C(6)-alkylation of 3-hydroxypiperidine via reductive and homolytic cleavage of N,S-acetals

Article abstract of DOI:10.1055/s-2004-832846

N-Carboxymethyl 3-hydroxypiperidine (6) undergoes substitution at C(6) via reductive cleavage of N,S-acetal 8a with lithium naphthalenide (LN) and trapping of the resulting carbanionic intermediate 9 with different electrophiles to give adducts 10. N,S-Acetal 8a also undergoes C-S homolysis and trapping of the resulting radical provides an alternative entry to 2-substituted-5- hydroxypiperidines.

Full text of DOI:10.1055/s-2004-832846

2636
LETTER
C(6)-Alkylation of 3-Hydroxypiperidine via Reductive and Homolytic
Cleavage of N,S-Acetals
C
(6)-Alkylation of 3-H
a
ydroxypipe
r
ridine c Bartels, Julian Zapico, Timothy Gallagher*
School of Chemistry, University of Bristol, Bristol BS8 1TS, UK
Fax +44(117)9298611; E-mail: t.gallagher@bristol.ac.uk
Received 20 June 2004
enables subsequent metalation by reductive lithiation
rather than deprotonation.
Abstract: N-Carboxymethyl 3-hydroxypiperidine (6) undergoes
substitution at C(6) via reductive cleavage of N,S-acetal 8a with
lithium naphthalenide (LN) and trapping of the resulting carbanion-
ic intermediate 9 with different electrophiles to give adducts 10.
N,S-Acetal 8a also undergoes C–S homolysis and trapping of the
resulting radical provides an alternative entry to 2-substituted-5-
hydroxypiperidines.
N-Carboxymethyl 3-hydroxypiperidine (6) was subjected
to anodic oxidation4 in methanol to provide a 53% yield of
the N,O-acetals 7a and 7b as an inseparable mixture of re-
gio and cis/trans diastereoisomers (Scheme 2).5 This reac-
tion was also investigated using the corresponding N-tosyl
and N-Boc piperidines, but these were less efficient sub-
strates for the oxidation process. In addition, the oxidation
of 6 was best carried out using graphite rather than plati-
num electrodes6 as the latter produced significant amounts
of overoxidized material.
Key words: piperidines, alkylation, anodic oxidation, radical
Hydroxylated pyrrolidines and piperidines1 are found
within a range of natural products and an ability to intro-
duce these fragments directly via functionalization of an
appropriate intact heterocyclic unit is a synthetically
attractive option. We have previously described the regio-
selective lithiation of N-Boc 3-hydroxypyrrolidine (1),
which undergoes preferential deprotonation at C(5) to
give 2, thereby providing access to 2-substituted 4-hy-
droxypyrrolidines 3 (Scheme 1).2
OH
OH
OH
a
MeO
N
N
OMe
N
CO2Me
CO2Me
CO2Me
6
7a
7b
not separable
OH
OH
b
OH
OLi
OH
s-BuLi
(2 equiv)
R-X
PhS
N
N
SPh
5
R
TMEDA/THF Li
N
CO2Me
N
CO2Me
N
Boc
Boc
Boc
8a
8b
3 : 1
1
2
3
Scheme 2 Reagents: (a) 5 mol% Et4NOTs, carbon electrodes,
MeOH, 8 h, r.t., 2.34 Fmol–1 (53%); (b) PhSH, p-TsOH, CH2Cl2, 2 h,
0 °C (3:1 ratio of 8 and 9, 80% total yield).
