Lipase-catalyzed kinetic resolution as key step in the synthesis of enantiomerically pure σ ligands with 2-benzopyran structure

Article abstract of DOI:10.1016/j.bmc.2017.04.042

In order to obtain enantiomerically pure σ₁ receptor ligands with a 2-benzopyran scaffold an Oxa-Pictet-Spengler reaction with the enantiomerically pure 2-phenylethanol derivatives (R)-4 and (S)-4 was envisaged. The kinetic resolution of racemic alcohol (±)-4 using Amano Lipase PS-C II and isopropenyl acetate in tert-butyl methyl ether led to the (R)-configured alcohol (R)-4 in 42% yield with an enantiomeric excess of 99.6%. The (S)-configured alcohol (S)-4 was obtained by Amano Lipase PS-C II catalyzed hydrolysis of enantiomerically enriched acetate (S)-5 (76.9% ee) and provided (S)-4 in 26% yield and 99.7% ee. The absolute configuration of alcohol (R)-4 was determined by exciton coupled CD spectroscopy of the bis(bromobenzoate) (R)-7. The next important step for the synthesis of 2-benzopyrans 2 and 3 was the Oxa-Pictet-Spengler reaction of the enantiomerically pure alcohols (R)-4 and (S)-4 with piperidone ketal 8 and chloropropionaldehyde acetal 12. The conformationally restricted spirocyclic 2-benzopyrans 2 revealed higher σ₁ affinity than the more flexible aminoethyl derivatives 3. The (R)- and (R,R)-configured enantiomers (R)-2 and (R,R)-3 represent the eutomers of this class of compounds with eudismic ratios of 4.8 (2b) and 4.5 (2c). High σ₁/σ₂ selectivity (>49) was found for the most potent σ₁ ligands (R)-2b, (R)-2c, (R)-2d, and (S)-2d (K_i(σ₁) 9–15?nM).

Full text of DOI:10.1016/j.bmc.2017.04.042

Accepted Manuscript
Lipase-catalyzed kinetic resolution as key step in the synthesis of enantiomeri-
cally pure σ ligands with 2-benzopyran structure
Inga Knappmann, Kirstin Lehmkuhl, Jens Köhler, Dirk Schepmann, Martin
Giera, Franz Bracher, Bernhard Wünsch
PII:
S0968-0896(16)31056-2
DOI:
http://dx.doi.org/10.1016/j.bmc.2017.04.042
BMC 13712
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
24 October 2016
31 March 2017
21 April 2017
Please cite this article as: Knappmann, I., Lehmkuhl, K., Köhler, J., Schepmann, D., Giera, M., Bracher, F.,
Wünsch, B., Lipase-catalyzed kinetic resolution as key step in the synthesis of enantiomerically pure σ ligands with
2-benzopyran structure, Bioorganic & Medicinal Chemistry (2017), doi: http://dx.doi.org/10.1016/j.bmc.
2017.04.042
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1
Lipase-catalyzed kinetic resolution as key step in the synthesis of
enantiomerically pure σ ligands with 2-benzopyran structure
Inga Knappmann,a Kirstin Lehmkuhl,a Jens Köhler,a Dirk Schepmann,a Martin Giera,b
Franz Bracher,c Bernhard Wünsch*a,d
a
Institut für Pharmazeutische und Medizinische Chemie der Universität Münster,
Corrensstraße 48, D-48149 Münster, Germany
Tel.: +49-251-8333311; Fax: +49-251-8332144; E-mail: wuensch@uni-muenster.de
b
Leiden University Medical Center, Center for Proteomics and Metabolomics,
Albinusdreef 2, 2300RC Leiden, the Netherlands
c Department Pharmazie, Ludwig-Maximilians-Universität München, Butenandtstr. 5,
D-81377 München, Germany
d
Cells-in-Motion Cluster of Excellence (EXC 1003 – CiM), Westfälische Wilhelms-
Universität Münster, Germany
Dedicated to Prof. Dr. Dr. h.c. Gottfried Blaschke on the occasion of his 80th birthday
Abstract
In order to obtain enantiomerically pure σ1 receptor ligands with a 2-benzopyran
scaffold an Oxa-Pictet-Spengler reaction with the enantiomerically pure 2-
phenylethanol derivatives (R)-4 and (S)-4 was envisaged. The kinetic resolution of
racemic alcohol ( )-4 using Amano Lipase PS-C II and isopropenyl acetate in tert-
butyl methyl ether led to the (R)-configured alcohol (R)-4 in 42 % yield with an
enantiomeric excess of 99.6 %. The (S)-configured alcohol (S)-4 was obtained by
Amano Lipase PS-C II catalyzed hydrolysis of enantiomerically enriched acetate (S)-

2
5 (76.9 % ee) and provided (S)-4 in 26 % yield and 99.7 % ee. The absolute
configuration of alcohol (R)-4 was determined by exciton coupled CD spectroscopy of
the bis(bromobenzoate) (R)-7. The next important step for the synthesis of 2-
benzopyrans 2 and 3 was the Oxa-Pictet-Spengler reaction of the enantiomerically
pure alcohols (R)-4 and (S)-4 with piperidone ketal 8 and chloropropionaldehyde
acetal 12. The conformationally restricted spirocyclic 2-benzopyrans 2 revealed
higher σ1 affinity than the more flexible aminoethyl derivatives 3. The (R)- and (R,R)-
configured enantiomers (R)-2 and (R,R)-3 represent the eutomers of this class of
compounds with eudismic ratios of 4.8 (2b) and 4.5 (2c). High σ1/σ2 selectivity (>49)
was found for the most potent σ1 ligands (R)-2b, (R)-2c, (R)-2d, and (S)-2d (Ki(σ1) 9-
15 nM).
Key words: σ ligands; lipases; kinetic resolution; Oxa-Pictet-Spengler reaction;
simulation; 2-benzopyrans; spirocyclic compounds; exciton coupled CD
spectroscopy; sterol ∆8/∆7-isomerase; late stage diversification.
1. Introduction
The class of σ receptors consists of two subtypes, which are termed σ1 and σ2
receptor.1-4 Cloning of the σ1 receptor provided a gene encoding for a 223 amino acid
protein with a molecular weight of 25.3 kDa. It is a unique protein, which does not
show any similarity to other mammalian proteins, but shares 30 % similarity with the
yeast enzyme sterol ∆8/∆7-isomerase. However, the σ1 receptor does not have any
sterol isomerase activity.5-7

3
The σ1 receptor is associated with several neuropsychiatric disorders including
depression, psychosis, neuropathic pain, cocaine and alcohol dependence as well as
memory and cognition deficits. Moreover, it plays a role in neurodegenerative
disorders such as Alzheimer’s disease, Parkinson’s disease and amyotrophic
lateralsclerosis.8-10 Therefore the modulation of the σ1 receptor activity is of major
interest for the treatment of these diseases. Moreover, several clinically used drugs
such as haloperidol11 (antipsychotic), fluvoxamine,12 sertraline12 (antidepressants),
opipramol13 (anxiolytic), pentazocine14-16 (analgesic) and donepezil17 (anti-Alzheimer
drug) show high affinity towards the σ1 receptor in addition to their main activity.
Very recently we have reported on the development of the fluorinated σ1 receptor
ligand fluspidine 1 (Figure 1).18,19 The 18-F-labeled (S)-configured fluspidine
enantiomer is currently investigated as positron emission tomography (PET) tracer
for non-invasive imaging of σ1 receptors in the central nervous system of humans
(clinical study). The (R)-configured enantiomer (R)-1 (Ki = 0.57 nM) has higher σ1
affinity than the (S)-enantiomer (S)-1 (Ki = 2.3 nM). However, in animal studies (R)-1
showed an unexpected quasi-irreversible binding to σ1 receptors in the brain without
washout over the time. In contrast, the less affine (S)-enantiomer (S)-1 displayed a
normal uptake into the central nervous system and a normal washout over the
time.18-23
Obviously, the enantiomers of σ1 receptor ligands of type 1, which were obtained by
resolution of the racemic mixture, show different pharmacodynamic and
pharmacokinetic behavior in animal experiments.20-22 In addition to the 2-benzofuran
fluspidine (1) the spirocyclic 2-benzopyrans 2 and the corresponding derivatives 3

