Stereospecific construction of substituted piperidines. Synthesis of (-)-paroxetine and (+)-laccarin

Article abstract of DOI:10.1039/b617260a

Short and efficient enantioselective syntheses of (-)-paroxetine and (+)-laccarin are described based on the highly stereospecific cleavage of C(3)-substituted 1,3-cyclic sulfamidates. The Royal Society of Chemistry.

Full text of DOI:10.1039/b617260a

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Stereospecific construction of substituted piperidines. Synthesis of (2)-
paroxetine and (+)-laccarin{
John F. Bower,a Thomas Riis-Johannessen,b Peter Szeto,c Andrew J. Whiteheadc and Timothy Gallagher*a
Received (in Cambridge, UK) 27th November 2006, Accepted 3rd January 2007
First published as an Advance Article on the web 18th January 2007
DOI: 10.1039/b617260a
Short and efficient enantioselective syntheses of (2)-paroxetine
and (+)-laccarin are described based on the highly stereospecific
cleavage of C(3)-substituted 1,3-cyclic sulfamidates.
Piperidines represent a class of heterocycles frequently associated
with biologically active natural products and often embedded
within scaffolds recognised as privileged by medicinal chemists.1
Accordingly, new entries to substituted piperidines are of wide-
spread interest and we have recently outlined methodologies based
on the reactivity of 1,2- and 1,3-cyclic sulfamidates towards
nucleophiles that provide a versatile approach to a range of
enantiomerically pure N-heterocycles, including piperidines and
piperidinones.2 A key feature of this is the high reactivity exhibited
Scheme 1 Structural relationships between 1,3-cyclic sulfamidates and
by the C(3)–O bond towards nucleophiles, and in this sense 1,2-
3,4-disubstituted piperidine scaffolds.
and 1,3-cyclic sulfamidates offer highly activated synthetic
equivalents of aziridines and azetidines respectively.3
alcohol 7, and two-step cyclic sulfamidate formation proceeded
Stereocontrol during nucleophilic displacement is an important
proviso in this regard, and the focus of this paper is the
stereospecific (i.e. clean SN2) cleavage of C(3)-substituted 1,3-
cyclic sulfamidates with synthetically useful C-nucleophiles.4 Here
we report on the reactivity of C(3)-aryl and C(3)-alkyl cyclic
sulfamidates 1 and 3 respectively towards enolate nucleophiles,
and the resulting reactivity profiles (efficient inversion at C(3))
provide concise, asymmetric entries to two biologically and
structurally interesting 3,4-disubstituted piperidines, (2)-paroxe-
tine 2 and (+)-laccarin 4 (Scheme 1).
smoothly to give the key intermediate 1 in excellent (81%) overall
yield from 6. The key step in this sequence involved the reaction of
1 with the sodium enolate of dimethyl malonate. Following
nucleophilic displacement, the crude product was immediately
subjected to mild acid hydrolysis (to cleave the intermediate
N-sulfate) and then thermolysis (to achieve ring closure) which
The therapeutically important antidepressant (2)-paroxetine 25
is representative of an emerging group of pharmacologically active
4-aryl piperidines and this target has received extensive attention
and a number of synthetic approaches to both racemic and
enantiomerically pure 2 have been described.6
Our approach to (2)-paroxetine is shown in Scheme 2, with the
key feature being the intermediacy of the C(3)-aryl substituted 1,3-
cyclic sulfamidate 1 and the high degree of stereochemical control
exercised during the reaction of 1 with the enolate of dimethyl
malonate. Commercially available keto ester 5 was reduced using a
[Ru][Cl-MeO-BIPHEP] catalyst system7 to give alcohol 6 in 95%
yield and 97% ee (as assessed by chiral GC).{ Aluminium-
mediated amidation, followed by amide reduction gave amino
Scheme 2 Reagents and conditions: i, [((S)-Cl-MeO-BIPHEP)Ru-
(cymene)Cl]Cl?CH2Cl2 (0.5 mol%), H2 (8 bar), MeOH, 60 uC (95%); ii,
AlMe3, BnNH2, PhMe, 0 uC to rt (100%); iii, LiAlH4, THF, reflux (98%);
iv, SOCl2, Et3N, imidazole, CH2Cl2, 220 uC to 0 uC (95%); v, RuCl3
(0.25 mol%), NaIO4, MeCN–H2O, 0 uC (87%); vi, dimethyl malonate,
NaH, DMF, 60 uC; then 5 M HCl; then PhMe, reflux (70%); vii, LiAlH4,
THF, reflux; viii, MsCl, Et3N, CH2Cl2; ix, sesamol, NaH, DMF, 90 uC
(52% from 8); x, 10% Pd/C (35%), H2 (6 bar), i-PrOH, AcOH, 50 uC, then
aq. HCl, i-PrOH (82%).
aSchool of Chemistry, University of Bristol, Bristol, UK BS8 1TS.
