1

The natural alkaloid isoanabasine: synthesis from

2,30-bipyridine, efficient resolution with BINOL, and assignment

of absolute configuration by Mosher
s method

Chuan-Qing Kang, Yan-Qin Cheng, Hai-Quan Guo, Xue-Peng Qiu and Lian-Xun Gao*

The State Key Laboratory of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry,

Chinese Academy of Sciences, Graduate School of Chinese Academy of Sciences, 5625 Renmin Street, Changchun 130022, PR China

Received 18 April 2005; accepted 10 May 2005

Abstract—A highly efficient and practical resolution of racemic 1-benzylisoanabasine, which was synthesized by reduction of the

benzyl salt of 2,30-bipyridine, has been achieved through molecular complexation with (R)-BINOL or (S)-BINOL to afford pure

enantiomers (100% ee). The two enantiomers of the natural alkaloid isoanabasine have been obtained by debenzylation of the corresponding

enantiomeric 1-benzylisoanabasine. Using Mosher
s method by NMR techniques, the absolute configuration of (
)-isoanabasine

has been assigned as the (R)-configuration for the first time. Moreover, an unexpected rotamer ratio of Mosher
s amide

was observed. The syn-form of two rotamers of (R)-MTPA-(R)-isoanabasine was predominant over the anti-form.

2005 Elsevier Ltd. All rights reserved.

1. Introduction

The rarely investigated minor natural alkaloid isoanabasine

1 (Fig. 1) has been discovered only in low quantities

in the desert plant Anabasis aphylla L., which can be

found in the northwest of China and Central Asia.1 Recently,

much attention has been paid to the synthesis

and biological activities of its skeleton analogues including

natural and synthetic alkaloids, such as anabasine,

anatabine, cytisine, and some other pyridyl piperidines.

2,3 These analogues displayed potential as therapeutic

agents for peripheral and central nervous

system diseases and other disorders because of their

affinity activities to neuronal nicotinic acetylcholine

receptors (nAChRs) but with lower toxicities than the

well-known alkaloid nicotine.3 Experience has shown

that important biological activities of chiral natural

products are related with their absolute configuration,

so that methods providing single enantiomers were

of utmost importance. However, to the best of our

knowledge, the enantiomers of 1 had never gained as

much attention, no report has ever clarified the absolute

configuration or focused on the preparation of its enantiomer.

Limited preliminary studies related to 1 were

conducted only on reactions of the pyridine or the piperidine

ring during the 1970s in the former Soviet Union.4

Herein, we report our approach to the two enantiomers

of 1 via synthesis and resolution, plus assignment of

absolute configuration of 1 by Mosher
s method.

2. Results and discussion

2.1. Synthesis of racemic alkaloids

Syntheses of substituted piperidines have been reported

using a variety of methods including stereoselective

N N

R

1 R = H

2 R = Bn

0957-4166/$ - see front matter
2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.tetasy.2005.05.012

* Corresponding author. Tel.: [phone]; fax: [phone]

5653; e-mail: [e-mail]

OH

OH OH

RO CO2H OH

RO CO2H

5 R = H

6 R = p-Toluoyl

(R)-7 (S)-7

Figure 1. Chiral reagents bearing double acidic groups for resolution.

Tetrahedron: Asymmetry 16 (2005) 2141–2147

Tetrahedron:

Asymmetry

approaches in the development of new drugs.5 However,

those methods involving enantioselective reaction usually

lead to a complex mixture of products, low yield,

or quantitative consumption of the chiral auxiliary

meaning that they are not practical. All reported preparations

of 1 were based on the reduction of 2,30-bipyridine

3 in two approaches. One involved the direct use

of a large amount of reductive metals in low to moderate

yields.6a,b The other method was the oxidation of

2,30-bipyridine prior to catalytic hydrogenation with

Pd/C.6c,d In view of the advantages of a pre-constructed

heterocyclic skeleton upon reduction of bipyridines to

pyridyl piperidines, we have utilized an improved reductive

approach to 1 from 3.

Formation of the piperidine ring of 1 was accomplished

on a large scale by selective reduction of 30-pyridyl

through benzylation on less hindered nitrogen atom of

3 (Scheme 1). Benzyl salt 4, obtained by heating a solution

of 3 and benzyl chloride in acetonitrile, was elaborated

to 2 via hydrogenation in the presence of

triethylamine catalyzed with PtO2 or Pd/C. Debenzylation

of 2 was performed by substituent exchange reaction

on nitrogen with benzyl chloroformate and then

hydrolysis with 37% HCl to afford racemic 1 in total

67% yield.

2.2. Enantiomerically pure alkaloids from resolution

Molecular complexation, based on molecular recognition

between host and guest molecules directed by steric

complementary action and specific intermolecular

forces (such as hydrogen bonding and second-order

interaction), has proven to be an effective method for

the resolution of organic molecules.7 Optically active

1,10-bi-2-naphthol 7 (BINOL) has shown excellent performances

in this field. A few successful examples on

resolution of alkaloids by forming host–guest complex

with resolving reagent bearing double carboxylic or

phenolic groups have been published.8 The selectively

formed diastereomeric salt in well-matched chirality, if

possible, would display marked differences in physical

or chemical properties such as solubility and stability

compared with the alkali guest and acidic host themselves

or the chirality-mismatched diastereomeric salt.

