EPC syntheses and structure-activity relationships of hypoglycaemic semicyclic amidines

Article abstract of DOI:10.1016/S0223-5234(00)00138-0

A series of homochiral sterically hindered mono- and bicyclic amidines was prepared as hypoglycaemic agents by lethargic reaction of O- methylcaprolactim and 3-ethoxy-2-azabicyclo[2.2.2]oct-2-ene, respectively, with homochiral cis-2-substituted cyclopentane amines provided by asymmetrical reductive amination of racemic 2-substituted cyclopentanones. All compounds, except the cyclohexylmethyl-isoquinuclidone derivative which inhibited secretion at 100 μM, significantly stimulated insulin secretion 2- 8-fold at 10 μM and 100 μM in INS-1 cells. The most potent activator was the 2-cyclopentyl-substituted caprolactam derivative 5e. The stimulatory effects on secretion increased with rising steric hindrance of both the amidine α-carbon and the bicyclic amidine moiety itself. Enantiomeric discrimination was observed for the 2-[(cis-2-bulkysubstituted cyclopentyl)imino]hexahydroazepine halides 5e and 5f and for the 3-[(cis-2- substituted cyclopentyl)imino]-2-azabicyclo[2.2.2]octane halides 6a and 6c. The amidines depolarized INS-1 cells and generated action potentials, accompanied by a decrease of membrane conductance. Simultaneously [Ca²⁺](i) increased, probably due to Ca²⁺-entry through voltage-dependent Ca²⁺- channels. At high concentrations, where inhibition of secretion was observed, [Ca²⁺](i) still rose upon application of the amidines, indicating an additional inhibitory pathway downstream to the elevation of [Ca²⁺](i). Even at high concentrations (100 μM), the amidines had no toxic effects on insulin secreting INS-1 cells. (C) 2000 Editions scientifiques et medicales Elsevier SAS.

Full text of DOI:10.1016/S0223-5234(00)00138-0

Eur. J. Med. Chem. 35 (2000) 377−392
© 2000 Éditions scientifiques et médicales Elsevier SAS. All rights reserved
S0223523400001380/FLA
377
Original article
EPC syntheses and structure–activity relationships
of hypoglycaemic semicyclic amidines
Susanne Hartmanna, Susanne Ullrichb,c, Charlotte Hupferb, August W. Frahma*
aChair of Pharmaceutical Chemistry, University of Freiburg, Hermann-Herder-Str. 9, D-79104 Freiburg, Germany
bDepartment of Physiology; University of Freiburg, D-79104 Freiburg, Germany
Received 1 June 1999; revised 19 August 1999; accepted 9 September 1999
Abstract – A series of homochiral sterically hindered mono- and bicyclic amidines was prepared as hypoglycaemic agents by lethargic
reaction of O-methylcaprolactim and 3-ethoxy-2-azabicyclo[2.2.2]oct-2-ene, respectively, with homochiral cis-2-substituted cyclopentane
amines provided by asymmetrical reductive amination of racemic 2-substituted cyclopentanones. All compounds, except the cyclohexylmethyl-
isoquinuclidone derivative which inhibited secretion at 100 µM, significantly stimulated insulin secretion 2–8-fold at 10 µM and 100 µM in
INS-1 cells. The most potent activator was the 2-cyclopentyl-substituted caprolactam derivative 5e. The stimulatory effects on secretion
increased with rising steric hindrance of both the amidine α-carbon and the bicyclic amidine moiety itself. Enantiomeric discrimination was
observed for the 2-[(cis-2-bulkysubstituted cyclopentyl)imino]hexahydroazepine halides 5e and 5f and for the 3-[(cis-2-substituted
cyclopentyl)imino]-2-azabicyclo[2.2.2]octane halides 6a and 6c. The amidines depolarized INS-1 cells and generated action potentials,
accompanied by a decrease of membrane conductance. Simultaneously [Ca2+]i increased, probably due to Ca2+-entry through voltage-
dependent Ca2+-channels. At high concentrations, where inhibition of secretion was observed, [Ca2+]i still rose upon application of the
amidines, indicating an additional inhibitory pathway downstream to the elevation of [Ca2+]i. Even at high concentrations (100 µM), the
amidines had no toxic effects on insulin secreting INS-1 cells. © 2000 Éditions scientifiques et médicales Elsevier SAS
asymmetric reductive amination / mono- and bicyclic amidines / insulin secretion / INS-1 cells / structure–activity relationship
1. Introduction
The phenolethanolamine derivative (R, R)-CL 316,243
[7, 8], a highly selective ꢀ3-receptor agonist, is a prom-
ising phase II candidate for the new target ꢀ3-receptors
[9] in treatment of diabetes mellitus type IIb. (R)-(+)-
Etomoxir, withdrawn from development due to myocar-
dial hypertrophy, lowers blood glucose via irreversible
inhibition of carnitin-palmitoyl-transferase I (CPT I),
while the (S)-(–)-enantiomer does not show any activity
[10, 11]. N-Palmitoylcarnitine and its phosphate ana-
logues are reversible inhibitors of CPT I [12]. The
naturally occurring, highly potent (R)-enantiomers are by
far superior to the corresponding (S)-enantiomers.
Although achiral sulfonylureas, which improve insulin
secretion from pancreatic ꢀ-cells, are widely used for the
treatment of type 2 diabetes (NIDDM or non-insulin-
dependent diabetes mellitus), enantiomerically pure oral
antidiabetics of different mechanisms of action are gain-
ing growing interest in the treatment of type
II-hyperglycaemia. The sulfonylamidopyrimidine gliflu-
mide [1, 2] was one of the first homochiral compounds
interacting with the sulfonylurea receptor [3], the (S)-
enantiomer being 20-fold more efficacious than the (R)-
enantiomer [4]. The insulinotropic non-sulfonylurea (+)-
repaglinide was recently launched and is twice as active
as its (–)-enantiomer [5, 6].
Semicyclic amidines are also known for their hypogly-
caemic effects, with racemic 2-[(cis-2-cyclohexyl-
cyclopentyl)imino]-hexahydroazepine hydrochloride as
the most efficacious representative developed as a result
of extensive structure–activity relationship studies [13].
Dose-dependent blood glucose lowering effects could be
observed in rats after oral application according to a
* Correspondence and reprints: awfrahm@ruf.uni-freiburg.de
cPresent address: Department of Neurophysiology, University
of Köln, Robert-Koch-Str. 39, D-50931 Köln, Germany
Dedicated to Prof. Dr. Woldemar Schneider on the occasion of his
80th birthday.

378
generally accepted method [14]. Increasing steric hin-
drance of the C-1 atom neighbouring the amidino func-
tion by spatially demanding 2-substituents correlated
with an increasing hypoglycaemic effect. However, chiral
differentiating efficacy of the enantiomerically pure com-
pounds of this new class of hypoglycaemic agents has yet
to be elucidated.
Therefore, we have developed the EPC (enantiomeri-
cally pure compounds) synthesis of homochiral semicy-
clic amidines derived from ε-caprolactam and have tested
their chiral differentiating insulinotropic activity in vitro
in INS-1 cells and by means of electrophysiological
methods. Besides introduction of various spatially de-
manding substituents in position 2 of the aminocyclopen-
tane moiety (isopropyl, phenyl, cyclohexyl, benzyl, cy-
clopentyl and cyclohexylmethyl) we were especially
interested in the influence of increasing steric hindrance
of the semicyclic amidine moiety. Thus, the hexahy-
droazepine system was also replaced by the bicyclic
isoquinuclidine in a second series of semicyclic amidines.
In 1989 Palfreyman et al. [15] reported lactamimide
analogues representing a new chemical class of calcium
antagonists with diltiazem-like properties in in vitro test
systems. However, the most potent member N-(2,2-
diphenylpentyl)azacyclotridecan-2-imine hydrochloride
contains a large 13-membered ring, whereas the racemic
forms of the 7-membered hexahydroazepines 5b and 5c
are only weak competitive antagonists of the calcium-
induced contractions.
2. Chemistry
The homochiral semicyclic amidines 5a–f HX and
6a–f HX were prepared from the racemic 2-substituted
cyclopentanones 1a–f by means of asymmetric reductive
amination [16] and subsequent lethargic reaction of the
homochiral 2-substituted cyclopentaneamine halides 4a–f
HX with O-methylcaprolactim and 3-ethoxy-2-
azabicyclo[2.2.2]oct-2-ene, respectively, in a four-step
procedure (figure 1).
2-Isopropylcyclopentanone (1a) [17, 18], 2-phenyl-
cyclopentanone (1b) [19–21], 2-benzylcyclopentanone
(1d) [22, 23], and 2-cyclopentylcyclopentanone (1e) [24,
25] were synthesized according to common literature.
2-Cyclohexylcyclopentanone (1c) and 2-cyclohexyl-
methylcyclopentanone (1f) were accessible from 1b and
1d, respectively, by hydrogenation of the aromatic ring.
Hydrogenation of unprotected 1b with Rh/γ-Al2O3 in
methanol/acetic acid [26] furnished a diastereomeric
mixture of 2-cyclohexylcyclopentanols and 1c in a 5:4
ratio, which was oxidized in situ to pure 1c with
pyridiniumchlorochromate on Al2O3 [27]. However, hy-
Figure 1. Four-step procedure for the preparation of amidines
5a–f HX and 6a–f HX.
drogenation of 1d yielded 97% 1f with only negligible
amounts of the respective cyclopentanols.
Condensation of the racemic 2-substituted cyclopen-
tanones 1a–f with optically active (R)-(+)- and (S)-(–)-
1-phenylethylamine (PEA), respectively, led to the imi-
nes 2a–f. The latter were hydrogenated in situ with Raney
nickel at room temperature in a Parr shaker for a period
of 5–11 days to give the optically active, diastereomeri-
cally pure secondary cis-2-substituted N-(1-phenylethyl)-
cyclopentaneamines 3a–f in yields up to 73%. Their
halides 3a–f HX were hydrogenolysed with palladium-
on-charcoal (10%) at 50 °C in a Parr shaker within 24 h
to give the enantiomerically high grade pure primary