E-mail: t.gallagher@bristol.ac.uk; Fax: +44 117 9298611;
Tel: +44 117 9288260
bStructural Chemistry, School of Chemistry, University of Bristol,
Bristol, UK BS8 1TS
cChemical Development, GlaxoSmithKline, Medicines Research Centre,
Stevenage, UK SG1 2NY
{ Electronic supplementary information (ESI) available: Full experimental
details and spectroscopic data. See DOI: 10.1039/b617260a
728 | Chem. Commun., 2007, 728–730
This journal is ß The Royal Society of Chemistry 2007

View Online
gave lactam 8 as essentially a single diastereomer (.95 : 5 dr as
1
determined from the H NMR spectrum of the crude reaction
this protection step is discussed below, but this was done to inhibit
premature lactam formation.
product) in 70% overall yield. Analysis of 8 by chiral HPLC (using
the corresponding racemate as a standard) showed that no
detectable erosion of enantiomeric purity had occurred during
nucleophilic ring cleavage.{ This demonstrates that the aryl-
substituted C(3) stereocentre of 1 undergoes highly efficient
inversion with malonate, which was confirmed by subsequent
conversion of 8 to (2)-paroxetine 1. The latter involved global
reduction followed by transformation to 9 in 52% yield over
3 steps. N-Debenzylation of 9 then gave (2)-paroxetine 1 which
was isolated and characterised as the HCl salt. This material was
correlated with the data reported for (2)-paroxetine, ([a]2D0 278.1
(c 1.0, MeOH); lit.6c [a]D20 280.8 (c 1.3, MeOH), lit.6l [a]2D0 286.5
(c 1.0, MeOH)). The overall sequence from 5 to (2)-2 (97% ee)
proceeds in 24.1% overall yield over ten synthetic operations.
(+)-Laccarin 4 was the second target chosen for exploring this
critical stereochemical aspect of cyclic sulfamidate-based meth-
odology. Laccarin is a fungal metabolite which was first isolated
and characterised from Laccaria vinaceoavellanea in 1996 by
Ohta, Nozoe et al.8a and is reported to have phosphodiesterase
inhibitory activity. Recently Yue and co-workers8b also isolated
laccarin, together with an unstable N-acylated variant laccarin A,
from Lactarius subplinthogalus. Laccarin is a densely functiona-
lised alkaloid based on a piperidine core, but the synthesis of
laccarin has yet to be reported. Here we report the first total (and
asymmetric) synthesis of laccarin, based on a stereochemically
efficient nucleophilic cleavage step involving a C(3)-alkyl
substituted cyclic sulfamidate 3. This serves to confirm the
assignment of the relative stereochemistry between C(4) and C(5)
of 4 and we are also able to assign for the first time the absolute
configuration of (+)-laccarin.
The functional array associated with laccarin demanded
amination at C(2) of 13, which was achieved using monochlor-
amine9§ under basic conditions to give 14 in 75% isolated yield
(and 98% yield based on recovered 13). N-Acylation of 14 using
diketene gave 15, which was not isolated but cyclised under Claisen
conditions (to give 16) followed by ester hydrolysis and
decarboxylation using aqueous acid. N-Boc cleavage from 17
(using TFA) then neutralisation and thermolysis gave N-benzyl
laccarin 18a and the cis diastereomer 18b as a 7 : 2 mixture" in
68% overall yield from 14; intermediates 15–17 were not isolated
during this sequence.I The structure of 18a was based on
detailed spectroscopic analysis and X-ray crystallography.**
Hydrogenation of N-benzyl laccarin 18a in acetic acid was very
slow, but in the presence of TFA, debenzylation was rapid,
efficient and chemoselective leading to (+)-laccarin 4 [mp . 250 uC;
[a]D20 +176.9 (c 0.5, CHCl3); lit.8a [a]D31 +188.0 (c 0.5, CHCl3)] in
93% yield.{{ Correlation of the identity of synthetic laccarin with
the natural product itself was achieved by direct comparison
(HPLC, 1H and 13C NMR, CD and optical rotation; for full
details see ESI{) using an authentic sample of the natural product.