Three readily available chiral reagents bearing double

acidic groups, l-tartaric acid 5, l-O,O0-ditoluoyl tartaric

acid 6 and (R)-7 (Fig. 1), were selected as candidates to

screen out better resolving process. As shown in Table 1,

fortunately, (R)-7 could form crystals with one enantiomer

of 2 in ethanol in a molar ratio of 1:2 [(R)-7:rac-2].

The solvent was then changed with methanol to give a

similar result. Acetonitrile and toluene were not suitable

for the resolution. Varying the ratio of the resolving reagent

and racemic alkaloid to 1:1 did not afford better

results. Hence, we selected (R)-7 and (S)-7 as resolving

reagents to optimize the resolution process.

The typical resolution process employed is as follows.

A solution of (R)-7 (0.5 equiv) and rac-2 (1.0 equiv) in

ethanol was stirred under reflux for 0.5 h. The crystals

formed during cooling to room temperature were collected

by filtration and purified by recrystallization.

The resulting crystals were characterized as a 1:1 molecular

complex of 2 and (R)-7 by 1H NMR and elemental

analysis.9 Decomposition of the resolving complex with

15% NaOH afforded the (
)-enantiomer of 2 (100%

ee,10 79% yield based on half of the starting rac-2),

which displayed ?a20

D ?
65.4 (c 2.0, ethanol). All mother

liquids from resolution and recrystallization, enriched

in the (+)-enantiomer of 2, were combined and washed

with 15% NaOH to remove (R)-7. The resulting residues

were applied to form molecular complex with (S)-7

(0.5 equiv) to afford (+)-2 (100% ee,10 83% yield based

on half of the starting rac-2), which displayed ?a20

D ?

t65.2 (c 2.0, ethanol). Both (R)- and (S)-7 were recovered

in >80% yield by acidification of the corresponding

alkali aqueous solution, respectively (Scheme 2).

Unfortunately, no crystals of the molecular complex of

(R)-7 and (
)-2 suitable for X-ray crystallography structure

analysis have been obtained until now. However, by

1H NMR (Fig. 2), we have observed that the hydroxy

N N

3

N N

4 Bn

Cl

N N

Bn

N NH

BnCl, MeCN

reflux, 4 hr

94%

H2, Et3N, EtOH

PtO2 at 1 atm

or Pd/C at 10 atm

95%

1) CbzCl, toluene

reflux, 15 hr

2) 37% HCl

reflux, 6 hr

76%

2

1

Scheme 1.

Table 1. Screening of resolving reagents by formation of precipitatesa

Alkaloids Resolving reagents

L-5 L-6 (R)-7

1

2

+

a Those formed precipitates after mixing were marked with +
,

otherwise with 

.

rac-2 + (R)-7

(1.0 equiv : 0.5 equiv)

(R)-7 (-)-2

(-)-2

100% ee

(R)-7 (-)-2

(+)-2

(+)-2, (-)-2

recycled

(R)-7

(S)-7

(0.5 equiv)

(+)-2 (S)-7 (+)-2

100% ee

recycled

(S)-7

(mother liquor)

Scheme 2. Schematic procedure of the resolution.

2142 C.-Q. Kang et al. / Tetrahedron: Asymmetry 16 (2005) 2141–2147

protons in the molecular complex (R)-7?(
)-2 have

shown a downfield shift from d 5.02 [in (R)-7] to d

6.80 ppm, while the protons on benzyl methylene

showed a double doublet peak with a large coupling constant

(J = 60.0 and 13.2 Hz). The 1:1 mixture of (S)-7

and (
)-2 in the same concentration11 to the molecular

complex (R)-7?(
)-2 exhibited great differences and that

the hydroxy protons also showed a downfield shift from

d 5.02 to d 6.50 ppm and the protons on benzyl methylene

showed double doublet with small split (J = 18.0

and 13.2 Hz).12 The larger split in 1H NMR spectra of

the molecular complex (R)-7?(
)-2 demonstrated the

stronger hindered rotation of benzyl around its C–N

bond, which should come from the stronger binding between

(R)-7 and (
)-2. These phenomena suggested that

the hydrogen bonding present in both diastereomeric

mixtures, and the stronger interactions in (R)-7?(
)-2

originated from the matching of chirality of the host

and guest in terms of chiral recognition, which was a

fundamental driver to the stereoselectivity of (R)-7 to

(
)-2 or (S)-7 to (+)-2 during resolution.

Based on the debenzylation procedure described in the

last section, both enantiomers of 2 were converted to

corresponding pure enantiomers of 1, which displayed

a specific rotation of 15.2 (c 1.0, ethanol) with the same

sign to their benzyl precursors.