379
Table I. Reaction times and yields of the lethargic reaction to the
semicyclic amidines 5a–f HX and 6a–f HX.
Compound
Reaction time (days)
Yield (%)
5a HCl
5b HCl
5c HBr
5d HCl
5e HBr
5f HBr
6a HCl
6b HCl
6c HBr
6d HCl
6e HBr
6f HBr
30
16
42
17
35
21
84
63
105
63
90
86
79
79
85
92
56
91
73
86
76
90
Figure 2. Configurations of 7a–f, 8a–f and 8p.
substituted cyclopentaneamines. The CD spectra are con-
gruent in their positive Cotton effects at 315 and 255 nm
and their negative Cotton effect at 225 nm. Divergence in
intensity is caused by differing conformative arrange-
ments of the 2-substituents. The CD curves with extrema
at comparable wavelength ranges, but with actually
slightly deviating τ-values are strongly indicative of the
identical absolute configuration of (1S, 2S)-7b and (1S,
2S)-7a, -7c–f.
154
42
cis-2-substituted cyclopentaneamine halides 4a–f HX in
yields of about 90%.
Lethargic reaction [13, 28] of 4a–f HX with
The enantiomeric excesses (ee%) of 4a–f and of the
auxiliary amines were determined by 19F-NMR spectros-
copy and HPLC via derivatization with (S)-(+)-α-
methoxy-α-trifluoro-methylphenylacetylchloride (Mosh-
er’s reagent) [32, 33] to the diastereomeric Mosher’s
amides 8a–f and 8p (figure 2). With more than 99.0% the
ee values of 8a–f exceeded those of the auxiliary amines,
97.9% for (R)-(+)-PEA and 97.0% for (S)-(–)-PEA,
which provides evidence for a recrystallization effect of
the secondary and primary amines 3a–f HX and 4a–f HX.
The 19F-NMR spectra of 8a–f and 8p show two singlets
for a couple of diastereomeric Mosher’s amides each,
which are slightly widened by lr-coupling to the methoxy
protons. The diastereotopic CF3 signals of 8a, 8b, 8d, 8f
and 8p range between 50 and 230 Hz, sufficient for
precise quantification. However, the corresponding sig-
nals of 8c and 8e diverge by only 14 and 17 Hz,
respectively. Reliable spectroscopic determination of ee
values was impossible in these two cases, even though the
signals of the major diastereomer did not reveal any
shoulder in the expanded spectra. The HPLC separation
of the Mosher’s amides 8a–f and 8p were in accordance
with results derived from 19F-NMR spectroscopy. The
subsidiary diastereomer of 8c and 8e, respectively, was
below the detection limit (< 0.5%) each time.
O-methylcaprolactim
and
3-ethoxy-2-azabicyclo-
[2.2.2]oct-2-ene, respectively, led to 5a–f HX in yields of
about 85% in a period of 17–42 days, and to 6a–f HX in
yields of about 85% (with one exception) within 6–15
weeks (table I). The extremely long reaction times for the
lethargic reactions could not be overcome by well-known
accelerating methods of organic synthesis, such as acid
catalysis, pressure, sonication or microwave, partly be-
cause of the instability of the lactimethers used [13]. 5a–f
HX and 6a–f HX show E/Z-equilibria in solution. Their
composition has been reported elsewhere [29]. In solid
state Z-configuration is observed [30]. The Z-configured
species is supposed to produce the biological effect, since
a complete shift of the equilibria towards Z-5a–f HX and
Z-6a–f HX was observed in polar solvents such as
methanol and dimethylsulfoxide.
Various 1-D and 2-D H- and 13C-NMR techniques,
1
including lr (longrange)-INEPT, COSY, HETCOR, GH-
SQCR and INADEQUATE, were used to elucidate the
structure and relative configuration of all compounds and
to analyse the E/Z-equilibria of the semicyclic amidines
in solution.
The absolute configuration of 4a–f was determined
indirectly by CD correlation spectroscopy of the corre-
sponding
cis-2-substituted
(1S, 2S)-cyclopentane-
aminesalicylidenes (1S, 2S)-7a–f (figure 2) with refer-
ence to cis-(1S, 2S)-2-phenylcyclopentaneamine-N-
salicylidene ((1S, 2S)-7b) [16] of known absolute
configuration. The results relate well to findings in
literature [16], where like-induction [31] was reported for
the reductive amination step, thus (R)-(+)-PEA leading to
secondary cis-(1R, 1’R)-2-substituted cyclopentaneam-
ines and (S)-(–)PEA to corresponding cis-(1S, 1’S)-2-
3. Physiology
3.1. Effects of semicyclic amidines
on insulin secretion in INS-1 cells
The insulin secreting cell line INS-1 was used to test
the insulinotropic effect of the semicyclic amidines 5a–f

380
Figure 3. Effects of caprolactame derivatives 5a–f HX on insulin secretion of INS-1 cells. INS-1 cells were incubated and insulin
secretion measured as described in Experimental protocols. The compounds were tested at 10 and 100 µM in the presence of 0.5 mM
glucose and 5 µM forskolin. Results are expressed as fold-stimulation over forskolin alone and are means ± SEM for the indicated
#
number of observations (resulting from up to five independent experiments). 50 µM. * and §; significant differences between the
enantiomers.
HX and 6a–f HX. Forskolin was present throughout
incubations since it potentiated the stimulatory effect of
amidines on secretion (data not shown). All compounds
increased insulin release at 10 µM (figures 3 and 4). The
effect varied between 43% and 369%. The most potent
activators of secretion were cyclohexyl-, benzyl-, and
cyclohexylmethyl substituted isoquinuclidone derivatives
6c HBr, 6d HCl and 6f HBr, respectively, and the
cyclopentyl substituted caprolactam derivate 5e HBr.
However, the caprolactam-derived cyclohexyl substituted
compound 5c HBr in contrast to in vivo data [13],
together with the isopropyl-derivates 5a HCl and 6a HCl
were the weakest stimulators of insulin secretion. Com-
paring the respective enantiomers, only (1’R, 2’R)-6c
HBr increased secretion more effectively than the (1’S,
2’S)-6c HBr distomer. All other enantiomers showed no
chiral differentiation at this concentration.
respectively, as well as cyclohexyl and cyclopentyl deri-
vates 5c HBr and 5e HBr further augmented insulin
release at the higher concentration. In contrast, cyclo-
hexyl and cyclopentyl derivates (1’R, 2’R)-6c HBr and
(1’R, 2’R)-6e HBr showed similar effects at 100 and
10 µM. Surprisingly, the cyclohexylmethyl derivate 6f
HBr inhibited secretion at 100 µM but had a stimulatory
effect at 10 µM. However, the cyclohexylmethyl derivate
5f HBr had a smaller activating effect at 100 µM than at
10 µM. The sulfonylurea tolbutamide was tested simul-
taneously and did not affect secretion at 10 µM but
increased insulin release 2.46 ± 0.13-fold (n = 20) at
50 µM and 2.33 ± 0.25-fold (n = 8) at 100 µM (cf. also
reference [34]). Thus, semicyclic amidines stimulate
insulin secretion in a similar concentration range to
tolbutamide. The semicyclic amidines 6c HBr, 6d HCl
and 6f HBr with the bicyclic lactam moiety were more
potent at 10 µM than the respective sterically less
hindered caprolactam compounds 5c HBr, 5d HCl and 5f
HBr.
The substances were also tested at 100 µM. Isopropyl,
phenyl and benzyl derivates of both series of amidines, 5a
HCl, 5b HCl, 5d HCl, 6a HCl, 6b HCl and 6d HCl,

381
Figure 4. Effects of isoquinuclidone derivatives 6a–f HX on insulin secretion of INS-1 cells. INS-1 cells were incubated and insulin
secretion measured as described in Experimental protocols. The compounds were tested at 10 and 100 µM in the presence of 0.5 mM
glucose and 5 µM forskolin. Results are expressed as fold-stimulation over forskolin alone and are means ± SEM for the indicated
number of observations (resulting from up to five independent experiments). * and §; significant differences between the enantiomers.
3.2. Effects of semicyclic amidines
on [Ca2+]i in INS-1 cells
At 10 µM, (1’R, 2’R)-6f HBr increased the fura-2 fluo-
rescence ratio 340/380 nm from 1.32 ± 0.08 (n = 12) to
1.57 ± 0.15 (n = 5), at 100 µM to 1.84 ± 0.09 (n = 7).
Thus, although (1’R, 2’R)-6f HBr stimulated secretion
only at the lower concentration (figure 4), it further
increased [Ca2+]i more consistently at the higher concen-
tration. In the case of 100 µM (1’R, 2’R)-6f HBr the
increase in [Ca2+]i was transient and returned to basal
level before the removal of the drug.
The observation that forskolin potentiated amidine-
induced secretion suggests that the amidines mainly act
by increasing [Ca2+]i. They might activate the cells at
least in part by inducing depolarization due to closure of
K+ATP-channels and by subsequent activation of voltage-
dependent Ca2+-channels.
Indeed, when measuring [Ca2+]i in fura-2 loaded INS-1
cells, [Ca2+]i increased upon amidine addition (figures 5
and 6). The effect was comparable to tolbutamide-
induced [Ca2+]i-increase (100 µM). [Ca2+]i returned to
basal levels after removal of (1’R, 2’R)-6a HCl with a
delay indicating that the wash-out of the compound was
slow. Since [Ca2+]i oscillated in 68% of INS-1 cells,
which probably followed membrane voltage oscillations
observed previously [35], the effects on [Ca2+]i were not
quantified.
3.3. Effects of semicyclic amidines on membrane voltage
and membrane conductance in INS-1 cells
As the increase of [Ca2+]i may be due to depolarization
of the plasma membrane, membrane voltage and conduc-
tance were measured using the standard whole cell mode
of the patch clamp method. Figure 7A shows a represen-
tative registration of membrane voltage [Vm] in INS-1
cells, depolarized with 100 µM (1’R, 2’R)-5a HCl. The
effects of (1’R, 2’R)-5a HCl on membrane voltage and
membrane conductance are summarized in figure 7B.
In another series of experiments the effects on [Ca2+]i
of 10 µM and 100 µM (1’R, 2’R)-6f HBr were compared.

382
Figure 5. Effects of tolbutamide (Tol), (1’S, 2’S)-6a HCl (S6a) and (1’R, 2’R)-6a HCl (R6a) on [Ca2+]i in INS-1 cells. Shown is a
representative trace of continuous [Ca2+]i recording. The compounds were added as indicated, each at a concentration of 100 µM.
Membrane conductance was decreased 80% by 100 µM
(1’R, 2’R)-5a HCl, whereas the sulfonylurea tolbutamide
reduced membrane conductance by 95% at the same
concentration. The depolarizing effect of (1’R, 2’R)-5a
HCl was also smaller than that of tolbutamide, 15 mV and
45 mV, respectively. These observations indicate that
Figure 6. Effects of tolbutamide (Tol), (1’R, 2’R)-5a HCl (R5a) and (1’S, 2’S)-5a HCl (S5a) on [Ca2+]i in INS-1 cells. Shown is a
representative trace of continuous [Ca2+]i recording. The compounds were added as indicated, each at a concentration of 100 µM.

383
Figure 7. Effect of (1’R, 2’R)-5a HCl on membrane voltage in INS-1 cells. A: Shown is a representative trace of continuous
membrane voltage recording. 100 µM (1’R, 2’R)-5a HCl and 100 µM tolbutamide (Tol) were added as indicated. B. Mean ± SEM of
seven experiments.
semicyclic amidines at least partially stimulate insulin
release by depolarizing the INS-1 cells, which in succes-
sion leads to the opening of voltage-sensitive Ca2+
channels and to an increase of [Ca2+]i.
spatially demanding cyclohexylmethyl substituted
amidines 5f HBr and 6f HBr stimulated insulin release
3–4-fold at 10 µM. At this concentration the sulfonylurea
tolbutamide had no significant effect on insulin release
(figure 3). At 100 µM, the most potent secretagogue was
the compound (1’R, 2’R)-5e HBr, increasing insulin
release 8-fold. In comparison, 100 µM tolbutamide in-
creased secretion 2–3-fold.
4. Results and discussion
Sulfonylureas improve insulin secretion by a receptor
mediated mechanism. Structurally unrelated racemic
amidines have been described as hypoglycaemic agents
with an unelucidated mode of action [13]. In this study
we have prepared enantiomerically pure semicyclic
amidines in order to test their ability to stimulate insulin
secretion and to examine their mode of action.
The compounds had no toxic effects in INS-1 cells at
the highest concentration tested (100 µM), since i) depo-
larization and increase of [Ca2+]i were reversible (fig-
ures 5–7), ii) insulin content and the secretory behaviour
of the cells were unchanged under conditions of cell
culturing for 24 h in the presence of amidines (not
shown), iii) the stimulatory effect of amidines was
inhibited by adrenaline, which is a physiological inhibitor
of insulin secretion (not shown). Thus, these substances
stimulate ꢀ-cell secretion via a controllable mechanism.
All semicyclic amidines increased insulin secretion of
INS-1 cells at 10 µM (figures 3 and 4). The flexible,
Significant differences between the potency of enanti-
omers to stimulate insulin release was observed for 5e
HBr, 5f HBr, 6a HCl and 6c HBr. Thus, the (1’R,
2’R)-eutomers of 5e HBr, 5f HBr, 6a HCl induced a 37%,
95% and 33% higher secretion than the respective (1’S,
2’S)-enantiomers. In contrast, the (1’R, 2’R)-enantiomer
of 6c HBr was 22% less potent than its (1’S, 2’S)-
enantiomer. These observations show the site of action to
possess only a weak stereoselectivity for these amidines.
In contrast to findings with other insulin secreting
cells, tolbutamide induced secretion in INS-1 cells only
in the presence of increased adenylyl cyclase activity
[34]. Synergistic effects of cAMP and [Ca2+]i on insulin
secretion are also well described for native ꢀ-cells [36].
Thus, the potentiation of amidine induced secretion by
cAMP suggests a site of action involving increases in
[Ca2+]i. Indeed, pronounced elevation of [Ca2+]i is ob-
served upon amidine addition (figures 5 and 6). This