OLi
OLi
OH
various
6
conditions
also ref.3
2
Li
N
N
Li
N
Boc
Boc
Boc
Exposure of 7a,b to PhSH under acidic conditions provid-
ed the corresponding N,S-acetals 8a,b, which were readi-
ly separable and the desired C(6)-regioisomer 8a was
isolated in 61% yield.7,8
4
5b
5a
Scheme 1
Reductive lithiation of 8a (as a mixture of cis/trans dia-
stereomers) was achieved using (i) BuLi (to generate the
corresponding alkoxide) followed by (ii) freshly prepared
lithium naphthalenide (LN, 4 equiv). Addition of an ap-
propriate electrophile to the putative a-lithiated piperidine
9 gave the N-Boc 2-substituted-5-hydroxypiperidines
10a–c in moderate yields (Scheme 3, Table 1).9
Generating a disubstituted piperidine via a similar ap-
proach is attractive but Beak3 has reported that N-Boc 3-
hydroxypiperidine (4) does not undergo lithiation at either
C(2) or C(6) to give 5a and 5b, respectively. In light of our
ability to produce and characterize 2, we have also exam-
ined deprotonation of 4 under a variety of conditions, but
without success. In this paper we describe an alternative
entry to an a-lithiated 3-hydroxypiperidine (cf. 5b) based
on a regioselective oxidation of the piperidine nucleus that
Products 10a–c were all obtained as cis and trans isomers,
which were separable. In each case, the major component
was the trans isomer (however, see below and ref.11) and
1H NMR (NOE and in particular the signal associated
with H6ax) was especially helpful in making this assign-
ment. In the case of cis/trans-10a, this product was con-
verted by hydrogenation to 11, the cis and trans isomers
SYNLETT 2004, No. 14, pp 2636–2638
2
2
.1
1
.2
0
0
4
Advanced online publication: 24.09.2004
DOI: 10.1055/s-2004-832846; Art ID: D16604ST
© Georg Thieme Verlag Stuttgart · New York

LETTER
C(6)-Alkylation of 3-Hydroxypiperidine
2637
These issues not withstanding, reductive cleavage of N,S-
acetal 8a does provide for the first time a route to a 3-hy-
droxypiperidine moiety nucleophilic at C(6). Intermediate
9 has been trapped with synthetically useful electrophiles,
and studies are underway to address the practical issues
that have been highlighted above. It is also appropriate to
recognize that the reactivity associated with 8a when
coupled to 9 complements the well known electrophilic
profile associated with N-acyl imimium ions, which
provides an alternative method for substitution adjacent to
nitrogen.11
OLi
OH
a
Li
N
PhS
N
CO2Me
CO2Me
9
8a
OH
b
E
N
CO2Me
10a–c
Scheme 3 Reagents: (a) n-BuLi, THF, –78 °C, 2 min then LN (1 M
In addition to undergoing reductive lithiation, N,S-acetal
8a undergoes homolytic cleavage. Keck allylation (allyl
tributylstannane, AIBN, PhMe, reflux) of 8a gave 10a in
33% yield but as a 5:1 mixture of cis and trans isomers.
Note under these conditions the cis isomer predominated.
Using Bu3SnH and AIBN, the resulting a-aza radical was
also trapped by a series of alkenes to provide the corre-
sponding 2-substituted-5-hydroxypiperidines 10d–f in
moderate yields and as cis/trans mixtures (Scheme 5).12,13
in THF, 4 equiv), 2 min; (b) electrophile (see Table 1).
Table 1 Reaction of the Electrophile with a-Lithiated Piperidine 9
Electrophile
2-Substituted-5-hydroxypiperidine Yield (%)
10
trans:cis
60
3.3:1
Br
OH
N
CO2Me
OH
OH
OH
10a
a
49
3.0:1
CO2Et
Br
OH
R
N
N
PhS
N
CO2Et
CO2Me
CO2Me
CO2Me
N
8a
10d R = CO2t-Bu (40%)
10e R = SO3Ph (45%)
10f R = Ph (20%)
CO2Me
10b
MeOCOCl
38
2.0:1
OH
Scheme 5 Reagents: (a) Bu3SnH, AIBN, PhMe, reflux, RCH=CH2.
MeO2C
N
In summary, N,S-acetal 8a provides access to nucleo-
philic reactivity using two complementary methods based
on either (i) reductive lithiation or (ii) C-S homolysis.
Both methods can be utilized to generate a range of N-pro-
tected 2-substituted-5-hydroxypiperidines.
CO2Me
10c
of which were individually characterized (Scheme 4).
Trans-11 is a known intermediate in the synthesis of N-
methyl pseudoconhydrine 12 and both cis- and trans-11
have been described previously.10
Acknowledgment
We thank the University of Bristol, DFG and AstraZeneca for finan-
cial support. We also acknowledge the support provided by the
EPSRC Mass Spectrometry Service Centre at the University of
Swansea.
