Professor Charles Marson

Research Overview

Welcome to the webpages of the Charles Marson Group based in the Christopher Ingold laboratories of UCL right in the heart of London. UCL has recently gained a 5* rating and is well equipped as an establishment for top rate research. The menu on the left will help you gain an idea on the diverse type of research done in the group which ranges from cancer chemotherapy to asymmetric synthesis.

Prof. Marson's research interests concern the invention of new synthetic reactions, particularly with stereocontrol. Major current research themes include: asymmetric synthesis of enantiopure C2 -symmetric compounds; catalytic asymmetric synthesis; and the design and synthesis of anti-cancer compounds; chiral organofluorine chemistry.

Research Interests

  1. Catalytic Asymmetric Synthesis

    Asymmetric synthesis, or the ability to synthesise a compound in one, or predominantly one enantiomerically pure form, is of great contemporary importance, partly because single enantiomers find application in numerous areas including pharmaceuticals, biological and medicinal chemistry, and materials science. The search is on for new chiral catalysts that will give compounds of high optical purity, preferably in large-scale processes. Within this site we disclose new amine catalysts that switch their enantiomeric bias simply upon N-methylation.

  2. Cancer Chemotherapy

    Much of conventional cancer chemotherapy depends upon chemical modification of DNA, a process that often damages healthy as well as abnormal cells. Recently, a more focussed approach has been to identify a cellular enzyme whose function is necessary for oncogenic transformation, and to attempt to find small-molecule inhibitors of that enzyme. Here, we discuss new lipidic anhydrides that are selective for GGPTase, an enzyme associated with K-Ras, cell signalling, and some cancers that have been difficult to treat.

  3. Pyranone Chemistry

    The construction of heterocylic rings, especially with stereocontrol, is frequently central to the synthesis of biologically active compounds, including natural products. Below you will find a new and stereospecific rearrangement discovered in this group. The dihydropyranone products are valuable intermediates in total synthesis.

  4. Asymmetric Fluorine Chemistry

    The special properties of organic compounds containing fluorine give rise to applications in many fields, including liquid crystals, agrochemistry, and biological and clinical chemistry. However, while one chiral centre containing fluorine is usually straightforward to create, the formation of compounds with two or more -CHF- centres of defined stereochemistry presents major challenges. Here you will see some of the first symmetrical and enantiomerically pure 1,2-difluoro compounds ever to be made. Interestingly, they can act as catalysts in asymmetric synthesis.