384
effect is reversible, but after removal of the drug, [Ca2+]i
returns only slowly to basal levels indicating a reduced
wash-out of the substance. The measurements of mem-
brane voltage and membrane conductance revealed that
amidines depolarized the cells and reduced the membrane
conductance by 80% under control conditions (figure 7).
Since the whole cell conductance under control condi-
multiplets were determined by the corresponding 1H-13C-
Hetcor spectra. H-NMR data are always listed in the
1
following order: multiplicity (s, singlet; d, doublet; t,
triplet; q, quartet; quint, quintet; m, multiplet; mc, centric
multiplet), number of nuclei, coupling constant(s) in Hz,
assignment. *, #, §, ◆ indicate protons/carbon atoms that
can be exchanged. GC MS spectra were recorded on a
Perkin Elmer QMASS 910 spectrometer on an Rtxt-35
column (Restek, Bellefonte, USA), crossbondt (35%
diphenyl-, 65% dimethylpolysiloxane) 0.25 µm, 35 m.
HPLC separation of the Mosher amides (enantiomeric
excesses (ee%)) was performed on a Lichrosorbt Si-60
column (5 µm, 250 × 4 mm, Bischoff, Leonberg, D) with
a Waters 600 multisolvent delivery system with a diode-
array detector Waters 990 or a Waters 486 Tunable
Absorbance Detector. Flash chromatography was carried
out on E. Merck silica gel 60 (0.040–0.068 mm). Thin
layer chromatography (TLC) was performed on silica gel
F254 plates from E. Merck, Darmstadt, D. Visualization
was done using ninhydrin reagent (Stahl) and heating.
Analyses (Analytical Laboratories, Chemical Institutes,
University of Freiburg, D) indicated by the symbols of
the elements or functions were within ± 0.4% of theo-
retical values.
tions has been shown to result mainly from K+
-
channel activity, the observed effects of amidines areAdTuPe
to inhibition of these channels [34].
Thus, the proposed mechanism of action involves the
closure of K+ATP-channels which leads to a decrease of
membrane conductance, depolarization and subsequent
opening of voltage-activated Ca2+-channels. Whether the
site of action is also the SUR-molecule [35] as for
sulfonylureas, remains obscure and further studies are
needed for clarification.
Since
the
flexible,
spatially
demanding
cyclohexylmethyl-substituted amidine 6f HBr inhibited
secretion at the highest concentration tested (100 µM)
while increasing [Ca2+]i, it is suggested that accumulation
of this compound leads to the interference with further
membrane proteins resulting in an inhibition of secretion.
In conclusion, this study shows that enantiomerically
pure semicyclic amidines induce insulin secretion with
only weak stereoselectivity, at least in part through
inhibition of K+ATP-channels and a subsequent rise in
[Ca2+]i.
5.1.1. 2-Cyclohexylcyclopentanone (1c)
To
a
solution of 3.2 g (20 mmol) 2-phenyl-
cyclopentanone (1b) in 40 mL methanol abs were added
0.6 g of Rh/γ-Al2O3 (5%) and 0.5 mL acetic acid [26].
The mixture was hydrogenated at an initial pressure of 5
bar and at room temperature in a Parr shaker. The
progress of the reaction was observed by GC/MS. After
10.5 h the reaction mixture was filtered (Celite 535), the
catalyst was washed with five portions of 20 mL metha-
nol each and the filtrate was evaporated after addition of
5 mL toluene for complete removal of the acetic acid. The
residue was oxidized in situ.
Hydrogenation mixture (3.2 g) was added dropwise to
22.6 g (17.6 mmol) pyridiniumchlorochromate-Al2O3 [27]
suspended in 50 mL toluene, and stirred at room tempera-
ture over night. The oxidant was filtered off and washed
with five portions of 30 mL diethylether each. The
organic phase was dried with sodium sulfate, filtered and
evaporated in vacuo. A 2.9 g (88%) yield of a colourless
oil, containing 96.5% 1c (GC-MS) was condensed in situ
with (R)-(+)- and (S)-(–)-phenylethylamine, respectively.
MS (EI) m/e 166 ([M]+); 1H-NMR (CDCl3) δ 0.97–1.32
(m, 5H, H-2’b, H-6’b, H-3’b, H-4’b, H-5’b), 1.42 (mc,
1H, H-6’a*), 1.58–1.82 (m, 7H, H-3’a, H-4’a, H-5’a,
H-3b, H-4b, H-2’a*, H-1’), 1.89–2.06 (m, 4H, H-4a, H-2,
H-3a, H-5b), 2.19–2.31 (m, 1H, H-5a); 13C-NMR (CDCl3)
δ 20.68 (C-4), 25.33 (C-3), 26.22 (C-3’*), 26.24 (C-4’),
5. Experimental protocols
5.1. Chemistry
Hydrogenation reactions were carried out in a Parr
shaker, constructed by the Pharmaceutical Institute of the
University of Freiburg using Rh/γ-Al2O3 (5%) type G
207 R, Raney-nickel suspended under water type B 113
W and Pd/C (10%) type E 10 N/D (all Degussa, Frankfurt
a. M., D). Melting points (m.p.) were determined on a
Mel Temp II capillary melting point apparatus (Labora-
tory Devices, USA) and are uncorrected. IR spectra were
recorded on a Perkin Elmer 841 infrared spectrometer.
Optical rotation was determined on a Perkin Elmer 241
polarimeter with c = 1.0, ethanol. CD spectra were
recorded on a CD 6 Dichrograph System, ISA Instru-
1
ments Jobin Yvon, Longjumeau, F. H-, 13C-, 19F- 1-D
and 2-D-NMR spectra were recorded on a Varian Unity
300 spectrometer operating at 300, 75 and 282 MHz,
respectively. Chemical shift values are referenced to
CHCl3, δ 7.24 in CDCl3 for 1H-NMR, to CDCl3, δ 77.00
for 13C-NMR and to 1,1,2-trichlorotrifluoroethane, δ
–68.03 for 19F-NMR. The chemical shifts of proton

385
26.44 (C-5’*), 28.74 (C-2’#), 31.56 (C-6’#), 37.49 (C-1’),
39.11 (C-5), 54.46 (C-2), 221.16 (C-1); IR (NaCl, cm–1)
1 737 (C=O).
one third and the sec amine hydrobromide was precipi-
tated by adding diethylether and recrystallized from
ethanol/ether to give 3e HBr and 3f HBr, respectively.
The ice-cooled solution of an aliquot amount of sec
cis-2-cyclohexylcyclopentaneamine 3c, rectified by col-
umn chromatography, was acidified with trifluoroacetic
acid. 3c TFA was recrystallized from n-hexane/di-
ethylether (2:1). 3c HBr (73%) was hydrogenated in situ.
5.1.2. 2-Cyclohexylmethylcyclopentanone 1f
To
a
solution of 8.7 g (50 mmol) 2-benzyl-
cyclopentanone (1d) in 100 mL methanol abs were added
2.4 g of Rh/γ-Al2O3 (5%) and 1.6 mL acetic acid. The
mixture was hydrogenated at an initial pressure of 5 bar
and at room temperature in a Parr shaker. The progress of
the reaction was observed by GC/MS. After 20.5 h the
reaction mixture was filtered (Celite 535), the catalyst
was washed with five portions of 20 mL methanol each
and the filtrate was evaporated after addition of 5 mL
toluene for complete removal of the acetic acid. An 8.8 g
(98%) yield of a colourless oil, containing 98.5% 1f
(GC-MS) was condensed in situ with (R)-(+)- and (S)-
(–)-phenylethylamine, respectively. MS (EI) m/e (M), 84
5.1.4.1. cis-2-Cyclohexyl-N-(1-phenylethyl)-
cyclopentaneamine trifluoroacetate 3c TFA
25
D
(1R, 2R)-3c TFA: m.p. 117–119 °C, ␣
–7.9°; (1S,
͓ ͔
25
1
2S)-3c TFA: m.p. 119–121 °C,
␣
͓ ͔
+8.2°; H-NMR
D
(CDCl3) δ 0.86 (mc, 2H, H-2’b, H-6’b), 1.05–1.21 (m,
1H, H-3b), 1.21–1.39 (m, 2H, H-3’b, H-5’b), 1.42–1.71
(m, 9H, H-2’a*, H-4b, H-1’, H-2, H-3’a, H-5’a, H-5b,
H-3a, H-4’b), 1.75 (d, 3H, 6.8, ꢀ-CH3), 1.71–1.92 (m,
4H, H-4’a, H-6’a*, H-5a, H-4a), 3.35 (s (broad (br)), 1H,
H-1), 4.35 (q, 1H, 6.8, α-CH), 7.37–7.47, (m, 3H, H-3″,
H-4″, H-5″), 7.50–7.55 (m, 2H, H-2″, H-6″), 7.71 (s (br),
1H, +NH-b), 10.02 (s (br), 1H, +NH-a); 13C-NMR (CDCl3)
δ 18.96 (ꢀ-CH3), 20.12 (C-4), 24.97 (C-3’, C-5’), 25.16
(C-4’), 25.55 (C-3), 29.95 (C-5), 31.24 (C-2’*), 32.50
(C-6’*), 34.91 (C-1’), 50.65 (C-2), 58.68 (C-1), 59.06
1
(100); H-NMR (CDCl3) δ 0.70–0.92 (m, 2H, H-2’b,
H-6’b), 0.97–1.29 (m, 5H, H-7’b, H-3’b, H-4’b, H-5’b,
H-1’), 1.33–1.46 (m, 1H, H-3b), 1.51–1.75 (m, 7H,
H-2’a*, H-7’a, H-6’a*, H-3’a, H-4’a, H-5’a, H-4b),
1.86–2.25 (m, 5H, H-4a, H-5b, H-2, H-3a, H-5a); 13C-
NMR (CDCl3) δ 20.54 (C-4), 25.99 (C-3’#), 26.11
(C-5’#), 26.35 (C-4’), 30.00 (C-3), 32.12 (C-2’*), 33.98
(C-6’*), 35.48 (C-1’), 37.27 (C-7’), 37.74 (C-5), 46.60
(C-2), 221.61 (C-1); IR (NaCl, cm–1) 1 739 (C=O).
1
(α-CH), 116.17 (q, JC, F = 294 Hz, CF3), 127.34 (C-2″,
C-6″), 128.63 (C-3″, C-5″), 128.67 (C-4″), 135.89 (C-1″),
161.35 (q, 2JC, F = 35 Hz, F3C–COO–). (1R, 2R)-3c TFA
anal. C21H30F3NO2 (H, N); C: calcd., 65.43; found 64.80;
(1S, 2S)-3c TFA anal. C21H30F3NO2 (C, H, N).
5.1.3. Preparation
of 2-substituted cyclopentaneimines 2a–f
The 2-substituted cyclopentaneimines 2a–f were pre-
pared according to a common procedure [16]. Toluene
was used as solvent. The optimized reaction times and
yields of imine condensation are: 25 h, 88% for 1c, 22 h,
80% for 1e and 14 h, 90% for 1f. The yields were
determined by 1H-NMR (α-H-signal imine 4.48 and 4.43
ppm/PEA 4.1 ppm). For the complete NMR data see the
supporting information.
5.1.4.2. cis-2-Cyclopentyl-N-(1-phenylethyl)-
cyclopentaneamine hydrobromide 3e HBr
(1R, 2R)-3e HBr: m.p. 165–167 °C, ␣ 25 +1.4°; (1S,
͓ ͔
D
25
1
2S)-3e HBr: m.p. 164–165 °C,
␣
–1.7°; H-NMR
͓ ͔
(CDCl3) δ 0.96–1.18 (m, 2H, H-2’b,DH-5’b), 1.48–1.75
(m, 8H, H-4b, 3’-CH2, 4’-CH2, H-3b, H-2’a*, H-2), 1.80
(mc, 1H, H-5b), 1.92–2.13 (m, 3H, H-5’a*, H-4a, H-3a),
2.09 (d, 3H, 6.9, ꢀ-CH3), 2.26–2.46 (m, 2H, H-5a, H-1’),
3.17 (s (br), 1H, H-1), 4.55 (s (br), 1H, α-CH), 7.36 (tt,
1H, 7.2, 1.3, H-4″), 7.43 (dt, 2H, 7.0, 1.8, H-3″, H-5″),
7.81 (td, 2H, 6.9, 1.5, H-2″, H-6″), 8.49 (s (br), 1H,
5.1.4. Preparation of cis-2-substituted
N-(1-phenylethyl)-cyclopentaneamine halides 3a–f HX
The cis-2-substituted N-(1-phenylethyl)-cyclopent-
aneamine halides 3a–f HX were prepared according to a
common procedure [16]. The optimized hydrogenation
conditions and yields (calculated from ketones over both
steps) are 11 days hydrogenation time, 73% (oil) for 2c,
11 days hydrogenation time, 54% for 2e and 5 days
hydrogenation time, 60% for 2f with 1.9 g Raney-nickel/
10 mmol 2-substituted cyclopentaneimines 2c, 2e and 2f,
respectively.
+
+NH-b), 9.27 (s (br), 1H, NH-a); 13C-NMR (CDCl3) δ
20.41 (ꢀ-CH3), 21.08 (C-4), 25.08 (C-4’#), 25.10 (C-3’#),
27.22 (C-3), 30.39 (C-5), 31.89 (C-5’*), 32.77 (C-3’*),
38.00 (C-1’), 49.38 (C-2), 59.98 (α-CH), 60.91 (C-1),
128.48 (C-2″, C-6″), 129.18 (C-4″), 129.25 (C-3″, C-5″),
136.34 (C-1″). Anal. C18H28NBr (C, H, N).
5.1.4.3. cis-2-Cyclohexylmethyl-N-(1-phenylethyl)-
cyclopentaneamine hydrobromide 3f HBr
The ice-cooled solution of the hydrogenation product
of 2c, 2e and 2f in ethanol was acidified with hydrobro-
mic acid 48%. The volume of the solution was reduced to
(1R, 2R)-3f HBr: m.p. 240–241 °C decomposition,
␣
͓ ͔
25 +54.9°; (1S, 2S)-3f HBr: m.p. 241–242 °C decom-
D