4
with a flexible 2-aminoethyl side chain in 1-postion represent promising σ1 receptor
ligands as well.24 A common feature of the lead compounds 1-3 is the side chain with
two C-atoms in 3-position adjacent to the ether O-atom of the ring system. 2-
Benzopyrans 2 and 3 were prepared by an Oxa-Pictet-Spengler reaction of a 2-
phenylethanol derivative with piperidone ketals or haloalkanal acetals.
Figure 1: Potent σ1 receptor ligands with 2-benzofuran and 2-benzpyran
substructure.
The synthesis of tetrahydroisoquinolines by condensation of 2-arylethylamines with
aldehydes or ketones is termed Pictet-Spengler reaction. In the Oxa-Pictet-Spengler
reaction a 2-arylethanol derivative is reacted with an aldehyde, ketone or derivatives
thereof to afford 3,4-dihydro-1H-2-benzopyrans.25-27 This transformation represents a
versatile method for the synthesis of substituted 2-benzopyrans and even larger
fused ring systems. Moreover, intramolecular versions of the Oxa-Pictet-Spengler
reaction have been described,24,28 and it has been used for the stereoselective
synthesis of complex natural products.29-31 Usually it is promoted by Brønsted or
.
Lewis acids, e.g. BF3 OEt2, TMSOTf, and Bi(OTf)3.32
Herein, we report the synthesis of enantiomerically pure spirocyclic and aminoethyl
substituted 2-benzopyrans 2 and 3 by an Oxa-Pictet-Spengler reaction of
enantiomerically pure 2-phenylethanol building blocks (R)-4 and (S)-4 as key step.

5
Synthesis of the enantiomeric alcohols (R)-4 and (S)-4 has already been reported by
enantiodifferentiating reduction of β-ketoesters using H2/tartaric acid modified Raney
nickel33 or a recombinant S. coelicolor A 3(2) strain.34 Very recently, the asymmetric
hydrogenation of β-ketoesters using chiral Ir-35 and Ru-catalyst36 was reported. In
order obtain both enantiomers of the secondary alcohol 4 with high enantiomeric
excess (>98 % ee) with easily accessible reagents, but avoiding metal reagents, a
lipase-catalyzed kinetic resolution of racemic secondary alcohol ( )-4 was envisaged.
2. Synthesis
Scheme 1: Acetylation of secondary alcohols 4 and hydrolysis of acetates 5 including
the definition of the rate constants k1-k4 of the involved reactions.
The starting point of this study was racemic methyl 3-hydroxy-4-phenylbutanoate ( )-
4,24,37,38 which was obtained by acylation of Meldrum’s acid with phenylacetyl
chloride, heating of the product in methanol39 and subsequent NaBH4 reduction.40,41
The racemic ester ( )-5 was prepared by acetylation of ( )-4 with Ac2O (Scheme 1).
2.1. Chiral HPLC
In order to analyze the results obtained by lipase-catalyzed reactions a chiral HPLC
method had to be developed for the simultaneous quantitative determination of (R)-4,
(S)-4, (R)-5, and (S)-5.

6
Figure 2: Chromatogram of a mixture of (R)-4, (S)-4, (R)-5, and (S)-5. Chiral HPLC:
Chiralpak® IB, n-hexane : isopropanol = 98 : 2, flow rate 1.0 mL/min, UV detection λ
= 210 nm, resolution: Rs ((R)-5:(S)-5) = 4.8; Rs ((S)-5:(R)-4) = 10.4; Rs ((R)-4:(S)-4) =
5.1.
In Figure 2 the chromatogram obtained from a mixture containing enantiomeric
alcohols (R)-4 and (S)-4 and enantiomeric acetates (R)-5, and (S)-5 is shown. With
the Chiralpak® IB column the separation of the four compounds was performed within
a very short analysis time of 18 min and excellent resolution of the four components.
This method allowed the simultaneous analysis of the ratio of enantiomers (R)-4 :
(S)-4 and (R)-5 : (S)-5 and moreover, the progress of the transformation during
acetylation of alcohol ( )-4 and hydrolysis of acetate ( )-5.

7
2.2. Lipase-catalyzed acetylation of alcohol 4
In a first screening nine lipases41 were tested in the solvent tert-butyl methyl ether
(TBME) at room temperature to catalyze the acetylation of racemic alcohol ( )-4. The
definition of the lipases used in this study is given in Table 4 in the experimental
section. Due to the higher activity of immobilized lipases (Chirazyme L-5, Lipozyme,
Amano Lipase PS-C II) only 200 mg were used instead of 400 mg of native lipases.
Isopropenyl acetate was employed as irreversible acetylating agent, as less side
reactions were expected compared to vinyl acetate. In order to avoid mechanical
damage of lipases a KPG stirrer was used instead of a magnetic bar.
Table 1:
Lipase-catalyzed enantioselective acetylation of racemic secondary alcohol ( )-4.
amount of
lipase [mg]
time
[h]
conv.a).
[%]
ee ((R)-4) ee ((S)-5)
Entry
Lipase
Solvent
Eb)
[%]
[%]
1
2
Chirazyme L-2
TBME
TBME
TBME
TBME
TBME
TBME
TBME
iPr2O
400
400
200
400
400
200
400
200
200
200
400
200
44
48
12.2
64.6
16
33
44
--
5.2
7.3
1.8
--
Chirazyme L-3
34.0
16.2
--
68.0
20.8
--
3
Chirazyme L-5c)
Chirazyme L-7
4.2
95
4
5
Candida rugose
Lipozymed)
24
3.8
44.2
94.6
96.0
97.1
96.0
87.2
92.8
97.4
7.9
20
20
22
2.4
7.0
13
5.8
2.7
42
62
89
50
16
31
81
6
192
95
23.4
24.4
28.0
2.4
7
Amano Lipase PS
Amano Lipase PS
Amano Lipase PS
Amano Lipase PS
Amano Lipase AK
Amano Lipase AK
8
192
144
671
443
97
toluene
(+H2O)g)
9
10
11
12
Acetone
TBME
iPr2O
6.6
13.5
6.0