Selected Publications

  1. C. Marson, "Oxygen-Directed Carbocyclizations of Epoxides. ", Tetrahedron 200056, 8779-8794.
  2. C. M. Marson, J. H. Pink, C. Smith, M. B. Hursthouse and K. M. A. Malik, “A Biomimetic Synthesis of the Pyrrolizidine Ring by Sequential Intramolecular Cyclizations.”, Tetrahedron Lett. 200041, 127-129.
  3. A. Rioja, A. R. Pizzey, C. M. Marson and N. S. B. Thomas, “Preferential Induction of Apoptosis of Leukaemic Cells by Farnesol”, FEBS Lett2000467, 291-295.
  4. C. M. Marson, "Str 3 May, 2010 " in The Chemistry of Natural Products, ed. K. J. Hale, Sheffield Academic Press, 2000, pp. 199-228.
  5. A. Khan, C. M. Marson and R. A. Porter, “Synthesis of Exocyclic Enamides via Stereocontrolled Allylic Isomerisation and 1,3-Transposition.”, Synth. Commun200131, 1753-1764.
  6. C. M. Marson , A. Khan, and R. A. Porter, “Stereocontrolled Formation of Epoxy Peroxide Functionality Appended to a Lactam Ring”, J. Org.Chem. 200166, 4771-4775.
  7. A. J. A. Cobb and C. M. Marson, “Reversal of Enantioselectivity using Catalysts Containing Multiple Stereogenic Centres.”, Tetrahedron: Asymmetry200112, 1547-1550.
  8. C. M. Marson , “Synthesis via N-Acyliminium Cyclisations of N-Heterocyclic Ring Systems Related to Alkaloids”, ARKIVOC (electronic heterocyclic chemistry site), 20012, U60-U75.
  9. C. M. Marson , A. Khan, R. A. Porter and A. J. A. Cobb, “Construction of Functionalised Medium Rings by Stereospecific Expansions of 2,3-Epoxy Alcohols under Mild Conditions.”, Tetrahedron Lett. 200243, 6637-6640.
  10. C. M. Marson , A. Rioja, D. M. Vigushin and R. C. Coombes, “Cyclic Acid Anhydrides as a New Class of Potent and Selective Inhibitors of Geranylgeranyl Transferase.”Bioorg. Med. ChemLett.200212, 255-259.
  11. R. A. Decréau, C. M. Marson and K. E. Smith, “Preparation of 3 a - and 3 b -Fluoro Compounds in the 5 a -Androst-16-ene and Androsta-5,16-diene Series.”, Synth. Commun200232, 2125-2135.
  12. C. M. Marson ,"Heterocyclic Chemistry.", in Encyclopaedia of Physical Science and Technology, ed. R. A. Meyers, Academic Press, 20027, 321-343.
  13. C. M. Marson , C. A. Oare, T. Walsgrove and H. Adams, “A Sequential Stereocontrolled Cyclopropane Ring Formation and Semi-Pinacol Rearrangement.”, Tetrahedron Lett.200344, 141-143.
  14. C. M. Marson , A. S. Rioja, K. E. Smith and J. M. Behan, “The First Syntheses of Multiply Hydroxylated 16,17-Ene-Sterols Possessing the Androst-16-ene-3 b ,5 a -diol Motif.”, Synthesis2003, 535-538.
  15. C. M. Marson , J. H. Pink, D. Hall, M. B. Hursthouse, A. Malik and C. Smith, “An Asymmetric Synthesis of Aza-Analogues of the Tricyclic Skeleton of Daphnane and the ABC Ring System of Phorbol.”, J. Org. Chem.,200368, 792-798.
  16. C. M. Marson , L. D. Farrand, R. Brettle and D. A. Dunmur, “Highly Efficient Syntheses of 3-Aryl-2-cycloalken-1-ones and an Evaluation of their Liquid Crystalline Properties.”, Tetrahedron200359, 4377-4381.
  17. C. M. Marson , J. H. Pink and C. Smith, “Synthesis of the First Monoaromatic B-Ring 13-Azasteroid Ring System by Sequential Angular Annulation.”, Tetrahedron200359, 10019-10023.
  18. R. A. Decréau, C. M. Marson, K. E. Smith and J. M. Behan, “Generation of Malodorous Steroids from Androsta-5,16-dienes and Androsta-4,16-dienes by Corynebacteria and other Human Axillary Bacteria”, J. Steroid Biochem. & Mol. Biol. 200387, 327-336.
  19. C. M. Marson , N. Serradji, A. S. Rioja, S. P. Gastaud, J. P. Alao, R. C. Coombes and D. M. Vigushin, “Stereodefined and Polyunsaturated Inhibitors of Histone Deacetylasebased on (2E,4E)-5-Arylpenta-2,4-dienoic Acid Hydroxyamides.”, Bioorg. Med. ChemLett.200414, 2477-2481.
  20. C. M. Marson and S. Pucci, "Three-Component Condensations of Aldehydes with N-Methoxycarboxamides.”, Tetrahedron Lett200445, 9007-9010.
  21. C. M. Marson and R. A. Decréau, “Preparation of 3,17- and 3,20-Difluoro-Derivatives of the Androst-5-ene and Pregn-5-ene Series.” Synth. Commun200434, 4368-4385.
  22. C. M. Marson and P. Savy, "Alkylnitrogen Compounds: Amines and their Salts.", Organic Functional Group Transformations, eds. A. R. Katritzky, and R. J. K. Taylor, Elsevier, New York, 2004, 255-300.
  23. A. J. A. Cobb and C. M. Marson, “Asymmetric Synthesis using Catalysts containing Multiple Stereogenic Centres and a trans-1,2-Diaminocyclohexane Core; Reversal of Predominant Enantioselectivity upon N-Alkylation.”, Tetrahedron 200561, 1269-1279.
  24. C. M. Marson , R. C. Melling, S. J. Coles and M. B. Hursthouse , “Synthesis of (3S,3S',4S,4S') -1,1'-Ethylenedipyrrolidine- 3,3',4,4'-tetraol and related Diamino Diols: Donor-Acceptor Hydrogen-Bonding Motifs of the C 2 Symmetric 3,4-Dihydroxypyrrolidine Unit.”, Tetrahedron: Asymmetry200516, 2799-2809.
  25. C. M. Marson and R. C. Melling, “Enantioselective Syntheses of trans-3,4-Difluoropyrrolidines and Investigation of their Applications in Catalytic Asymmetric Synthesis.”, J. Org. Chem., 200570, 9771-9779.
  26. C. M. Marson , P. Savy, A. S. Rioja, T. Mahadevan, C. Mikol, A. Veerupllai, E. Nsubuga, A. Chahwan, S. P. Joel “Aromatic Sulfide Inhibitors of Histone Deacetylase based on Arylsulfinyl-2,4-dienoic Acid Hydroxyamides.”, J. Med. Chem., 200649, 800-805.
  27. C. M. Marson and R. C. Melling, “Syntheses of Enantiopure 3,4-Diamino-1-Substituted Pyrrolidines.”,Synthesis2006, 247-256.
  28. C. M. Marson and M. Saadi, “ Synthesis of the penta-oxazole core of telomestatin in a convergent approach to poly-oxazole macrocycles.”, Org. Biomol. Chem.,20064, 3892-3893.
  29. C. M. Marson , T. Mahadevan, J. Dines, S. Sengmany, J. M. Morrell, J. P. Alao, S. P. Joel, D. M. Vigushin and R. C. Coombes, “Structure-Activity Relationships of Aryloxyalkanoic acid Hydroxyamides as Potent Inhibitors of Histone Deacetylase.”, Bioorg. Med. ChemLett.,200717, 136-141.
  30. C. M. Marson , E. Edaan, S. J. Coles, M. B. Hursthouse and D. T. Davies, “A Catalytic Asymmetric Protocol for the Enantioselective Synthesis of 3(2H)-Furanones.”, Chem. Commun2007, 2494-2496.
  31. J. M-M. Kwok, S. S. Myatt, C. M. Marson, G. Constantinidou, and E. W-F. Lam, “Thiostrepton selectively targets breast cancer cells through inhibition of FOXM1 expression.”, Molecular Cancer Therapeutics,2008, 7, 2022-2032.
  32. K. H. Peh, B. Y. Wan, E-S. K. Assem and C. M. Marson, “Effect of a Selection of Histone Deacetylase Inhibitors on Mast Cell Activation and Airway and Colonic Smooth Muscle Contraction.” International Immunopharmacology2008, 8, 1793-1801.
  33. C. M. Marson, “Histone Deacetylase Inhibitors: Design, Structure-Activity Relationships and Therapeutic Implications for Cancer.”, Anti-Cancer Agents in Medicinal Chemistry 2009, 9, 661-692.
  34. E-S. K. Assem, K. H. Peh, B. Y. Wan, J. Dines, J. B. Middleton, and C. M. Marson, “Effects of a Selection of Histone Deacetylase Inhibitors on the Antigen- and Agonist(s)-Induced Tracheal Smooth Muscle Contraction.”, Inflammation Research 2009, 58, S22-S23.
  35. E-S. K. Assem, S. Mann, B. Y. C. Wan and C. M. Marson, “Effect of Antioxidants on Airway Smooth Muscle Contraction: Action of Lipoic Acid and Some of its Novel Derivatives on Guinea-Pig Tracheal Smooth Muscle.”, Inflammation Research2010, 59, 235-237.
  36. B. Y. Wan, S. Mann, E-S. K. Assem and C. M. Marson, “Effect of Endogenous and Synthetic Antioxidants on Hydrogen Peroxide-Induced Guinea-Pig Colon Contraction.”, Inflammation Research2010, 59, 231-233

Prof Marson is the author of over 110 publications including one book: "Synthesis using Vilsmeier Reagents", co-authored with P. R. Giles, CRC Press, 1994, a research text of 247 pages.

Research Interests

  • New synthetic methodology
  • Anti-cancer compounds
  • Enzyme inhibitors
  • Catalytic asymmetric synthesis
  • Enantiopure C2 -symmetric compounds

Selected Publication

  1. K. H. Peh, B. Y. Wan, E-S. K. Assem and C. M. Marson, “Effect of a Selection of Histone Deacetylase Inhibitors on Mast Cell Activation and Airway and Colonic Smooth Muscle Contraction.” International Immunopharmacology, 2008, 8, 1793-1801.
  2. C. M. Marson, “Histone Deacetylase Inhibitors: Design, Structure-Activity Relationships and Therapeutic Implications for Cancer.”, Anti-Cancer Agents in Medicinal Chemistry 2009, 9, 661-692.
  3. E-S. K. Assem, K. H. Peh, B. Y. Wan, J. Dines, J. B. Middleton, and C. M. Marson, “Effects of a Selection of Histone Deacetylase Inhibitors on the Antigen- and Agonist(s)-Induced Tracheal Smooth Muscle Contraction.”, Inflammation Research 2009, 58, S22-S23.
  4. E-S. K. Assem, S. Mann, B. Y. C. Wan and C. M. Marson, “Effect of Antioxidants on Airway Smooth Muscle Contraction: Action of Lipoic Acid and Some of its Novel Derivatives on Guinea-Pig Tracheal Smooth Muscle.”, Inflammation Research, 2010, 59, 235-237.
  5. B. Y. Wan, S. Mann, E-S. K. Assem and C. M. Marson, “Effect of Endogenous and Synthetic Antioxidants on Hydrogen Peroxide-Induced Guinea-Pig Colon Contraction.”, Inflammation Research, 2010, 59, 231-233

Research Summary

Charles Marson Research Image

Research in the Marson group involves both the uncovering of new synthetic organic methodology and the synthesis of novel organic compounds, either for use in chemical biology or medicinal chemistry. In the latter category are potent inhibitors of histone deacetylase (HDAC), as depicted in gold and magenta, docked into HDAC1 enzyme, and possessing in vitro IC50values down to 1 nM.