2.3. Assignment of absolute configuration of (
)-1

After pioneering work in the 1970s, Mosher
s method

has been one of the most useful techniques amongst

those employed to determine the absolute configuration

of organic compounds.13 Hoye and Renner extended

and modified Mosher
s method to cyclic secondary

amines, which have a stereogenic center on the ring.14

It has been proven that Hoye
s method was the most efficient

technique for the assignment of the absolute

configuration of substituted cycloamines15a–f even

though the inconveniences presented in MTPA amide

synthesis and NMR spectra analysis might limit its

application.15g,h

In principle, the most stable conformation of the

Mosher amides derived from the secondary amines

(such as those derived from piperidine) has the trifluoromethyl

group coplanar with the carbonyl group

syn-periplanar and with the nitrogen anti-periplanar,

the phenyl group on MTPA moiety of the amides

shielded one of four possible quadrants of the molecule

depending on the specific rotamer and absolute stereochemistry.

Furthermore, it was reasonable to assume

that the larger pyridyl group at the 3-position in the

equatorial position was favorable over that at the axial

position. For Mosher
s amide isomers 8 and 9 (Scheme

3), one rotamer of 8 in which the axial protons shielded

most at the 2- and 6-positions, should take the same

rotation conformation around the amide bond to that

of 9 in which most-shielded axial protons at 3- and 5-

positions and vice versa.16

Based on the above principles and in light of the 1H

NMR, 13C NMR, 1H–1H COSY, and 1H–13C HMQC,

the chemical shifts of the rotamers were assigned and

we easily knew that the minor of 8 and the major of 9

were a pair of rotamers, which had taken the syn-form

while another pair were anti-rotamers (Table 2). Hence,

the molar ratio17 of syn-8 to anti-8 was 1:3.1 similar to

what has generally been reported14,15a–e and that of

syn-9 to anti-9 was unexpectedly 2.8:1, which had been

presented only once in the literature.15f From the differences

of chemical shifts (DdS–R), the axial 20-proton in

the syn-8 and the axial 30-proton in syn-9 should take

syn-form with phenyl on corresponding MTPA moiety

based on the carbonyl plane. Thus, the absolute stereochemistry

on stereogenic center of (
)-1 was deduced to

be an (R)-configuration. The rendered 3D representations

have clearly exhibited the absolute configuration

Figure 2. Partial 1H NMR spectra of diastereomeric 1:1 mixtures of

(
)-2 with (R)-7 and (S)-7.

(-)-1

(R)-MTPACl

(S)-MTPACl

N

N O

CF3

Ph

OMe

N

N O

CF3

OMe

Ph

anti-8

anti-9

O

Cl

CF3

Ph

OMe

O

Cl

CF3

OMe

Ph

N

N

O

Ph CF3

OMe

syn-9

N

N

O

MeO CF3

Ph

syn-8

+

+

1

3 2

4 6

5

Scheme 3. Conversion of (
)-1 to corresponding Mosher amides.16

Table 2. Assignment of chemical shifts (ppm) of protons on piperidinyl

ring of Mosher amides

Ha syn-8

(minor)

anti-8

(major)

syn-9

(major)

anti-9

(minor)

DdS–R

syn anti

2a 2.41 2.96 3.17 2.99
0.76
0.03

2e 4.10 4.83 4.08 4.80 0.02 0.05

3 2.86 2.78 1.67 3.05 1.19
0.24

4a 1.70 1.75 1.87 1.76
0.18
0.01

4e 2.05 1.84 1.76 2.02 0.29
0.18

5a 1.63 0.48 1.50 1.55 0.13
1.07

5e 1.92 1.18 1.79 1.55 0.13
0.37

6a 2.57 2.91 2.62 2.30
0.05 0.61

6e 4.85 4.01 4.78 3.84 0.07 0.17

a Protons at axial position were marked with a
and those at equatorial

position marked with e
.

C.-Q. Kang et al. / Tetrahedron: Asymmetry 16 (2005) 2141–2147 2143

and the shielding circumstance in Mosher amides (Fig.

3).

3. Conclusions

In conclusion, we have synthesized the natural alkaloid

isoanabasine 1 via a facile and practical route in which

the intermediate 1-benzylisoanabasine 2 was resolved

to both enantiomers in 100% ee efficiently by forming

a molecular complex with (R)- and (S)-1,10-bi-2-naphthol

7. The preliminary investigation to the chiral recognition

involving in resolution, studied by 1H NMR,

revealed that the matching of chirality between (R)-7

and (R)-(
)-2 [or (S)-7 and (S)-(+)-2] strengthened the

hydrogen bonding of the host–guest complex. Enantiomerically

pure 1 was obtained by debenzylation of corresponding

enantiomeric 2. The absolute configuration

of (
)-1 was assigned as an (R)-configuration using

Mosher
s method by NMR techniques. During the

deduction, it was observed from unexpected rotamer

ratio of Mosher amide derived from (R)-(
)-1 and (S)-

MTPACl that the syn-rotamer predominated over the

anti-one.

4. Experiments

Generally, chiral HPLC measurements were carried out

on a Shimadzu Class-VP workstation through CHIRALCEL

 OD-H column (Daicel Chemical Industries,

Ltd) with a UV detector at 254 nm. Optical rotations

were measured on a Perkin Elmer 341LC polarimeter.