8
toluene
(+H2O)g)
13
14
15
16
17
18
Amano Lipase AK
Amano Lipase AK
200
200
200
100
100
100
121
407
17
6.8
4.0
94.0
92.0
93.6
94.4
98.1
98.7
6.7
4.2
50
35
25
Acetone
TBME
Amano Lipase PS-C
IIe)
92.4
54.0
20.2
41.6
101
60
Amano Lipase PS-C
IIe)
iPr2O
18
36
Amano Lipase PS-C
IIe)
Acetone
TBME
480
257
17
127
230
Amano Lipase PS-C
IIe)f)
30
Standard reaction conditions; ( )-4 (192 mg), isopropenyl acetate (0.5 mL, 4.5
equiv.), native lipase (400 mg) or immobilized lipase (200 mg), solvent (10 mL), rt,
KPG stirrer.
a) conv. [%] = conversion [%] = { ee ((R)-4) / [ee ((R)-4) + ee ((S)-5)]} x 100 %.41-43
.
.
b)
E = enantioselectivity = ln [1 – conv (1 + ee ((S)-5))] / ln [1 – conv (1 - ee ((S)-
5))].41-43
c) immobilized on plastic particles.
d) immobilized on anion exchange resin.
e) immobilized on ceramic particles.
f) reaction was performed at -15 °C.
g) the amount of water was approx. 3-5 %.
The results in Table 1 clearly indicate that the lipases Chirazyme L-2, L-3 and, L-5 as
well as Candida rugosa lipase did not provide high enantioselectivity (E-values < 8),
the lipase Chirazyme L-7 did not lead to a conversion of racemic alcohol ( )-4
(entries 1-5).
For optimization of the conversion and enantioselectivity Amano Lipase PS, Amano
Lipase AK and Amano Lipase PS-C II were investigated. At first the conversion and
E-values in different solvents were analyzed. In dry toluene a conversion could not be
observed. Although addition of one drop of water resulted in clean acetylation of (S)-

9
4, the reaction times in the solvents toluene (+ H2O) and acetone were very long
(entries 9, 10, 13, 14, 17). Therefore, these solvents were not considered any longer
even though the enantioselectivity was partly higher than in TBME. The conversion
rates in iPr2O and TBME were very similar. Since for the most promising Amano
Lipase PS-C II the enantioselectivity in TBME was higher than in iPr2O (entries 15,
16), TBME was selected as preferred solvent for the acetylation of ( )-4. All lipases
led to a preferred acetylation of (S)-4.
The highest enantioselectivity (E = 101) and conversion (conv. = 50 %) were
achieved with the Amano Lipase PS-C II in TBME at rt (entry 15). In order to further
improve the E-value the temperature was reduced to -15 °C (entry 18). Although an
excellent enantioselectivity was achieved at -15 °C (E = 230) the very long reaction
time (257 h) and the low conversion (conv. = 30 %) were not appropriate to produce
enantiomerically pure alcohol (R)-4 or acetate (S)-5. Nevertheless, it is remarkable
that the immobilized Amano Lipase PS-C II is still active at such a low temperature.
In literature only few examples are reported, where lipases were used at such low
temperature.44-46
In conclusion, under optimized reaction conditions (Amano Lipase PS-C II,
isopropenyl acetate, TBME, 20 °C) (R)-configured alcohol (R)-4 and (S)-configured
acetate (S)-5 were formed in >90 % ee at a conversion of 50 % (entry 15). Therefore,
upon non-enzymatic hydrolysis of (S)-5 into (S)-4 it should be possible to obtain both
alcohols (R)-4 as well as (S)-4 with >90 % ee and yields of almost 50 % via chiral
resolution of ( )-4. However, it was our aim to produce both alcohols in >99.5 % ee.
A simulation47 of the reaction course indicates that the ee value of alcohol (R)-4

10
should increase significantly during progress of the reaction, whereas the yield
decreases only to a very low extent (Figure 3, see also Figure SI1 in Supporting
Information). Therefore, the preparative scale conversion of ( )-4 was monitored by
HPLC and continued until (R)-4 reached >99.5 % ee. The reaction was carried out for
145 h and after workup (R)-4 was isolated with a yield of 42 % and 99.6 % ee.
Additionally, (S)-5 was obtained with 52 % yield and 76.9 % ee, which should be a
valuable starting material for the production of (S)-4. Hydrolysis of the
enantiomerically enriched acetate (S)-5 should result in a higher ee of (S)-4 than
hydrolysis of racemic ( )-5 assuming that the investigated lipases keep their
enantiopreference in aqueous media.
(R)-4
(R)-4
(S)-5
(S)-5
(R)-4
(R)-4
(S)-5
(S)-5
100
80
100
80
60
60
n [%]
% ee
40
20
0
40
20
0
0
50
100
150
200
0
50
100
150
200
reaction time [h]
reaction time [h]
Figure 3: Simulation of the lipase-catalyzed acetylation of racemic alcohol ( )-4 and
comparison with experimental results. Left: amount of compounds (R)-4 and (S)-5;
right: ee of (R)-4 and (S)-5. The reaction was carried out using ( )-4 (1.70 g), Amano
lipase PS-C II (1.70 g) and isopropenyl acetate (8.5 mL) in TBME (120 mL) and was
stopped after 145 h leading to (R)-4 in 42 % yield with 99.6 % ee (dotted line). Rate
constants used for the simulation: k1 = 45.5, k2 = 1, k3 = 0.00455, k4 = 0.0001
(definition of k1-k4 see Scheme 1).
2.3. Lipase-catalyzed hydrolysis of acetate 5
The enantioselective hydrolysis of the racemic acetate ( )-5 in phosphate buffer pH
7.0 with methanol as cosolvent was investigated using the five most promising

11
lipases from the acetylation reaction. Table 2 displays clearly that Amano lipase AK
and Amano Lipase PS-C II converted ( )-5 with high enantioselectivity (E = 108 and
E = 263). However, the reaction rate of both transformations was rather low (entries 4
and 5). After 123 h, Amano Lipase PS-C II provided (R)-configured acetate (R)-5 and
(S)-configured alcohol (S)-4 in 83.8 % ee and 98.0 % ee, respectively (entry 5).
Table 2
Lipase-catalyzed enantioselective hydrolysis of racemic acetate ( )-5.
conv.a)
[%]
ee ((R)-5) ee ((S)-4)
[%]
amount of
lipase [mg]
Time
[h]
Eb)
Entry
Lipase
[%]
1
2
3
4
Chirazyme L-5c)
Lipozymed)
100
100
200
200
3
1.6
6.6
20
8.3
30
20
1.2
2.0
61
48
74
95
2.8
31.0
95.2
97.7
Amano Lipase PS
Amano Lipase AK
40.7
23.6
108
Amano Lipase PS-C
IIe)
5
100
123
83.8
98.0
46
263
Standard reaction conditions; ( )-5 (118 mg), phosphate buffer pH 7.0 (5 mL), lipase
(200 mg) or immobilized lipase (100 mg), cosolvent methanol (2 mL), rt, magnetic bar
for native lipases (entries 3 and 4), KPG stirrer for immobilized lipases (entries 1, 2,
and 5).
a) conv. [%] = conversion [%] = { ee ((R)-5) / [ee ((R)-5) + ee ((S)-4)]} x 100 %.41-43
.
.
b)
E = enantioselectivity = ln [1 – conv (1 + ee ((S)-4))] / ln [1 – conv (1 - ee ((S)-
4))].41-43
c) immobilized on plastic particles.
d) immobilized on anion exchange resin.
e) immobilized on ceramic particles.
For the preparative scale production of (S)-4 the enantiomerically enriched acetate
(S)-5 (76.9 % ee) obtained from the Amano Lipase PS-C II catalyzed acetylation (see
above) was hydrolyzed in phosphate buffer pH 7.4 using the same enzyme. Again, it