Reversal of Enantioselectivity using Catalysts containing Multiple Stereogenic Centres

Professor Charles Marson


To develop organic reactions, including carbonyl and Michael additions, and Strecker reactions so that they give predominantly one enantiomer, by the use of a catalytic quantity of a new ligand that contains four or more chiral centres.

Summary of Results

Ligands containing four or more chiral centres are indeed effective catalysts for asymmetric reactions including carbonyl and Michael additions. Moreover, in some cases, the sense of asymmetric induction can be switched (e.g. R to S) simply by converting the NH groups of the catalyst into NMe ones. Such reversal of enantioselectivity suggests that it may not be necessary to prepare both enantiomeric forms of a catalyst in order to make the desired product in both of its enantiomerically pure forms.


(a) We seek to understand the fundamental principles of catalysis as a function of additional chiral centres, and therefore have prepared new ligand systems such as 2, 3 and 4 in enantiopure forms.1,2 We have shown that despite being only a secondary amine (rather than tertiary), 2a is an effective catalyst for asymmetric alkylation.1 This establishes the use of potentially quadridentate catalysts with multiple chiral centres (and several heteroatoms), such compounds being essentially without precedented use in asymmetric synthesis.1,2 Their likelihood of success and biological relevance is exemplified by zinc finger protein regions, in which multicoordination of zinc to nitrogen (and oxygen) in a highly stereodefined chiral pocket induces catalytic activity of crucial enzymes, including protein phosphatases and metallo-enzymes in general, in which several different metals can form polyligated catalytic centres. Such catalysts impinge upon enzyme mimicry, but also probe new catalytic phenomena: dimer formation may be suppressed, (since the highly organised single ligand array can satisfy metal coordination requirements) leading to efficient catalysis via a monomer. A practical aim is the achievement of higher e.e'.s and on a wider range of compounds and reactions.

Synthesis of ligands with multiple stereogenic centres

Scheme 1. Synthesis of ligands with multiple stereogenic centres

Asymmetric addition of diethylzinc to aldehydes

Previous work has described only low e.e.'s with the secondary amine ligands, so it is noteworthy that the secondary amine catalyst 10a affords 80% e.e. (Table 1, entry 5). For all of the catalysts 8b, 9b and 10b the effective of N-methylation is to favour formation of (S)-1-phenylpropan-1-ol, the effect being greatest with 10b, a catalyst with six stereogenic centres. With a trans-1,2-diaminocyclohexane as part of a salen ligand, e.e.'s of 30-70% were obtained for the addition of diethylzinc to benzaldehyde;3 our ligands 9 compare well, and show that systems based on a trans-1,2-diaminocyclohexane but with extended chirality can deliver significantly high e.e.'s in the enantioselective addition to carbonyl compounds. In contrast, the amines 7a and 7b, lacking additional chirality, did not give satisfactory results.

Catalysts with multiple stereogenic centres for asymmetric alkylation

Scheme 2. Catalysts with multiple stereogenic centres for asymmetric alkylation.1

In the present work, the absolute configurations obtained with catalysts 10a and 10b, as well as the fact that 10b favours the (S)-configuration of 1-arylpropan-1-ol (compared with 10a) might be accounted for by a model in which the aldehyde presents to the pocket of the catalyst defined by the flanking wall of the aminocyclohexanol ring and the basal plane that includes two zinc and two oxygen atoms (Scheme 3). For 10a, the NH group is sufficiently small to allow the aryl ring to reside nearby, leading to the attack of the aldehyde on its Re-face. Conversely, for 10b, the bulk of the N-methyl group is presumed to be sufficient to hinder location of the aryl group as above, thereby leading to Si-face addition and predominantly the (S)-1-arylpropan-1-ol. However, further investigations are needed before these tentative proposals can be regarded as having been substantiated.

A rationale for enantioselective switching upon N-methylation

Scheme 3. A rationale for enantioselective switching upon N-methylation.

Catalytic asymmetric transfer hydrogenation

Multiple co-ordination of a metal to nitrogen is frequently observed in metalloproteins, e.g. Cu, Zn superoxide dismutase in which Cu(II) is bound tetragonally bound to four histidine ligands. In this project, tetradentate coordination as part of an amide network were sought that might possess some essential features of metalloprotein catalysts. Since metalloproteins are commonly involved in redox processes, we selected the contemporary area of asymmetric transfer hydrogenation4 as a probe reaction for the design of catalysts and their optimisation. The use of amides in catalytic asymmetric synthesis has been described only recently,5 but we have found that the potentially tetradentate ligands 3 and 4 containing amides can act as catalysts for asymmetric transfer hydrogenation (Scheme 4).2,6

Catalytic asymmetric transfer hydrogenation

Scheme 4. Catalytic asymmetric transfer hydrogenation.2

Ruthenium catalysed asymmetric transfer hydrogenation takes advantage of low redox potentials and the strong affinities of ruthenium for heteroatoms, and is an efficient system for the asymmetric reduction of carbonyl groups.4 Consequently, it was used to test the amide ligands 3 and 4 (Table 1) under standard conditions in which the catalysts are preformed by heating the amide (2 mol%) with RuCl2(PPh3)3 (1 mol%)7 in propan-2-ol at 80 oC for 1 h. After cooling to 20 oC, KOH (10 mol%) was added, then acetophenone (0.1 M) or a derivative. The major conclusions to be drawn are: first, that the oxamide ligand 3a, possessing terminal NH2 groups, afforded e.e.'s (39-48%) that depended little on the p-substituent R of the aryl ketone (Scheme 4). Secondly, compared with NH2 as the terminal group, N-benzyl substitution generally furnishes higher yields (e.g. entries 3 and 4), although in only low e.e.'s.

Entry Ketone R1 Ligand Ligand linker X Terminal Group R Reaction time (h) Conversion (%)b e.e. (%)c Configuration
1 H 3a - H 24 29 48 (S)
2 H 3a - H 48 44 44 (S)
3 H 3b - CH2Ph 24 64 16 (R)
4 H 3b - CH2Ph 48 71 15 (R)
5 H 4a (CH3)2C H 24 8 62 (R)
6 H 4a (CH3)2C H 48 8 61 (R)
7 Cl 3a - H 72 28 39 (S)
8 Cl 3b - CH2Ph 72 49 13 (R)
9 F 3a - H 72 29 46 (S)
10 F 3b - CH2Ph 72 71 9 (S)
11 OMe 3a - H 72 22 43 (S)
12 OMe 3b - CH2Ph 72 13 12 (S)

a Standard conditions of KOH (10 mol%), ligand (2 mol%) and RuCl2(PPh3)3 (1 mol%) were used.

b Conversions were determined by 1H NMR spectroscopy.