ESI-MS measurements were conducted with a LCQ

instrument. Unless otherwise noted, 1H spectra were recorded

on a Bruker 600 MHz spectrometer and the proton

chemical shifts referenced to TMS as internal

standard at 0.00 ppm. 13C NMR chemical shifts were reported

in parts per million relative to the solvent CDCl3

at 77.0 ppm or DMSO-d6 at 39.5 ppm. Melting points

were measured with a hot-stage microscope XT-4. Elemental

analysis was carried out with a VarioEL instrument.

TLC was performed on aluminum TLC-layers

Silica gel GF-254. Detection was done by UV light

(254 and 365 nm). All materials were available commercially

without further purification. All products were

dried in vacuum prior to further reaction and analysis.

4.1. Preparation of 10-benzyl-2,30-bipyridinium chloride 4

A stirring solution of 2,30-bipyridine 3 (15.60 g,

0.10 mol) and benzyl chloride (13.8 mL, 0.12 mol) in

acetonitrile (100 mL) was heated to reflux for 4 h. After

being cooled to room temperature, the precipitates were

collected by filtration, washed with acetonitrile (30 mL),

and dried in vacuum to give 26.56 g of 10-benzyl-2,30-

bipyridinium chloride 4 as buff granule (94% yield).

Mp 78–80 C. 1H NMR (600 MHz, DMSO-d6): d

10.15 (s, 1H), 9.38 (d, 1H, J = 6.0 Hz), 9.27 (d, 1H,

J = 8.4 Hz), 9.79 (d, 1H, J = 4.2 Hz), 8.39 (d, 1H,

J = 7.8 Hz), 8.30 (dd, 1H, J = 7.8, 6.0 Hz), 8.08 (dt,

1H, J = 7.8, 1.8 Hz), 7.68 (d, 2H, J = 6.6 Hz), 7.59 (dd,

1H, J = 7.2, 4.8 Hz), 7.44 (m, 3H), 6.10 (s, 2H). 13C

NMR (150 MHz): d 150.1, 149.6, 144.4, 143.0, 142.5,

138.4, 138.0, 134.3, 129.2, 129.1, 128.9, 128.5, 125.1,

122.0, 63.2. ESI-MS (m/z): [M
I]+ 247.0.

4.2. Preparation of racemic 1-benzylisoanabasine [i.e., 1-

benzyl-3-(pyridin-2-yl)piperidine] rac-2

A suspension of 4 (11.28 g, 0.040 mol), triethylamine

(6.2 mL, 0.044 mol), and 10% Pd/C (0.6 g) in ethanol

(100 mL) was stirred in an autoclave under hydrogen

atmosphere at 10 atm for 6–8 h at room temperature.

Upon deflation of the hydrogen, the insoluble species

were removed by filtration through Celite. After

evaporation of the ethanol, the residues were dissolved

in dichloromethane (100 mL) and washed with 10%

NaOH (30 mL · 2) and water. The organic layer was

dried over magnesium sulfate. The solvent was removed

by evaporation to give 9.57 g of 1-benzylisoanabasine 2

as colorless oil (95% yield). 1H NMR (600 MHz,

CDCl3): 8.52 (d, 1H, J = 4.2 Hz), 7.57 (dt, 1H, J = 7.8,

1.8 Hz), 7.35 (d, 2H, J = 7.2 Hz), 7.30 (t, 2H, J = 7.2

Hz), 7.24 (t, 1H, J = 7.2 Hz), 7.16 (d, 1H, J = 7.8 Hz),

7.09 (dd, 1H, J = 7.8, 4.8 Hz), 3.59 (s, 2H), 3.07

(m, 2H), 2.94 (d, 1H, J = 11.4 Hz), 2.28 (t, 1H, J = 10.8

Hz), 2.07 (m, 1H), 1.97 (dt, 1H, J = 12.6, 1.8 Hz),

1.78 (m, 2H), 1.61 (m, 1H). 13C NMR (150 MHz):

d 163.6, 149.1, 138.1, 136.3, 129.2, 128.2, 127.0, 122.0,

121.3, 63.4, 59.0, 53.6, 44.6, 30.5, 25.3. ESI-MS (m/z):

[M+H]+ 253.1.

An alternative hydrogenation using PtO2 as catalyst was

carried out at ambient pressure in a similar procedure.

4.3. Preparation of racemic isoanabasine [i.e., 3-(pyridin-

2-yl)piperidine] rac-1

A stirred solution of 2 (2.52 g, 0.010 mol) and benzyl

chloroformate (1.7 mL, 0.012 mol) in toluene (30 mL)

was heated to reflux for 10 h under a nitrogen atmosphere.

After being cooled to room temperature, the

Figure 3. 3D representations of rotamers of Mosher amides 8 and 9

from minimizing energy.

2144 C.-Q. Kang et al. / Tetrahedron: Asymmetry 16 (2005) 2141–2147

reaction mixture was filtered through silica gel (200

mesh) and eluted with ethyl acetate (100 mL) to remove

tar and decolorizing. The filtrate was then washed with

saturated NaHCO3 (40 mL · 2) and water. The solvent

was removed by evaporation and the residues dissolved

in a mixture of acetic acid (20 mL) and 37% HCl

(10 mL). The acidic solution was stirred and heated to

reflux for 6 h and the acids then distilled off. The

residues were suspended in 10% NaOH (30 mL) and

extracted with dichloromethane (50 mL · 2). The organic

layer was dried over magnesium sulfate and evaporated

to give 1.23 g of rac-1 as viscous oil (76% yield).