12
was an important feature to find the optimal end point to stop the transformation. A
comparison of the screening reactions shows that the enantioselectivity E of the
Amano Lipase PS-C II for the hydrolysis in phosphate buffer is even higher than for
the acetylation in TBME (Table1 entry 15, Table 2 entry 5). Simulation47 of the
reaction course demonstrates that hydrolysis of enantiomerically enriched acetate
(S)-5 should result in a high ee of alcohol (S)-4, which slightly decreases over the
time, whereas the yield increases during the progress of the reaction (Figure 4, see
also Figure SI2 in Supporting Information). However, the activity of the lipase was
very low leading to very long reaction times. The hydrolysis was stopped after 311 h
leading to (S)-4 in 26 % yield upon workup, but with an excellent enantiopurity of 99.7
% ee.
(S)-4
(S)-4
(S)-5
(S)-5
(S)-4
(S)-4
(S)-5
(S)-5
100
80
100
80
60
60
n [%]
% ee
40
20
0
40
20
0
0
200
400
600
800
1000
0
200
400
600
800
1000
reaction time [h]
reaction time [h]
Figure 4: Simulation of the lipase-catalyzed hydrolysis of enantiomerically enriched
acetate (S)-5 and comparison with experimental results. left: amount of compounds
(S)-4 and (S)-5; right: ee of (S)-4 and (S)-5. The reaction was carried out using (S)-5
(1.06 g, 76.9 % ee), Amano lipase PS-C II (1.0 g) and methanol (22 mL) in
phosphate buffer pH 7.4 (55 mL) and was stopped after 311 h leading to (S)-4 in 26
% yield with 99.7 % ee (dotted line). rate constants used for the simulation: k1 =
0.012, k2 = 0.0001, k3 = 120, k4 = 1 (definition of k1-k4 see Scheme 1).
As an alternative strategy for the production of alcohol (S)-4, enantiomerically
enriched acetate (S)-5 could be hydrolyzed non-enzymatically to afford

13
enantiomerically enriched alcohol (S)-4 followed by a second lipase-catalyzed
acetylation and a second non-enzymatic hydrolysis. This kind of kinetic resolution,
however, comprises two additional reaction steps, i.e. non-enzymatic hydrolysis
using H2SO4 in methanol and provided (S)-4 in an overall yield of only 7 % with 99.2
% ee. These results demonstrate that the lipase-catalyzed hydrolysis of
enantiomerically enriched acetate (S)-5 is the most advantageous strategy.
2.4. Determination of the absolute configuration
According to the rule of Kazlauskas42,48 the size of the two additional substituents at
the secondary alcohol define the selectivity of the lipase during acetylation and
hydrolysis. On condition the larger substituent has higher priority according to the CIP
rules than the smaller substituent the (R)-enantiomer is transformed preferentially. It
is assumed that the benzyl moiety of alcohol 4 and acetate 5 is larger than the
methoxycarbonylmethyl moiety. Thus, the (S)-configured enantiomers (S)-4 and (S)-5
should be acetylated and hydrolyzed preferentially, since the larger benzyl moiety
has a lower CIP priority than the smaller methoxycarbonylmethyl group. The results
of our experiments correlate nicely with this rule.
The synthesis of the enantiomeric alcohols (R)-4 and (S)-4 has been reported (see
Introduction).33-36 In the first paper a specific optical rotation [α]D = -8.6 (c = 1.2,
20
CHCl3, optical yield 88 %). was given for (R)-configured alcohol (R)-4.33 The second
paper reports [α]D20 = +1.2 (c = 0.9, CH2Cl2) for (S)-4.34 An enantiomeric excess of 85
% was given for this sample.34 In the very recent publication,35 the (R)-configured β-
hydroxyester (R)-4 was obtained by asymmetric hydrogenation of the corresponding
β-ketoester. The enantiomeric excess of (R)-4 was determined by chiral HPLC (98 %
25
ee, [α]D = -10.0 (c = 1.0, CHCl3). The enantiomeric alcohols (R)-4 and (S)-4

14
prepared by lipase-catalyzed kinetic resolution in this work possess the following
20
20
specific rotations: (R)-4: [α]D = +9.1 (c = 1.0, CH3OH, 99.6 % ee); (S)-4: [α]D = -
8.8 (c = 1.0, CH3OH, 99.7 % ee). The sign of the optical rotation is opposite to the
reported one. Since the specific optical rotation depends considerably on the solvent,
the measurement of the optical rotation of the (S)-configured alcohol (S)-4 was
20
repeated in the solvent CH2Cl2, which resulted in [α]D = +10.3 (c = 1.63, CH2Cl2,
99.7 % ee). The plus-sign of prepared (S)-4 correlates with the plus-sign reported for
(S)-4 (CH2Cl2)34 and the minus-sign determined for (R)-4 (CH2Cl2).35 However, a
proof of the absolute configuration of the prepared alcohols is missing in the reports.
Scheme 2: Synthesis of bis(4-bromobenzoate) (R)-7.
Reagents and reaction conditions: (a) LiBH4, THF, rt, 16 h, 87 %. (b) 4-bromobenzoyl
chloride, pyridine, CH2Cl2, rt, 16 h, 76 %.
In order to prove the absolute configuration of the enantiomeric alcohols (R)-4 and
(S)-4 unequivocally, exciton-coupled CD spectroscopy of a bis(4-bromobenzoate)
should be used. For this purpose (R)-4 was reduced with LiBH4 to afford the diol (R)-
6, which was acylated with 4-bromobenzoyl chloride to provide bis(4-bromobenzoate)

15
(R)-7 (Scheme 2). The UV- and CD-spectra of the bis(4-bromobenzoate) (R)-7 were
recorded in n-hexane (Figure 5). The exciton-coupled CD spectrum of (R)-7 displays
a negative Cotton effect at longer wave length (λ = 252 nm) and a positive Cotton
effect at shorter wave length (λ = 235 nm). The position of the positive and negative
Cotton effects in this exciton coupled CD spectrum proves unequivocally (R)-
configuration of the center of chirality of (R)-7 and thus (R)-configuration of the
starting alcohol (R)-4 as well.49
Figure 5: CD and UV spectra (n-hexane) of the bis(4-bromobenzoate) (R)-7.
2.5. Synthesis of enantiomerically pure σ ligands by Oxa-Pictet-Spengler
reaction
The Oxa-Pictet-Spengler reaction of racemic alcohol ( )-4 with piperidone ketal 8 and
the subsequent transformation of ( )-9 into racemic spirocyclic σ ligands ( )-2 have

16
already been reported.24 For the synthesis of the enantiomerically pure spirocyclic
compounds (R)-2 and (S)-2 the same reaction conditions were applied (Scheme 3).
.
The BF3 OEt2 catalyzed Oxa-Pictet-Spengler reaction of enantiomerically pure
alcohols (R)-4 and (S)-4 with piperidone dimethyl ketal 8 led to the spirocyclic
piperidines (R)-9 and (S)-9. The ester moiety was reduced with LiBH4 and the N-
acetyl protecting group of the resulting alcohols (R)-10 and (S)-10 was cleaved off by
NaOH. The piperidine derivatives (R)-11 and (S)-11 represent valuable building
blocks, which allow the introduction of diverse substituents at the very end of the
synthesis.50,51 Herein the secondary amines (R)-11 and (S)-11 were alkylated with
bromoalkanes or reductively alkylated with aldehydes and NaBH(OAc)3 to yield the
enantiomerically pure spirocyclic ligands (R)-2a-d and (S)-2a-d.
Scheme 3: Synthesis of enantiomerically pure spirocyclic σ ligands.
.
Reagents and reaction conditions: (a) BF3 OEt2, CH2Cl2, rt, 7 d, 78 % ((R)-9), 91 %
((S)-9). (b) LiBH4, THF, rt, 16 h, 47 % ((R)-10), 37 % ((S)-10). (c) NaOH, 100 °C, 3 h,
53 % ((R)-11), 78 % ((S)-11). (d) CH3(CH2)7-Br or Ph(CH2)3-Br or C6H11CH2-Br,

