crystal data for 11: C6H6N6O2S, M = 226.23, monoclinic, space group
P21/n, a = 7.522(7), b = 9.121(2), c = 13.383(7) Å, b = 97.73(6)°,
U
NHMe
NO
N
NHMe
NO2
N
[O]
N
N
= = 4, Dc = = 464, (Cu-
910(1) Å3, Z 1.65 g cm23, F(000)
S
S
NH
NH
Ka) = 1.54180 Å, m = 3.148 mm21, A lath 0.65 3 0.40 3 0.24 mm grown
from Me2SO–Et2O was mounted on an Enraf-Nonius CAD 4 dif-
fractometer. 1734 unique reflections were collected by w¯ 2 2q for 2° @ q
@ 75° and phased by direct methods.8 Full-matrix least-squares refine-
ment10 on F2 with anisotropic thermal parameters for non hydrogen atoms
and all hydrogen atom positions determined from difference Fourier
synthesis. At convergence, R = 0.076, Rw = 0.22 and GOF = 1.105 for
1479 observed reflections [I > 2 dI]. Atomic coordinates, bond lengths and
angles, and thermal parameters have been deposited in the Cambridge
Crystallographic Data Centre (CDDC). See information for Authors, Issue
No. 1. Any requests to the CDDC for this material should quote the full
literature citation and the reference number 182/301.
–HNO2
8
9
CONH2
CONH2
Me
N
N
NO+
–H+
N
NHMe
N
N
N
N
N
S
10
S
O
11
Scheme 3
This new synthesis of 11 in two steps from the aminoimidazole
4 is notable for the high yield and mild conditions and may be
adapted for the synthesis of other examples of this intriguing
new bicyclic system.
This work was supported by the Cancer Research Campaign,
UK. We thank collaborators at Aston Molecules Ltd,
Birmingham, UK and Schering-Plough Research Institute,
Kenilworth, New Jersey, USA for helpful discussions.
References
1 S. M. O’Reilly, E. S. Newlands, M. G. Glaser, M. Brampton, J. M. Rice-
Edwards, R. D. Illingworth, P. G. Richards, C. Kennard, I. R.
Colquhoun, P. Lewis and M. F. G. Stevens, Eur. J. Cancer, 1993, 29A,
940.
2 N. M. Bleehen, E. S. Newlands, S. M. Lee, N. Thatcher, P. Selby,
A. H. Calvert, G. J. S. Rustin, M. Brampton and M. F. G. Stevens,
J. Clin. Oncol., 1995, 13, 910.
3 M. F. G. Stevens, J. A. Hickman, R. Stone, N. W. Gibson, G. U. Baig,
E. Lunt and C. G. Newton, J. Med. Chem., 1984, 27, 196.
4 G. Ege and K. Gilbert, Tetrahedron Lett., 1979, 4253.
5 Y. Wang, M. F. G. Stevens, W. T. Thomson and B. P. Schutts, J. Chem.
Soc., Perkin Trans. 1, 1995, 2783.
6 G. Ege, K. Gilbert and K. Maurer, Chem. Ber., 1987, 120, 1375.
7 J. G. Erickson, in The Chemistry of Heterocyclic Compounds, ed. J. G.
Erickson, P. F. Wiley and V. P. Wystrach, Interscience Publishers, Inc.,
New York, 1956, vol. 10, pp. 1–43; M. F. G. Stevens, Progr. Med.
Chem., 1976, 13, 205.
8 C. C. Beard, USP 3 979 404/1976 (Chem. Abstr., 1977, 86, P29815).
9 G. M. Sheldrick, SHELX86, Program for Crystal Structure Solution,
Acta Crystallogr., Sect. A, 1990, 46, 467.
10 G. M. Sheldrick, SHELX93, University of Go¨ttingen, Germany,
1993.
Footnotes
† General experimental method for ureas 5 and thiourea 7. A solution or
suspension of 5-aminoimidazole-4-carboxamide hydrochloride (0.5 g) and
dry triethylamine (1 ml) in dry Me2SO or acetonitrile (10 ml) was treated
dropwise (1 h) with the isocyanate or isothiocyanate (1.2 equiv.) at 10 °C
(210 °C in the case of acetonitrile). The mixture was stirred overnight at
25 °C, quenched with water (25 ml), and products collected and washed
successively with water and ethyl acetate. Yields of ureas were 5a (85%), 5b
(75%), 5c (70%), 5d (95%) and the thiourea 7 (85%). Satisfactory
microanalytical data were obtained for new compounds.
‡ Selected physical data for ureas 5, thiourea 7 and imidazothiadiazole 11.
5-Amino-1-(N-methylcarbamoyl)imidazole-4-carboxamide 5a: mp 170 °C
(decomp.); nmax (KBr) 3409, 1718, 1661, 1535, 1453, 1311, 1241, 947
cm21; dH [(CD3)2SO] 8.46 (q, 1 H, NH), 7.62 (s, 1 H, H-2), 6.93 (br s, 1 H,
NH), 6.83 (br s, 1 H, NH), 6.39 (br s, 2 H, NH2), 2.78 (d, 3 H, CH3); dC
[(CD3)2SO] 167.2, 151.6, 144.3, 127.0, 112.1, 27.5; m/z 184 (M+ + 1);
Received, 6th September 1996; Com. 6/06159A
364
Chem. Commun., 1997