A new synthesis of novel tricyclic 2(5H)-furanone heterocycles from 3,4,5-trichloro-2(5H)-furanone

Article abstract of DOI:10.3987/COM-10-11923

A novel class of five-six-six tricyclic 2(5H)-furanone heterocycles was synthesized from 3,4,5-trichloro-2(5H)-furanone and bifunctional o-nucleophiles in a single step. In addition, 2-(chloromethyl)quinoxaline was obtained through this method, and the distinctive formation mechanism of this compound is discussed.

Full text of DOI:10.3987/COM-10-11923

HETEROCYCLES, Vol. 81, No. 6, 2010
1427
HETEROCYCLES, Vol. 81, No. 6, 2010, pp. 1427 - 1434. © The Japan Institute of Heterocyclic Chemistry
Received, 8th February, 2010, Accepted, 12th April, 2010, Published online, 13th April, 2010
DOI: 10.3987/COM-10-11923
A NEW SYNTHESIS OF NOVEL TRICYCLIC 2(5H)-FURANONE
HETEROCYCLES FROM 3,4,5-TRICHLORO-2(5H)-FURANONE
Jian Ren, Dan-Dan Ma, Yu Sha, Feng Li, and Mao-Sheng Cheng*
The Key Laboratory of Structure-Based Drug Design & Discovery of Ministry of
Education, Shenyang Pharmaceutical University, School of Pharmaceutical
Engineering, Shenyang, 110016, PR China
*E-mail address: mscheng@syphu.edu.cn
Abstract – A novel class of five-six-six tricyclic 2(5H)-furanone heterocycles
was synthesized from 3,4,5-trichloro-2(5H)-furanone and bifunctional
o-nucleophiles in a single step. In addition, 2-(chloromethyl)quinoxaline was
obtained through this method, and the distinctive formation mechanism of this
compound is discussed.
INTRODUCTION
2(5H)-Furanone derivatives are a large well-known family of important compounds in medicinal
chemistry. A large number of derivatives containing the 2(5H)-furanone moiety have been widely used as
antitumor,1 anti-inflammatory,2 antibacterial,3 antifungal4 and antiplatelet5 agents. Moreover,
2(5H)-furanone derivatives have been used in a variety applications in organic synthesis, and there have
been many studies reported on their chemistry, synthesis and synthetic uses.6 The potential applications of
the 2(5H)-furanone derivatives have attracted considerable attention. 3,4,5-Trichloro-2(5H)-furanone is a
versatile synthon for the synthesis of different 2(5H)-furanone heterocyclic skeletons due to its “pseudo
acid chloride” that possesses three potentially reactive chlorides.7 We report herein on an efficient and
general method for the synthesis of a novel class of five-six-six tricyclic 2(5H)-furanone heterocycles by
reaction of 3,4,5-trichloro-2(5H)-furanone with bifunctional o-nucleophiles under mild conditions
(Scheme 1).
RESULTS AND DISCUSSION
Scheme 1 outlines the synthetic sequences for the preparation of the five-six-six tricyclic 2(5H)-furanone
heterocycle 3-chlorobenzo[b]furo[3,2-e][1,4]dioxin-2(9aH)-one (3) and its analogues 4 and 5. Initially,

1428
HETEROCYCLES, Vol. 81, No. 6, 2010
commercially available mucochloric acid (1) was used as a starting material. After treatment of 1 with
thionyl chloride, the desired important intermediate 3,4,5-trichloro-2(5H)-furanone (2) was prepared in
fair yield. Compound 2 was reacted with catechol and potassium carbonate in dimethylformamide at
room temperature for one hour, the corresponding novel 2(5H)-furanone heterocyclic compound 3 was
obtained in a single step.
Scheme 1
Thus, a convenient method to synthesize novel five-six-six tricyclic 2(5H)-furanone heterocycles has
been developed. Following this procedure, we attempted to synthesize the 2(5H)-furanone heterocyclic
analogues 4 and 5 using 3,4,5-trichloro-2(5H)-furanone (2) as a reaction material. Treatment of 2 with
o-aminothiophenol offered the desired product 3-chloro-4,9a-dihydro-2H-benzo[b]furo[3,2-e][1,4]-
thiazin-2-one (4) at a 91% yield at room temperature in less than one hour. In a similar way, 3-chloro-
4,9a-dihydro-2H-benzo[b]furo[3,2-e][1,4]oxazin-2-one (5) was smoothly cyclized by treatment with 2
and o-aminophenol in good yield (88%). By reviewing the relevant literature, we found one report8 on the
preparation of compound 5, which employed 4-methoxycarbonyloxy-2,3-dichloro-2-buten-4-olide,
o-aminophenol and CsF as reagents. However, this approach has some drawbacks, such as a tedious
workup, unclear yield and a long reaction time (more than 41 hours).
1
The structures of compounds 4 and 5 were confirmed by their spectral data. In the H NMR spectra of
compounds 4 and 5, the methine and NH protons were both signal peaks, and the NH protons of 4 and 5
appeared at δ 11.19 and 11.31 ppm, respectively; these downfield signals indicate that the existence of a
1
nitrogen atom between the phenyl group and the ethylenic bond. In addition, the H-1H COSY, NOESY