Thirdly, an oxamide linker gave the best yields but low e.e.'s (9-15%, S or R); in contrast, a malonamide spacer gave poorer yields but substantially higher e.e.'s (e.g. entries 1 and 2 compared with entries 5 and 6). Lastly, and most significantly, N-benzylation leads in every case for the oxamide catalysts to a greatly increased proportion of the (R)-enantiomer, and for the examples R = H and Cl, even results in a reversal of the absolute configuration of the major product. Thus, for R = H, entries 1 and 3, and entries 2 and 4 show switching to enantiomeric excess in favour of the (R)-configuration, using the N-benzyl catalysts, as do the runs for R = Cl (entries 7 and 8). These examples of reversal8 of enantiomeric excess designation are apparently the first to be reported for asymmetric transfer hydrogenation. Thus, although rhodium catalysed reduction of ketones with diurea ligands containing four chiral centres exhibited matched and mismatched effects, depending on the diastereoisomerism of the ligands, N-methylation did not lead to reversal of enantioselectivity.9 The potential of the same ligand skeleton to provide both enantiomers, thus obviating the need to prepare both enantiomeric ligands, could be a major advantage of these new tetra-aza catalysts, and deserves further study.

Although tetradentate ligation of these catalysts through four nitrogen atoms is possible, it has yet to be investigated. Amide coordination to metals is known in both the neutral amide and the anionic amide forms, and the possibility of coordination through either nitrogen or oxygen atoms presents further options. The present study shows that significant enantiomeric excesses can also be obtained for asymmetric transfer hydrogenation using catalysts containing amide groups in the ligand, and such catalysts hold promise for the design of new amidic catalysts that mimic redox processes carried out by enzyme complexes.

Support from the EPSRC for a studentship (to A.J.C.) and from the Deutsche Forschungsgemeinschaft (to I. S.) is gratefully acknowledged.


  1. Cobb, A. J.; Marson, C. M. Tetrahedron: Asymmetry 2001, 12, 1547.
  2. Marson, C. M.; Schwarz, I. S. Tetrahedron Lett. 2000, 41, 8999.
  3. Cozzi, P. G.; Papa, A.; Umani-Ronchi, A. Tetrahedron Lett. 1996, 37, 4613.
  4. (a) Noyori, R.; Hashiguchi, S. Acc. Chem. Res. 1997, 30, 97. (b) Naota, T.; Takaya, H.; Murahashi, S.-I. Chem. Rev. 1998, 98, 2599. (c) Palmer, M. J.; Wills, M. Tetrahedron: Asymmetry 1999, 10, 2045.
  5. Amides have been used mainly in catalytic asymmetric oxidations: (a) End, N.; Pfaltz, A. J. Chem. Soc., Chem. Commun. 1998, 589. (b) End, N.; Macko, L.; Zehnder, M.; Pfaltz, A. Chem. Eur. J. 1998, 4, 818.
  6. Halle, R.; Breheret, A.; Schulz, E.; Pinel, C.; Lemaire, M. Tetrahedron: Asymmetry 1997, 8, 2101. Here, amide tetradentate ligands were used in the reduction of ketones but gave a maximum of 22% e.e. (5% conversion).
  7. Hallman, P. S; Stephenson, T. A.; Wilkinson, G. Inorg. Synth. 1970, 12, 238.
  8. In the asymmetric addition of diethylzinc to benzaldehyde, N-methylation can lead to reversal of enantioselectivity: Kimura, K.; Sugiyama, E.; Ishizuka, T.; Kunieda, T. Tetrahedron Lett. 1992, 33, 3147.
  9. Gamez, P.; Dunjic, B.; Lemaire, M. J. Org. Chem. 1996, 61, 5196.

Cancer Chemotherapy : New, Selective Anhydride Inhibitors of Geranylgeranyl-Protein Transferase

Professor Charles Marson


The optimisation of a novel lead compound (anhydride 1d) that is a specific inhibitor of geranylgeranyl-protein transferase-I (GGPTase-I), with a longer term goal of a clinical trial on refractory cancers, especially those involving mutation of K-Ras protein such as some colon, lung and breast cancers and pancreatic carcinomas.


Non-peptidic inhibitors specific for geranylgeranyl-protein transferase-I (GGPTase-I) were recently synthesised in our laboratories (e.g. acid anhydride 1d).1 This is an important finding because it has emerged that prevention of post-translational modification of K-Ras by geranylgeranylation is actually much more relevant as a potential therapy than prevention of the corresponding farnesylation (Scheme 1). Protein prenylation (C-terminal addition of a farnesyl or geranylgeranyl isoprenoid unit) is effected by one of three cellular enzymes: farnesyl-protein transferase (FPTase), geranylgeranyl-protein transferase-I (GGPTase-I) or GGPTase-II. The extent of Ras prenylation has important consequences for signal transduction. However, prenylation (especially geranylgeranylation) of proteins other than Ras, notably those of the Rho family,2 is now considered to be central to anti-transforming mechanisms.3 A GGPTase-I inhibitor blocked K-Ras4B processing and oncogenic signaling; in contrast, K-Ras4B, the protein encoded by the commonly mutated ras gene in human cancers,4 is not very sensitive to farnesyl-protein transferase (FPTase) inhibitors. Since mutations of the ras oncogene are found in 50% of colon and 90% of pancreatic carcinomas, as well as in lung cancers, and in 25% of all human tumours, inhibition of the prenylation of Ras protein offers a new approach to cancer therapy.5,6

A new anhydride inhibitor that is selective for geranylgeranyl-protein transferase I (GGPTase-I)

Figure 1. A new anhydride inhibitor that is selective for geranylgeranyl-protein transferase-I (GGPTase-I).

Our novel GGPTase-I inhibitors are cyclic acid anhydrides such as 1d, which was shown to revert the phenotype of a Ki-Ras-transformed rat kidney fibroblast cell line in culture (Figure 3).7 Competitive inhibition of GGPTase-I with respect to geranylgeranyl diphosphate (GGPP) and non-competitive inhibition with respect to Ras-CVLL have been demonstrated.7 Cyclic acid anhydrides have not to our knowledge been described previously as specific inhibitors of any enzyme. Moreover, anhydrides of type 1 are unusual in that (a) they are non-peptidic inhibitors of (GGPTase-I) (b) they incorporate a lipid chain and (c) they may be the first examples in which an anhydride group mimics the diphosphate group, so prevalent in key bioenergetic pathways.

