An improved synthesis of the broad spectrum matrix metalloprotease inhibitor marimastat

Article abstract of DOI:10.1016/S0040-4039(00)01376-9

A considerably shortened synthesis of the broad spectrum matrix metalloproteinase (MMP) inhibitor marimastat has been achieved by a direct hydroxylamino ring opening of a key acetonide intermediate. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01376-9

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 7983–7986
An improved synthesis of the broad spectrum matrix
metalloprotease inhibitor marimastat
Richard J. Davenport* and Robert J. Watson
Celltech Chiroscience Ltd, Granta Park, Abington, Cambridge CB1 6GS, UK
Received 24 July 2000; accepted 10 August 2000
Abstract
A considerably shortened synthesis of the broad spectrum matrix metalloproteinase (MMP) inhibitor
marimastat has been achieved by a direct hydroxylamino ring opening of a key acetonide intermediate.
©
2000 Elsevier Science Ltd. All rights reserved.
Keywords: matrix metalloproteinase (MMP); hydroxylamino ring opening; marimastat synthesis.
1
The hydroxamic acid marimastat (1) (Fig. 1) is a broad spectrum inhibitor of the matrix
metalloproteinase (MMP) family of enzymes. Matrix metalloproteinase enzymes (including
collagenases, gelatinases, stromelysins, and matrilysin) are involved in tissue remodelling and
2
destruction. Over-expression of members of this enzyme family has been implicated in human
3
4
5
conditions such as rheumatoid and osteoarthritis, asthma, and a variety of different cancers.
Matrix metalloproteinase inhibitors such as marimastat are already reported to be in clinical
6
,7
trials for the treatment of some of the above conditions.
Figure 1.
1
The synthesis of marimastat reported by British Biotechnology Ltd starts with a stereocon-
8
trolled alkylation of an (S)-diisopropyl malate dianion. There is evidence to suggest that the
stereocontrol achieved in this reaction arises from a cyclic intermediate (2) formed by chelation
*
Corresponding author. Tel: 01223 896491; fax: 896400; e-mail: richarddavenport@chiroscience.com
0
040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01376-9

7984
between the lithium enolate and the adjacent hydroxyl group (Fig. 2). This gives rise to the
diisopropyl esters hindering one approach of the electrophile, and stereoinduction as high as
1
0:1 (R:S) has been achieved.
Figure 2.
Hydrogenation of the alkene (3) (Scheme 1), followed by saponification of the esters and
acetonide formation yielded the carboxylic acid (4). Carbodiimide coupling with pentafluorophe-
nol gave the activated ester in 58% yield, which was displaced with -tert-leucine methylamide
L
to give amide (5). Acetonide deprotection, followed by EDC coupling with O-benzylhydroxyl-
amine and protecting group removal gave the hydroxamic acid marimastat (1).
Scheme 1. (a) LDA, THF, methallyl iodide 49%; (b) EtOH, Pd–C, H , 98%; (c) KOH, dioxane, 99%; (d)
2
2,2-dimethoxypropane, tosic acid, 100%; (e) EDC, DCM, pentafluorophenol, 58%; (f)
L-tert-leucine methylamide,
DMF; (g) THF, 2 M HCl; (h) EDC, BnOꢀNH , DCM; (i) EtOH, Pd–C, H2
2
In our improved synthesis (S)-diisopropylmalate was again used as the starting material, with
conversion to the acid 4 proceeding smoothly in comparable yields to those reported in the
1
literature. Our first modification (Scheme 2) to the synthesis was to couple acid 4 directly with

