Furanyl cyclic amines: A diastereoselective synthesis of 2,6-syn-disubstituted piperidines under thermodynamic control

Article abstract of DOI:10.1039/c2ob07008a

Utilising the propensity of the 2-furanyl group to facilitate equilibration of an adjacent tosylamide chiral centre, a diastereoselective route to 2,6-syn-piperidines was developed that proceeds with high levels of thermodynamic stereocontrol. X-ray crystallography structures suggest that, as seen in similar systems, pseudo-allylic strain between the N-tosyl group and the substituents at the 2 and 6 positions dominates stereochemical preference, overriding 1,3 diaxial interactions. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2ob07008a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 2392
www.rsc.org/obc
COMMUNICATION
Furanyl cyclic amines: a diastereoselective synthesis of 2,6-syn-disubstituted
piperidines under thermodynamic control†‡
Matthew O’Brien,*a Andrew Leach,a Roly J. Armstrong,a Keting Chonga and Ross Sheridanb
Received 29th November 2011, Accepted 6th January 2012
DOI: 10.1039/c2ob07008a
2,6-disubstituted piperidines.9 Whilst the thermodynamic prefer-
Utilising the propensity of the 2-furanyl group to facilitate
ence with 2,6-disubstituted tetrahydropyrans is for a syn
configuration with the two ring substituents adopting equatorial
positions (presumably to minimise steric hindrance), literature
reports suggest that, with an electron withdrawing substituent on
nitrogen, pseudo-allylic strain can dominate to favour a 2,6-syn-
diaxial arrangement.10
equilibration of an adjacent tosylamide chiral centre, a dia-
stereoselective route to 2,6-syn-piperidines was developed
that proceeds with high levels of thermodynamic stereo-
control. X-ray crystallography structures suggest that, as
seen in similar systems, pseudo-allylic strain between the
N-tosyl group and the substituents at the 2 and 6 positions
dominates stereochemical preference, overriding 1,3 diaxial
interactions.
The general strategy is outlined in Scheme 1. The mixture of
syn and anti piperidines 1 and 2 (which should equilibrate at the
2 position under acidic conditions to the thermodynamically
most stable stereoisomer) would result from the acid catalysed
dehydrative cyclisation of the 1,5-amino alcohol 3. This could
be formed, as a mixture of C-2 epimers, by reducing the furanyl
ketone which would result from displacement of the morpholine
in amide 4 with furanyl lithium.11 If the stereochemically stable
amine chiral centre at C-6 was installed using a Mitsunobu reac-
tion of the alcohol 5, this would lead back to the δ-lactones 6,
several of which, as racemates, are commercially available. The
toluene–sulfonyl group was chosen to protect the nitrogen as it
would facilitate the Mitsunobu reaction12 and also be compatible
with both the acidic conditions required for the cyclisation/equi-
libration and the use of furanyl lithium.
The 2,6-syn-piperidine moiety is seen in numerous natural pro-
ducts of biological and pharmacological significance. Examples
(shown in Fig. 1) include (−)-lobeline,1 piclavine A3,2 cis-sole-
nopsin A3 and (−)-pinidine.4
We have recently utilised the ability of the 2-furyl group to
facilitate acid catalysed epimerisation at an adjacent ether chiral
centre5 in diastereoselective syntheses of doubly anomeric 6,6-
spiroketals6 (and thence 1,9-anti-diols7) and 2,6-syn-disubsti-
tuted tetrahydropyrans.8 We sought to extend this approach to
the synthesis of nitrogen heterocycles, working initially towards
The hydroxyl amide ( ) 5a was synthesised from δ-hexalac-
tone as previously reported using morpholine and aluminium
chloride (Scheme 2). The hydroxyl group was then displaced by
the Boc-Tosyl-amine group using the Mitsunobu protocol with
DIAD and PPh3 to afford the doubly protected amine 7a. Displa-
cement of the morpholine group by furanyl lithium (generated
by reacting the corresponding furan with n-butyllithium in THF)
and the subsequent reduction with sodium borohydride could be
Fig. 1 Some naturally occurring 2,6-syn-piperidines.
aUniversity Chemical Laboratory, University of Cambridge, Lensfield
Road, Cambridge CB2 1EW, UK. E-mail: mo263@cam.ac.uk
bSchool of Chemistry, Trinity College Dublin, College Green, Dublin,
D2, Republic of Ireland
†Dedicated to Professor E. James Thomas on the occasion of his 65th
birthday.
‡Electronic supplementary information (ESI) available: Experimental
procedures and 1H/13C NMR data for the sequence ( )-6a→ ( )−10a.
CCDC reference number 85663. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c2ob07008a
Scheme 1 Retrosynthesis of 2,6-disubstituted piperidines.
2392 | Org. Biomol. Chem., 2012, 10, 2392–2394
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 2 X-ray structure of ( )-(syn)-10a. Thermal ellipsoids are shown
at the 50% probability level.