Inhibition of the geranylgeranylation of Ras protein as a target for cancer therapy

Recently, the importance of GGTase inhibitors has been accepted; even in the case of a very effective in vivo FPTase inhibitor, cellular Ras is still prenylated by switching its requirement from a farnesyl group to a geranylgeranyl one (Scheme 1). Indeed, (oncogenic) K-Ras assumes transforming activity when a geranylgeranyl group becomes linked to a terminal cysteine residue in the Ras protein (Scheme 1 but with CVLL terminus). K-Ras4B can be prenylated by both GGPTase-I and FPTase, and this suggests that a combination of GGPTase-I and FPTase inhibitors may be required for the inhibition of tumour growth that is dependent upon K-Ras; indeed, a combination antitumour therapy of GGPTase-I inhibitors with taxol or cisplatin has been shown to be effective. There is good evidence that inhibition of K-Ras prenylation requires both an FPTase and a GGPTase-I inhibitor.

Prenylation of Ras protein

Scheme 1. Prenylation of Ras protein.

In the mevalonic acid pathway, prenyl groups (C5 units) undergo condensation to give compounds comprised of multiples of C5 units, (e.g. a geranylgeranyl group comprises four prenyl groups: Scheme 1, n=1 is one prenyl unit). In this way, a wide variety of terpenoids are biosynthesised, some of which, especially the monoterpenoids (C10 compounds such as limonene, perillyl alcohol and farnesol) exhibit anticancer effects. Both limonene8 and perillyl alcohol have been the study of clinical trials against cancer. For over ten years, the mode of action of such terpenoids was the subject of much speculation, but Gould has recently shown that in NIH3T3 cells perillyl alcohol inhibits in vivo prenylation of GGPTase types I and II, though not of FPTase.9 The same is likely to be true of farnesol (also a monoterpenoid), that we have shown preferentially inhibits proliferation and induces apoptosis of blasts derived from patients with acute myeloid leukaemia (AML), but which does not kill non-transformed cells, including primary monocytes and T-lymphocytes.10

The recent recognition of the importance of inhibitors selective for GGPTase makes them an essentially unexplored and novel target. Our anhydrides 1 are among the first selective and non-peptidic compounds, and have moderate potency but close to clinical usefulness. Consequently, we seek to optimise the molecular design of 1 with regard to GGPTase activity, cellular and clinical potency.

Outline of Results

Determination of in vitro inhibition of GGPTase-I, enzyme kinetics and anti-transforming properties of the new anhydrides.7 These were conducted by Dr David Vigushin of the Department of Cancer Medicine in the CRC unit of the Hammersmith hospital headed by Prof. Charles Coombes. The longest chain tested (anhydride 1d) gave excellent inhibition of GGPTase-I; in contrast, the lack of an alkyl (including prenyl) chain, as in itaconic anhydride 1a, excludes inhibition of prenyl-protein transferase activity. Moreover, 1d is highly selective for GGPTase-I: FPTase (48:1), and 1b also shows good selectivity (11:1). Interestingly, a 2:1 E:Z mixture of 1c shows excellent selectivity (56:1) for FPTase. Such selectivity for GGPTase-I has not, to our knowledge, been previously reported for non-peptidomimetic inhibitors, and is also a rare feature for peptidomimetic compounds.

Synthetic inhibitors of GGPTase-I

Figure 2. Synthetic inhibitors of GGPTase-I.

The anhydrides 1b - 1d are much more efficient inhibitors of GGPTase-I than are the corresponding succinic acids (obtained by hydrolysis). That implies a special role for the anhydride group, and indicates that rapid hydrolysis to the corresponding diacid did not occur in vitro. That the anhydride moiety is not merely acting as a delivery system is shown by the activity of 1d (GGPTase-I IC50 = 21 µM) versus the inactivity of the corresponding dimethyl ester. This may support an acylating role for the anhydrides. As expected, the length and nature of the lipid chain markedly influences the prenyl-protein transferase activity, since the inhibitor must dock into the pocket usually occupied by GGPP.

Kinetics of GGPTase-1 Inhibition. Inhibition of GGPTase-I by 1b was competitive with respect to GGPP (Ki 314 µM) and non-competitive with respect to Ras-CVLL (Ki 297 µM). The pattern of inhibition for 1d was similar, competitive with respect to GGPP (Ki 127 µM) and non-competitive with respect to Ras-CVLL (Ki 72.9 µM, Appendix I).

Cell Morphology Studies. Cell morphology studies confirmed that 1d could revert the phenotype of a Ki-Ras-transformed rat kidney fibroblast cell line in culture (Appendix II). The untransformed cells are long and spindle-shaped with typical fibroblast morphology that cease to grow when confluent (contact inhibition). In contrast, Ki-Ras-transformed KNRK cells are rounded, have a retracted spindle shape, and are not contact-inhibited (Appendix IIA). Treatment with 50 µM 1d for 48 hours caused partial reversion of the transformed morphology (Appendix IIB). Exposure to 500 µM 1d induced many of the cells to undergo more complete phenotypic reversion to the flattened, spindle-shaped appearance of normal fibroblasts (Appendix IIC). Phenotypic reversion is most likely due to restoration of the normal actin cytoskeleton which is severely disrupted in transformed cells.11

Outline of Future Work

Synthesis of new small molecule inhibitors of GGPTase-I. In our programme to develop new terpenoid-based anticancer agents,1,7,14 design and synthesis of compounds was conducted in order to evaluate a model for PTase inhibitors that required a bulky polar 'head' group and a lipophilic 'tail' region, to which anhydrides 1 conform. This model takes into account both the polar 'head' region (the ring) found in some PTase inhibitors such as perillyl alcohol, and also the molecular and spatial requirements of prenyl-to-protein transfer induced by the enzyme (Scheme 1). Since the 'tail' region of anhydride 1 possesses some structural resemblance to the diterpenoid GGPP (a C20 compound) it was thought that 1 might mimic GGPP, bind to the PTase, and thereby be a competitive inhibitor, which indeed we have recently demonstrated.1,7