1H NMR (600 MHz, CDCl3): d 8.49 (d, 1H, J = 4.8

Hz), 7.56 (dt, 1H, J = 7.8, 1.8 Hz), 7.12 (d, 1H, J =

7.8 Hz), 7.07 (dd, 1H, J = 7.2, 4.8 Hz), 3.23 (m, 1H), 3.15

(s, 1H), 3.08 (d, 1H, J = 12.8 Hz), 2.86 (m, 2H), 2.66

(dt, 1H, J = 12.2, 2.4 Hz), 2.00 (dd, 1H, J = 12.6,

3.0 Hz), 1.76 (m, 2H), 1.60 (m, 1H). 13C NMR (150

MHz): d 163.4, 149.0, 136.3, 121.6, 121.3, 51.7, 46.1,

45.2, 30.6, 26.1. ESI-MS (m/z): [M+H]+ 163.1.

4.4. Resolution of racemic 1-benzylisoanabasine to enantiomers

(R)-(
)-2 and (S)-(+)-2

A solution of rac-2 (5.56 g, 0.022 mol) and (R)-1,10-bi-2-

naphthol (R)-7 (3.15 g, 0.011 mol) in ethanol (25 mL)

was stirred at 70 C for 0.5 h and cooled naturally to

room temperature. After 1 h, the precipitates were

collected by filtration and purified by recrystallization

in ethanol (15 mL). The resultant molecular complex

was suspended in 15% NaOH (12 mL) and extracted

with dichloromethane (30 mL · 3). The extract was

washed with saturated NaHCO3, dried over magnesium

sulfate, and evaporated to dryness to give 2.20 g of

(R)-(
)-2 (79% yield). All filtrates from the separation

of the crystals and recrystallization were combined and

evaporated to remove ethanol. The residues were

dissolved in dichloromethane (60 mL) and washed with

15% NaOH (15 mL · 2) to remove (R)-7. All aqueous

solutions containing sodium salt of (R)-7 were combined

and acidified with 10% HCl to form precipitates, which

were collected by filtration, washed with water, and

recrystallized in toluene to give 2.57 g of (R)-7 (82%

yield). The organic phase enriched with (S)-(+)-2 was

then dried with magnesium sulfate and evaporated to

give an oil, which was complexed with (S)-7 (3.15 g,

0.011 mol) in ethanol to give 2.31 g of (S)-(+)-2

(83% yield) in similar procedure as described above.

(S)-7 was recovered in 80% yield and the terminal

residual oil, usually >0% ee, could be used for resolution

again. (R)-(+)-2: ?a20

D ?
65.4 (c 2.0, ethanol), 100% ee

[tR(R) = 5.5 min]. (S)-(
)-2: ?a20

D ? t65.2 (c 2.0,

ethanol), 100% ee [tR(S) = 6.6 min].9,10

4.5. (R)-(
)-Isoanabasine (R)-(
)-1

Prepared from (R)-(
)-2 in a similar procedure to rac-1

from rac-2. ?a20

D ?
15.2 (c 1.0, ethanol).

4.6. (S)-(+)-Isoanabasine (S)-(+)-1

Prepared from (S)-(+)-2 in a similar procedure to rac-1

from rac-2. ?a20

D ? t15.2 (c 1.0, ethanol).

4.7. Preparation of Mosher
s amide (3R,20S)-1-(2-methoxy-

2-trifluoromethyl-phenylaceto)-3-(pyridin-2-yl)piperidine

anti- and syn-8

To a solution of (R)-(
)-1 (21.0 mg, 0.12 mmol) and triethylamine

(18.4 lL, 0.13 mmol) in dichloromethane

(2.0 mL) was added dropwise a solution of (R)-(
)-2-

methoxy-2-trifluoromethyl-phenylacetyl chloride [(R)-

MTPACl, 24.5 lL, 0.13 mmol] in dichloromethane

(1 mL) at 0 C. The reaction mixture was stirred for

another 4 h at room temperature and diluted with

dichloromethane (5 mL) followed by quenching with

water (3 mL). The organic layer was separated and

washed with water (3 mL) again, dried over magnesium

sulfate, and evaporated to dryness. The residues were

purified with silica-gel chromatography column to give

40.0 mg of rotameric mixture of 8 as oil (86% yield),

which showed a single Rf value on TLC. The molar ratio

of syn-8 to anti-8 was 1:3.1 characterized by 1H NMR.

