A new asymmetric route to substituted piperidines: Synthesis of N-alkyl- 3,4-dihydroxy-5-alkylpiperidines

Article abstract of DOI:10.1016/S0040-4039(00)00267-7

A facile asymmetric synthesis of N-alkyl-3,4-dihydroxy-5- alkylpiperidines has been developed, which is based on an asymmetric alkylation reaction of the SAMP hydrazone of N-BOC piperidinone. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00267-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 2801–2804
A new asymmetric route to substituted piperidines: synthesis of
N-alkyl-3,4-dihydroxy-5-alkylpiperidines
Hongbin Sun, Kim M. Millar, Jingsong Yang, Khalil Abboud and Benjamin A. Horenstein ∗
Department of Chemistry, University of Florida, Gainesville, FL 32611-7200, USA
Received 12 January 2000; revised 8 February 2000; accepted 9 February 2000
Abstract
A facile asymmetric synthesis of N-alkyl-3,4-dihydroxy-5-alkylpiperidines has been developed, which is based
on an asymmetric alkylation reaction of the SAMP hydrazone of N-BOC piperidinone. © 2000 Elsevier Science
Ltd. All rights reserved.
Keywords: piperidinone; alkylation; asymmetric synthesis; hydrazones.
Substituted piperidines are common structures in natural and synthetic compounds with an extensive
range of biological activities. Among them, polyhydroxylated piperidines have recently received much
attention due to their utility as glycosidase inhibitors.1 For example, isofagomine 1 is a potent β-
glucosidase inhibitor.2 The 3,4,5-trisubstituted piperidine 2 was found to be a potent α-fucosidase
inhibitor with a Ki of 6.4 µM.3 Glucuronic acid analogue 3 has a Ki versus β-glucuronidase of 79 nM.4 In
the course of our study on the development of electronic analogues of the oxocarbenium ion derived from
N-acetylneuraminic acid,5 we became interested in developing a synthesis of 1,3,4,5-tetrasubstituted
piperidines 4 and 5 (Scheme 1). Although a number of asymmetric routes to substituted piperidines
have been developed,6 asymmetric synthesis of 3,4,5-trisubstituted piperidines is still a challenge since
carbohydrate-based approaches can vary in complexity.2,4,7 The target compounds 4 and 5 contain
branched propyl or benzyl C-chains which are not commonly found in carbohydrate building blocks nor
would they be readily installed, so we considered a non-carbohydrate approach. This report describes a
chiral–hydrazone based asymmetric synthesis of 1,3,4,5-tetrasubstituted piperidines 4 and 5.
As depicted in Fig. 1, the chiral auxiliary was introduced by condensation of the commercially
available ketone 6 with the Enders hydrazine (SAMP) to give the hydrazone 7.8 The lithium aza-
enolate of hydrazone 7 was prepared with n-butyllithium in the presence of N,N0-dimethylpropyleneurea
(DMPU) in THF at −78°C, and was allowed to react with iodopropane or benzyl bromide at −78°C to
afford hydrazone 8, which was unstable on silica gel and was used without any purification for the next
step. Cleavage of hydrazone 8 was satisfactorily carried out in a two-phase system containing a saturated
∗
Corresponding author. Tel: (352) 392-9859; fax: (352) 846-2095; e-mail: horen@chem.ufl.edu (B. A. Horenstein)
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00267-7
tetl 16547

2802
Scheme 1.
aqueous oxalic acid solution and diethyl ether to give ketone 9 in high yield (9a: 90% for two steps;
9b: 87% for two steps).9 The configuration of the ketone 9 was assigned to be R-based on the well-
established stereochemical model for alkylation via SAMP/RAMP hydrazones.8 Kinetic deprotonation
of ketone 9 was performed with lithium diisopropylamide (LDA) in THF at −78°C, and reaction of
the resulting lithium enolate with TMSCl gave silyl enol ether 10. The hydroboration–oxidation of
silyl enol ether 1010 and then desilylation with tetrabutylammonium fluoride (TBAF) gave a mixture
of trans-diols 11 and 12 (48% from 9a, 11a:12a≈1:1 by NMR; 43% from 9b, 11b:12b≈1:1.3 by NMR)
together with monoalcohol 13 as the by-product. Diols 11 and 12 could be separated by careful column
chromatography. Alternatively, we used a mixture of diols 11 and 12 for the next reactions without
further purification, and very easily carried out the purification procedure by column chromatography
of the benzyl esters 15 and 17 (Fig. 2). The relative stereochemistry of diols 11 and 12 was determined
by NMR spectroscopy and the structure of 12a had been unambiguously determined by single-crystal
X-ray analysis of its corresponding derivative 17a. Deprotection of 11 with trifluoroacetic acid (TFA),
and then N-alkylation with benzyl 2-bromoacetate, gave benzyl esters 15 (64% from 11a; 55% from
11b). In the same way, benzyl esters 17 were prepared (61% from 12a; 49% from 12b). Compound
17a gave fine crystals which allowed us to confirm its relative configuration by X-ray analysis. The
enantiomeric excesses (ees) of compounds 15a and 15b were assigned to be 78 and 63%, respectively by
chiral shift experiments with Eu(tfc)3 as the chiral shift reagent. Debenzylation of 15 and 17 by catalytic
hydrogenolysis over palladium on carbon was a clean reaction, affording the target compounds 4 and 5
in nearly quantitative yield, respectively.11 Considering that the series of reactions from silyl enol ether
Fig. 1. Reagents and conditions: (a) SAMP, toluene, MgSO4, 90%; (b) (i) BuLi, THF–DMPU, −78°C, 2 h; (ii) 1-iodopropane
or benzyl bromide, −78°C–rt; (c) sat. oxalic acid, ethyl ether, rt; (d) (i) LDA, THF, −78°C; (ii) TMSCl, −78°C–rt; (e) (i)
BH3·SMe2, THF, 0°–rt; (ii) H2O2, NaOH; (iii) TBAF, THF, rt