  1. Marson, C. M.; Rioja, A. S.; Brooke, G.; Coombes, R. C.; Vigushin, D. M. Bioorg. Med. Chem. Lett. 2002, 12, 0000. Initial communication (accepted) concerning anhydrides as inhibitors of prenyl-protein transferases.
  2. (a) Du, W.; Lebowitz, P. F.; Prendergast, G. C. Mol. Cell Biol. 1999, 19, 1831-1840. (b) Prendergast, G. C.; Lebowitz, P. F. Oncogene 1998, 17, 1439-1435. (b) Prendergast, G. C.; Khosravi-Far R.; Solski, P. A.; Kurzawa, H.; Lebowitz, P. F.; Der, C. J. Oncogene 1995, 10, 2289-2296. (c) Qiu, R. G.; Chen, J.; McCormick, F.; Symons, M. Proc. Natl. Acad. Sci. USA 1995, 92, 11781-11785. (d) Khosravi-Far, R.; Solski, P. A.; Clark, G. J.; Kinch, M. S.; Der, C. J. Mol. Cell Biol. 1995, 15, 6443-6453.
  3. Prendergast, G. C.; Davide, J. P.; deSolms, S. J.; Giuliani, E. A.; Graham, S. L.; Gibbs, J. B.; Oliff, A.; Kohl, N. E. Mol. Cell Biol. 1994, 14, 4193-4202.
  4. Emmanuel, P. D.; Snyder, R. C.; Wiley, T.; Gopurala, B.; Castleberry, R. P. Blood 2000, 95, 639-645.
  5. Sebti, S. M.; Hamilton, A. D. Pharmacol. Ther. 1997, 74, 103. (b) Leonard, D. M. J. Med. Chem. 1997, 40, 2971-2990.
  6. (a) Gibbs, J. B. Cell 1991, 65, 1-4. (b) Barbacid, M. Annu. Rev. Biochem. 1987, 56, 779.
  7. Marson, C. M.; Rioja, A. S.; Brooke, G.; Coombes, R. C.; Vigushin, D. M., in preparation.
  8. Donaldson, M. J., Skoumas, V.; Watson, M.; Ashworth, P. A.; Ryder, H.; Moore, M.; Coombes, R. C. Eur. J. Cancer 1999, 35, 1014.
  9. Ren, Z. B.; Elson, C. E.; Gould, M. N. Biochem. Pharmacol. 1997, 54, 113-120.
  10. Rioja, A.; Pizzey, A. R.; Marson, C. M.; Thomas, N. S. B. FEBS Lett., 2000, 467, 291-295.
  11. Pompliano, D. L.; Rands, E.; Schaber, M. D.; Mosser, S. D.; Anthony, N. J.; Gibbs, J. B. Biochemistry 1992, 31, 3800-3807.

A New Rearrangement for the Enantioselective Synthesis of Pyranoid Systems with Antitumor and Immunosuppressant Activity

Professor Charles Marson


The stereocontrolled construction of natural products by a new rearrangement of epoxy alkynols to give dihydropyranones. The methodology is to be extended so that spiroketals as well as isolated ring pyranoid fragments are accessible enantioselectively. Synthetic approaches to immunosuppressants, anticancer agents and inhibitors of signal transduction are planned.

Examples of biologically active tetrahydropyrans systems and lactones accessible through the rearrangement of epoxy alkynols to 2,3-dihydropyran-4-ones

Figure 1. Examples of biologically active tetrahydropyrans systems and lactones accessible through the rearrangement of epoxy alkynols to 2,3-dihydropyran-4-ones (see Scheme 1).


Our group has uncovered an unprecedented stereospecific rearrangement of epoxy alkynols to give dihydropyranones (Scheme 1, eq. i).1 This rearrangement involves suprafacial migration of hydride, leading to ring opening of the epoxide with inversion solely at the alpha-position. As a consequence, the geometry of the alkene exclusively determines the relative configuration of the 2,3-disubstituted dihydropyranone (Scheme 1, eqs. ii and iii); the cis-alkyl groups in (E)-2-methylpent-2-enal results in the cis-disubstituted dihydropyranone 3a (eq. ii), whereas trans-substituents on the enal afford the trans-dihydropyranone 3b (eq. iii), a stereospecific process. In the general case, it appears that both epimers of 1 contribute to the formation of the single diastereoisomer 3. No intermediates have been detected, although it would seem probable that alkynone 2 is involved. It appears that catalytic Hg(II) induces activation of the alkyne bond and also ring opening of the epoxide with hydride shift.

A new stereospecific rearrangement of epoxy alkynols to 2,3-dihydropyran-4-ones

Scheme 1. A new stereospecific rearrangement of epoxy alkynols to 2,3-dihydropyran-4-ones.

Reagents: i, Propargyl alcohol, 2 n-BuLi ii, t-BuOOH, VO(acac)2 or m-CPBA iii, 1 mM Hg in 0.5 mM aq. H2SO4 iv, hept-1-yne, n-BuLi.

An important extension was to show that an asymmetric synthesis of the dihydropyranones could be achieved by rearrangement of an optically active epoxy alkynol derived from Sharpless epoxidation2 of the enynol under conditions of kinetic resolution (Scheme 2, eq. iv). No measurable loss of enantiomeric excess was detected. This indicates that if the epoxy alkynols 2 can be prepared enantiopure, that the corresponding dihydropyranones 3 should also be obtained enantiopure. The ability to prepare 2,3-disubstituted dihydropyranones (eqs. ii, iii and iv) with stereocontrol is important, since they cannot be prepared from standard Danishefsky diene components3 which usually furnish only the 2-monosubstituted pyranoid.

Extensions of the new rearrangement of epoxy alkunols to 2,3-dihydropyran-4-ones

Scheme 2.Extensions of the new rearrangement of epoxy alkynols to 2,3-dihydropyran-4-ones.

Reagents: i, t-BuOOH, Ti(OiPr)4, (+)-DET ii, 1 mM Hg in 0.5 mM aq. H2SO4

Outline of strategies for natural product synthesis

I. Enantioselective Synthesis of the Discodermolide Lactone Ring. Discodermolide, a marine natural product that inhibits the proliferation of T cells (IC50 = 9 nm),4 is a potent immunosuppressant and also possesses high antitumour activity, probably through microtubule stabilisation.5 Its limited availability increases the importance of synthetic efforts5-7 towards such lactonic natural products. Total syntheses of discodermolide and its enantiomer have taken several approaches, including Evans oxazolidinone-mediated aldol condensations,5 and more recently, allenyltin additions to aldehydes.6 A route to the lactone portion of (+)-discodermolide is shown in Scheme 3.

The ester 3, prepared from propenal and a stabilised ylid,8 is subjected to a standard reduction followed by Sharpless asymmetric epoxidation and Swern oxidation to give aldehyde 4 (Scheme 3). The conversion of 4 into 6 is secured according to the cyclisation established1 in eq. v of Scheme 2, the ethoxy group being an effective terminus for the alkyne unit. Subsequent hydrolysis to the keto lactone would be followed by Luche reduction, at the less hindered face, and to give the equatorial alcohol 7. Alkylation (with protection of the hydroxyl oxygen as necessary) would deliver the trans-dimethyl diastereoisomer 8; there is good precedent for the introduction of the methyl group at C-3, and trans to the existing C-5 methyl group. Such stereoselective alkylation was accomplished in total syntheses of venturicidins A and B by using LDA/HMPA and MeI at -78 oC. There, the desired isomer was obtained in 95% yield (80:1 diastereoselectivity).9 A hindered hydroboration followed by selective oxidative work-up would give 9, containing the correct assembly for discodermolide (Figure 1).