386
25
1
(C-5), 32.26 (C-5’*), 32.32 (C-2’*), 39.99 (C-1’), 50.14
(C-2), 56.06 (C-1). Anal. C10H20NBr (C, H, N).
position,
␣
͓ ͔
–55.0°; H-NMR (CDCl3) δ 0.86–1.00
(m, 1H, H-6’bD*), 1.06–1.32 (m, 5H, H-2’b*, H-1’, H-3’b,
H-4’b, H-5’b), 1.36–1.53 (m, 4H, H-4b, 7’-CH2, H-3b),
1.58–1.80 (m, 6H, H-3a, H-3’a, H-4’a, H-5’a, H-2’a#,
H-4a), 1.80–1.92 (m, 1H, H-6’a#), 1.99–2.30 (m, 3H,
H-2, 5-CH2), 2.03 (d, 3H, 6.9, ꢀ-CH3), 2.99 (quint, 1H,
7.0, H-1), 4.30 (quint, 1H, 6.5, α-CH), 7.31–7.46 (m, 3H,
H-3″, H-4″, H-5″), 7.75 (dt, 2H, 7.4, 1.4, H-2″, H-6″),
9.11 (s (br), 1H, +NH-b), 9.66 (s (br), 1H, +NH-a);
13C-NMR (CDCl3) δ 19.80 (C-4), 21.32 (ꢀ-CH3), 26.15
(C-3’#), 26.29 (C-5’#), 26.50 (C-4’), 27.12 (C-5), 27.64
(C-3), 32.23 (C-6’*), 34.54 (C-2’*), 35.21 (C-7’), 35.37
(C-1’), 35.97 (C-2), 58.88 (α-C), 59.80 (C-1), 128.19
(C-2″, C-6″), 129.10 (C-4″), 129.25 (C-3″, C-5″), 136.25
(C-1″). Anal. C20H32NBr (C, H, N).
5.1.5.3. cis-2-Cyclohexylmethylcyclopentaneamine
hydrobromide 4f HBr
(1R, 2R)-4f HBr: m.p. 212–213 °C decomposition,
25
D
␣
͓ ͔
+6.5°; (1S, 2S)-4f HBr: m.p. 213–214 °C decom-
25
position, ␣ –7.3°; 1H-NMR (CDCl3) δ 0.77–1.01 (m,
͓ ͔
D
2H, H-6’b, H-2’b), 1.03–1.40 (m, 5H, H-3’b, H-4’b,
H-5’b, H-1’, H-7’b), 1.51 (mc, 1H, H-7’a), 1.55–1.73 (m,
6H, H-3’a, H-4’a, H-5’a, H-3b, H-4b, H-2’a*), 1.73–1.90
(m, 2H, H-6’a*, H-3a), 1.90–2.14 (m, 4H, H-4a, 5-CH2,
H-2), 3.60 (mc, 1H, H-1), 7.92 (s (br), 3H, +NH3);
13C-NMR (CDCl3) δ 21.60 (C-4), 26.09 (C-3’*), 26.19
(C-5’*), 26.54 (C-4’), 28.72 (C-3), 30.80 (C-5), 32.79
(C-2’#), 33.74 (C-6’#), 35.92 (C-1’), 37.04 (C-7’), 40.02
(C-2), 56.22 (C-1). Anal. C12H24NBr (C, H, N).
5.1.5. Preparation of cis-2-substituted
cyclopentaneamine halides 4a–f HX
The cis-2-substituted cyclopentaneamine halides 4a–f
were prepared according to a common procedure [16].
The optimized hydrogenolysis conditions and yields are:
0.8 g Pd/C / 10 mmol 3c HBr, 24 h, 61% 4c HBr, 1.0 g
Pd/C / 10 mmol 3e HBr, 24 h, 93% 4e HBr and 0.9 g
Pd/C / 10 mmol 3f HBr, 24 h, 81% 4f HBr.
5.1.6. Preparation of 2-[(cis-2-substituted
cyclopentyl)imino]hexahydroazepine halides 5a–f HX
and 3-[(cis-2-substituted cyclopentyl)imino]-2-
azabicyclo[2.2.2]octane halides 6a–f HX
Five mmol of the primary cis-2-substituted cyclopen-
taneamine halide 4a–f HX in 1 mL ethanol abs and
5.5 mmol of O-methylcaprolactim (0.7 g) and 3-ethoxy-
2-azabicyclo[2.2.2]oct-2-ene (0.84 g), respectively, were
carefully dispersed in an ultrasonic device. The reaction
mixture was stirred at room temperature. The progress of
the lethargic reaction [13] was observed by TLC (ethyl
acetate/ethanol/ammonia conc.: 90/10/1, ninhydrin re-
agent). When necessary, further ethanol abs was added to
keep the suspension liquid and further lactimether was
added to achieve complete turnover. Reaction times and
yields are listed in table I. The reaction mixture was then
resolved in methanol, filtered and the filtrate was evapo-
rated. The white residues were recrystallized from
methanol/acetone to give the semicyclic amidines 5a–e
HX and 6a–f HX. Analytically pure 5f HBr was collected
as residue after extraction of the raw 5f HBr with ethyl
acetate.
5.1.5.1. cis-2-Cyclohexylcyclopentaneamine
hydrobromide 4c HBr
(1R, 2R)-4c HBr: m.p. 275–278 °C decomposition,
␣
25 –16.5°; (1S, 2S)-4c HBr: m.p. 277–280 °C decom-
͓ ͔
D
25
1
position,
␣
+16.1°; H-NMR (CDCl3) δ 0.88 (mc,
͓ ͔
D
2H, H-2’b, H-6’b), 1.02–1.30 (m, 2H, H-3’b, H-5’b),
1.35 (mc, 1H, H-1’), 1.43–1.71 (m, 6H, H-2, H-3’a,
H-5’a, H-4’b, H-4b, H-3b), 1.71–1.83 (m, 2H, H-3b,
H-6’a*), 1.83–2.05 (m, 4H, H-5b, H-3a, H-2’a*, H-4a),
2.11–2.23 (m, 1H, H-5a), 3.71 (s (br), 1H, H-1), 7.86 (s
(br), 3H, +NH3); 13C-NMR (CDCl3) δ 20.70 (C-4), 25.45
(C-4’), 26.04 (C-3’#), 26.07 (C-3), 26.14 (C-5’#), 30.67
(C-5), 32.25 (C-2’*), 32.65 (C-6’*), 36.88 (C-1’), 50.44
(C-2), 54.53 (C-1); C-3, C-3’, C-4’, C-5’ were assigned
using lr INEPT spectroscopy. Anal. C11H22NBr (C, H,
N).
5.1.5.2. cis-2-Cyclopentylcyclopentaneamine
hydrobromide 4e HBr
5.1.6.1. 2-[(cis-2-Isopropylcyclopentyl)-
imino]hexahydroazepine hydrochloride 5a HCl
(1R, 2R)-4e HBr: m.p. 266–267 °C decomposition,
25 –21.9°; (1S, 2S)-4e HBr: m.p. 270–271 °C decom-
(1’R, 2’R)-5a HCl: m.p. 256–258 °C decomposition,
␣
͓ ͔
D
25
D
25
1
␣
͓ ͔
–122.8°; (1’S, 2’S)-5a HCl: m.p. 259–263 °C
position,
␣
+21.6°; H-NMR (CDCl3) δ 0.95–1.23
͓ ͔
D
(m, 2H, H-5’b, H-2’b), 1.42–2.04 (m, 12H, 3’-CH2,
4’-CH2, H-2, H-4b, H-1’, 3-CH2, H-2’b*, H-5b, H-4a),
2.07–2.26 (m, 2H, H-5a, H-5’b*), 3.65 (mc, 1H, H-1),
7.95 (s (br), 3H, +NH3); 13C-NMR (CDCl3) δ 21.24
(C-4), 24.81 (C-4’#), 25.13 (C-3’#), 27.65 (C-3), 30.85
decomposition, ͓␣͔25 +115.9°; 1H-NMR (CDCl3) δ 0.81
(d, 3H (Z), 6.4, 3″-CDH3*), 0.85 (d, 3H (E), 6.4, 3″-CH3*),
0.87 (d, 3H (Z), 6.4, 2″-CH3*), 0.90 (d, 3H (E), 6.4,
2″-CH3*), 1.40–2.05 (m, 14H, H-4’b, H-2’, 6-CH2,
H-3’b, H-5’b, 4-CH2, 5-CH2, H-4’a, H-1″, H-3’a, H-5’a),

