Synthesis of 2-acetamido-1,2-dideoxy-D-galacto-nojirimycin [DGJNAc] from D-glucuronolactone: the first sub-micromolar inhibitor of a-N-acetylgalactosaminidases Daniel Best a , Phoom Chairatana a , Andreas F. G. Glawar a,b , Elizabeth Crabtree a,b , Terry D. Butters b , Francis X. Wilson e , Chu-Yi Yu d , Wu-Bao Wang d , Yue-Mei Jia d , Isao Adachi c , Atsushi Kato c , George W. J. Fleet a, * a Chemistry Research Laboratory, Department of Chemistry, University of Oxford, Mansfield Road, Oxford OX1 3TA, UK b Oxford Glycobiology Institute, University of Oxford, South Parks Road, Oxford OX1 3QU, UK c Department of Hospital Pharmacy, University of Toyama, 2630 Sugitani, Toyama 930-0194, Japan d CAS Key Laboratory of Molecular Recognition and Function, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China e Summit PLC, 91, Milton Park, Abingdon, Oxon OX14 4RY, UK article info Article history: Received 24 December 2009 Revised 2 February 2010 Accepted 12 February 2010 Available online 23 February 2010 abstract 2-Acetamido-1,2-dideoxy-D-galacto-nojirimycin [DGJNAc], prepared in 20% overall yield from D-glucu- ronolactone, is the first potent competitive sub-micromolar inhibitor of a-N-acetyl-galactosaminidases (K i 0.081 lM from chicken liver, K i 0.136 lM from Charonia lampas). DGJNAc is a good competitive— whereas the enantiomer L-DGJNAc is a very weak but non-competitive—inhibitor of b-hexosaminidases. Ó 2010 Published by Elsevier Ltd. Iminosugars—in which the ring pyranose or furanose oxygen has been replaced by nitrogen—are the archetypes for interaction with carbohydrate processing enzymes. 1 However, among the myriad of sugar mimics reported, there is not a single example of efficient inhi- bition of a-N-acetyl-galactosaminidases (GalNAcases). This Letter reports DGJNAc [2-acetamido-1,2-dideoxy-D-galacto-nojirimycin] 1D as the first potent, specific and competitive inhibitor of GalNA- cases; D-glucuronolactone 2D, a well-established chiron for the syn- thesis of many homochiral targets including amino acids 2 and iminosugars, 3 is the starting material for an efficient synthesis of DGJNAc 1D in an overall yield of 20%. The L-enantiomers of many imi- nosugars have surprising biological activities compared to their D-natural products. 4 The synthesis of L-DGJNAc 1L from the readily available 5 L-glucuronolactone 2L is also reported. The only previous synthesis of 1D starts from 1-deoxynojirimycin 6 and a racemic mix- ture of 1D and 1L has also been prepared; 7 no investigations of the glycosidase inhibitory properties of 1D have hitherto been reported. The synthesis of DGJNAc 1D requires introduction of nitrogen at C5 of the glucuronolactone with inversion of configuration (Scheme 1), epimerization of the hydroxy group at C3 and formation of the piper- idine ring by introduction of nitrogen between C6 and C2 (with inversion of configuration). Selective inhibition of b-hexosaminidases has potential in the study of osteoarthritis, 8 allergy, 9 Alzheimer’s disease, 10 O-GlcNA- case inhibition, 11 cancer metastasis, 12 type II diabetes, 13 genetic diseases such as Tay-Sachs and Sandhoff diseases, 14 and of plant regulation. 15 The synthetic piperidine analogue of N-acetylglucosa- mine DNJNAc 3 16 and its N-alkyl derivatives 17 are potent inhibitors of b-hexosaminidases. The natural product nagstatin 4, 18 with a galacto-configuration, is not reported to inhibit GalNAcases even though it is a potent inhibitor of b-hexosaminidases. 19 The synthetic analogue with a gluco-configuration 5 20 together with PUG derivative 6 21 and GlcNAc–thiazoline 7 22 are very potent inhibitors of b-hexosaminidases. A rare example of a potent pyrrolidine hexosaminidase inhibitor is LABNAc 8; 23 the first pyrr- olizidine b-hexosaminidase inhibitor, pochonicine 9 [or its enan- tiomer], was isolated from a fungal strain Pochonia suchlasporia var. suchlasporia TAMA 87 (Fig. 1). 24 Some seven-membered ring iminosugars also display potent inhibition. 25 In contrast to the diversity of structure of b-hexosaminidase inhibitors, there are no compounds which show significant inhibi- tion of GalNAcases. Studies on reversible binding to exo-GalNA- cases may allow the design of chaperones for the treatment of Schindler–Kanzaki disease. 26 Inhibition of exo-GalNAcases pro- vides a strategy for the treatment of cancer by the protection of macrophage activating factor. 27 endo-GalNAcases may have che- motherapeutic potential in the modification of a number of pathogens. 28 For the synthesis of DJGNAc 1D from D-glucuronolactone 2D, the acetonide 10 5 was esterified with trifluoromethanesulfonic (triflic) anhydride in dichloromethane in the presence of pyridine and the resulting crude triflate was treated with sodium azide in DMF to give the ido-azide 11, mp 112–114 °C; ½a 25 D +261.4 (c 1.0, 0040-4039/$ - see front matter Ó 2010 Published by Elsevier Ltd. doi:10.1016/j.tetlet.2010.02.063 * Corresponding author. E-mail address: george.fleet@chem.ox.ac.uk (G.W.J. Fleet). Tetrahedron Letters 51 (2010) 2222–2224 Contents lists available at ScienceDirect Tetrahedron Letters journal homepage: www.elsevier.com/locate/tetlet