OH
Ref.10
OH
OH
a
N
N
N
Me
CO2Me
cis and trans 11
CO2Me
10a
N-Methyl
References
pseudoconhydrine
(1) (a) Buffat, M. G. P. Tetrahedron 2004, 60, 1701.
(b) Weintraub, P. M.; Sabol, J. S.; Kane, J. A.; Borcherding,
D. R. Tetrahedron 2003, 59, 2953. (c) Laschat, S.; Dickner,
T. Synthesis 2000, 1781. (d) Bailey, P. D.; Millwood, P. A.;
Smith, P. D. J. Chem. Soc., Chem. Commun. 1998, 633.
(2) (a) Sunose, M.; Peakman, T. M.; Charmant, J. P. H.;
Gallagher, T.; Macdonald, S. J. F. J. Chem. Soc., Chem.
Commun. 1998, 1723. (b) For earlier work which
incorrectly assigned the regiochemistry of the lithiation of 1,
see: Pandey, G.; Chakrabarti, D. Tetrahedron Lett. 1996, 37,
2285. See also: Pandey, G.; Lakshmalah, G. Synlett 1994,
277.
Scheme 4 Reagents: (a) H2, 5% Pd/C, MeOH (96%).
Two issues currently present themselves. Firstly, an ex-
cess of LN is required to achieve fast conversion of 8a,
and this limits the nature of the electrophile that can be
used to trap 9. Attempts to capture 9 with, for example,
PhCHO failed as benzyl alcohol was formed by rapid
competitive reduction. Also PhS– is generated by reduc-
tion of 8a, and this can lead to the formation of by-prod-
ucts; when allyl bromide was used, phenyl allyl sulfide
was also detected.
(3) Beak, P.; Lee, W. K. J. Org. Chem. 1993, 58, 1109.
Synlett 2004, No. 14, 2636–2638 © Thieme Stuttgart · New York

2638
M. Bartels et al.
LETTER
(4) (a) Shono, T. In Best Synthetic Methods; Katritzky, A. R.;
Meth-Cohn, O.; Rees, C. W., Eds.; Academic Press:
London, 1991, 63. (b) Matsumura, Y.; Kanda, Y.; Shirai, K.;
Onomura, O.; Maki, T. Tetrahedron 2000, 56, 7411.
(c) Barrett, A. T.; Pilipauskas, D. J. Org. Chem. 1991, 56,
2787. (d) Shono, T.; Matsumura, Y.; Tsubata, K. J. Am.
Chem. Soc. 1981, 103, 1172. (e) Shono, T.; Hamaguchi, H.;
Matsumura, Y. J. Am. Chem. Soc. 1975, 97, 4264.
(5) The anodic oxidation of N-carboxymethyl 3-hydroxy-
pyrrolidine has been described and exploited for the
synthesis of pyrrolizidine alkaloids: (a) Thaning, M.;
Wistrand, L.-G. Acta Chem. Scand. 1989, 53, 290. (b) A
1:1 mixture of regioisomers was obtained, analogous to
oxidation of 6 leading to 7a and 7b. See: Thaning, M.;
Wistrand, L.-G. J. Org. Chem. 1990, 55, 1406.
doubling due to rotameric populations): d = 7.72–7.19 (10 H,
m, Ar), 6.15–5.49 (2 H, 4 × br s, 2-H), 4.24–3.51 (6 H, m, 3-
H and 6-H2), 3.23 (6 H, br s, OCH3) and 2.15–1.48 (8 H, m,
4-H2 and 5-H2). 13C NMR (75 MHz, CDCl3): d = 156.4,
155.5, 135.0, 134.6, 133.3, 132.9, 129.3, 129.0, 128.9,
127.5, 72.2, 70.7, 69.3, 67.3, 52.9, 52.4, 39.3, 38.3, 29.4,
23.9, 23.4. Compound 8b also gave satisfactory HRMS data.