HETEROCYCLES, Vol. 81, No. 6, 2010
1429
and HMBC spectrums were also certificated the structures of above two compounds, and the key NOESY
and HMBC correlations of 4 and 5 were collected in Figure 1.
Figure 1. Key NOESY and HMBC correlations of 4 and 5
Before starting the above two experiments, we supposed that when 3,4,5-trichloro-2(5H)-furanone (2)
reacted with o-aminothiophenol or o-aminophenol, two isomeric compounds (4, 4′ or 5, 5′, as noted in
Scheme 1) might be generated. Interestingly, in the actual experiments, compounds 4 and 5 were formed
as the sole products, and the isomeric compounds 4′ and 5′ were not detected. Therefore, we attempted to
use a molecular modeling method9 to calculate the charge values of the substrate 3,4,5-trichloro-2(5H)-
furanone in the preliminary analysis of these results. The result of the calculation is shown in Figure 2. As
noted in Figure 2, the charge value of the 5-position carbon atom was 0.226, which was slightly higher
than that of the 4-position (0.215). This indicated that the electron cloud density of the 5-position was
lower than that of the 4-position, and in theory, the 5-position of compound 2 would be attacked first by
sulfhydryl or hydroxyl moieties under alkaline conditions.
Figure 2
In the following step, we did experiments in an effort to synthesize the tricyclic heterocyclic compound
3-chloro-9,9a-dihydrofuro[2,3-b]quinoxalin-2(4H)-one (6) from 3,4,5-trichloro-2(5H)-furanone by
reacting with o-phenylenediamine (Scheme 1). The desired product 6 was not initially obtained; instead,
the decyclized compound 7 was produced. Based on spectral analysis, the structure shown in the Scheme
1 was assigned to compound 7. Although we attempted to convert 7 to 6 by further reaction under
different conditions, such as changing the base, solvent, and reaction time. However, no desired product 6
was generated. After extending the reaction time at room temperature or increasing the temperature, the

1430
HETEROCYCLES, Vol. 81, No. 6, 2010
decyclized compound 7 was converted into a surprising product, 2-(chloromethyl)quinoxaline (8), which
was identified by spectral analysis and further comparison with the reported data.10
2-(Chloromethyl)quinoxaline is an available building block for the construction of many bioactive
substances11 with diverse biological activities, including antiarrhythmic, leukotriene biosynthesis
inhibitor, phosphodiesterase inhibitor and GABAA receptor agonist activities. There have only been a few
synthetic approaches established previously for the preparation of this compound.10 Herein, we
discovered a simple and practical method for synthesizing this kind of compound.
Scheme 2
Regarding this transformation, the detailed mechanism was not clear at this stage. A possible reaction
mechanism accounting for the formation of product 8 is proposed in Scheme 2. The mechanism might
involve the following. o-Phenylenediamine reacted with 3,4,5-trichloro-2(5H)-furanone to generate
decyclized compound 7. The NH moiety of compound 7 was deprotonated with one equivalent base, the
furanone ring was opened, and another amino group reacted with 4-position chloride moiety to give i.
Protonation and tautomerization formed the corresponding “β-keto acid” transition state ii, which
underwent decarboxylation via a six-membered transition state to give iii. After further tautomerization,
the stable product 8 was obtained.
In order further examine this method, similar reactions were carried out on 4-nitro-1,2-diaminobenzene
and 4-bromo-1,2-diaminobenzene.
Scheme 3