387
2.66 (mc, 2H (E), 3-CH2), 2.91 (dd, 1H (Z), 14.0, 7.7,
H-3b), 3.19 (dd, 1H (Z), 14.0, 9.6, H-3a), 3.32–3.39 (m,
2H (E), 7-CH2), 3.39–3.51 (m, 1H (Z), H-7b), 3.51–3.64
(m, 1H (Z), H-7a), 3.94 (mc, 1H (E), H-1’), 4.40 (mc, 1H
(Z), H-1’), 9.30 (d, 1H (Z), 9.3, NHexo), 9.68 (d, 1H (E),
9.3, NHexo), 9.72 (t, 1H (Z), 5.6, NHendo), 10.43 (t, 1H
(E), 5.6, NHendo); 13C-NMR (CDCl3) δ Z-5a HCl: 22.03
(C-2″, C-3″), 22.10 (C-4’), 23.97 (C-4), 27.68 (C-3’),
27.76 (C-1″), 28.29 (C-6), 29.70 (C-5), 31.16 (C-3),
33.15 (C-5’), 43.59 (C-7), 51.61 (C-2’), 55.80 (C-1’),
168.02 (C-2); E-5a HCl: 21.03 (C-4’), 21.75 (C-3″*),
21.83 (C-2″*), 22.95 (C-4), 26.37 (C-3), 27.58 (C-3’),
27.89 (C-1″), 28.29 (C-6), 29.70 (C-5), 34.74 (C-5’),
43.74 (C-7), 52.16 (C-2’), 57.37 (C-1’), 171.46 (C-2); IR
(KBr, cm–1) 3 434 (ν (NH)), 1 641 (ν (C=N)). Anal.
C14H27N2Cl (C, H, N).
1.03–1.31 (m, 3H, H-3″b, H-4″b, H-5″b), 1.42–2.07 (m,
19H, H-6b, H-4’b, H-2″a, H-1″, H-4b, H-4″a, H-5b,
H-3″a, H-5″a, H-2’, H-5’b, H-6a, H-4a, H-6″a, H-3’b,
H-4’a, H-5a, H-3’a, H-5’a), 2.66 (mc, 2H (E), 3-CH2),
2.90 (dd, 1H (Z), 13.8, 10.1, H-3b), 3.24 (dd, 1H (Z),
13.8, 7.9, H-3a), 3.38 (mc, 2H (E), 7-CH2), 3.46 (mc, 1H
(Z), H-7b), 3.61 (mc, 1H (Z), H-7a), 3.99 (s (br), 1H (E),
H-1’), 4.47 (s (br), 1H (Z), H-1’), 8.64 (d, 1H (Z), 7.3,
+
+NHexo), 9.09 (d, 1H (E), 7.3, NHexo), 9.20 (s (br), 1H
(Z), NHendo), 10.06 (s (br), 1H (E), NHendo); 13C-NMR
(CDCl3) δ Z-5c HBr: 21.89 (C-4’), 23.99 (C-4), 25.91
(C-3″*), 26.00 (C-5″*), 26.36 (C-4″), 27.25 (C-3’), 28.23
(C-6), 29.66 (C-5), 31.71 (C-3), 32.26 (C-2″#), 32.56
(C-6″#), 32.89 (C-5’), 37.20 (C-1″), 43.61 (C-7), 50.25
(C-2’), 55.63 (C-1’), 168.10 (C-2); E-5c HBr: 20.80
(C-4’), 23.03 (C-4), 25.69 (C-3″*), 25.73 (C-5″*), 26.23
(C-4″), 26.66 (C-3), 27.61 (C-3’), 28.23 (C-6), 29.66
(C-5), 32.13 (C-2″#), 32.34 (C-6″#), 34.58 (C-5’), 36.90
(C-1″), 43.83 (C-7), 50.74 (C-2’), 57.06 (C-1’), 171.44
(C-2); IR (KBr, cm–1) 3 432 (ν (NH)), 1 643 (ν (C=N)).
Anal. C17H31N2Br (C, H, N).
5.1.6.2. 2-[(cis-2-Phenylcyclopentyl)-
imino]hexahydroazepine hydrochloride 5b HCl
25
(1’R, 2’R)-5b HCl: m.p. 184–185 °C, ␣
͓ ͔
+113.4°;
–114.1°;
D
25
(1’S, 2’S)-5b HCl: m.p. 182–184 °C,
␣
͓ ͔
D
1H-NMR (CDCl3) δ 0.82 (mc, 1H, H-4b), 0.99 (mc, 1H,
H-6b), 1.11–1.67 (m, 5H, H-5b, H-6a, H-4a, H-5a,
H-4’b), 1.98–2.12 (m, 2H, H-4’a, H-3’b), 2.13–2.23 (m,
1H, H-3’a), 2.28 (mc, 2H (Z) + 1H (E), 5’-CH2 (Z), H-3b
(E)), 2.43 (dd, 1H (Z), 14.2, 11.2, H-3b), 2.82 (dd, 1H
(Z), 14.2, 7.8, H-3a), 3.05–3.18 (m, 1H (Z) + 2H (E),
H-7b (Z + E), H-2’ (E)), 3.18–3.25 (m, 2H (E), H-3a,
H-7a), 3.25–3.37 (m, 1H (Z), H-7a), 3.69 (ddd, 1H (Z),
J2’-H/3’a-H = 11.5, J2’-H/3’-b = J2’-H/1’-H = 7.5, H-2’), 4.04
(mc, 1H (E), H-1’), 4.85 (mc, 1H (Z), H-1’), 7.18 (tt, 1H
(Z), 7.1, 1.2, H-4″), 7.24–7.34 (m, 2H (Z) + 5H (E), H-3″,
H-5″ (Z), H-2″, H-3″, H-4″, H-5″, H-6″ (E)), 7.59 (td, 2H
(Z), 7.1, 1.2, H-2″, H-6″), 9.29 (t, 1H (Z), 5.0, NHendo),
9.59 (d, 1H (Z), 8.8, NHexo), 10.06 (s (br), 1H (E),
NHexo), 10.09 (s (br), 1H (E), NHendo); 13C-NMR (CDCl3)
δ Z-5b HCl: 22.83 (C-4), 24.33 (C-4’), 27.08 (C-6),
29.37 (C-5), 31.10 (C-3), 31.91 (C-5’), 32.29 (C-3’),
43.26 (C-7), 49.58 (C-2’), 56.07 (C-1’), 126.19 (C-4″),
127.54 (C-3″, C-5″), 129.92 (C-2″, C-6″), 139.81 (C-1″),
167.08 (C-2); E-5b HCl: 21.94 (C-4), 24.33 (C-4’), 25.83
(C-3), 27.34 (C-6), 28.62 (C-5), 31.63 (C-3’), 31.91
(C-5’), 43.66 (C-7), 51.88 (C-2’), 59.65 (C-1’), 127.10
(C-4″), 128.38 (C-3″, C-5″), 129.16 (C-2″, C-6″), 138.58
(C-1″), 171.77 (C-2); IR (KBr, cm–1) 3 430 (ν (NH)),
1 650 (ν (C=N)). Anal. C17H25N2Cl (C, H, N).
5.1.6.4. 2-[(cis-2-Benzylcyclopentyl)-
imino]hexahydroazepine hydrochloride 5d HCl
25
(1’R, 2’R)-5d HCl: m.p. 164–166 °C,
(1’S, 2’S)-5d HCl: m.p. 164–166 °C,
␣
␣
͓ ͔
+30.4°;
–30.3°;
͓ ͔
D
25
D
1H-NMR (CDCl3) δ 1.33–1.69 (m, 9H, H-4’b, 6-CH2,
4-CH2, 5-CH2, 3’-CH2), 1.77 (mc, 1H, H-4’a), 2.02 (mc,
2H, 5’-CH2), 2.35 (m, 1H (E), H-2’), 2.51 (part B, 1H,
J
BA = 13.0, JBX = 9.6, H-7″b), 2.63 (mc, part X, 1H (Z),
H-2’), 2.72 (mc, 2H, 3-CH2), 2.87 (part A, 1H, JAB
=
13.0, JAX = 5.4, H-7″a), 3.27 (mc, 2H (E), 7-CH2), 3.36
(s (br), 2H (Z), 7-CH2), 3.55 (s (br), 1H (E), H-1’), 4.39
(mc, 1H (Z), H-1’), 7.02–7.09 (m, 1H, H-4″), 7.10–7.19
(m, 4H, H-2″, H-3″, H-5″, H-6″), 9.60 (s (br), 1H (Z),
NHendo), 9.75 (d, 1H (Z + E), 6.4, NHexo), 10.08 (s (br),
1H (E), NHendo); 13C-NMR (CDCl3) δ Z-5d HCl: 21.92
(C-4’), 23.93 (C-4), 28.16 (C-6), 29.75 (C-5), 29.85
(C-3’), 30.93 (C-5’), 31.41 (C-3), 35.45 (C-7″), 43.16
(C-2’), 43.71 (C-7), 56.44 (C-1’), 125.55 (C-4″), 128.04
(C-2″, C-6″), 128.83 (C-3″, C-5″), 141.10 (C-1″), 168.18
(C-2); E-5d HCl: 21.65 (C-4’), 22.19 (C-4), 26.15 (C-3),
27.67 (C-6), 29.42 (C-5), 29.59 (C-3’), 33.56 (C-5’),
35.45 (C-7″), 43.71 (C-7), 45.19 (C-2’), 57.36 (C-1’),
126.03 (C-4″), 128.35 (C-2″, C-6″), 128.60 (C-3″, C-5″),
140.33 (C-1″), 171.49 (C-2); IR (KBr, cm–1) 3 436 (ν
(NH)), 1 644 (ν (C=N)). Anal. C18H27N2Cl (C, H, N).
5.1.6.3. 2-[(cis-2-Cyclohexylcyclopentyl)-
imino]hexahydroazepine hydrobromide 5c HBr
5.1.6.5. 2-[(cis-2-Cyclopentylcyclopentyl)-
imino]hexahydroazepine hydrobromide 5e HBr
25
(1’R, 2’R)-5c HBr: m.p. 218–220 °C,
(1’S, 2’S)-5c HBr: m.p. 217–218 °C,
␣
␣
͓ ͔
–75.0°;
+71.8°;
͓ ͔
D
25
(1’R, 2’R)-5e HBr: m.p. 227–229 °C decomposition,
D
25
D
1H-NMR (CDCl3) δ 0.88 (mc, 2H, H-2″b, H-6″b),
␣
͓ ͔
–55.4°; (1’S, 2’S)-5e HBr: m.p. 227–229 °C de-