(9) Typical Experimental Procedure: N,S-Acetal 8a (149 mg,
0.56 mmol) in THF (5 mL) was cooled to –78 °C and BuLi
(0.28 mL, 2 M in hexanes, 0.56 mmol) was added slowly
followed after 2 min by freshly prepared lithium
naphthalenide (2.23 mL, 1 M in THF, 2.23 mmol) [prepared
from Li metal (70.0 mg, 10.0 mmol) and naphthalene (1.28
g, 10.0 mmol) in THF (10 mL)]. After 2 min, allyl bromide
(0.24 mL, 338 mg, 2.79 mmol) was added and the mixture
was stirred for 1 h at –78 °C, then slowly warmed to r.t. After
addition of 10 mL of H2O, the organic layer was separated
and the aqueous layer extracted with EtOAc (3 × 10 mL).
The combined organic layers were dried (Na2SO4),
(6) (a) Chiba, T.; Takata, Y. J. Org. Chem. 1977, 42, 2973.
(b) Andreades, S.; Zahnow, E. W. J. Am. Chem. Soc. 1969,
91, 4181. (c) Ross, S. D.; Finkelstein, M.; Petersen, R. C. J.
Am. Chem. Soc. 1966, 88, 4657.
(7) Procedure for Electrochemical Oxidation of 6 to 7a,b and
Conversion of 7a,b to 8a,b:
concentrated and the residue was purified by flash
chromatography (silica gel, hexane–EtOAc 5:1) to give
trans-10a (52 mg, 46%) and cis-10a (16 mg, 14%). trans-
10a: 1H NMR (300 MHz, CDCl3): d = 2.08–1.38 (4 H, m, 3-
H2 and 4-H2), 2.24 (1 H, m, 1¢-H) 2.38 (1 H, m, 1¢-H), 3.04
(1 H, dd, J = 14.1 and 1.2 Hz, 6-Hax), 3.69 (3 H, s, OCH3),
3.96 (1 H, br s, 5-H), 4.06 (1 H, br d, J = 14.6 Hz, 6-Heq),
4.32 (1 H, br s, 2-H), 5.10–5.01 (2 H, m, 3¢-H2) and 5.73 (1
H, ddd, J = 17.2, 10.5 and 7.1 Hz, 2¢-H). 13C NMR (100
MHz, CDCl3): d = 26.0, 28.5, 34.1, 45.6, 49.4, 52.6, 67.2,
117.2, 134.8, 156.4. MS (CI): m/z calcd for C10H18NO3
[MH+]: 200.1287; found: 200.1277. cis-10a: 1H NMR (300
MHz, CDCl3): d = 1.92–1.42 (4 H, m, 3-H2 and 4-H2), 2.25
(1 H, m, 1¢-H), 2.39 (1 H, m, 1¢-H), 2.62 (1 H, dd, J = 13.1
and 10.9 Hz, 6-Hax), 3.60 (1 H, m, 5-H), 3.68 (3 H, s, OCH3),
4.31–4.08 (2 H, m, 6-Heq and 2-H), 5.14–4.98 (2 H, m, 3¢-H2)
A solution of N-methoxycarbonyl-3-hydroxypiperidine (6,
2.0 g, 12.60 mmol) and tetraethylammonium tosylate (189
mg, 0.63 mmol, 5 mol%) in dry MeOH (20 mL) was placed
into a beaker-type undivided electrolysis cell equipped with
a graphite anode and cathode. A constant current of 0.1 A
(10–12 V) was passed through the solution at 15 °C until
2.34 Fmol–1 of electricity had passed (approx. 8 h). The
electrolyzed solution was concentrated in vacuo and the
crude product purified by flash chromatography (silica gel;
hexane–EtOAc 3:1) to give 7a,b (1.26 g, 53%) as a pale
yellow oil.
A mixture of 7a,b (1.0 g, 5.40 mmol) in CH2Cl2 (35 mL) was
cooled to 0 °C and TsOH·H2O (1.1 g, 5.90 mmol) was added
followed by thiophenol (0.74 mL, 7.00 mmol). After 2 h at 0
°C, the mixture was quenched with H2O (30 mL) and
extracted with CH2Cl2 (3 × 15 mL). The organic layer was
dried (Na2SO4) and concentrated in vacuo and the residue
was purified by flash chromatography (hexane–EtOAc 3:2)
to afford 2-phenylthio-N-carbomethoxy-3-hydroxy-
piperidine (8b, 275 mg, 19%) as a mixture of diastereo-
isomers and as colorless oil. Continued elution gave 2-
phenylthio-N-carbomethoxy-5-hydroxypiperidine (8a, 880
mg, 61%) as a mixture of diastereoisomers and as a colorless
oil.