Scheme 2 Reagents and conditions: (a) morpholine, AlCl3, DCM; (b)
iPrO2CN = NCO2iPr, PPh3, BocNHTs, THF; (c) i. furan, n-BuLi, THF,
ii. NaBH4, MeOH; (d) K2CO3, MeOH, Δ; (e) TsOH·H2O, CDCl3.
Scheme 2, is plausible. However, epimerisation of either of the
C-2 or C-6 chiral centres would lead to the same diastereomer of
product. In order to establish that only the furanyl chiral centre
was epimerising, the synthetic sequence was repeated with an
enantio-enriched starting material.
carried out without isolating the intermediate ketone to afford
the secondary alcohol 8a as an approximately 1 : 1 mixture of
diastereomers. Initial attempts at an acidic ‘one-pot’ Boc depro-
tection and cyclisation resulted in decomposition. Instead, the
Boc group was first cleanly removed using potassium carbonate
in refluxing methanol to afford the tosyl-amino-alcohol 9a
(Scheme 2). Upon treatment with toluenesulfonic acid in deuter-
ated chloroform, the tosyl-amino-alcohol underwent rapid dehy-
drative cyclisation to yield the piperidines 10a. The reaction was
The racemic alcohol ( )-5a was selectively acylated in high ee
with vinyl acetate using Candida Antarctica lipase B, to afford
(S)-5a. This was then taken through the same sequence of reac-
tions shown in Scheme 2, resulting in a furanyl piperidine (6R)-
syn-10a which had a measurable optical rotation. The furan
group was oxidatively cleaved to the carboxylic acid with cataly-
tic ruthenium trichloride and sodium periodate, followed by
borane mediated reduction to the alcohol 11 (Scheme 3). This
was then coupled to (R)- and (S)-acetoxymandelic acids to afford
the esters 12 and 13. Pleasingly, these were both obtained as
1
most conveniently monitored by H NMR spectroscopy. Loss of
starting material was essentially complete after around 2 min at
room temperature. The initial ratio of product diastereomers at
this time was found to be around 1 : 3. Equilibration was slower
than the initial cyclisation, taking around 48 h at room tempera-
ture. The final thermodynamic diastereomeric ratio was 15 : 1
(the thermodynamically favoured product was the least favoured
kinetically). The major product was isolated by direct column
chromatography on silica gel, avoiding aqueous workup.
Whilst NMR spectroscopy had assisted in determining that
10a was formed with high levels of diastereoselectivity, it did
not readily reveal what the stereochemistry actually was.
However, as the major isomer which formed was crystalline, it
was possible to obtain a crystal structure by X-ray diffraction
which provided the relative stereochemistry (shown in Fig. 2).
As can be seen, the piperidine is 2,6-syn-diaxially substituted.
Presumably due to the lone pair delocalising into the electron
withdrawing tosyl group, the nitrogen is more or less planar. The
other substituents seemingly prefer a 2,6-syn configuration
which allows them to adopt a diaxial conformation, thereby
minimising steric clashing with the bulky tosyl group (pseudo-
allylic strain).
1
single diastereomers (whose H NMR spectra had several non-
overlapping peaks) thus establishing the homochirality of the
Scheme 3 Reagents and conditions: (a) i. RuCl3 (cat.), NaIO4, H2O,
DCM, MeCN, ii. BH3·SMe2, THF (b) (R)-acetoxymandelic acid, EDCI,
DMAP, DCM; (c) (S)-acetoxymandelic acid, EDCI, DMAP, DCM.
Given the electron rich nature of the furan group, a mechanism
involving formation of a furanyl cation at C-2, as shown in
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 2392–2394 | 2393

View Article Online
Table 1 Results of diastereoselective piperidine formation
under acidic conditions with thermodynamic control. It makes
use of the electron rich nature of the 2-furyl group and its ability
to facilitate epimerisation of an adjacent tosylamine chiral centre.
An X-ray crystal structure obtained from one of the products
reveals a preference for 2,6-syn-diaxial substitution whereby the
substituents can avoid an apparent steric clash with the bulky
tosyl group on nitrogen (pseudo-allylic strain). Work is under-
way to further investigate the scope of the reaction, including
ring size, and the results will be published in due course.
Entry SM
R
R′ Prod.
Yielda (%) drb
1
2
3
4
5
6
7
8
9
( )-9a Me
H
H
H
H
H
( )-syn-10a
( )-syn-10b
( )-syn-10c
(6S)-syn-10d 91
(6R)-syn-10e 88
80
89
91
15 : 1
( )-9b n-C6H11
( )-9c n-C9H17
(S)-9d CH2OH
(R)-9e (CH2)3OH
(R)-9a′ Me
( )-9b′ n-C6H11
( )-9c′ n-C9H17
(S)-9d′ CH2OH
25 : 1
40 : 1
18 : 1
20 : 1c
13 : 1
25 : 1
20 : 1
29 : 1
28 : 1c
Acknowledgements
Me (R)-syn-10a′ 93
Me ( )-syn-10b′ 87
Me ( )-syn-10c′
We thank the following for funding: Kinerton/Ipsen, EPSRC
(MO’B). We thank Professor Steven V. Ley for supporting this
work. We thank the EPSRC for financial assistance towards the
purchase of the Nonius-Kappa X-ray diffractometer. We thank
Dr John E. Davies for performing the X-ray crystallography
work.