388
25
+55.4°; 1H-NMR (CDCl3)
δ
5.1.6.7. 3-[(cis-2-Isopropylcyclopentyl)imino]-
2-azabicyclo[2.2.2]octane hydrochloride 6a HCl
composition,
␣
͓ ͔
1.00–1.22 (m, 2H, DH-2″b, H-5″b), 1.37–2.12 (m, 20H,
3″-CH2, 4″-CH2, H-4’b, H-4b, H-5″a, 6-CH2, H-5b,
H-2″a, H-3’b, H-4a, H-2’, H-4’a, H-5’b, H-5a, H-3’a,
H-1″, H-5’a), 2.64 (mc, 2H (E), 3-CH2), 2.92 (dd, 1H (Z),
14.2, 9.3, H-3b), 3.15 (dd, 1H (Z), 14.3, 7.9, H-3a), 3.37
(dd, 2H (E), 10.0, 5.4, 7-CH2), 3.42–3.65 (m, 2H (Z),
7-CH2), 3.92 (mc, 1H (E), H-1’), 4.44 (mc, 1H (Z), H-1’),
8.76 (d, 1H (Z), 9.3, +NHexo), 9.16 (t, 1H (Z), 5.0,
NHendo), 9.23 (d, 1H (E), 8.3, +NHexo), 10.03 (s (br), 1H
(E), NHendo); 13C-NMR (CDCl3) δ Z-5e HBr: 22.38
(C-4’), 23.92 (C-4), 25.27 (C-4″#), 25.40 (C-3″#), 28.28
(C-6), 29.13 (C-3’), 29.74 (C-5), 31.85 (C-5″*), 31.83
(C-3), 32.40 (C-2″*), 32.66 (C-5’), 39.79 (C-1″), 43.72
(C-7), 49.88 (C-2’), 56.85 (C-1’), 168.01 (C-2); E-5e
HBr: 21.45 (C-4’), 23.04 (C-4), 25.11 (C-4″#), 25.14
(C-3″#), 26.69 (C-3), 27.71 (C-6), 28.12 (C-3’), 29.74
(C-5), 31.74 (C-5″*), 32.26 (C-2″*), 34.46 (C-5’), 39.94
(C-1″), 43.87 (C-7), 50.90 (C-2’), 58.60 (C-1’), 171.37
(C-2); IR (KBr, cm–1) 3 433 (ν (NH)), 1 644 (ν (C=N)).
Anal. C16H29N2Br (C, H, N).
(1’R, 2’R)-6a HCl: m.p. 229–231 °C decomposition,
25
D
␣
͓ ͔
–103.3°; (1’S, 2’S)-6a HCl: m.p. 226–228 °C
25
decomposition, ␣ +107.5°; 1H-NMR (CDCl3) δ 0.84
͓ ͔
(d, 3H (Z), 6.6, 3″-CDH3*), 0.86 (d, 3H (E), 6.8, 3″-CH3*),
0.90 (d, 3H (Z), 6.4, 2″-CH3*), 0.93 (d, 3H (E), 6.4,
2″-CH3*), 1.48–2.10 (m, 16H, H-4’b, H-2’, H-8a#, H-5’b,
#
5-CH2 , 6-CH2, 7-CH2, H-8b#, 3’-CH2, H-4’a, H-1″,
H-5’a), 3.14 (mc, 1H (E), H-4), 3.80 (s (br), 1H (Z) + 1H
(E), H-4 (Z), H-1 (E)), 3.99 (mc, 1H (E), H-1’), 4.00 (s
(br), 1H (Z), H-1), 4.37 (mc, 1H (Z), H-1’), 9.50 (d, 1H
(Z), 9.3, NHexo), 9.77 (d, 1H (E), 9.0, NHexo), 10.35 (s
(br), 1H (E), NHendo), 10.52 (d, 1H (Z), 6.0, NHendo);
13C-NMR (CDCl3) δ Z-6a HCl: 22.01 (3″-CH3*), 22.06
(C-4’), 22.09 (2″-CH3*), 22.87 (C-5#), 23.54 (C-8#),
26.78 (C-6, C-7), 27.77 (C-3’), 28.06 (C-1″), 31.33 (C-4),
33.19 (C-5’), 45.97 (C-1), 51.78 (C-2’), 55.42 (C-1’),
167.66 (C-3), E-6a HCl: 21.02 (C-4’), 21.74 (3″-CH3*),
21.83 (2″-CH3*), 23.02 (C-5#), 23.07 (C-8#), 25.91
(C-7§), 26.35 (C-6§), 26.99 (C-3’), 28.06 (C-1″), 28.36
(C-4), 34.81 (C-5’), 46.32 (C-1), 52.08 (C-2’), 56.54
(C-1’), 170.52 (C-3); IR (KBr, cm–1) 3 435 (ν (NH)),
1 665 (ν (C=N)). Anal. C15H27N2Cl (C, H, N).
5.1.6.6. 2-[(cis-2-Cyclohexylmethylcyclopentyl)-
imino]hexahydroazepine hydrobromide 5f HBr
(1’R, 2’R)-5f HBr: m.p. 217–218 °C, (1’R, 2’R)-5f:
25
D
␣
͓ ͔
+27.3° (free base); (1’S, 2’S)-5f HBr: m.p.
25
D
1
216–217 °C, ␣
͓ ͔
–41.0°; 5f HCl: H-NMR (CDCl3) δ
5.1.6.8. 3-[(cis-2-Phenylcyclopentyl)imino]-
0.66–0.90 (m, 2H, H-2″b, H-6″b), 0.98–1.28 (m, 6H,
H-3″b, H-4″b, H-5″b, H-1″, 7″-CH2), 1.36–1.89 (m, 16H,
H-4’b, H-5b, H-3″a, H-4″a, H-5″a, H-3’b, 4-CH2, 6-CH2,
H-2″a, H-6″a, H-5a, H-3’a, H-4’a, H-5’b), 1.96 (mc, 1H,
H-5’a), 2.19–2.32 (m, 1H, H-2’), 2.58–2.65 (m, 2H (E),
3-CH2), 2.73–2.86 (m, 1H (Z), H-3b), 3.04–3.17 (m, 1H
(Z), H-3a), 3.32–3.47 (m, 1H (Z) + 2H (E), H-7b (Z),
2-azabicyclo[2.2.2]octane hydrochloride 6b HCl
(1’R, 2’R)-6b HCl: m.p. 255–257 °C decomposition,
25
D
␣
+96.3°; (1’S, 2’S)-6b HCl: m.p. 256–258 °C
͓ ͔
25
1
decomposition, ␣
͓ ͔
–94.2°; H-NMR (CDCl3) δ 0.72
(mc, 1H, H-8b#), 0.D92 (mc, 1H, H-7b*), 1.34 (mc, 2H,
H-8a#, H-7a*), 1.46–1.72 (m, 5H, 5-CH2, 6-CH2, H-4’b),
1.98–2.16 (m, 2H, H-3’b, H-4’a), 2.16–2.41 (m, 3H,
H-3’a, 5’-CH2), 2.55 (s (br), 1H (E), H-4), 3.16 (mc, 1H
(E), H-2’), 3.26 (s (br), 1H (Z), H-4), 3.65 (ddd, 1H (Z),
11.7, 7.5, 7.5, H-2’), 3.72 (s (br), 1H (Z + E), H-1), 4.04
(mc, 1H (E), H-1’), 4.78 (mc, 1H (Z), H-1’), 7.12–7.37
(m, 3H (Z) + 5H (E), H-3″, H-4″, H-5″ (Z), H -2″, H-3″,
H-4″, H-5″, H-6″ (E)), 7.62 (d, 2H (Z), 7.1, H-2″, H-6″),
9.84 (d, 1H (Z), 9.3, NHexo), 9.92 (d, 1H (E), 3.8,
NHendo), 10.05 (d, 1H (E), 10.6, NHexo), 10.16 (d, 1H (Z),
3.8, NHendo); 13C-NMR (CDCl3) δ Z-6b HCl: 22.52
(C-8#), 23.07 (C-5#), 24.25 (C-4’), 26.28 (C-7*), 26.37
(C-6*), 31.19 (C-4), 32.00 (C-3’), 32.08 (C-5’), 45.51
(C-1), 49.86 (C-2’), 55.73 (C-1’), 126.24 (C-4″), 127.54
(C-3″, C-5″), 129.93 (C-2″, C-6″), 139.84 (C-1″), 166.74
(C-3), E-6b HCl: 21.91 (C-8#), 24.25 (C-4’), 26.05
(C-7*), 26.20 (C-6*), 28.62 (C-4), 30.95 (C-3’), 32.64
(C-5’), 46.32 (C-1), 51.83 (C-2’), 59.02 (C-1’), 127.10
(C-4″), 128.44 (C-3″, C-5″), 129.14 (C-2″, C-6″), 138.74
7-CH (E)), 3.49–3.61 (m, 1H (Z), H-7a), 3.86 (mc, 1H
2
(E), H-1’), 4.31 (mc, 1H (Z), H-1’), 9.44 (d, 1H (Z), 8.6,
+
+NHexo), 9.53 (d, 1H (E), 8.6, NHexo), 9.59, (t, 1H (Z),
5.0, NHendo), 10.15 (t, 1H (E), 5.0, NHendo); 13C-NMR
(CDCl3) δ Z-5f HBr: 22.00 (C-4’), 24.11 (C-4), 26.22
(C-3″*), 26.31 (C-5″*), 26.52 (C-4″), 28.32 (C-6), 29.69
(C-3’), 29.83 (C-5), 31.14 (C-5’), 31.41 (C-3), 32.69
(C-2″#), 34.39 (C-6″#), 35.97 (C-1″), 36.45 (C-7″), 39.44
(C-2’), 43.66 (C-7), 56.72 (C-1’), 167.93 (C-2); E-5f
HBr: 21.23 (C-4’), 22.93 (C-4), 26.22 (C-3″*), 26.31
(C-5″*), 26.39 (C-4″, C-3), 27.69 (C-6), 28.66 (C-3’),
30.07 (C-5), 32.85 (C-2″#), 33.31 (C-5’), 34.03 (C-6″#),
36.06 (C-1″), 37.01 (C-7″), 41.16 (C-2’), 43.81 (C-7),
58.81 (C-1’), 171.39 (C-2); 5f HBr: IR (KBr, cm–1) 3 440
(ν (NH)), 1 650 (ν (C=N)); (1’R, 2’R)-5f HBr Anal.
C18H33N2Br (H, N); C: calcd, 60.50; found, 59.67; (1’S,
2’S)-5f HBr Anal. C18H33N2Br (H, N); C: calcd, 60.50;
found, 59.77.

389
(C-1″), 170.54 (C-3); C-5# (E) hidden; IR (KBr, cm–1)
3 435 (ν (NH)), 1 656 (ν (C=N)). Anal. C18H25N2Cl (C,
H, N).
(C-4), 29.08 (C-3’), 33.87 (C-5’), 35.33 (C-7″), 45.17
(C-2’), 46.39 (C-1), 56.58 (C-1’), 126.06 (C-4″), 128.23
(C-2″, C-6″), 128.42 (C-3″, C-5″), 140.32 (C-1″), 170.61
(C-3); IR (KBr, cm–1) 3 440 (ν (NH)), 1 642 (ν (C=N)).
Anal. C19H27N2Cl (C, H, N).
5.1.6.9. 3-[(cis-2-Cyclohexylcyclopentyl)imino]-
2-azabicyclo[2.2.2]octane hydrobromide 6c HBr
(1’R, 2’R)-6c HBr: m.p. 227–228 °C decomposition,
5.1.6.11. 3-[(cis-2-Cyclopentylcyclopentyl)imino]-
2-azabicyclo[2.2.2]octane hydrobromide 6e HBr
25
␣
–59.6°; (1’S, 2’S)-6c HBr: m.p. 225–227 °C de-
͓ ͔
D
25
25
1
(1’R, 2’R)-6e HBr: m.p. 211–212 °C,
(1’S, 2’S)-6e HBr: m.p. 209–211 °C,
␣
␣
͓ ͔
–47.6°;
+49.2°;
͓ ͔
composition,
␣
͓ ͔
+61.1°; H-NMR (CDCl3) δ 0.87
D
(mc, 2H, H-2″b, DH-6″b), 1.01–1.28 (m, 3H, H-3″b,
H-4″b, H-5″b), 1.38–2.05 (m, 21H, H-1″, H-4’b, H-2″a,
H-8b, H-2’, H-3″a, H-4″a, H-5″a, 5-CH2, 6-CH2, 7-CH2,
H-5’b, H-6″a, H-8a, H-3’b, H-4’a, H-3’a, H-5’a), 3.12 (s
(br), 1H (E), H-4), 3.83 (s (br), 1H (Z) + 1H (E), H-4 (Z),
H-1 (E)), 3.92 (s (br), 1H (E), H-1’), 4.00 (s (br), 1H (Z),
H-1), 4.40 (mc, 1H (Z), H-1’), 8.86 (d, 1H (Z), 9.8,
25
D
1H-NMR (CDCl3) δ 1.02–1.23 (m, 2H, H-2″b, H-5″b),
1.36–2.10 (m, 22H, 3″-CH2, 4″-CH2, H-4’b, H-5″a*,
#
H-8b#, 5-CH2 , 6-CH2, 7-CH2, H-2″a*, H-5b, H-8a#,
3’-CH2, H-2’, H-4’a, H-1″, H-5’a), 3.13 (s (br), 1H (E),
H-4), 3.79 (s (br), 1H (Z) + 1H (E), H-4 (Z), H-1 (E)),
3.94 (mc, 1H (E), H-1’), 4.01 (s (br), 1H (Z), H-1), 4.36
(mc, 1H (Z), H-1’), 8.97 (d, 1H (Z), 9.6, +NHexo), 9.34 (d,
1H (E), 7.9, +NHexo), 9.92 (s (br), 1H (Z), NHendo), 9.97
(s (br), 1H (E), NHendo); 13C-NMR (CDCl3) δ Z-6e HBr:
22.34 (C-4’), 22.99 (C-5§), 23.47 (C-8§), 25.20 (C-4″#),
25.34 (C-3″#), 26.76 (C-6, C-7), 29.10 (C-3’), 31.82 (C-4,
C-5″*), 32.36 (C-2″*), 32.67 (C-5’), 40.03 (C-1″), 46.01
(C-1), 50.05 (C-2’), 56.39 (C-1’), 167.56 (C-3); E-6e
HBr: 21.31 (C-4’), 23.08 (C-5, C-8), 25.02 (C-3″, C-4″),
26.06 (C-7#), 26.37 (C-6#), 27.90 (C-3’), 28.57 (C-4),
31.67 (C-5″*), 32.12 (C-2″*), 34.55 (C-5’), 40.03 (C-1″),
46.49 (C-1), 50.73 (C-2’), 57.73 (C-1’), 170.43 (C-3); IR
(KBr, cm–1) 3 444 (ν (NH)), 1 663 (ν (C=N)). Anal.
C17H29N2Br (C, H, N).
+NHexo), 9.20 (d, 1H (E), 8.8, NHexo), 9.94 (s (br), 1H
+
(Z), NHendo), 10.02 (s (br), 1H (E), NHendo); 13C-NMR
(CDCl3) δ Z-6c HBr: 21.89 (C-4’), 22.76 (C-5(), 23.71
(C-8◆), 25.86 (C-3″§), 26.05 (C-5″§), 26.33 (C-4″), 26.68
(C-6#), 26.78 (C-7#), 27.30 (C-3’), 31.83 (C-4), 32.24
(C-2″*), 32.61 (C-6″*), 32.79 (C-5’), 37.57 (C-1″), 46.00
(C-1), 50.40 (C-2’), 55.24 (C-1’), 167.69 (C-3); E-6c
HBr: 20.73 (C-4’), 23.04 (C-5◆), 23.12 (C-8◆), 25.67
(C-3″#), 25.71 (C-5″ #), 26.19 (C-4″), 28.65 (C-4), 32.07
(C-2″*), 32.37 (C-6″*), 34.39 (C-5’), 37.03 (C-1″), 46.46
(C-1), 50.62 (C-2’), 56.19 (C-1’), 170.46 (C-3); C-6 (E),
C-7 (E), C-3’ (E) hidden; IR (KBr, cm–1) 3 429 (ν (NH)),
1 657 (ν (C=N)). Anal. C18H31N2Br (C, H, N).
5.1.6.10. 3-[(cis-2-Benzylcyclopentyl)imino]-
2-azabicyclo[2.2.2]octane hydrochloride 6d HCl
5.1.6.12. 3-[(cis-2-Cyclohexylmethylcyclopentyl)imino]-
2-azabicyclo[2.2.2]octane hydrobromide 6f HBr
25
(1’R, 2’R)-6d HCl: m.p. 244–246 °C decomposition,
(1’R, 2’R)-6f HBr: m.p. 233–234 °C,
(1’S, 2’S)-6f HBr: m.p. 233–234 °C,
␣
␣
͓ ͔
+37.9°;
–37.8°;
͓ ͔
D
25
D
25
␣
͓ ͔
+5.5°; (1’S, 2’S)-6d HCl: m.p. 243–245 °C de-
composition, ␣ –5.5°; 1H-NMR (CDCl3) δ 1.32–1.72
1H-NMR (CDCl3) δ 0.78–0.89 (m, 2H, H-2″Db, H-6″b),
1.00–1.32 (m, 6H, H-3″b, H-4″b, H-5″b, H-1″, H-7″b,
H-7″a), 1.49 (mc, 1H, H-4’b), 1.49–1.89 (m, 17H, H-3’b,
H-3″a, H-4″a, H-5″a, H-8b*, H-2″a, H-6″a, 5-CH2*,
6-CH2, 7-CH2, H-8a*, H-3’a, H-4’a, H-5’b), 1.99 (mc,
1H, H-5’a), 2.22 (mc, 1H, H-2’), 3.08 (s (br), 1H (E),
H-4), 3.71 (s(br), 1H (Z), H-4), 3.80 (s (br), 1H (E), H-1),
3.87 (mc, 1H (E), H-1’), 3.99 (s (br), 1H (Z), H-1), 4.27
(mc, 1H (Z), H-1’), 9.03 (d, 1H (Z), 9.3, +NHexo), 9.26 (d,
25
͓ ͔
D
(m, 11H, H-4’b, H-7b*, 6-CH2*, H-7a*, 3’-CH2, 5-CH2,
8-CH2), 1.72–1.86 (m, 1H, H-4’a), 1.92 (mc, 1H, H-5’a),
2.03 (mc, 1H, H-5’b), 2.34 (m, 1H (E), H-2’), 2.49 (mc,
part B, 1H (Z + E) + 1H (Z), JAB = 17.7, H-7″b (Z + E),
H-2’ (Z)), 2.80 (m, 1H (E), H-4), 2.92 (part A, 1H (Z +
E), JAB = 17.7, H-7″a), 3.63 (s (br), 1H (Z), H-4), 3.74 (s
(br), 2H (E), H-1, H-1’), 3.90 (s (br), 1H (Z), H-1), 4.35
(mc, 1H, H-1’), 7.02–7.09 (m, 1H, H-4″), 7.09–7.21 (m,
4H, H-2″, H-3″, H-5″, H-6″), 9.86 (d, 1H (Z + E), 8.8,
NHexo), 10.13 (s (br), 1H (E), NHendo), 10.48 (d, 1H (Z),
4.0, NHendo); 13C-NMR (CDCl3) δ Z-6d HCl: 22.49
(C-4’), 23.12 (C-8#), 23.42 (C-5#), 26.66 (C-7*), 26.75
(C-6*), 29.56 (C-3’), 31.87 (C-5’), 31.49 (C-4), 35.21
(C-7″), 44.10 (C-2’), 46.02 (C-1), 56.05 (C-1’), 125.56
(C-4″), 128.04 (C-2″, C-6″), 128.71 (C-3″, C-5″), 141.08
(C-1″), 167.79 (C-3); E-6d HCl: 21.49 (C-4’), 22.20
(C-8#), 23.12 (C-5#), 25.80 (C-7*), 26.47 (C-6*), 28.32
+
1H (E), 9.0, NHexo), 9.92 (d, 1H (Z + E), 3.7, NHendo);
13C-NMR (CDCl3) δ Z-6f HBr: 22.09 (C-4’), 23.04
(C-5*), 23.66 (C-8*), 26.28 (C-3″, C-5″), 26.53 (C-4″),
26.69 (C-7#), 26.84 (C-6#), 29.79 (C-3’), 31.27 (C-5’),
31.78 (C-4), 32.82 (C-6″ §), 34.28 (C-2″ §), 35.98 (C-1″),
36.63 (C-7″), 39.88 (C-2’), 45.98 (C-1), 56.62 (C-1’),
167.72 (C-3); E-6f HBr: 21.28 (C-4’), 23.04 (C-5*),
23.66 (C-8*), 26.28 (C-3″, C-5″), 26.40 (C-4″), 26.69
(C-7#), 26.84 (C-6#), 28.57 (C-4), 28.72 (C-3’), 32.94