Construction of the lactone portion of (+)-discodermolide

Scheme 3. Construction of the lactone portion of (+)-discodermolide. Reagents: i, DIBAH ii, t-BuOOH, Ti(OiPr)4, (+)-DET iii, TFAA, DMSO, Et3N, -78 degC iv, , n-BuLi v, 1 mM Hg in 0.5 mM aq. H2SO4 vi, 2M HCl; then NaBH4, CeCl3 vii, 2 LDA, HPMA; MeI viii (C6H11)2BH, then alkaline H2O2; (COCl)2, DMSO, CH2Cl2, -78 degC.10

II. Enantioselective Synthesis of the Spiroketal Subunit of Cytovaricin. Cytovaricin is a 22-membered lactone that incorporates an annealed spiroketal unit.11 The spirocyclic portion is essential for pharmacological activity. Recently, cytovaricin B has been isolated from a Streptomyces fermentation broth,12 but in amounts of less than 1 mg. Cytovaricin B (which possesses an OMe group at C-17 in place of the OH group in cytovaricin itself) has been identified as an inhibitor of signal transduction.12 The finding that cytovaricin B inhibits STAT5 phosphorylation in cell lines without affecting JAK2 phosphorylation (IC50 = 32 µm),12 together with the great scarcity of cytovaricins, prompts further synthetic methodological studies in this area.

Construction of the lactone ring of discodermolide would provide the basis for assembling the substitution pattern in the cyctovaricins. This approach is to be combined with the successful spontaneous spiroketalisation strategy we have shown is effective for the spirocyclisation leading to a [5,4] spiroketal system,1 and should deliver the key spiroketal portion of the cytovaricins (see Figure 1). The formation of a [5,5] spiroketal unit, with its locked double chair conformation and well-defined anomeric effects will favour the spiroketal form more than the ring-chain isomer (as compared with the [5,4] spiroketal analogue, e.g. Scheme 4).

Construction of a [5,4] spiroketal subunit

Scheme 4. Construction of a [5,4] spiroketal subunit.1 Reagent: i, 1 mM Hg in 0.5 mM aq. H2SO4


  1. C. M. Marson, S. Harper, C. A. Oare and T. Walsgrove, J. Org. Chem., 1998, 63, 3798.
  2. R. A. Johnson and K. B. Sharpless in 'Comprehensive Organic Synthesis', eds. B. M. Trost and I. Fleming, Pergamon, Oxford, 1991, vol. 2, p. 1083.
  3. S. J. Danishefsky, D. F. Harvey, G. Quallich and B. J. Uang, J. Org. Chem., 1984, 49, 392.
  4. D. T. Hung, J. B. Nerenberg and S. L. Schreiber, J. Am. Chem. Soc., 1996, 118, 11054.
  5. S. P. Gunasekera, M. Gunasekera, and R. E. Longley, J. Org. Chem., 1991, 56, 1346.
  6. J. A. Marshall and B. A. Johns, J. Org. Chem., 1998, 63, 7885.
  7. A. B. Smith, III, Y. Qui, D. R. Jones and K. Kobayashi, J. Am. Chem. Soc., 1995, 117, 12011.
  8. E. Piers and E. H. Ruediger, J. Org. Chem., 1980, 45, 1725.
  9. H. Akita, H. Yamada, H. Matsukura, T. Nakata and T. Oishi, Tetrahedron Lett., 1990, 31, 1735.
  10. M. Nakata, T. Ishiyama, Y. Hirose, H. Maruoka and K. Tatsuta, Tetrahedron Lett., 1993, 34, 8439.
  11. D. A. Evans, S. W. Kaldor, T. K. Jones, J. Clardy and T. J. Stout, J. Am. Chem. Soc., 1990, 112, 7001.
  12. N. Yamashita, K. Shin-Ya, M. Kitamura, H. Wakao, K. Furihata, K. Furihata, Y. Hayakawa, A. Miyajama and H. Seto, J. Antibiotics, 1997, 50, 440.

Stereoselective Synthesis of Chiral Organofluorine Compounds and their Applications in Catalysis and Materials

Professor Charles Marson


The stereocontrolled and chiefly enantioselective synthesis of new chiral fluorine compounds, especially compounds containing one or more -CHF- units, and an investigation into their catalytic and materials properties.


Recently, our group achieved the enantioselective syntheses of vicinal difluoropyrrolidines 5 (Scheme 1), perhaps the first enantioselective synthesis of any vicinally substituted difluoro compound of C2 symmetry.1 Remarkably, the difluoropyrrolidines were found to be catalysts for asymmetric synthesis (Scheme 3). This is the first time that asymmetric induction mediated by a -CHF-CHF- unit been demonstrated, and hence the first example of catalysis by a compound whose chirality depends upon organofluorine asymmetry. The 3,4-difluoropyrrolidine system also appears to be the first catalyst lacking hydroxy groups that mediates enantioselective epoxidation under Sharpless conditions.


(a) Uses and synthesis of chiral organofluorine compounds. The stereoselective synthesis of organofluorine compounds2,3,4 is of major importance in many fields, for example pharmaceuticals,2,3,5 nucleoside and carbohydrate chemistry,2b biochemistry,5,6 and liquid crystals7 and polymers.8 Many fluorinated alpha-amino acids are potent antitumour and antiviral agents.2a,5 Additionally, organofluoro ligands can be as powerful, or more so, than oxgyen ligands in coordinating metals.9 Whereas enantiocontrolled syntheses of monofluoro-organic compounds is well established, synthesis of an enantiopure vicinal difluoro compound, especially of C2 symmetry, has not to our knowledge been reported,10 prior to our communication.1 Generally, molecular fluorine adds to alkenes with syn-stereoselection, thereby precluding the formation of C2 symmetric difluorides;11 where trans-addition is observed, yields are usually low.12 For example diethylaminosulfur trifluoride (DAST),13 one of the most commonly used reagents for the conversion of alcohols into fluorides, gives merely a trace of 1,2-difluorocyclohexanes, and with loss of stereointegrity compared with the initial cyclohexane-1,2-diol.14 Those failures emphasise the significance of the enantiocontrolled introduction of fluorine at two adjacent carbon stereocentres in a single operation which proved possible for the enantiopure vicinal difluorides 5.

Synthesis of enantiopure trans-3,4-difluoropyrrolidines

Scheme 1. Synthesis of enantiopure trans-3,4-difluoropyrrolidines.

(b) Some physical properties of organofluorine compounds. Ferroelectric liquid crystals enable flat panel technology for large screens to be developed. The introduction of fluorine of defined absolute configuration radically alters the polarizability (crucial to high-speed switching devices) of ferroelectric liquid crystals (Scheme 2),15,16 and such organofluorine compounds have been used to validate a 'bent cylinder' model for the molecular origins of ferroelectric polarization P in liquid crystals arising from ordered packing in a chiral smectic (C*) phase.15 Thus, for 6a (Scheme 2) a small value of P is predicted (and found), but for epimer 6b, a large value is predicted (and observed). Both the absolute and relative configuration of organic molecules can be predicted by measurement of the sign and magnitude of P (when the molecule is added as a dopant to an aryl ester host of low polarization).15 However, these fundamental studies involving polarizability as a result of molecular recognition are hampered by a general inability to create the sp3C-F bond with enantiocontrol. Thus, single enantiomer alcohols or epoxy alcohols have been found to give scalemic and epimeric mixtures, respectively (Scheme 2), and with no predictable rationale. Even in such simple systems, controlling the stereochemistry at the benzylic position (crucial to materials properties) is problematic. We seek new methodology that will deliver enantiocontrolled introduction of F at sp3C, whether in an alkyl chain, or at allylic or benzylic positions.