anti-8: 1H NMR (600 MHz, CDCl3): d 8.53 (d, 1H,

J = 4.5 Hz), 7.62 (d, 2H, J = 7.4 Hz), 7.61 (m, 1H),

7.41 (m, 2H), 7.39 (d, 1H, J = 7.3 Hz), 7.18 (d, 1H,

J = 7.8 Hz), 7.13 (dd, 1H, J = 7.3, 5.0 Hz), 4.83 (d, 1H,

J = 12.9 Hz), 4.01 (d, 1H, J = 13.5 Hz), 3.68 (s, 3H),

2.96 (t, 1H, J = 12.4 Hz), 2.91 (dt, 1H, J = 13.6,

2.8 Hz), 2.78 (tt, 1H, J = 11.4, 3.7 Hz), 1.84 (d, 1H,

J = 12.3 Hz), 1.75 (dt, 1H, J = 12.5, 4.0 Hz), 1.18 (dt,

1H, J = 13.6, 3.1 Hz), 0.48 (m, 1H). 13C NMR

(150 MHz): d 163.7, 161.6, 149.1, 136.5, 134.1, 129.1,

128.1, 126.5, 123.6 (q, J = 288 Hz), 122.0, 121.7, 84.7

(q, J = 24.9 Hz), 55.3, 47.2, 45.1, 43.6, 30.3, 23.5. syn-

8: 1H NMR (600 MHz, CDCl3): d 8.43 (d, 1H,

J = 4.1 Hz), 7.55 (dt, 1H, J = 7.7, 1.5 Hz), 7.44 (dd,

2H, J = 8.0, 3.2 Hz), 7.30 (m, 3H), 7.08 (dd, 1H,

J = 7.3, 5.0 Hz), 6.97 (d, 1H, J = 7.8 Hz), 4.85 (overlap

with a peak of anti-8, 1H), 4.10 (dd, 1H, J = 13.5,

1.7 Hz), 3.85 (s, 3H), 2.86 (tt, 1H, J = 11.8, 3.7 Hz),

2.57 (dt, 1H, J = 13.0, 2.7 Hz), 2.41 (t, 1H,

J = 12.5 Hz), 2.05 (d, 1H, J = 13.4 Hz), 1.92 (dt, 1H,

J = 13.3, 2.5 Hz), 1.70 (m, 1H), 1.63 (m, 1H). 13C

NMR (150 MHz): d 164.0, 161.3, 149.2, 136.5, 133.8,

129.0, 128.2, 126.3, 123.5 (q, J = 288 Hz), 121.8, 121.4,

84.9 (q, J = 24.9 Hz), 56.2, 50.5, 44.8, 42.9, 30.6, 25.2.

ESI-MS (m/z): [M+H]+ 378.3.

4.8. Preparation of Mosher
s amide (3R,20R)-1-(2-methoxy-

2-trifluoromethyl-phenylaceto)-3-(pyridin-2-yl)piperidine

anti- and syn-9

The synthesis of 9 was performed by amidation of (R)-

(
)-1 with (S)-MTPACl to give 41.8 mg of rotameric

mixture (90% yield) in similar procedure to that of 8.

The molar ratio of syn-9 to anti-9 was 2.8:1. syn-9: 1H

NMR (600 MHz, CDCl3): d 8.54 (d, 1H, J = 4.5 Hz),

7.58 (d, 2H, J = 7.6 Hz), 7.53 (dt, 1H, J = 7.7, 1.6 Hz),

7.43 (m, 3H), 7.16 (m, 1H), 6.21 (d, 1H, J = 7.8 Hz),

4.78 (dt, 1H, J = 13.0, 1.9 Hz), 4.08 (dt, 1H, J = 13.0,

1.9 Hz), 3.62 (s, 3H), 3.17 (dd, 1H, J = 11.7, 13.5 Hz),

2.62 (dt, 1H, J = 13.0, 2.4 Hz), 1.87 (m, 1H), 1.78 (m,

2H), 1.67 (tt, 1H, J = 11.9, 3.7 Hz), 1.50 (m, 1H). 13C

NMR (150 MHz): d 163.5, 160.3, 147.8, 137.8, 134.5,

129.1, 128.3, 126.8, 123.6 (q, J = 288 Hz), 123.3, 122.3,

84.5 (q, J = 24.9 Hz), 55.4, 49.3, 43.0, 42.8, 29.7, 24.7.

C.-Q. Kang et al. / Tetrahedron: Asymmetry 16 (2005) 2141–2147 2145

anti-9: 1H NMR (600 MHz, CDCl3): d 8.64 (d, 1H,

J = 4.5 Hz), 7.76 (dt, 1H, J = 7.7, 1.6 Hz), 7.47 (m,

2H), 7.36 (m, 3H), 7.29 (m, 2H), 4.80 (overlap with a

peak of syn-9, 1H), 3.84 (d, 1H, J = 13.4 Hz), 3.78 (s,

3H), 3.05 (m, 1H), 2.99 (m, 1H), 2.30 (dt, 1H,

J = 12.0, 4.5 Hz), 2.02 (d, 1H, J = 13.4 Hz), 1.76 (overlap

with a peak of syn-9, 1H), 1.55 (m, 2H). 13C NMR

(150 MHz): d 164.2, 160.8, 147.9, 138.3, 133.9, 129.1,

128.2, 126.4, 123.7 (q, J = 288 Hz), 122.6, 122.5, 84.9

(q, J = 24.9 Hz), 56.1, 46.7, 45.9, 43.5, 30.2, 25.0. ESIMS

(m/z): [M+H]+ 378.3.