17
CH3CN, K2CO3, reflux. (e) PhCH=O, NaBH(OAc)3), CH2Cl2, rt. In the Scheme the
transformations are shown for (R)-configured compounds, the corresponding (S)-
configured enantiomers were prepared in the same manner.
The Oxa-Pictet-Spengler reaction of (R)-4 with chloropropionaldehyde acetal 12 and
.
BF3 OEt2 provided the 1-(chloroethyl) substituted 2-benzopyran (R,R)-13 with high
diastereoselectivity,24 only the thermodynamically more stable cis-configured
diastereomer (R,R)-13 could be isolated (Scheme 4). The last step of the synthesis
comprises nucleophilic substitution of (R,R)-13 with various amines leading to diverse
1-(aminoethyl) substituted 2-benzopyrans (R,R)-3a and (R,R)-3b. The enantiomeric
2-benzopyrans (S,S)-3a and (S,S)-3b were prepared in the same manner starting
with (S)-configured hydroxyester (S)-4.
Scheme 4
Scheme 4: Synthesis of enantiomerically pure aminoethyl substituted 2-benzopyrans.
.
Reagents and reaction conditions: (a) BF3 OEt2, CH2Cl2, rt, 5 d, 32 % ((R,R)-13), 27
% ((S,S)-13). (b) piperidine or 4-phenylppiperidine, Bu4NI (cat.), K2CO3, CH3CN,
reflux, 20 h.
The enantiomeric purity of the final products was analyzed exemplarily for the benzyl
substituted spirocyclic piperidine enantiomers (R)-2b and (S)-2b. Among the different

18
chiral stationary phases, only the column Chiralcel OJ-H allowed the separation of
the enantiomers. The chromatograms of both enantiomers did not show significant
signals for the other enantiomer ((ee > 98 %, see Supporting Information) indicating
negligible racemization during the synthesis. Comparable enantiopurity results for the
test compounds 2a-d and 3a-b, since they were prepared according to the same
method. The opposite values for the specific rotation of enantiomeric pairs (see
Supporting Information) confirm the high enantiomeric purity of the test compounds.
3. Biological activity
The σ receptor affinity of the enantiomerically pure spirocyclic 2-benzopyrans (R)-2a-
d and (S)-2a-d as well as the aminoethyl substituted benzopyrans (R,R)-3a,b and
(S,S)-3a,b was investigated with radioligand receptor binding studies. In the σ1 assay
guinea pig brain homogenates were used as receptor material and [3H]-(+)-
pentazocine as σ1 selective radioligand. Rat liver homogenates served as receptor
source in the σ2 assay. Since a σ2 selective radioligand for labeling of σ2 receptors is
not commercially available, [3H]-di-o-tolylguanidine was used as radioligand. In order
to achieve σ2 selectivity an excess of (+)-pentazocine was added to mask σ1
receptors.52-54

19
Table 3
σ1 and σ2 receptor affinity of enantiomerically pure 2-benzopyrans.
Ki SEM (nM)a)
σ1/σ2
selectivity
eudismic
ratio
(σ2 affinity)
eudismic
ratio
(σ1 affinity)
compd.
R
σ1
σ2
(R)-2a
(S)-2a
(R)-2b
(S)-2b
(R)-2c
(S)-2c
(R)-2d
(S)-2d
(R,R)-3a
(S,S)-3a
n-octyl
24
4
6
4
68 29
260 50
21 % *
2.9
4.6
2.3
4.8
4.5
0.9
1.5
3.8
-
n-octyl
56
10
PhCH2
>100
>20
79
PhCH2
48 15
15
68 14
9 % *
Ph(CH2)3
Ph(CH2)3
C6H11CH2
C6H11CH2
piperidin-1-yl
piperidin-1-yl
4
1220 445
519 122
510 224
614 241
13 % *
2.4
1.2
-
7.6
10
1
49
9.0
2
69
556 184
860 188
>1.7
>1.6
39 %*
4-phenyl-
(R,R)-3b
(S,S)-3b
33
9
525 130
32 19
16
piperidin-1-yl
4-phenyl-
piperidin-1-yl
1.3
0.06
43 17
0.7
(+)-pentazocine
Haloperidol
4.2 1.1
3.9 1.5
61 18
-
78 2.0
42 8.1
di-o-tolylguanidine
a) The reported Ki-values are mean values of three independent experiments (n =
3). For low affinity compounds, values in % (marked with an asterisk*) are given,
indicating the inhibition of specific radioligand binding at a test compound
concentration of 1 µM.

20
Table 3 displays clearly that (R)-configured spirocyclic 2-benzopyrans (R)-2 and
(R,R)-configured aminoethyl substituted 2-benzopyrans (R,R)-3 show higher σ1
affinity than their enantiomers (S)-2 and (S,S)-3, although the low eudismic ratios of
2d, 3a and 3b are not significant. The highest eudismic ratios were found for the
most potent benzyl and phenylpropyl substituted spirocyclic 2-benzopyrans 2b (4.8)
and 2c (4.5). They result from the high σ1 affinity of (R)-2b (Ki = 10 nM) and (R)-2c
(Ki = 15 nM). The cyclohexylmethyl substituted spirocyclic compound (R)-2d has high
σ1 affinity (Ki = 10 nM) as well, but due to the increased σ1 affinity of the (S)-
configured enantiomer (S)-3d (Ki = 9 nM) the eudismic ratio is reduced.
In general, the more rigid 2-benzopyrans 2 show higher σ1 affinity than the 2-
benzopyrans 3 with a flexible 2-aminoethyl side chain. In particular the spirocyclic
compound (R)-2c with a phenylpropyl substituent (Ki = 15 nM) and the 2-aminoethyl
derivative with a 4-phenylpiperidine moiety (R,R)-3b (Ki = 33 nM) represent
constitutional isomers considering the amino moiety bearing substituents at 1-
position of the 2-benzopyran ring, but differ considerably in their σ1 affinity.
With exception of (S,S)-3b (Ki(σ2) = 32 nM) all ligands show high selectivity (e.g. (R)-
2b, (R)-2c, (R)-2d, (S)-2d) or at least a preference (e.g. (R)-2a, (S)-2a, (R,R)-3a) for
the σ1 receptor over the σ2 subtype. The highest σ1/σ2 selectivity was found for the
very potent σ1 ligand (R)-2b: (Ki(σ1) = 10 nM, σ1/σ2 = 100.
Since some ligands for σ1 receptors and the phencyclidine (PCP) binding site within
the channel pore of the N-methyl-D-aspartate (NMDA) receptor are often closely
related,55-57 the PCP affinity of the 2-benzopyrans 2 and 3 was also investigated in