HETEROCYCLES, Vol. 81, No. 6, 2010
1431
From Scheme 3, it can be seen that the reaction of 3,4,5-trichloro-2(5H)-furanone (2) with
4-nitro-1,2-diaminobenzene or 4-bromo-1,2-diaminobenzene formed the two desired isomers 9a, 9b or
10a, 10b respectively. The structures of above four compounds were identified by analysis of the 1H and
13C NMR spectroscopic data with the aid of HMBC data. These preliminary results proved that this
method could be applied to the synthesis of 2-(chloromethyl)quinoxaline derivatives.
In conclusion, three five-six-six tricyclic 2(5H)-furanone heterocyclic compounds and five quinoxaline
derivatives were synthesized from 3,4,5-trichloro-2(5H)-furanone through a novel method. With respect
to o-aminothiophenol and o-aminophenol, selective reaction with 3,4,5-trichloro-2(5H)-furanone
generated compound 4 and 5, and we calculated the charge value of the substrate in the preliminary
analysis of this reaction phenomenon. Furthermore, the distinctive reaction mechanism of
2-(chloromethyl)quinoxaline, which was generated by the reaction of 3,4,5-trichloro-2(5H)-furanone and
o-phenylenediamine, was proposed.
EXPERIMENTAL
Melting points were obtained using a Büchi B-540 micro melting point apparatus and were uncorrected.
Infrared (IR) spectra were recorded as KBr disks, using Brucker IFS55 spectrometer. All NMR spectra
were recorded on a Brucker AM-300 (300 MHz) or AM-600 (600 MHz) spectrometers in dimethyl
sulfoxide-d6 solution at room temperature. The chemical shifts were given in ppm relative to
tetramethylsilane as the internal reference standard. The EI mass spectra were performed using an Agilent
6890N-5975i GC-MS. High-resolution mass spectra (HRMS) were determined using a Brucker
micrOTOFQ 125.
3,4,5-Trichloro-2(5H)-furanone (2): A solution of mucochloric acid (2.0 g, 12.0 mmol) in thionyl
chloride (15 mL) was heated under reflux with stirring for 12 h. After the mixture was cooled to room
temperature, the solvent was removed under reduced pressure and the crude product was subjected to
column chromatography using silica-gel as an absorbent and Petrol/EtOAc = 10:1 as eluent to offer 2.1 g
1
(95%) of 2. bp 201-202 °C; IR (KBr) ν max cm-1: 2993, 1809, 1630, 1309, 1226; H NMR (300 MHz,
13
DMSO-d6): δ 7.33 (s, 1H); C NMR (75 MHz, DMSO-d6): δ 87.20, 123.21, 151.08, 163.26; MS m/z:
185.9 [M]+, 187.9 [M+2]+, 189.9 [M+4]+, 191.9 [M+6]+.
General procedure for the synthesis of compound 3, 4 and 5: To a stirring suspension of 2 (2.0 g, 10.7
mmol) in DMF (15 mL), the catechol (1.30 g, 11.8 mmol) and K2CO3 (1.63 g, 11.8 mmol) was added.
After stirring at room temperature for 1 h, ice water was added. The resulting precipitate was collected by
filtration and the crude product was subjected to column chromatography using silica-gel as an absorbent
and Petrol/EtOAc = 5:1 as eluent to offer 2.04 g 3 (85%). mp 137-138 °C; IR (KBr) ν max cm-1: 3068,
1
13
1789, 1487, 1366, 1110, 762; H NMR (300 MHz, DMSO-d6): δ 6.50 (s, 1H), 7.18-7.49 (m, 4H); C