Importance of F in ferroelectric liquid crystals, and the difficulties in obtaining enantiopure fluorides

Scheme 2. Importance of F in ferroelectric liquid crystals, and the difficulties in obtaining enantiopure fluorides.

The synthesis of 5a and 5c are particularly suitable as subunits for liquid crystalline properties, and the scope of applications of difluoropyrrolidines 5 and their derivatives are currently being evaluated for use as liquid crystals and other new materials.

Catalytic features of difluoropyrrolidines

Catalysis of the epoxidation of allylic alcohols by difluorides 5 was investigated; reactions were conducted in dichloromethane using 15 mol% of Ti(OiPr)4 and 10 mol% of catalyst (Table 1, entries 2-7). In the absence of a catalyst, racemic 7 was obtained (81% yield). The diol 3a, derived (from (2R,3R)-(+)-tartaric acid, the natural isomer) afforded 2,3-epoxygeraniol (7) (97%), in 25% e.e. in favour of the (2S, 3S)-enantiomer (entry 2); however, the difluorinated catalyst 5b afforded a 90% yield of 2,3-epoxygeraniol (7) in 66% e.e., in favour of the opposite i.e. (2R, 3R)-enantiomer (Table 1, entry 5). Entries 3-5 suggest that fluoro groups can provide as good as or greater enantioselection than hydroxyl groups (entry 2), at least in cases where the catalyst incorporates a C2 -CHF-CHF- unit that is part of a heterocyclic ring.

Catalytic asymmetric epoxidation with 3,4-difluoropyrrolidines

Scheme 3. Catalytic asymmetric epoxidation with 3,4-difluoropyrrolidines, and comparison with 3a.1

The reversal of the major enantiomer of 2,3-epoxygeraniol when using catalyst 3 compared with catalyst 5 would be expected if the modes of binding of the hydroxyl and fluoro catalysts had important features in common. However, it is clear that the key catalytic species must be very different from the accepted model for Sharpless asymmetric epoxidation using tartrates (Scheme 4). In the catalytic asymmetric Sharpless epoxidation,17 free hydroxyl groups on the catalyst (dialkyl tartrate) are a prerequisite for enantioselectivity. In marked contrast to such Sharpless catalysts, the difluorides 5 lack hydroxyl groups and are incapable of deprotonation that could lead to ligand exchange, in yet 5 are viable catalysts for asymmetric epoxidation. Catalysis involving the difluorides 5 would appear to involve (a) co-ordination of titanium to the ring nitrogen atom (b) probably a monomeric catalytic assembly and (c) asymmetric induction that requires the fluorine atoms to have a significant spatial volume, a feature that is surprising in view of the generally accepted effective size of fluorine (attached to carbon) being smaller than a methyl group.

Sharpless asymmetric epoxidation

Scheme 4. Sharpless asymmetric epoxidation [using (2R,3R)-(+)-tartaric acid esters].

The presence of fluorine ligands in organic reactions mediated by catalysis is an emerging area of importance.18 In only one other study has chiral catalysis been a consequence of the spatial arrangement of the fluorine atoms.19 Consequently, the present examples are, to our knowledge, the first examples of asymmetric synthesis catalyzed by a C2 symmetric vicinal difluoro compound. Extensions in this area will contribute to our research programme concerning the syntheses and uses of enantiopure organofluorine compounds.1,20

Support from the EPSRC for a studentship (to R.C.M.) under the ROPA initiative is gratefully acknowledged.


  1. C. M. Marson and R. C. Melling, J. Chem. Soc., Chem. Commun., 1998, 1224.
  2. (a) V. Aoloshonok in Biomedical Frontiers of Fluorine Chemistry, eds. I. Ojima, J. R. McCarthy and J. T. Welch, American Chemical Society Symposium Series, Washington, DC, 1996, vol. 639 pp. 26-41 (b) J. A. McCarthy et al., ibid., pp. 246-264; L. W. Hertek et al., ibid, pp. 265-278; M. Namchuk et al., ibid, pp. 279-293.
  3. P. Bravo, G. Resnati, Tetrahedron: Asymmetry, 1990, 1, 661.
  4. a) J. T. Welch, S. Eswarakrishnan, Fluorine in Bio-organic Chemistry, Wiley, New York, 1991; b) J. A. Wilkinson, Chem. Rev., 1992, 92, 505.
  5. D. Bouzard, P. Dicesare, M. Essiz, J. P. Jacquet, J. R Kiechel, P. Remuzon, A. Weber, T. Oki, M. Masuyoshi, R. E. Kessler, J. Fungtomc, J. Desiderio, J. Med. Chem., 1990, 33, 1344.
  6. E. Kun, E. Kirsten, M. L. Sharma, Proc. Natl. Acad. Sci. USA, 1977, 74, 4942.
  7. H. Liu and H. Nohira, Liquid Crystals, 1996, 581.
  8. C. K. Chen, Y. L. Hu, M. Spears, J. W. Hodby, B. M. Wanklyn, A. V. Narlikar, S. B. Samanta, J. Mater. Sci. Lett., 1996, 15, 886.
  9. H. Plenio, R. Diodone, D. Badura, Angew. Chem., Int. Ed. Engl., 1997, 36, 156.
  10. For a single example of a racemic vic-difluoro-2,3-dihydrobenzo[b]furan see: R. Ruzziconi, G. V. Sebastiani, J. Heterocycl. Chem., 1980, 17, 1147.
  11. S. Rozen, M. Brand, J. Org. Chem., 1986, 51, 3607.
  12. M. Sato, T. Hirokawa, A. Hattori, A. Toyota, C. Kaneko, Tetrahedron: Asymmetry, 1994, 5, 975.
  13. a) W. J. Middleton, J. Org. Chem., 1975, 40, 574; b) M. Hudlicky, Org. React., 1988, 35, 513.
  14. D. F. Shellhamer, D. T. Austine, K. M. Gallego, B. R. Ganesh, A. A. Hanson, K. A. Hanson, R. D. Henderson, J. M. Prince, V. L. Heasley, J. Chem. Soc., Perkin Trans. 2, 1995, 861.
  15. D. M. Walba, H. A. Razavi, N. A. Clark and D. S. Parmar, J. Am. Chem. Soc., 1988, 110, 1779.
  16. C. Loubser and J. W. Goodby, J. Mater. Chem., 1995, 5, 1107.
  17. T. Katsuki and K. B. Sharpless, J. Am. Chem. Soc., 1980, 102, 5974.
  18. A. Armstrong and B. R. Hayter, J. Chem. Soc., Chem. Commun., 1988, 621.
  19. R. O. Duthaler and A. Hafner, Angew. Chem., Int. Ed. Engl., 1997, 36, 43.
  20. R. A. Decréau, C. M. Marson and K. E. Smith, Synth. Commun., 2002, accepted.