21
this study. However, at a concentration of 10 µM the compounds did not compete
considerably with the radioligand [3H]-(+)-MK-80155,58 indicating IC50 values higher
than 10 µM. Obviously the 2-benzopyrans 2 and 3 with a basic substituent are highly
selective for the σ1 receptor over the PCP binding site of the NMDA receptor.
Due to the structural similarity of the σ1 receptor with the yeast enzyme sterol-∆8/∆7-
isomerase,5-7 the inhibition of this enzyme by racemic 2-benzopyrans ( )-2a, ( )-2c
and ( )-3b was investigated exemplarily in a cellular assay.59 Even at a concentration
of 4 µg/mL these compounds did not at all inhibit the growth or change the sterol
pattern of the model yeast Yarrowia lipolytica. This finding indicates the high
selectivity for the σ1 receptor over the fungal sterol-∆8/∆7-isomerase. Further, the
compounds were subjected to a cellular assay for inhibition of post-squalene
enzymes in cholesterol biosynthesis.60 In HL-60 cells no changes in sterol pattern
were detected at a test concentration of 1 µM. At an extremely high concentration of
50 µM, a slight accumulation of cholesta-8,14-dienol was observed (see Supporting
Information, Figure SI3), indicating a very weak inhibition of human sterol-∆8/∆7-
isomerase.
4. Conclusion
The Oxa-Pictet-Spengler reaction of enantiomerically pure alcohols (R)-4 and (S)-4
was employed for the synthesis of enantiomerically pure spirocyclic and flexible
aminoethyl substituted σ1 ligands (R)-2a-d, (S)-2a-d, (R,R)-3a,b, and (S,S)-3a,b
based on a 2-benzopyran scaffold. The synthesis of the enantiomerically pure
alcohols (R)-4 and (S)-4 was performed by lipase-catalyzed kinetic resolution. The
absolute configuration of prepared (R)-4 was assigned by exciton-coupled CD-
spectroscopy of its bis(4-bromobenzoate) derivative (R)-7. In general (R)-configured

22
spirocyclic compounds (R)-2 and (R,R)-configured aminoethyl substituted derivatives
(R,R)-3 reveal higher σ1 affinity than their enantiomers. The very potent σ1 ligand (R)-
2c (Ki = 15 nM) is 4.5-fold more affine than its enantiomer (S)-2c and shows high
selectivity over the σ2 subtype, the PCP binding site of the NMDA receptor and the
sterol-∆8/∆7-isomerases in both yeast and human cells.
5. Experimental
5.1. Chemistry, General Methods
Unless otherwise noted, reactions were conducted under dry nitrogen in absolute
solvents. THF was dried with sodium/benzophenone and was freshly distilled before
use.
5.2. Instruments
Thin layer chromatography (tlc): Silica gel 60 F254 plates (Merck). Flash
chromatography (fc): Silica gel 60, 40–64 µm (Merck); parentheses include: diameter
of the column, eluent, fraction size, Rf value. Melting point: Melting point apparatus
SMP 3 (Stuart Scientific), uncorrected. MS: MAT GCQ (Thermo-Finnigan); IR: IR
1
spectrophotometer 480Plus FT-ATR-IR (Jasco). H NMR (400 MHz): Unity Mercury
Plus 400 spectrometer (Varian); δ in ppm related to tetramethylsilane; coupling
constants are given with 0.5 Hz resolution. Optical rotation: Polarimeter 341 (Perkin
Elmer); 1.0 dm tube; concentration c in g/100 mL; T = 20 °C; wavelength 589 nm (D-
T
line of Na light); the unit of the specific rotation ([α]D deg.mL.dm-1.g-1) is omitted for
clarity.

23
5.3. HPLC method for the determination of the purity
Merck Hitachi Equipment; UV detector: L-7400; autosampler:L-7200; pump: L-7100;
degasser: L-7614; column: LiChrospher® 60 RP-select B (5 µm); LiCroCART® 250-4
mm cartridge; flow rate: 1.00 mL/min; injection volume: 5.0 µL; detection at λ = 210
nm; solvents: A: water with 0.05% (v/v) trifluoroacetic acid; B: acetonitrile with 0.05%
(v/v) trifluoroacetic acid: gradient elution: 0.0 min: 90.0 % of A, 10.0 % of B; 4.0 min:
90.0 % of A, 10.0 % of B; 29.0 min: 0.0 % of A, 100.0 % of B; 31.0 min: 0.0 % of A,
100.0 % of B; 31.5 min: 90.0 % of A, 10.0 % of B; 40.0 min: 90.0 % of A, 10.0 % of B.
5.4. HPLC method for the determination of enantiomeric excess
Merck Hitachi Equipment; pump: L-6200A; UV detector, autosampler, degasser: as
described in Part 5.3; column: Chiralpak® IB (5 µm); temperature: 20 °C; flow rate:
1.000 mL/min; injection volume: 20.0 µL; detection at λ = 210 nm; isocratic elution,
eluent: n-hexane : isopropanol 98 : 2.
5.5. Lipases used in this project
Table 4: Characterization of lipases used in this project
Name
Chirazyme L-2
Supplier
Roche
origin
preparation
lyophilized
lyophilized
Candida antarctica
fraction B (fungus)
Candida rugosa
(fungus)
Chirazyme L-3, pur.
Chirazyme L-5
Roche
immobilized on plastic
particles
Candida antarctica
fraction A (fungus)
Roche
Pig pancreas
(mammal)
Chirazyme L-7
Roche
solid
solid
Candida rugosa
(fungus)
Candida rugose Lipase
Lipozyme
Fluka
immobilized on anion
exchange resin
Mucor miehei
((thermophilic fungus)
Fluka
Burkholderia cepacia
(bacterium)
Amano Lipase PS
Amano Lipases Inc.
solid

24
Pseudomonas
fluorescens (bacterium)
Amano Lipase AK
Amano Lipases Inc.
solid
immobilized on
ceramic particles
Burkholderia cepacia
(bacterium)
Amano Lipase PS- C II
Sigma-Aldrich
5.6. Synthetic procedures
5.6.1. ( )-Methyl 3-acetoxy-4-phenylbutanoate (( )-5)
A solution of racemic alcohol ( )-4 (50 mg, 0.26 mmol) NEt3 (0.125 mL, 0.9 mmol),
Ac2O (0.055 mL, 0.56 mmol) and 4-(dimethylamino)pyridine (5.1 mg, 0.042 mmol) in
CH2Cl2 abs (2 mL) was stirred at rt for 3 h. Et2O (2 mL) was added and the mixture
was washed with 0.5 M HCl (2 mL) and H2O (2 mL). The aqueous layer was
extracted with Et2O (2 x 2 mL) and the combined organic layer was dried (Na2SO4),
concentrated in vacuo and the residue was purified by fc (2 cm, cyclohexane : ethyl
acetate = 9 : 1, 20 mL, Rf = 0.16). Colorless oil, yield 33 mg (54 %). C13H16O4
(236.3). MS (ESI): m/z = 236 [M]. IR: ν [cm-1] = 2952 (C-H), 1735 (C=O), 1027 (C-O),
1
700 (C-H). H NMR (CDCl3): δ [ppm] = 2.00 (s, 3H, COCH3), 2.53 (dd, J = 15.7/5.7
Hz, 1H, CH2CO2CH3), 2.57 (dd, J = 15.7/7.2 Hz, 1H, CH2CO2CH3), 2.87 (dd, J =
13.7/6.8 Hz, 1H, PhCH2), 2.98 (dd, J = 13.7/6.4 Hz, 1H, PhCH2) , 3.65 (s, 3H,
CO2CH3), 5.37 – 5.45 (m, 1H, CHOAc), 7.18 – 7.32 (m, 5H, arom.).
5.6.2. Methyl (R)-3-hydroxy-4-phenylbutanoate ((R)-4) and
Methyl (S)-3-acetoxy-4-phenylbutanoate ((S)-5)
A mixture of racemic alcohol ( )-4 (1.70 g, 8.75 mmol), isopropenyl acetate (8.5 mL)
and Amano Lipase PS-C II (1.70 g) in tert-butyl methyl ether (120 mL) was shaken at
rt for 145 h. The immobilized enzyme was filtered off and the solvent was removed in
vacuo. Alcohol (R)-4 and acetate (S)-5 were separated by fc (3 cm, at first
cyclohexane : ethyl acetate = 90 : 10, 20 mL, Rf = 0.16 (S)-5, then cyclohexane :
ethyl acetate = 80 : 20, 20 mL, Rf = 0.29 (R)-4).