Based on this, we assign the absolute configuration of natural
laccarin as shown in 4. This represents the first synthesis of this
natural product and the route outlined in Scheme 3 proceeds in
18.4% overall yield. This route is efficient and is convergent in
terms of the functionality that can be incorporated within the
bicyclic framework, which opens the way to explore more fully the
biological profile of this interesting natural product.
One other aspect of our strategy to laccarin merits comment.
The initial malonate adduct 12 underwent facile base-mediated
lactamisation to give 19 (13 : 2 dr, major diastereomer shown) in
52% yield from 3.{{ The enantiomeric purity of 19 was assessed by
chiral HPLC as .98% ee{ and this determination provided the
basis for our conclusions concerning the enantiomeric purity of 13
and ultimately (+)-laccarin 4. Amination (using monochloramine)
of 19, followed by decarboxylation and acylation (with diketene)
produced 20 (2 : 1 dr), but despite extensive efforts we were unable
to convert 20 to N-benzyl laccarin 18a (or 18b) (Scheme 4).
Commercially available ethyl (3R)-hydroxybutyrate 10 (99% ee)
was converted to N-benzyl amide 11, followed by reduction and
conversion of the resulting amino alcohol to the target cyclic
sulfamidate 3 in 62% overall yield. The sodium enolate of diethyl
malonate reacted efficiently with cyclic sulfamidate 3 and,
following ring cleavage (and N-sulfate cleavage), the resulting
adduct 12 was protected as its N-Boc derivative 13. The reason for
Scheme 3 Reagents and conditions: i, AlMe3, BnNH2, PhMe, 0 uC to rt (88%); ii, LiAlH4, THF, reflux (98%); iii, SOCl2, Et3N, imidazole, CH2Cl2, 220 uC
to 0 uC (84%); iv, RuCl3 (0.25 mol%), NaIO4, MeCN–H2O, 0 uC (85%); v, diethyl malonate, NaH, DMF, 110 uC; then 5 M HCl; vi, Boc2O, NaHCO3,
MeCN (81% from 3); vii, NH2Cl (ca. 0.15 M in Et2O), t-BuOK, THF, 0 uC to rt (75% + 23% recovered 13); viii, diketene, cat. DMAP, THF, 60 uC; ix,
NaOEt, EtOH, 60 uC; x, H2O, then 5 M HCl; xi, TFA, CH2Cl2 then Et3N; xii, PhMe, 80 uC (68% of 18a and 18b from 14); xiii, 10% Pd/C (65 wt%), H2
(5.5 bar), TFA (93%).
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 728–730 | 729

View Online
1 For recent reviews on the synthesis of piperidines, see: (a) S. Laschat and
T. Dickner, Synthesis, 2000, 1781–1813; (b) P. M. Weintraub, J. S. Sabol,
J. M. Kane and D. R. Borcherding, Tetrahedron, 2003, 59, 2953–2989;
(c) M. G. P. Buffat, Tetrahedron, 2004, 60, 1701–1729.
2 (a) A. J. Williams, S. Chakthong, D. Gray, R. M. Lawrence and
ˇ
T. Gallagher, Org. Lett., 2003, 5, 811–814; (b) J. F. Bower, J. Svenda,
A. J. Williams, J. P. H. Charmant, R. M. Lawrence, P. Szeto and
T. Gallagher, Org. Lett., 2004, 6, 4727–4730; (c) J. F. Bower, P. Szeto
and T. Gallagher, Chem. Commun., 2005, 5793–5795; (d) J. F. Bower,
ˇ
S. Chakthong, J. Svenda, A. J. Williams, R. M. Lawrence, P. Szeto and
T. Gallagher, Org. Biomol. Chem., 2006, 4, 1868–1877.
3 For a comprehensive review on the chemistry of cyclic sulfamidates,
including the use of stabilised enolates as nucleophiles, see:
R. E. Mele´ndez and W. D. Lubell, Tetrahedron, 2003, 59, 2581–2616.
4 Diastereoselective cleavages involving 1,3-cyclic sulfamidates with
heteroatom nucleophiles have been reported, but these are subject to
internal stereocontrol mechanisms: (a) C. G. Espino, P. M. Wehn,
J. Chow and J. Du Bois, J. Am. Chem. Soc., 2001, 123, 6935–6936; (b)
P. M. Wehn and J. Du Bois, J. Am. Chem. Soc., 2002, 124,
12950–12951.