7985
L
-tert-leucine methylamide using EDC to give a 60% yield of pure amide (5), after trituration in
diethyl ether. This removed the need to proceed via the activated pentafluorophenyl ester, and
any extra purification associated with that coupling step, hence improving the yield.
Scheme 2. (a) DCM, EDC, L-tert-leucine methylamide, 60%; (b) H NOH, THF, 93%
2
In our hands previous syntheses of marimastat had suffered from variable yields in the final
9
three steps, leading to poor overall yields and relatively poor purity of the final product. We
found the overall yields for these three steps from acetonide (5) to marimastat (1) varied from
1
0
7
to 15%. Literature examples suggested that the acetonide (5) could provide a reactive
electrophilic centre for direct attack of hydroxylamine. A small scale experiment using aqueous
hydroxylamine in DMF provided marimastat directly from 5, product isolation being achieved
by dilution of the reaction mixture with water, followed by filtration. The product purity was
1
1
excellent by LC, LCMS and high field NMR, and the chemical yield (35%) was better than
1
that achieved using the reported route. A significant improvement in yield was achieved by
using THF as the reaction solvent, followed by evaporation of the reaction mixture to low
volume, before collection of the product by filtration and washing with water. In this case, the
1
2
yield was increased to 93%, with a purity of 99.6–99.8%.
A stereospecific synthesis of marimastat has been achieved. The first improvement employs a
direct coupling between acid 4 and -tert-leucine methylamide removing the need to proceed via
L
the activated pentafluorophenyl ester. The second improvement relies on a key acetonide ring
opening step which has been optimised to give a 93% yield, with excellent purity of the final
compound. This adaptation of the previously reported route reduces the number of synthetic
steps from nine to six. This not only significantly reduces the amount of time required to
perform this synthesis, but also increases the overall yield of marimastat from amide (5) from
7
–15 to 93%.

7986
Acknowledgements
We thank Dr. Neil Henderson for his contribution, and the Pharmaceutical Science Depart-
ment for their rapid analysis of our samples.
References
1
2
3
4
. Dickens, J. P.; Crimmen, M. J.; Beckett, P. R.; British Biotechnology Ltd; WO 9402447 A1 19940203.
. Whittaker, M.; Floyd, C. D.; Brown, P.; Gearing, A. J. H. Chem. Rev. 1999, 99, 2735–2776.
. Cawston, T. E. Pharm. Ther. 1996, 70, 163.
. King, G.; Tamm, M.; Woolcock, A.; Black, J.; Roth, M.; ‘Expression of Matrix Metalloproteases-2 and -9 in
Serum and Bronchoalveolar Lavage Fluid of Asthmatic Patients’ presented at ATS conference on 10:05:2000 in
Toronto.
5
6
. Wang, X.; Fu, X.; Brown, P. D.; Crimmen, M. J.; Hoffman, R. M. Cancer Res. 1995, 87, 1546.
. Nemunaites, J.; Poole, C.; Primrose, J.; Rosemurgy, A.; Malfetario, J.; Prown, P.; Berrington, A.; Cornish, A.;
Lynch, K.; Rasmussen, H.; Kerr, D.; Cox, D.; Miller, A. Clin. Cancer Res. 1998, 4, 1101.
. Wood, N. D.; Aitken, M.; Durston, S.; Harris, S.; McClelland, G. R.; Sharp, S. Agents Actions Suppl. 1998, 49,
7
4
9.
8. Seebach, D.; Wasmuth, D. Helv. Chim. Acta 1980, 63, 197.
9. Yields for conversion from acetonide (4) to marimastat are not reported in Ref. 1.
1
1
1
0. Bailey, S.; Bolognese, B.; Faller, A.; Louis-Flamberg, P.; MacPherson, D.; Mayer, R.; Marshall, L.; Milner, P.;
Mistry, J.; Smith, D.; Ward, J. Bioorg. Med. Chem. Lett. 1999, 9, 3165–3170.
1. Purity determined on a PHENOMENEX LUNA 5m C18 column, using acetonitrile and pH 2.2 water as the
mobile phase, and detection at 210 nm.
2. Optimised method: Acetonide (5) (12 g, 34 mmol) was taken up in tetrahydrofuran (50 ml) and aqueous
hydroxylamine (12 ml, 170 mmol) was added. The reaction was heated to reflux for 1 hour before being allowed
to cool, and diluted with water (20 ml). The tetrahydrofuran was removed in vacuo and the slurry filtered. The
filter pad was washed with water (10 ml), and then dried in vacuo to yield the desired product as a white solid
(10.1 g, 93%).
.
.