25
(S)-5: Colorless oil, yield 1.065 g, (52 %); ee = 76.9 %.
20
(R)-4: Colorless oil, yield 719.6 mg, (42 %); [α]D = +9.13 (c = 1.00, CH3OH); ee =
99.6 %.
5.6.3. Methyl (S)-3-hydroxy-4-phenylbutanoate ((S)-4)
Enantiomerically enriched acetate (S)-5 (76.9 % ee, 1.06 g, 4.5 mmol) was dissolved
in MeOH (22 mL) and phosphate buffer pH 7.4 (55 mL). Amano Lipase PS-C II (1.0
g) was added and the mixture was shaken at rt for 13 d. The immobilized enzyme
was filtered off and the filtrate was extracted with Et2O (5 x 25 mL). The combined
organic layer was dried (Na2SO4), concentrated in vacuo and the residue was
purified by fc (3 cm, at first: cyclohexane : ethyl acetate = 90 : 10, then cyclohexane :
20
ethyl acetate = 80 : 20, 20 mL, Rf = 0.29). Colorless oil, yield 224 mg, (26 %); [α]D
= -8.79 (c = 1.00, MeOH); [α]D20 = +10.3 (c = 1.63, CH2Cl2); ee = 99.7 %.
5.6.4. (R)-4-Phenylbutane-1,3-diol ((R)-6)
Under N2 and cooling with ice, 2 M LiBH4 (1 mL in THF, 2 mmol) was added to a
solution of (R)-4 (97.1 mg, 0.5 mmol) in THFabs (8 mL). The mixture was stirred at rt
for 16 h. Then it was concentrated in vacuo and the residue was dissolved in H2O (8
mL) and 0.5 M HCl (2 mL). The solution was extracted with CH2Cl2 (3 x 8 mL). The
combined organic layer was dried (Na2SO4), concentrated in vacuo and the residue
was purified by fc (2.5 cm, cyclohexane : ethyl acetate = 50 : 50, 10 mL, Rf = 0.29).
Colorless oil, yield 72.0 mg (87 %). C10H14O2 (166.2). MS (ESI): m/z = 166 [M]. IR: ν
[cm-1] = 3366 (O-H); 2917 (C-H); 1330, 1280 (O-H). 1H NMR (CDCl3): δ [ppm] = 1.72
– 1.80 (m, 2H, CH2CH2OH), 2.38 (s, broad, 2H, OH), 2.76 (dd, J = 13.3/7.7 Hz, 1H,
PhCH2), 2.82 (dd, J = 13.3/5.2 Hz, 1H, PhCH2), 3.78 – 3.92 (m, 2H, CH2CH2OH),
20
4.05 – 4.13 (m, 1H, PhCH2CHOH), 7.20 – 7.35 (m, 5H, arom.). [α]D = -15.3 (c =

26
1.10, MeOH).
5.6.5. (R)-(4-Phenylbutane-1,3-diyl) bis-(4-bromobenzoate) ((R)-7)
A solution of diol (R)-6 (63.8 mg, 0.38 mmol), 4-bromobenzoyl chloride (301 mg, 1.2
mmol) and pyridine (0.01 mL, 0.1 mmol) in CH2Cl2 abs (8 mL) was stirred at rt for 16 h.
0.5 M HCl was added (pH 2), the organic layer was separated and the aqueous layer
was extracted with CH2Cl2 (3 x 10 mL). The combined organic layer was dried
(Na2SO4), concentrated in vacuo and the residue was purified by fc (3 cm,
cyclohexane : ethyl acetate = 90 : 10, 10 mL, Rf = 0.35). Colorless oil, yield 153 mg
(76 %). C24H20Br2O4 (532.2). MS (EI): m/z = 532 [M], 185/183 COC6H4Br. IR: ν [cm-1]
1
= 3027, 2922 (C-H), 1714 (CO). H NMR (CDCl3): δ [ppm] = 2.11 - 2.18 (m, 2H,
CH2Ph), 2.97 - 3.13 (m, 2H, CHCH2CH2), 4.32 - 4.47 (m, 2H, CH2CH2O), 5.48 - 5.55
(m, 1H, CH2CHCH2), 7.19 - 7.31 (m, 5H, Ph), 7.49 - 7.55 (m, 4H, arom.), 7.77 - 7-83
(m, 4H, arom.). [α]D20 = -17.7 (c = 1.32, CH2Cl2).
5.6.6. Methyl (R)-2-(1'-acetyl-3,4-dihydrospiro[[2]benzopyran-1,4'-piperidin]-3-
yl)-acetate ((R)-9)
Under N2, a solution of 8 (1.87 g, 10 mmol) in tolueneabs (2 mL) was added to a
solution of (R)-4 (388 mg, 2.0 mmol) in CH2Cl2 abs (15 mL) and the mixture was stirred
at rt for 30 min. The solution was cooled to 0 °C, BF3•Et2O (2.6 mL, approx. 16 mmol)
was added and the mixture was stirred at rt for 7 d. 10 % NaHCO3 solution (15 mL)
was added, the organic layer was separated and the aqueous layer was extracted
with CH2Cl2 (3 x 10 mL). The combined organic layer was dried (Na2SO4),
concentrated in vacuo and the residue was purified by fc (3 cm, cyclohexane : ethyl
20
acetate = 1 : 1, 20 mL, Rf = 0.18). Pale yellow oil, yield 491.1 mg (78 %). [α]D
=
+28.6 (c = 0.96, MeOH).