5 A number of structurally related and pharmacologically relevant
piperidine derivatives are also known, including femoxetine and a series
of renin inhibitors: (a) J. B. Lassen, B. K. Skrumsager and
J. A. Christensen, US Patent 4442113, 1984; (b) C. Oefner,
A. Binggeli, V. Breu, D. Bur, J.-P. Clozel, A. D’Arcy, A. Dorn,
W. Fischli, F. Gru¨ninger, R. Gu¨ller, G. Hirth, H. P. Ma¨rki, S. Mathews,
M. Mu¨ller, R. G. Ridley, H. Stadler, E. Vieira, M. Wilhelm,
F. K. Winkler and W. Wostl, Chem. Biol., 1999, 6, 127–131; (c)
M. G. Bursavich, C. W. West and D. H. Rich, Org. Lett., 2001, 3,
2317–2320.
6 For leading references to earlier asymmetric syntheses of (2)-paroxetine,
see: (a) R. D. B. Barnes, M. W. Wood-Kaczmar, I. R. B. Lynch,
P. C. B. Buxton and A. D. B. Curzons, European Patent 0223403, 1986;
(b) M. S. Yu, I. Lantos, Z.-Q. Peng, J. Yu and T. Cacchio, Tetrahedron
Lett., 2000, 41, 5647–5651; (c) M. Amat, J. Bosch, J. Hidalgo, M. Canto´,
M. Pe´rez, N. Llor, E. Molins, C. Miravitlles, M. Orozco and J. Luque,
J. Org. Chem., 2000, 65, 3074–3084; (d) J. Cossy, O. Mirguet,
D. G. Pardo and J.-R. Desmurs, Tetrahedron Lett., 2001, 42,
5705–5707; (e) T. A. Johnson, M. D. Curtis and P. Beak, J. Am.
Chem. Soc., 2001, 123, 1004–1005; (f) L. T. Liu, P.-C. Hong, H.-L.
Huang, S.-F. Chen, C.-L. J. Wang and Y.-S. Wen, Tetrahedron:
Asymmetry, 2001, 12, 419–426; (g) D. A. Greenhalgh and N. Simpkins,
Synlett, 2002, 2074–2076; (h) G. de Gonzalo, R. Brieva, V. M. Sa´nchez,
M. Bayod and V. Gotor, J. Org. Chem., 2003, 68, 3333–3336; (i)
K. S. K. Murthy, A. W. Rey and M. Tjepkema, Tetrahedron Lett.,
2003, 44, 5355–5358; (j) M. S. Taylor and E. N. Jacobsen, J. Am. Chem.
Soc., 2003, 125, 11204–11205; (k) G. Hughes, M. Kimura and
S. L. Buchwald, J. Am. Chem. Soc., 2003, 125, 11253–11258; (l)
L. Czibula, A. Nemes, F. Sebo¨k, C. Sza´ntay and M. Ma´k, Eur. J. Org.
Chem., 2004, 3336–3339; (m) S. Yamada and I. Jahan, Tetrahedron
Lett., 2005, 46, 8673–8676; (n) S. Brandau, A. Landa, J. Franze´n,
M. Marigo and K. A. Jørgensen, Angew. Chem., Int. Ed., 2006, 45,
4305–4309; (o) P. K. Koech and M. J. Krische, Tetrahedron, 2006, 62,
10594–10602. For a synthesis of (+)-paroxetine, see: (p) M. Amat,
J. Hidalgo and J. Bosch, Tetrahedron: Asymmetry, 1996, 7, 1591–1594.
7 (a) C. Laue, G. Schroder and D. Arlt, US Patent 5801261, 1998; (b)
C. Laue, G. Schroder and D. Arlt, US Patent 5710339, 1998. The
reduction of 5 (in 89% ee) using a Ru[BINAP] catalyst system has
previously been reported ; (c) V. Ratovelomanana-Vidal, C. Girard,
R. Touati, J.-P. Tranchier, B. Ben Hassine and J.-P. Geneˆt, Adv. Synth.
Catal., 2003, 345, 261–274.
Scheme 4 Reagents and conditions: i, NaOEt, EtOH, reflux (52% from
3); ii, NH2Cl (ca. 0.15 M in Et2O), NaH, THF, 0 uC to rt (63% ¡ 30%
recovered 19); iii, KOH, dioxane, H2O, reflux, then 5 M HCl (92%); iv,
diketene, Et3N, THF (86%).