390
(C-6″ §), 33.36 (C-5’), 34.02 (C-2″ §), 36.11 (C-1″), 37.22
(C-7″), 41.15 (C-2’), 46.52 (C-1), 58.28 (C-1’), 170.53
(C-3); IR (KBr, cm–1) 3 440 (ν (NH)), 1 657 (ν (C=N)).
Anal. C19H33N2Br (C, H, N).
H-5b), 1.81–2.00 (m, 3H, H-3a, H-4a, H-5a), 2.00–2.14
(m, 1H, H-2), 3.63 (dt, 1H, 4.9, 1.6, 1-H), 6.85 (dt, 1H,
7.4, 1.1, H-5″), 6.93 (dd, 1H, 8.2, 0.55, H-3″), 7.25 (dt,
1H, 8.2, 1.6, H-4″), 7.29 (dd, 1H, 8.6, 1.6, H-6″), 8.27 (s,
1H, H-α), 13.71 (s, 1H, OH); 13C-NMR (CDCl3) δ 22.39
(C-4), 26.37 (C-3’#), 26.39 (C-5’#), 26.65 (C-4’), 29.86
(C-3), 33.23 (C-2’*), 34.06 (C-6’*), 34.36 (C-5), 36.17
(C-1’), 38.09 (C-7’), 42.78 (C-2), 73.04 (C-1), 116.93
(C-3″), 118.31 (C-5″), 118.86 (C-1″), 130.98 (C-4″),
131.82 (C-6″), 161.36 (C-2″), 162.23 (C-α); IR (KBr,
cm–1) 3 447 (ν (OH)), 1 629 (ν (C=N)). Anal. C19H27NO
(C, H, N).
5.1.7. Preparation of cis-2-substituted
cyclopentaneaminesalicylidenes 7a–f
The cis-2-substituted cyclopentaneaminesalicylidenes
7a–f were prepared according to a common procedure
[16].
5.1.7.1. (1S, 2S)-cis-Cyclohexylcylo-
pentaneaminesalicylidene (1S, 2S)-7c
25
88%; m.p. 105–106°; ␣ +215°; 1H-NMR (CDCl3)
5.1.8. Correlation of absolute configuration
͓ ͔
δ 0.81–0.99 (m, 2H, H-6’b,DH-2’b), 0.99–1.26 (m, 4H,
H-3’b, H-4’b, H-5’b, H-1’), 1.48–1.79 (m, 9H, H-3b,
H-3’a, H-4’a, H-5’a, H-2, H-5b, H-6’a*, H-4b, H-2’a*),
1.82–2.01 (m, 3H, H-4a, H-3a, H-5a), 3.79 (t, 1H, 3.8,
H-1), 6.85 (dt, 1H, 7.5, 1.1, H-5″), 6.94 (d, 1H, 8.2,
H-3″), 7.25 (dt, 1H, 8.3, 1.6, H-4″), 7.29 (dd, 1H, 8.6, 1.8,
H-6″), 8.30 (s, 1H, H-α), 13.87 (s, 1H, OH); 13C-NMR
(CDCl3) δ 21.82 (C-4), 25.89 (C-5’#), 26.03 (C-3’#),
26.54 (C-4’), 27.55 (C-3), 31.71 (C-6’*), 32.38 (C-2’*),
34.73 (C-5), 37.89 (C-1’), 52.29 (C-2), 70.71 (C-1),
116.94 (C-3″), 118.31 (C-5″), 118.90 (C-1″), 130.99
(C-4″), 131.85 (C-6″), 161.36 (C-2″), 162.20 (C-α); IR
(KBr, cm–1) 3 431 (ν (OH)), 1 630 (ν (C=N)). Anal.
C18H25NO (C, H, N).
A solution of the cis-2-substituted cyclopentaneamine-
salicylidenes 7a–f in acetonitrile (c = 10–3 M) was
measured CD-spectroscopically. λ (nm) (θ). (1S, 2S)-7a:
313 (+1 725), 260 (+3 113), 228 (–2 381) [16]; (1S,
2S)-7b: 314 (+6 981), 252 (+12 929), 228 (+7 657) [16];
(1S, 2S)-7c: 312 (+7 276), 259 (+11 415), 225 (–3 763);
(1S, 2S)-7d: 311 (+3 469), 252 (+7 811), 220 (–1 649)
[16]; (1S, 2S)-7e: 311 (+2 864), 255 (+4 077), 225
(–3 664); (1S, 2S)-7f: 312 (+2 878), 256 (+3 747), 226
(–2 235).
5.1.9. Determination of enantiomeric excess
The diastereomeric Mosher amides: cis-2-substituted
cyclopentaneamine-α-methoxy-α-trifluoro-methylphenyl-
acetic acid amides (8a–f) and 1-phenylethyl-amine-α-
methoxy-α-trifluoromethyl-phenylacetic acid amides (8p)
were prepared according to a common procedure [16].
The residue was directly used for HPLC and 19F-NMR
spectroscopy.
5.1.7.2. (1S, 2S)-cis-Cyclopentylcylo-
pentaneaminesalicylidene (1S, 2S)-7e
25
D
1
87%; m.p. 84–85°; ␣
͓ ͔
+242°; H-NMR (CDCl3) δ
1.01–1.19 (m, 2H, H-5’b*, H-2’b*), 1.33–1.48 (m, 2H,
H-3’b, H-4’b), 1.48–1.80 (m, 9H, H-3’a, H-4’a, H-1’,
H-5’a*, H-3b, H-5b, H-2, H-4b, H-2’a*), 1.83–2.00 (m,
3H, H-3a, H-4a, H-5a), 3.69 (t, 1H, 4.1, H-1), 6.85 (dt,
1H, 7.6, 1.1, H-5″), 6.93 (d, 1H, 8.2, H-3″), 7.25 (dt, 1H,
7.9, 1.6, H-4″), 7.29 (dd, 1H, 8.2, 1.6, H-6″), 8.30 (s, 1H,
H-α), 13.77 (s, 1H, OH); 13C-NMR (CDCl3) δ 22.45
(C-4), 25.09 (C-4’#), 25.21 (C-3’#), 29.10 (C-3), 30.90
(C-5’*), 32.46 (C-2’*), 34.61 (C-5), 41.03 (C-1’), 53.01
(C-2), 72.58 (C-1), 116.95 (C-3″), 118.34 (C-5C″), 118.87
(C-1″), 130.99 (C-4″), 131.86 (C-6″), 161.38 (C-2″),
162.06 (C-α); IR (KBr, cm–1) 3 452 (ν (OH)), 1 631 (ν
(C=N)). Anal. C17H23NO (C, H, N).
5.1.9.1. HPLC
Twenty microlitres of a 10–20 mg/mL solution of the
Mosher amides in ethyl acetate were injected on a
Lichrosorbt Si-60 column (5 µm). The flow rate was 2
and 3 mL/min, respectively. The Mosher amides were
detected at 260 nm (λmax): retention time (min) major
diastereomer/subsidiary diastereomer, α (ee%). (R)-8p
17.06/22.52, α = 1.32 (98.3%); (S)-8p: 21.49/18.76, α =
1.15 (95.8%); (1R, 2R)-8a: 9.73–10.26/–* (> 99.0%);
(1S, 2S)-8a: 12.87–13.40/–* (> 99.0%); (1R, 2R)-8a +
(1S, 2S)-8a#: 9.87–10.40/13.07, α = 1.26; (1R, 2R)-8b◆:
23.11/17.68, α = 1.31 (99.9%); (1S, 2S)-8b◆: 16.80/–*
(> 99.0%); (1R, 2R)-8c: 2.77–3.54/–* (> 99.0%); (1S,
2S)-8c: 4.90/–* (> 99.0%); (1R, 2R)-8c + (1S, 2S)-8c#:
3.40/4.70, α = 1.38; (1R, 2R)-8d◆: 12.54/–* (> 99.0%);
(1S, 2S)-8d◆: 18.19/14.59, α = 1.25 (98.8%); (1R,
2R)-8e: 4.90–6.61/–* (> 99.0%); (1S, 2S)-8e:
7.64–9.53/–* (> 99.0%); (1R, 2R)-8e + (1S, 2S)-8e#:
5.40–6.50/8.07–8.30, α = 1.24; (1R, 2R)-8f: 5.04–7.14/–*
5.1.7.3. (1S, 2S)-cis-Cyclohexylmethylcylo-
pentaneaminesalicylidene (1S, 2S)-7f
25
D
1
87%; m.p. 66–67°; ␣
͓ ͔
+216°; H-NMR (CDCl3) δ
0.79–0.92 (m, 2H, H-6’b*, H-2’b*), 1.02–1.29 (m, 6H,
H-7’b, H-7’a, H-3’b, H-4’b, H-5’b, H-1’), 1.43–1.79 (m,
8H, H-3b, H-3’a, H-4’a, H-5’a, H-6’a*, H-2’a*, H-4b,