27
Methyl
(S)-2-(1'-acetyl-3,4-dihydrospiro[[2]benzopyran-1,4'-piperidin]-3-yl)-
acetate ((S)-9)
As described for the synthesis of (R)-9, (S)-4 (220 mg, 1.13 mmol) was reacted with
8 (1.07 g, 5.7 mmol and BF3•Et2O (1.5 mL, approx. 9 mmol) in CH2Cl2 abs (1 mL).
Pale yellow oil, yield 327.8 mg (91 %). [α]D20 = -30.7 (c = 1.60 , MeOH).
Spectroscopic data of (R)-9 and (S)-9 are identical with the data for ( )-9.24
C18H23NO4 (317.4). MS (EI): m/z = 317 [M], 274 [M-COCH3]. IR: ν [cm-1] = 2925 (C-
1
H), 1736 (C=O ester), 1634 (C=O amide). H NMR (CDCl3): δ [ppm] = 1.64 – 1.75
(m, 2H, N(CH2CH2)2), 1.99 (td, J = 13.2/4.9 Hz, 1H, N(CH2CH2)2), 2.13 (s, 3H,
NCOCH3), 2.09 – 2.20 (m, 1H, N(CH2CH2)2), 2.59 – 3.00 (m, 5H, CH2CO2CH3 (2H),
ArCH2 (2H), N(CH2CH2)2 (1H)), 3.40 – 3.56 (m, 1H, N(CH2CH2)2), 3.59 – 3.68 (m, 1H,
N(CH2CH2)2), 3.73 (s, 3H, CO2CH3), 4.20 – 4.31 (m, 1H, ArCH2CH), 4.50 – 4.58 (m,
1H, N(CH2CH2)2), 7.03 – 7.09 (m, 2H, arom.), 7.14 – 7.21 (m, 2H, arom.).
5.6.7.
(R)-2-(1'-Acetyl-3,4-dihydrospiro[[2]benzopyran-1,4'-piperidin]-3-
yl)ethanol ((R)-10)
Under N2 at 0 °C, LiBH4 solution (2 M in THF, 3 mL 6 mmol) was added to a solution
of (R)-9 (468.5 mg, 1.5 mmol) in THFabs (25 mL). The mixture was stirred at rt for 20
h. Then it was concentrated in vacuo, the residue was diluted with H2O (10 mL), 0.5
M HCl was added (pH 2) and the solution was extracted with CH2Cl2 (4 x 10 mL).
The combined organic layer was dried (Na2SO4), concentrated in vacuo and the
residue was purified by fc (2 cm, CH2Cl2 : MeOH = 95 : 5, 10 mL, Rf = 0.24).
Colorless oil, yield 204.4 mg (47 %). [α]D20 = +46.5 (c = 0.90, MeOH).

28
(S)-2-(1'-Acetyl-3,4-dihydrospiro[[2]benzopyran-1,4'-piperidin]-3-yl)ethanol
((S)-10)
As described for the synthesis of (R)-10, (S)-9 (327 mg, 1.04 mmol) was reduced
with LiBH4 (2 M in THF, 2.1 mL, 4.2 mmol) in THFabs. Colorless oil, yield 111.3 mg
(37 %). [α]D20 = -46.0 (c = 0.71, MeOH).
Spectroscopic data of (R)-10 and (S)-10 are identical with the data for ( )-10.24
C17H23NO3 (289.4). MS (EI): m/z = 289 [M], 246 [M-COCH3]. IR: ν [cm-1] = 3385 (O-
1
H), 2923 (C-H), 1617 (C=O amide). H NMR (CDCl3): δ [ppm] = 1.63 – 2.03 (m, 6H,
CH2CH2OH (2H), N(CH2CH2)2 (4H)), 2.08 (s, 3H, NCOCH3), 2.60 (d, J = 15.9 Hz, 1H,
ArCH2), 2.77 (dd, J = 15.9/11.1 Hz, 1H, ArCH2), 2.81 – 2.92 (m, 1H, N(CH2CH2)2),
3.37 – 3.49 (m, 1H, N(CH2CH2)2), 3.66 – 3.71 (m, 1H, N(CH2CH2)2), 3.79 – 3.91 (m,
2H, CH2OH), 3.92 – 4.04 (m, 1H ArCH2CH), 4.49 – 4.57 (m, 1H, N(CH2CH2)2), 6.97 –
7.06 (m, 2H, arom.), 7.07 – 7.15 (m, 2H, arom.). A signal for the OH proton is not
seen in the spectrum.
5.6.8. (R)-2-(3,4-Dihydrospiro[[2]benzopyran-1,4'-piperidin]-3-yl)ethanol ((R)-11)
A solution of (R)-10 (190.2 mg, 0.66 mmol) in 2 M NaOH (20 mL) was heated under
reflux for 3 h. The mixture was cooled to rt and extracted with CH2Cl2 (3 × 10 mL).
The combined organic layer was dried (K2CO3), concentrated in vacuo and the
residue was used directly without further purification. Colorless oil, yield 131.1 mg,
(53 %), Rf = 0.03 (CH2Cl2 : MeOH = 95 : 5). [α]D20 = +45.5 (c = 0.74, MeOH).
(S)-2-(3,4-Dihydrospiro[[2]benzopyran-1,4'-piperidin]-3-yl)ethanol ((S)-11)
As described for the synthesis of (R)-11, (S)-10 (111 mg, 0.38 mmol) was hydrolyzed
with 2 M NaOH (20 mL). Colorless oil, yield 73.3 mg (78 %). [α]D20 = -47.1 (c = 0.72,

29
MeOH).
Spectroscopic data of (R)-11 and (S)-11 are identical with the data for ( )-11.24
C15H21NO2 (247.3). MS (EI): m/z = 247 [M], 228 [M-H3O+], 199 [M-H3O+-NHCH2], 184
[M-H3O+-NH2CH2CH2]. IR: ν [cm-1] = 3345 (O-H) (N-H), 2937 (C-H), 1488, 1422 (C-
1
H), 1043 (C-O). H NMR (CDCl3): δ [ppm] = 1.62 – 2.11 (m, 6H, NH(CH2CH2)2 (4H),
CH2CH2OH (2H)), 2.55 (dd, J = 15.8/2.6 Hz, 1H, PhCH2), 2.75 (dd, J = 15.8/11.2 Hz,
1H, PhCH2), 2.85 – 3.15 (m, 4H, NH(CH2CH2)2), 3.81 – 3.88 (m, 2H, CH2CH2OH),
3.92 – 4.01 (m, 1H, PhCH2CH), 7.0 (d, J = 7.4 Hz, 1H, arom.) 7.05 – 7.17 (m, 3H,
arom.). Signals for the OH and NH protons are not seen in the spectrum.
5.6.9. (R)-2–(1'-Octyl-3,4-dihydrospiro[[2]benzopyran-1,4'-piperidin]-3-yl)ethanol
((R)-2a)
1-Bromooctane (34.7 mg, 0.18 mmol) and K2CO3 (30.3 mg, 0.22 mmol) were added
to a solution of (R)-11 (36.0 mg, 0.15 mmol) in acetonitrileabs (8 mL). The mixture was
heated under reflux for 4 h. Then it was filtered and the solution was concentrated in
vacuo. The residue was purified by fc (1 cm, cyclohexane : ethyl acetate = 1 : 1 + 0.1
% NH3, 5 mL, Rf = 0.30). Colorless oil, yield 44.9 mg, (83 %). [α]D20 = +36.2 (c = 1.76,
MeOH). Purity determined by HPLC: 98.7 %.
(S)-2–(1'-Octyl-3,4-dihydrospiro[[2]benzopyran-1,4'-piperidin]-3-yl)ethanol
((S)-2a)
As described for the synthesis of (R)-2a, (S)-11 (21.7 mg, 0.09 mmol) was alkylated
with 1-bromooctane (23.2 mg, 0.12 mmol) in the presence of K2CO3 (20.7 mg, 0.15
20
mmol). Colorless oil, yield 22.9 mg (71 %). [α]D = -36.1 (c = 0.88, MeOH). Purity
determined by HPLC: 99.4 %.