Acknowledgments

We appreciate greatly Dr. Zheng Bian and Dr. Xiaoli

Bai
s valuable suggestions and corrections to this paper.

We also express our thanks to Mr. Yijie Wu and Ms.

Fengying Jing for their help on NMR data acquirement.

References

1. Duke, J. A. Handbook of Phytochemical Constituents of

GRAS Herbs and Other Economic Plants; CRC Press:

Boca Raton, FL, 1992, cited in online database (http://

www.ars-grin.gov:8080): USDA, ARS, National Genetic

Resources Program, Phytochemical and Ethnobotanical

Databases, National Germplasm Resources Laboratory,

Beltsville, Maryland, 2005.

2. Selected synthesis published recently: (a) Danieli, B.;

Lesma, G.; Passarella, D.; Sacchetti, A.; Silvani, A.;

Virdis, A. Org. Lett. 2004, 6, 493–496; (b) Amat, M.;

Canto′ , M.; Llor, N.; Bosch, J. Chem. Commun. 2002, 526–

527; (c) Stehl, A.; Canto, M.; Schulz, K. Tetrahedron 2002,

58, 1343–1354; (d) Romanelli, M. N.; Manetti, D.;

Scapecchi, S.; Borea, P. A.; Dei, S.; Bartolini, A.;

Ghelardini, C.; Gualtieri, F.; Guandalini, L.; Varani, K.

J. Med. Chem. 2001, 44, 3946–3955; (e) Felpin, F. X.;

Girard, S.; Vo-Thanh, G.; Robins, R. J.; Villie′ras, J.;

Lebreton, J. J. Org. Chem. 2001, 66, 6305–6312; Synlett

2000, 1646–1648; (f) O
Neill, B. T.; Yohannes, D.;

Bundesmann, M. W.; Arnold, E. P. Org. Lett. 2000, 2,

4201–4204; (g) Andre′s, J. M.; Herra′iz-Sierra, I.; Pedrosa,

R.; Pe′rez-Encabo, A. Eur. J. Org. Chem. 2000, 1719–1726;

(h) Balasubramanian, T.; Hassner, A. Tetrahedron: Asymmetry

1998, 9, 2201–2205; (i) Weymann, M.; Pfrengle, W.

Synthesis 1997, 1151–1160; (j) Yang, C. M.; Tanner, D. D.

Can. J. Chem. 1997, 95, 616–620; (k) Wanner, M. J.;

Koomen, G.-J. J. Org. Chem. 1996, 61, 5581–5586; (l)

Genisson, Y.; Mehmandoust, M.; Marazano, C. Heterocycles

1994, 39, 811–818.

3. For biological activities research, recent reviews: (a)

Toma, L.; Barlocco, D.; Gelain, A. Expert Opin. Ther.

Patents 2004, 14, 1029–1040; (b) Romanelli, M. N.;

Gualtieri, F. Med. Res. Rev. 2003, 23, 393–426; (c)

Bunnelle, W. H.; Decker, M. W. Expert Opin. Ther.

Patents 2003, 13, 1003–1021.

4. (a) Yunusov, T. K.; Akbarov, K.; Otroshchenko, O. S.

Uzb. Khim. Zh. 1980, 77–90; (b) Sadikov, A. S.;

Otroshchenko, O. S.; Yunusov, T. K. Pol. J. Chem.

1979, 53, 367–377; (c) Kurbatov, Y. V.; Zalyalieva, S. V.

Khim. Geterotsikl. Soedin. 1977, 1535–1537.

5. For reviews, see: (a) Buffat, M. G. P. Tetrahedron 2004, 60,

1701–1729; (b) Laschat, S.; Dickner, T. Synthesis 2000,

1781–1813; (c) Bailey, P. D.; Millwood, P. A.; Smith, P. D.

J. Chem. Soc., Chem. Commun. 1998, 633–640; (d) Baliah,

V.; Jeyraman, R.; Chandrasekaran, L. Chem. Rev. 1983,

83, 379–423.

6. (a) Lunn, G. J. Org. Chem. 1992, 57, 6317–6320; (b)

Smith, C. R. J. Am. Chem. Soc. 1931, 53, 277–283; (c)

Plaquevent, J.-C.; Chichaoui, I. Bull. Soc. Chim. Fr. 1996,

133, 369–380; (d) Plaquevent, J.-C.; Chichaoui, I. Tetrahedron

Lett. 1993, 34, 5287–5288.

7. (a) Cheng, X.; Hou, G. H.; Xie, J. H.; Zhou, Q. L. Org.

Lett. 2004, 6, 2381–2383; (b) Liao, J.; Sun, X. X.; Cui, X.;

Yu, K. B.; Zhu, J. Chem. Eur. J. 2003, 9, 2611–2615; (c)

Kaupp, G. Angew. Chem., Int. Ed. Engl. 1994, 33, 728–

729; (d) Toda, F.; Tohi, Y. J. Chem. Soc., Chem. Commun.

1993, 1238–1240; (e) Toda, F. Top. Curr. Chem. 1987, 140,

43–69.

8. (a) Bortolini, O.; Fogagnolo, M.; Fantin, G.; Medici, A.