1432
HETEROCYCLES, Vol. 81, No. 6, 2010
NMR (75 MHz, DMSO-d6): δ 89.67, 96.83, 117.42, 118.90, 124.61, 126.29, 138.52, 140.52, 160.58,
164.39; MS m/z: 224.0 [M]+, 226.0 [M+2]+; HRMS [M+Na]+ calc for C10H5ClNaO4: 246.9769, found:
246.9768.
3-Chloro-4,9a-dihydro-2H-benzo[b]furo[3,2-e][1,4]thiazin-2-one (4): This compound was obtained as
solid with 2.34 g (91%). mp 201-202 °C; IR (KBr) ν max cm-1: 3240, 2922, 1743, 1473, 1387, 1058, 752;
1
1H NMR (300 MHz, DMSO-d6): δ 6.44 (s, 1H), 7.02-7.36 (m, 4H), 11.19 (s, 1H); H NMR (600 MHz,
DMSO-d6, deuterium-exchangeable): δ 6.37 (s, 1H), 7.04-7.33 (m, 4H); 13C NMR (75 MHz, DMSO-d6):
δ 70.27, 85.44, 115.16, 118.55, 123.93, 128.35, 129.45, 135.65, 154.01, 167.20; MS m/z: 239.0 [M]+,
241.0 [M+2]+; HRMS [M+Na]+ calc for C10H6ClNNaO2S: 261.9700, found: 261.9698.
3-Chloro-4,9a-dihydro-2H-benzo[b]furo[3,2-e][1,4]oxazin-2-one (5): This compound was obtained as
solid with 2.11 g (88%). mp 206-207 °C; IR (KBr) ν max cm-1: 3227, 2910, 1760, 1495, 1399, 1095, 754;
1
1H NMR (300 MHz, DMSO-d6): δ 6.25 (s, 1H), 7.05-7.15 (m, 4H), 11.31 (s, 1H); H NMR (600 MHz,
DMSO-d6, deuterium-exchangeable): δ 6.18 (s, 1H), 7.02-7.12 (m, 4H); 13C NMR (75 MHz, DMSO-d6):
δ 86.18, 91.41, 117.28, 117.92, 124.02, 124.21, 127.36, 140.40, 150.25, 166.20; MS m/z: 223.0 [M]+,
225.0 [M+2]+; HRMS [M+Na]+ calc for C10H6ClNNaO3: 245.9928, found: 245.9926.
5-(2-Aminophenylamino)-3,4-dichlorofuran-2(5H)-one (7): 3,4,5-Trichloro-2(5H)-furanone (2.0 g,
10.7 mmol) in DMF (10 mL), K2CO3 (1.63 g, 11.8 mmol) and o-phenylenediamine (1.28 g, 11.8 mmol)
were added. After the reaction mixture was stirred for 15 min, the solvent was treated with ice/water. The
solid deposit was collected by filtration and chromatographed eluting with Petrol/EtOAc = 2:1 to give
2.49 g of 7 (90%). mp 133-134 °C; IR (KBr) ν max cm-1: 3378, 3239, 2939, 1704, 1502, 1463, 1145, 745;
1H NMR (300MHz, DMSO-d6): δ 5.18 (s, 2H), 5.60-5.63 (d, 1H, J = 9.0 Hz), 6.51-7.06 (m, 4H),
13
7.06-7.09 (d, 1H, J = 9.0 Hz); C NMR (75 MHz, DMSO-d6): δ 82.35, 115.40, 115.67, 119.17, 125.13,
129.13, 130.43, 144.55, 146.49, 161.16; MS m/z: 258.0 [M]+, 260.0 [M+2]+, 262.0 [M+4]+; HRMS
[M+Na]+ calc for C10H8Cl2N2NaO2: 280.9855, found: 280.9856.
General procedure for the synthesis of compound 8, 9a, 9b, 10a and 10b: A mixture of
3,4,5-trichloro-2(5H)-furanone (2.0 g, 10.7 mmol), K2CO3 (1.63 g, 11.8 mmol) and o-phenylenediamine
(1.28 g, 11.8 mmol) in DMF (10 mL) was stirred at 60 °C for 0.5 h or at room temperature for 3 h. The
solvent was treated with ice/water. After that the solid deposit was collected by filtration and purified by
chromatography on a silica-gel column using Petrol/EtOAc = 5:1 to give 1.76 g of 8 (92%). mp 46-47 °C;
1
IR (KBr) ν max cm-1: 3020, 1637, 1558, 1493, 757; H NMR (300 MHz, DMSO-d6): δ 5.07 (s, 2H),
13
7.88-8.15 (m, 4H), 9.10 (s, 1H); C NMR (150 MHz, DMSO-d6): δ 45.07, 129.36, 129.38, 131.10,
131.25, 141.33, 141.67, 146.23, 152.60; MS m/z: 178.0 [M]+, 180.0 [M+2]+; HRMS [M+H]+ calc for
C9H8ClN2: 179.0371, found: 179.0365.
2-(Chloromethyl)-7-nitroquinoxaline (9a): This compound was obtained as solid with 0.91 g (38%).