391
(> 99.0%); (1S, 2S)-8f: 10.00–10.50/–* (> 99.0%); (1R,
2R)-8f + (1S, 2S)-8f#: 5.91–7.17/9.51–10.90, α = 1.33;
*below detection limit (< 0.5%), #artificial mixture;
◆3 mL/min.
and 125I-labelled insulin (CIS Diagnostik GmbH, Dreie-
ich, Germany).
5.2.3. Patch clamp measurements
For patch clamp measurements, cells (0.2 × 106 cells)
were seeded onto glass coverslips coated with poly-L-
Orn (10 mg/L, Sigma, Munich, Germany) and used after
2–5 days of culture in standard culture medium. The
coverslip was mounted on the stage of an inverted
microscope (Axiovert 10, Zeiss, Germany) and perifused
continuously at 37 °C with a solution containing (in
mM): 140 NaCl, 5.6 KCl, 1.2 MgCl2, 2.6 CaCl2, 0.5
Glucose and 10.0 Hepes, pH 7.4. A flowing KCl-electrode
served as a reference and appropriate corrections for
liquid junction voltages were made. The patch clamp
pipettes (Clark-Medical, Reading, GB) were pulled auto-
matically (DMZ Universal Puller, Zeitz, Augsburg, Ger-
many) and had an input R of 6–8 MX. They were filled
with a solution containing in (mM): 30 KCl, 95 potas-
sium gluconate, 1.0 MgCl2, 1.2 NaH2PO4, 4.8 Na2HPO4,
5.0 Na2ATP, 0.5 Na2GTP, pH 7.15. After a GX seal was
achieved, the standard whole cell configuration was
obtained by sucking of the membrane underneath the
pipette. Membrane voltage (Vm) was measured continu-
ously using the current-clamp-mode of the patch-clamp
amplifier (Fröbe, U. and Busche, R., Institute of Physi-
ology of the University of Freiburg, Germany) and was
displayed by a pen recorder. After pipette capacitance was
compensated, the access conductance (Ga) was measured
as previously described [39]. Whole cell conductance
(Gm) was calculated from the total conductance (Gt)
measured in the voltage clamp mode using the equation
Gm = Ga (Gt /(Ga – Gt).
5.1.9.2. 19F-NMR
CDCl3, standard: 1,1,2-trichlorotrifluoroethane, δ ma-
jor diastereomer/subsidiary diastereomer (ee%). 8-(R)-
PEA: –64.60/–64.42 (97.9%); 8-(S)-PEA: –64.42/–64.60
(97.0%); (1R, 2R)-8a: –64.72/–* (> 99.0%); (1S, 2S)-8a:
–64.55/–64.71 (97.4%); (1R, 2R)-8b: –65.21/–64.38
(99.0%); (1S, 2S)-8b: –64.38/–* (> 99.0%); (1R, 2R)-8c:
–64.63/–* (> 99.0%); (1S, 2S)-8c: –64.58/–* (> 99.0%);
(1R, 2R)-8d: –64.49/–* (> 99.0%); (1S, 2S)-8d: –64.26/
–64.49 (98.8%); (1R, 2R)-8e: –64.59/–* (> 99.0%); (1S,
2S)-8e: –64.53/–* (> 99.0%); (1R, 2R)-8f: –64.60/–64.40
(99.6%); (1S, 2S)-8f: –64.40/–* (> 99.5%); *below
detection limit (< 0.5%).
5.2. Physiology
5.2.1. Cell culture
The effects of semicyclic amidines on insulin secre-
tion, [Ca2+]i and membrane voltage were tested in the
insulin secreting cell line INS-1 [37]. Cells were cultured
at a cell density of 250 × 106 cells/L in RPMI 1640
culture medium (GIBCO BRL, Eggenstein, Germany)
containing 11 mM glucose and supplemented with 50 µM
2-mercaptoethanol, 1 mM sodiumpyruvate, 10 mM
Hepes, 10 0000 IU/L penicillin, 100 mg/L streptomycin
and 100 g/L decomplemented foetal calf serum (Se-
romed, Berlin, Germany). The medium was changed
twice a week. Every week cells were trypsinized and
re-seeded at the same cell density.
5.2.4. [Ca2+]i-Measurements
5.2.2. Insulin secretion
For measurements of insulin secretion cells (0.25 × 106
cells per well) were cultured for 2 days in 24-well dishes.
After washing the cells with a modified Krebs-Ringer-
hydrogen carbonate (KRB) buffer containing (in mM):
136 NaCl, 4.7 KCl, 1.2 MgSO4, 1.0 CaCl2, 1.2 KH2PO4,
5.0 NaHCO3, 0.5 glucose, 10.0 Hepes, pH 7.4 and 5 g/L
BSA (fraction V, Sigma, Munich, Germany), they were
pre-incubated at 37 °C for 30 min in the same buffer and
incubated for another 30 min in buffer containing the
tested compounds as indicated in each experiment. Insu-
lin released into the supernatant and insulin content of the
cells after acidic ethanol (1.5% (v/v) HCl in 75% (v/v)
ethanol) extraction was measured by a radio-immunoassay
as described previously [38] using rat insulin (Novo
Nordisk, Bagsvaerd, Denmark) as standard, rabbit anti-
insulin porcine antiserum (Linco, Biotrend, Germany)
For [Ca2+]i measurements cells were incubated at room
temperature with 2 µmol/L fura-2 AM for 30 min. The
fluorescence was monitored with a photomultiplier sys-
tem at excitation wavelengths of 340, 360 and 380 nm,
with a filter rotation rate of 10 per second. As a measure
of [Ca2+]i the fluorescence emission ratio at 340/380 nm
excitation was calculated after subtraction of the auto-
fluorescence as described previously [40].
5.2.5. Statistical analysis
Data are presented as mean ± SEM. Student’s paired
t-test (for patch clamp data) or Mann-Whitney U- Wil-
coxon Rank Sum W Test (for insulin secretion) (P < 0.05)
was used for statistical analyses.
Supporting information is available (4 pages).

392
[15] Palfreyman M.G., Dudley M.W., Cheng H.C., Mir A.K., Yamada S.,
Roeske W.R., Obata T., Yamamura H.I., Biochem. Pharm. 38 (1989)
2459–2465.
Acknowledgements
S. Hartmann would like to thank the ‘Dr. Hilmer-
Stiftung im Stifterverband für die Deutsche Wissen-
schaft’ for the award of a scholarship. The authors are
indebted to Mrs M. Wagner and Mr V. Brecht for
conscientious CD and NMR measurements. We are
gratefully indebted to the medical students Mrs S. Lehr
and Mr M. Herbst for electrophysiological and [Ca2+]i
measurements. This work was supported by the grant
‘interdisziplinäre Projekte’ of the University of Freiburg
(Germany). Financial and substantial support from ‘Fonds
der Chemischen Industrie’ and Degussa, Frankfurt a. M.
and a generous gift of 3-ethoxy-2-azabicyclo[2.2.2]oct-
2-ene by Prof. Dr W. Schneider, Freiburg (Germany) is
gratefully acknowledged.
[16] Wiehl W., Frahm A.W., Chem. Ber. 119 (1986) 2668–2677.
[17] Wiehl W., Thesis, Bonn 1985.
[18] Cornubert R., Borrel C., Chim. Fr. 47 (1930) 301–322 modified
according to Wiehl.
[19] Kötz A., Blendermann K., Kárpáti E., Rosenbusch R., Liebigs Ann.
Chem. 400 (1913) 47–51.
[20] Wanzlick H.W., Gollmer G., Milz H., Chem. Ber. 88 (1955) 69–77.
[21] Arnold R.T., Buckley J.S., Dodson R.M., J. Am. Chem. Soc. 72
(1950) 3153–3155.
[22] Forward G.C., Whiting D.A., J. Chem. Soc. C, (1969) 1868–1873.
[23] Stork G., Brizzolara A., Landesman H., Szmuszkovicz J., Terrell R.,
J. Am. Chem. Soc. 85 (1963) 207–222.
[24] Hückel W., Maier M., Jordan E., Seeger W., Liebigs Ann. Chem. 616
(1958) 46–47.
[25] Mayer R., Chem. Ber. 89 (1965) 1443–1454.
[26] Stocker J.H., J. Org. Chem. 27 (1962) 2288–2289.
[27] Cheng Y.S., Liu W.L., Chen S.H., Synthesis (1980) 223–224.
[28] Pearson D.E., Keaton O.D., J. Org. Chem. 28 (1963) 1557–1558.
References
[1] Castañer J., Arrigoni-Martelli E., Drugs Future 1 (1976) 119–122.
[2] Schillinger E., Biochem. Pharmacol. 27 (1978) 651–653.
[29] Hartmann S., Brecht V., Frahm A.W., Magn. Reson. Chem. 37
(1999) 69–72.
[3] Sturgess N.C., Ashford M.L.J., Cook D.L., Hales C.N., Lancet 2
[30] Hartmann S., Weckert E., Frahm A.W., Acta Cryst. C 55 (1999)
(1985) 474–475.
806–808.
[4] Rufer C., Biere H., Ahrens H., Loge O., Schröder E., J. Med. Chem.
17 (1974) 708–721.
[31] Seebach D., Prelog V., Angew. Chem. 94 (1982) 696–702.
[32] Dale J.A., Dull D.L., Mosher H.S., J. Org. Chem. 34 (1969)
[5] Graul A., Castañer J., Drugs Future 21 (1996) 694–699.
2543–2549.
[6] Fuhlendorff J., Shymko R., Carr R.D., Kofod H., Diabetes 44
(Suppl. 1) (1995) Abstr. 848.
[33] Dale J.A., Mosher H.S., J. Am. Chem. Soc. 90 (1968) 3732–3738.
[34] Ullrich S., Abel K.B., Lehr S., Greger R., Pflügers Arch. 432 (1996)
630–636.
[7] Bloom J.D., Dutia M.D., Johnson B.D., J. Med. Chem. 35 (1992)
3081–3084.
[35] Aguilar-Bryan L., Nichols C.G., Wechsler S.W., Clement J.P., Boyd
[8] Bloom J.D., Claus T.H., Drugs Future 19 (1994) 23–26.
A.E., González G. et al., Science 268 (1995) 423–426.
[9] Emorine L.J., Marullo S., Briend-Sutren M., Patey G., Tate K.,
Delavier-Klutchko C., Strosberg A.D., Science 245 (1989)
1118–1121.
[36] Ullrich S., Wolheim C.B., J. Biol. Chem. 259 (1984) 4111–4115.
[37] Asfari M., Janjic D., Meda P., Li G., Halban P.A., Wollheim C.B.,
Endocrinol. 130 (1992) 167–178.
[10] Eistetter K., Wolf H.P.O., J. Med. Chem. 25 (1982) 109–113.
[11] Eistetter K., Wolf H.P.O., Drugs Future 11 (1986) 1034–1036.
[38] Herbert V., Lau K.S., Gottlieb C.W., Bleicher S.J., J. Clin. Endo-
crinol. Metab. 25 (1965) 1375–1384.
[12] Anderson R.C., Balestra M., Bell P.A., Deems R.O., Fillers W.S.,
[39] Kerst G., Fischer K.G., Normann C., Kramer A., Leipziger J.,
Foley J.E. et al., J. Med. Chem. 38 (1995) 3448–3450.
Greger R., Pflügers Arch. 430 (1995) 253–265.
[13] Grisar J.M., Claxton G.P., Carr A.A., Wiech N.L., J. Med. Chem. 16
[40] Nietschke R., Fröbe U., Greger R., Pflügers Arch. 417 (1991)
622–632.
(1973) 679–683.
[14] Gerritsen G.C., Dulin W.E., Diabetes 14 (1965) 507–515.