In summary, we have demonstrated that C(3)-aryl and -alkyl
substituted 1,3-cyclic sulfamidates undergo stereochemically and
chemically efficient nucleophilic ring cleavage with stabilised
enolates to provide the basis of an effective and stereocontrolled
entry to substituted piperidines. This represents an important
advance that significantly enhances the utility of cyclic sulfami-
dates as effective, synthetically useful electrophiles. The scope of
this methodology has been demonstrated using two biologically
active heterocyclic targets. In the laccarin case, this serves an
additional purpose of establishing, for the first time, the absolute
stereochemistry of this natural product, based on the use of (3R)-
10 as the starting point for the synthetic scheme.
We acknowledge the EPSRC and GSK for financial support,
and thank Professor J.-M. Yue for a sample of natural laccarin.
Notes and references
{ Chiral GC and HPLC were used extensively and in all cases analyses
were performed using the corresponding racemate as a standard (see ESI{).
§ The use of a glycine enolate (rather than malonate) as the nucleophilic
component would provide a direct entry to amino esters such as 14,
obviating the need to carry out a separate amination step. An extensive
range of glycine enolates (or equivalents) were examined and, while we
have achieved the synthesis of a-amino lactams using the Stork glycine-
derived imine,10 these transformations were neither efficient nor tolerant of
sterically demanding cyclic sulfamidates such as 3.
" We have compared our spectroscopic data for laccarin 4 to that reported
earlier by both Ohta, Nozoe et al.8a and Yue et al.8b Using the numbering
system shown on 4 in Scheme 3, the coupling constants observed for the
3
N-benzyl diastereomers 18a and 18b are diagnostic: 18a JH(5)–H(4)
=
10.5 Hz; 18b 3JH(5)–H(4) = 7 Hz. Laccarin shows JH(5)–H(4) = 10.5 Hz.
3
I The intermediacy of 16 and 17 is assumed. These intermediates could be
isolated but were difficult to analyse because of the presence of
diastereomers/enol tautomers/rotamers. Accordingly, we are unable to
determine the isomer ratio associated with 17.
** Full details of the crystal data for 18a are available in the ESI.{ CCDC
621099.ForcrystallographicdatainCIFformatseeDOI:10.1039/b617260a
{{ Debenzylation of 18a using acetic acid as solvent gave 4 in 27% yield
after 3 days. Using 20% Pd(OH)2 on carbon as catalyst in acetic acid (H2,
5.5 bar, 12 h) gave 4 in 52% yield. Attempted debenzylation of 18a using
dissolving metal conditions (Na–liquid ammonia) was unsuccessful and
resulted in reduction of the enamine (C(6)/C(7)) double bond. An approach
based on benzylic bromination11 resulted in decomposition. Interestingly,
to date, we have been unable to deprotect N-benzyl 5-epi-laccarin 18b
under acidic hydrogenation conditions. Isomerisation of 18a (Et3N, PhMe,
85 uC, 36 h) gave a 14 : 1 mixture of 18a and 18b.
8 For the first isolation of laccarin, see: (a) M. Matsuda, T. Kobayashi,
S. Nagao, T. Ohta and S. Nozoe, Heterocycles, 1996, 43, 685–690. For
the isolation of laccarin and laccarin A, see: (b) Y. Wang, S.-P. Yang,
Y. Wu and J.-M. Yue, Nat. Prod. Res., 2004, 18, 159–162.
9 For the preparation of ethereal monochloramine solution, see: (a)
J. Hynes, W. W. Doubleday, A. J. Dyckman, J. D. Godfrey,
J. A. Grosso, S. Kiau and K. Leftheris, J. Org. Chem., 2004, 69,
1368–1371. For the amination of a stabilised enolate, see: (b) P. Dowd
and C. Kaufman, J. Org. Chem., 1979, 44, 3956–3957.
10 G. Stork, A. Y. W. Leong and A. M. Touzin, J. Org. Chem., 1976, 41,
3491–3493.
11 S. R. Baker, A. F. Parsons and M. Wilson, Tetrahedron Lett., 1998, 39,
331–332.
{{ Lactamisation of 12 to generate 19 occurs relatively readily and it was to
block this cyclisation that we carried out immediate Boc protection of 12 to
give 13 (see Scheme 3).
730 | Chem. Commun., 2007, 728–730
This journal is ß The Royal Society of Chemistry 2007