HETEROCYCLES, Vol. 81, No. 6, 2010
1433
1
mp 89-90 °C; IR (KBr) ν max cm-1: 3430, 3045, 1618, 1536, 1347, 1079, 796; H NMR (600 MHz,
DMSO-d6): δ 5.13 (s, 2H), 8.33-8.34 (d, 1H, J = 9.0 Hz), 8.58-8.60 (dd, 1H, J = 9.0 Hz, J = 2.5 Hz),
13
8.93-8.94 (d, 1H, J = 2.5 Hz), 9.29 (s, 1H); C NMR (150 MHz, DMSO-d6): δ 44.78, 124.57, 125.45,
131.39, 140.47, 143.91, 148.32, 148.64, 155.85; MS m/z: 223.1 [M]+, 225.1 [M+2]+; HRMS [M-H]- calc
for C9H5ClN3O2: 222.0076, found: 222.0060.
2-(Chloromethyl)-6-nitroquinoxaline (9b): This compound was obtained as solid with 1.07 g (45%).
1
mp 96-97 °C; IR (KBr) ν max cm-1: 3431, 3050, 1618, 1530, 1349, 1076, 831; H NMR (600 MHz,
DMSO-d6): δ 5.14 (s, 2H), 8.37-8.38 (d, 1H, J = 9.0 Hz), 8.58-8.60 (dd, 1H, J = 9.0 Hz, J = 2.5 Hz),
13
8.89-8.90 (d, 1H, J = 2.5 Hz), 9.30 (s, 1H); C NMR (150 MHz, DMSO-d6): δ 44.30, 123.98, 124.98,
131.06, 139.86, 143.64, 148.11, 149.08, 154.67; MS m/z: 223.1 [M]+, 225.1 [M+2]+; HRMS [M-H]- calc
for C9H5ClN3O2: 222.0076, found: 222.0081.
2-(Chloromethyl)-7-bromoquinoxaline (10a): This compound was obtained as solid with 0.99 g (36%).
1
mp 62-63 °C; IR (KBr) ν max cm-1: 3425, 3055, 1582, 1451, 1385, 1091, 750; H NMR (600 MHz,
DMSO-d6): δ 5.06 (s, 2H), 8.02-8.05 (m, 2H), 8.37-8.38 (d, 1H, J = 2.5 Hz), 9.12 (s, 1H); 13C NMR (150
MHz, DMSO-d6): δ 44.50, 123.73, 130.85, 131.05, 134.04, 139.82, 141.84, 146.84, 152.76; MS m/z:
256.0 [M]+, 258.0 [M+2]+, 260.0 [M+4]+; HRMS [M-H]- calc for C9H5N2ClBr: 254.9319, found:
254.9312.
2-(Chloromethyl)-6-bromoquinoxaline (10b): This compound was obtained as solid with 1.15 g (42%).
1
mp 66-67 °C; IR (KBr) ν max cm-1: 3427, 3045, 1597, 1479, 1362, 1059, 726; H NMR (600 MHz,
DMSO-d6): δ 5.04 (s, 2H), 8.00-8.04 (m, 2H), 8.31-8.32 (d, 1H, J = 2.5 Hz), 9.10 (s, 1H); 13C NMR (150
MHz, DMSO-d6): δ 44.43, 123.90, 130.87, 130.95, 133.87, 140.12, 141.57, 146.33, 153.21; MS m/z:
256.0 [M]+, 258.0 [M+2]+, 260.0 [M+4]+; HRMS [M-H]- calc for C9H5N2ClBr: 254.9319, found:
254.9322.
ACKNOWLEDGEMENTS
The authors would like to thank The Key Laboratory of Structure-Based Drug Design & Discovery of
Ministry of Education for generous financial support.
REFERENCES
1. Y. Kim, N. Nam, Y. You, and B. Ahn, Bioorg. Med. Chem. Lett., 2002, 12, 719; N. Nam, Y. Kim, Y.
You, D. Hong, H. Kim, and B. Ahn, Bioorg. Med. Chem. Lett., 2002, 12, 1955; U. Marsteinstredet,
R. Wiger, G. Brunborg, J. Hongslo, and J. Holme, Chem. Boil. Interact., 1997, 106, 89; S. Bang, Y.
Kim, M. Yun, and B. Ahn, Arch. Pharmacal. Res., 2004, 27, 485; R. Melnick, G. Boorman, and V.
Dellarco, J. Natl. Cancer. I., 1997, 89, 832.