2803
10a to compounds 4a and 5a should be racemization free, we estimated that the enantiomeric excess of
compounds 4a/5a was 78% and of compounds 4b/5b was 63%. Further, ees for alkylated piperidinones
9a and 9b must have been at least 78 and 63%, respectively.12
Fig. 2. Reagents and conditions: (a) TFA, CH2Cl2; (b) BrCH2CO2Bn, K2CO3, n-Bu4NI (cat.), DMF, rt; (c) H2, Pd/C, EtOAc
In conclusion, we have described a new asymmetric route to 1,3,4,5-tetrasubstituted piperidines which
allows for flexible 5-alkyl-substitution with fair to good enantiomeric excesses. The availability of this
new entry to polysubstituted piperidines should assist in the further development of piperidine derivatives
as a biologically important class of compounds. Further application of this methodology to the synthesis
of sialyltransferase and other glycosyltransferase inhibitors is underway in our laboratory.
Acknowledgements
We wish to thank the University of Florida and National Science Foundation (CAREER Award MCB-
9501866 to B.A.H.) for support of this work.
References
1. (a) Winchester, B.; Fleet, G.W. J. Glycobiology 1992, 2, 199. (b) Legler, G. Adv. Carbohydr. Chem. Biochem. 1990, 48, 319.
2. Jespersen, T. M.; Dong, W.; Skrydstrup, T.; Sierks, M. R.; Lundt, I.; Bols, M. Angew. Chem. 1994, 33, 1778.
3. Hansen, A.; Tagmose, T. M.; Bols, M. J. Chem. Soc., Chem. Commun. 1996, 2649.
4. Bols, M. Acc. Chem. Res. 1998, 31, 1.
5. Parr, I. B.; Horenstein. B. A. J. Org. Chem. 1997, 62, 7489.
6. Bailey, P. D.; Millwood, P. A.; Smith, P. D. J. Chem. Soc., Chem. Commun. 1998, 633, and references cited therein.
7. Ichikawa, Y.; Igarashi, Y.; Ichikawa, M.; Suhara, Y. J. Am. Chem. Soc. 1998, 120, 3007.
8. Enders, D. In Asymmetric Synthesis; Morrison, J. D., Ed.; Academic Press: London, 1984; Vol. 3, Part B, pp. 275–339.
9. Enders, D.; Hunertmark, T.; Lazny, R. Synlett. 1998, 721.
10. Klein, J.; Levene, R.; Dunkelblum, E. Tetrahedron Lett. 1972, 28, 2845.
11. Selected data for 15a: 1H NMR (300 MHz, CDCl3) δ 7.31 (m, 5H), 5.14 (d, J=2.1 Hz, 2H), 3.4–3.8 (m, 3H), 3.3 (br s, 2H),
2.9–3.1 (m, 3H), 2.17 (t, J=10.5 Hz, 1H), 1.96 (t, J=11.1 Hz, 1H), 1.62–1.76 (m, 2H), 1.03–1.37 (m, 4H), 0.89 (t, J=6.9 Hz,
3H); 13C NMR (75 MHz, CDCl3) δ 170.14, 135.49, 128.54, 128.35, 78.18, 72.13, 66.38, 58.64, 57.85, 56.81, 40.79, 31.79,
1
19.99, 14.33; HRMS calcd for C17H26NO4 (M++H): 308.1862; found: 308.1861. For 4a: H NMR (300 MHz, d-DMSO)

2804
δ 3.28–3.36 (m, 1H), 3.14 (d, J=2.4 Hz, 2H), 2.93–3.00 (m, 2H), 2.75 (t, J=9.6 Hz, 1H), 2.12 (t, J=10.8 Hz, 1H), 2.01 (t,
J=11.4 Hz, 1H), 1.65 (m, 1H), 1.46 (m, 1H), 1.33 (m, 1H), 1.16 (m, 1H), 1.01 (m, 1H), 0.85 (t, J=7.2 Hz, 3H); 13C NMR
(75 MHz, d-DMSO) δ 170.19, 76.58, 70.75, 58.70, 57.49, 55.98, 39.50, 31.46, 19.57, 14.34; HRMS calcd for C10H20NO4
(M++H): 218.1392; found: 218.1392.
12. We tried to use various chiral shift reagents to directly determine the enantiomeric purity of 9a, 9b, 17a and 17b, but were
unable to obtain satisfactory results.