and 5.17 (1 H, ddt, J = 17.2, 10.5 and 7.2 Hz, 2¢-H). 13
C
NMR (100 MHz, CDCl3): d = 21.3, 25.5, 34.0, 45.1, 50.8,
52.7, 64.4, 117.1, 135.0, 157.0. MS (CI): m/z calcd for
C10H18NO3 [MH+]: 200.1287; found: 200.1283. The
environment (chemical shift and coupling constants)
associated with 6-Hax is a useful diagnostic probe for cis/
trans stereochemistry in this and related disubstituted
piperidines.10
(10) (a) Plehiers, M.; Hootelé, C. Can. J. Chem. 1996, 74, 2444.
(b) Herdeis, C.; Held, W. A.; Kirfel, A.; Schwabenländer, F.
Liebigs Ann. Chem. 1995, 1295. (c) Shono, T.; Matsumura,
Y.; Onomura, O.; Sato, M. J. Org. Chem. 1988, 53, 4118.
(11) N,S-Acetal 8a also provided access to the corresponding N-
acyl imimium ion under Lewis acid-mediated conditions.
Accordingly, reaction of 8a with allyl trimethylsilane in the
presence of TMSOTf gave 10a as a 6:1 mixture of cis and
trans isomer in a combined yield of 77%. Note that under
these conditions, cis-10a predominated.
(12) Radical addition adducts 10d–f were obtained as approx. 1:1
mixtures of cis and trans isomers which were separable by
chromatography. Styrene is generally a poor trap for
nucleophilic radicals, and the major byproduct in this case
was 6.
(13) The N,Se-acetal corresponding to 8a is generated from 7a,b
under similar conditions to those used for 8a and 8b and has
also been used as a source of a nucleophilic a-aza radical by
C-Se homolysis to provide 10d–f in similar yields to those
obtained from 8a. The use of an N,Se-acetal as a source of
nucleophilic radical reactivity adjacent to nitrogen within a
pyrrolidine framework has been described. See: Barrett, A.
G. M.; Pilipauskas, D. J. Org. Chem. 1991, 56, 2787.
(8) 1H NMR and 13C NMR data for the cis and trans isomers of
8a are presented here, and 1H NMR assignments are based
on a COSY analysis. While it was possible to separate the cis
and trans isomers of 8a, but we have been unable to
individually and unambiguously assign configurations to
these compounds. Isomer A: 1H NMR (300 MHz, CDCl3,
some doubling due to rotameric populations): d = 1.72–2.12
(4 H, m, 3-H2 and 4-H2), 3.10–3.32 (4 H, m, 6-H and OMe),
3.61 (1 H, m, 5-H), 3.99, 4.31 (1 H, br s, 6-H), 5.66, 6.08 (1
H, br s, 2-H), 7.21–7.56 (5 H, m). 13C NMR (75 MHz,
CDCl3): d = 28.8, 29.2, 44.8, 45.5, 52.5, 61.1, 62.1, 66.8,
128.8, 132.6, 133.8, 135.3, 155.5. Isomer B: 1H NMR (300
MHz, CDCl3, some doubling due to rotameric populations):
d = 1.47–2.50 (4 H, m, 3-H2 and 4-H2), 3.02–3.81 (5 H, m,
6-H and OMe), 3.89, 4.12 (1 H, br s, 6-H), 5.79, 6.10 (1 H,
br s, 2-H), 7.26–7.51 (5 H, m). 13C NMR (75 MHz, CDCl3):
d = 24.4, 25.9, 44.4, 44.9, 52.4, 62.1, 63.1, 63.8, 127.9,
128.8, 131.9, 132.7, 156.4. Both isomers of 8a gave
satisfactory HRMS data.
1H NMR and 13C NMR data for 8b (mixture of
diastereomers): 1H NMR (300 MHz, CDCl3, also some
Synlett 2004, No. 14, 2636–2638 © Thieme Stuttgart · New York