Chem. Lett. 2003, 32, 206–207; (b) Nishikawa, K.;

Tsukada, H.; Abe, S.; Kishimoto, M.; Yasuoka, N.

Chirality 1999, 11, 166–171; (c) Toda, F.; Shinyama, T.

J. Chem. Soc., Perkin Trans. 1 1997, 1759–1762; (d)

Zaderenko, P.; Lopez, P.; Ballesteros, P.; Takumi, H.;

Toda, F. Tetrahedron: Asymmetry 1995, 6, 381–384; (e)

Hatano, K.; Takeda, T.; Saito, R. J. Chem. Soc., Perkin

Trans. 2 1994, 579–584.

9. The integrals of all proton peaks in 1H NMR spectra and

element contents were matched with the formula of

molecular complex (R)-7?(
)-2. 1H NMR (53.8 mg in

0.5 mL of CDCl3): d 8.37 (d, 1H, J = 4.2 Hz), 7.84 (d, 2H,

J = 9.0 Hz), 7.81 (d, 2H, J = 8.4 Hz), 7.52 (dt, 1H,

J = Hz), 7.29 (d, 4H, J = 9.0 Hz), 7.22 (m, 9H), 7.06 (m,

2H), 3.36 (dd, 2H, J = 60.0, 13.2 Hz), 3.02 (m, 1H), 2.97

(d, 1H, J = 11.4 Hz), 2.81 (d, 1H, J = 10.8 Hz), 2.13 (t, 1H,

10.8 Hz), 1.95 (dt, 1H, J = 11.4, 3.0 Hz), 1.86 (d, 1H,

J = 11.4 Hz), 1.68 (m, 2H), 1.47 (qd, 1H, J = 12.0, 7.5 Hz).

Elemental analysis result: calculated for C37H34N2O2:

C, 82.50; H, 6.36; N, 5.20. Found: C, 82.41; H, 6.45; N,

5.19.

10. Enantiomeric excess analysis was performed with nhexane/

isopropanol/diethylamine (in volume ratio of 70/

30/0.1) as mobile phases. The retention time of (
)-2 was

5.5 min, while that of (+)-2 was 6.6 min under a flow

velocity of 1.0 mL/min.

11. The 1H NMR spectra of the molecular complex of (R)-7

and (
)-2 were measured in concentration of 53.8 mg of

the molecular complex in 0.6 mL of CDCl3, and that of

(S)-7 and (
)-2 in concentration of 25.2 mg of (
)-2 and

28.6 mg of (S)-7 in 0.6 mL of CDCl3.

12. Besides protons on benzyl methylene, few of other protons

also showed a few differences in chemical shifts between

two diastereomeric complexes.

13. (a) Sullivan, G. R.; Dale, J. A.; Mosher, H. S. J. Org.

Chem. 1973, 38, 2143–2147; (b) Dale, J. A.; Dull, D. L.;

Mosher, H. S. J. Org. Chem. 1969, 34, 2543–2549; Review:

(c) Seco, J. M.; Quin?oa′ , E.; Riguera, R. Chem. Rev. 2004,

104, 17–117, and references cited therein.

14. (a) Hoye, T. R.; Renner, M. K. J. Org. Chem. 1996, 61,

8480–8495; (b) Hoye, T. R.; Renner, M. K. J. Org. Chem.

1996, 61, 2056–2064.

15. (a) Kanger, T.; Kriis, K.; Pehk, T.; Mu¨u¨ risepp, A.-M.;

Lopp, M. Tetrahedron: Asymmetry 2002, 13, 857–865; (b)

Davis, H. M. L.; Hodges, L. M. J. Org. Chem. 2002, 67,

5683–5689; (c) Itoh, T.; Nagata, K.; Miyazaki, M.;

Kameoka, K.; Ohsawa, A. Tetrahedron 2001, 57, 8827–

8839; (d) Beshore, D. C.; Dinsmore, C. J. Tetrahedron

Lett. 2000, 41, 8735–8739; (e) Dinsmore, C. J.; Zartman,

C. B. Tetrahedron Lett. 2000, 41, 6309–6312; (f) Davies, H.

M. L.; Matasi, J. J.; Hodges, L. M.; Huby, N. J. S.;

Thornley, C.; Kong, N.; Houser, J. H. J. Org. Chem. 1997,

62, 1095–1105; (g) Huang, X.; Fujioka, N.; Pescitelli, G.;

Koehn, F. E.; Williamson, R. T.; Nakanishi, K.; Berova,

2146 C.-Q. Kang et al. / Tetrahedron: Asymmetry 16 (2005) 2141–2147

N. J. Am. Chem. Soc. 2002, 124, 10320–10335; (h)

Chalard, P.; Bertrand, M.; Canet, I.; The′ry, V.; Remuson,

R.; Jeminet, G. Org. Lett. 2000, 2, 2431–2434.

16. Both amides were composed of two inseparable rotamers

from hindered rotation around amido C–N bond. The

rotamers showed two different groups of peaks and one

predominated over another.

17. The rotamer ratios of Mosher amides measured by NMR

after purification were consistent with those before

purification.

C.-Q. Kang et al. / Tetrahedron: Asymmetry 16 (2005) 2141–2147 2147