1434
HETEROCYCLES, Vol. 81, No. 6, 2010
2. V. Weber, P. Coudert, C. Rubat, E. Duroux, D. Vallee-Goyet, D. Gardette, M. Bria, E. Albuisson, F.
Leal, and J. Gramain, Bioorg. Med. Chem., 2002, 10, 1647; R. Silva, G. de Souza, A. da Silva, V. de
Souza, A. Pereira, V. Royo, Bioorg. Med. Chem. Lett., 2005, 15, 1033; S. Rowland, P. Clark, R.
Gordon, A. Mullen, J. Guay, L. Dufresne, C. Brideau, B. Cote, Y. Ducharme, and J. Mancini, Eur. J.
Pharmacol., 2007, 560, 216; G. Grossmann, M. Poncioni, M. Bornand, B. Jolivet, M. Neuburger,
and U. Sequin, Tetrahedron, 2003, 59, 3237; C. Chan, S. Boyce, C. Brideau, S. Charleson, W.
Cromlish, D. Ethier, J. Evans, A. Ford-Hutchinson, M. Forrest, and J. Gauthier, J. Pharmacol. Exp.
Ther., 1999, 290, 551.
3. E. Lattmann, S. Dunn, S. Niamsanit, and N. Sattayasai, Bioorg. Med. Chem. Lett., 2005, 15, 919.
4. T. Paulitz, B. Nowak-Thompson, P. Gamard, E. Tsang, and J. Loper, J. Chem. Ecol., 2000, 26, 1515.
5. H. Yang, G. Hu, J. Chen, Y. Wang, and Z. Wang, Bioorg. Med. Chem. Lett., 2007, 17, 5210.
6. M. De Souza, Mini-Rev. Org. Chem., 2005, 2, 139; A. Hashem, A. Senning, and A. Hamad, Org.
Prep. Proced. Int., 1998, 30, 401; N. Carter, A. Nadany, and J. Sweeney, J. Chem. Soc., Perkin
Trans. 1, 2002, 2324.
7. K. Jisch, J. Prakt. Chem., 1990, 332, 233.
8. P. G. Blazecka, and J. Zhang, U.S. Patent 20050059831, (03. 17. 2005) (Chem. Abstr., 2005, 142,
316685u).
9. Molecular modeling studies were performed with SYBYL 6.91 software package on SGI Fuel
workstation. The structure of compound 2 was built with SYBYL Sketch module and minimized
using the Tripos force field and assigned charges using the Gasterger-Hückel method; S.H.R.
SYBYL 6.91 Manual. Tripos Inc., St. Louis, MO, USA. 2004.
10. G. Krippner, and M. Harding, Tetrahedron: Asymmetry, 1994, 5, 1793; G. Jeromin, W. Orth, B.
Rapp, and W. Wei, Chem. Ber., 1987, 120, 649.
11. I. Starke, G. Sarodnick, V. Ovcharenko, K. Pihlaja, and E. Kleinpeter, Tetrahedron, 2004, 60, 6063;
T. Kolasa, D. Gunn, P. Bhatia, A. Basha, R. Craig, A. Stewart, J. Bouska, R. Harris, K. Hulkower,
and P. Malo, J. Med. Chem., 2000, 43, 3322; T. Kolasa, D. Gunn, P. Bhatia, K. Woods, T. Gane, A.
Stewart, J. Bouska, R. Harris, K. Hulkower, and P. Malo, J. Med. Chem., 2000, 43, 690; J. Butera,
W. Spinelli, V. Anantharaman, N. Marcopulos, R. Parsons, I. Moubarak, C. Cullinan, and J. Bagli, J.
Med. Chem., 1991, 34, 3212; P. R. Verhoest, C. J. Helal, D. J. Hoove, and J. M. Humphrey, WO.
Patent 2006072828, (07. 13. 2006) (Chem. Abstr., 2006, 145, 145559); M. Russell, R. Carling, J.
Atack, F. Bromidge, S. Cook, P. Hunt, C. Isted, M. Lucas, R. Mckernan, and A. Mitchinson, J. Med.
Chem., 2005, 48, 1367.