89
Me (6S)-syn-10d′ 93
10
(R)-9e′ (CH2)3OH Me (6R)-syn-10e′ 85
a Isolated yield after chromatography on silica gel. b As determined from
1H NMR spectra of reaction mixture. c The starting material (which was
derived from L-pyroglutamic acid) and product have the opposite
configuration to that shown.
Notes and references
1 F. X. Felpin and J. Lebreton, Tetrahedron, 2004, 60, 10127–10153.
2 M. F. Raub, J. H. Cardellina and T. F. Spande, Tetrahedron Lett., 1992,
33, 2257–2260.
3 S. Leclercq, J. C. Braekman, D. Daloze, J. M. Pasteels and R. K. Van der
Meer, Naturwissenschaften, 1996, 83, 222–225.
4 W. H. Tallent and E. C. Horning, J. Am. Chem. Soc., 1956, 78, 4467–
4469.
5 S. J. Danishefsky, S. DeNinno and P. Lartey, J. Am. Chem. Soc., 1987,
109, 2082–2089; I. Maeba, K. Iwata, F. Usami and H. Furukawa, J. Org.
Chem., 1983, 48, 2998–3002.
6 S. Cahill and M. O’Brien, Tetrahedron Lett., 2006, 47, 3665–3668.
7 S. Cahill, L. A. Evans and M. O’Brien, Tetrahedron Lett., 2007, 48,
5683–5686.
8 M. O’Brien, S. Cahill and L. A. Evans, Chem. Commun., 2008, 5559–
5561.
9 (a) P. R. Girling, T. Kiyoi and A. Whiting, Org. Biomol. Chem., 2011, 9,
3105–3121; (b) J. P. A. Harrity and O. Provoost, Org. Biomol. Chem.,
2005, 3, 1349–1358; (c) M. G. P. Buffat, Tetrahedron, 2004, 60, 1701–
1729; (d) P. M. Weintraub, J. S. Sabol, J. M. Kane and
D. R. Borcherding, Tetrahedron, 2003, 59, 2953–2989; (e) F.-X. Felpin
and J. Lebreton, Eur. J. Org. Chem., 2003 (19), 3693–3712;
(f) S. Laschat and T. Dickner, Synthesis, 2000 (13), 1781–1813;
(g) V. Baliah, R. Jeyaraman and L. Chandrasekaran, Chem. Rev., 1983,
83, 379–423.
10 (a) Y. L. Chow, C. J. Colon and J. N. S. Tam, Can. J. Chem., 1968, 46,
2821; (b) R. R. Fraser and T. B. Grindley, Tetrahedron Lett., 1974, 15
(47), 4169–4172; (c) J. Quick, C. Mondello, M. Humora and T. Brennan,
J. Org. Chem., 1978, 43, 2705–2708; (d) P. Beak and W. K. Lee, J. Org.
Chem., 1993, 58, 1109–1117.
piperidine (6R)-syn-10a (and indirectly establishing that both the
enzymatic resolution and Mitsunobu reaction proceeded with
complete stereoselectivity). As the tosyl-amino-alcohol (R)-8a
was stereodefined at C-6 and a mixture at C-2, it must be the C-2
centre that underwent epimerisation. Stereochemical scrambling
at both centres would result in a racemic product.
Having established both the relative and absolute stereo-
chemistry of the piperidine product (6R)-syn-10a, a range of 1,5-
tosylamino-alcohols were cyclised under the same conditions.
The results are shown in Table 1. The starting materials were
diastereomeric mixtures (approximately 1 : 1 in all cases). The
yields were generally high and the diastereoselectivities excel-
lent. Hexyl and nonyl derivatives 9b, 9b′, 9c and 9c′ were syn-
thesised from the corresponding δ-lactones according to
Scheme 2 and used as racemates. The hydroxymethyl and hydro-
xypropyl precursors 9d, 9d′, 9e and 9e′ were synthesised from
D-serine and L-pyroglutamic acid respectively. In these cases the
stereo-random furanyl alcohol chiral centre resulted from
addition of lithiated furan (or methyl-furan) to the corresponding
aldehyde.
Conclusions
11 R. Martin, P. Romea, C. Tey, F. Urpi and J. Vilarrasa, Synlett, 1997, 12
(12), 1414–1416.
12 J. R. Henry, L. R. Marcin, M. C. McIntosh, P. M. Scola, G. Davis Harris
Jr and S. M. Weinreb, Tetrahedron Lett., 1989, 30, 5709–5712.
In summary, a highly diastereoselective synthesis of 2,6-syn-
disubstituted piperidines has been developed which proceeds
2394 | Org. Biomol. Chem., 2012, 10, 2392–2394
This journal is © The Royal Society of Chemistry 2012